U.S. patent application number 15/792673 was filed with the patent office on 2018-05-24 for system and method for implementing meal selection based on vitals, genotype and phenotype.
This patent application is currently assigned to Habit, LLC. The applicant listed for this patent is Habit, LLC. Invention is credited to Jon Allen, Joshua Anthony, Erin Barrett, Heather Cutter, Neil Grimmer, Matt Town, Matt Van Horn, Angie Westbrock, Barbara Winters, Ryan Yockey.
Application Number | 20180144820 15/792673 |
Document ID | / |
Family ID | 62023987 |
Filed Date | 2018-05-24 |
United States Patent
Application |
20180144820 |
Kind Code |
A1 |
Grimmer; Neil ; et
al. |
May 24, 2018 |
SYSTEM AND METHOD FOR IMPLEMENTING MEAL SELECTION BASED ON VITALS,
GENOTYPE AND PHENOTYPE
Abstract
Systems and methods for recommending foods to a user based on
health data, includes a database, a memory and a processor. The
database stores user health data for each user within a plurality
of users, including vitals, genotypic and phenotypic data, user
food preference data and foods data that includes macronutrient and
micronutrient data for foods that may be recommended to a user. The
memory stores program instructions, including program instructions
that are capable of (i) classifying user health data into
predetermined diet types and micronutrient recommendations, (ii)
filtering the food data to determine available foods for a user;
(iii) a ranking available meals for the user based on the
micronutrient recommendations and the food data, and (iv)
translating micronutrient recommendations and/or food data for the
available foods for the user into specific food recommendations for
the user.
Inventors: |
Grimmer; Neil; (Emeryville,
CA) ; Anthony; Joshua; (Princeton, NJ) ;
Yockey; Ryan; (Emeryville, CA) ; Van Horn; Matt;
(Emeryville, CA) ; Allen; Jon; (Emeryville,
CA) ; Barrett; Erin; (Emeryville, CA) ;
Winters; Barbara; (Emeryville, CA) ; Cutter;
Heather; (Emeryville, CA) ; Westbrock; Angie;
(Emeryville, CA) ; Town; Matt; (Emeryville,
CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Habit, LLC |
Emeryville |
CA |
US |
|
|
Assignee: |
Habit, LLC
Emeryville
CA
|
Family ID: |
62023987 |
Appl. No.: |
15/792673 |
Filed: |
October 24, 2017 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
62412114 |
Oct 24, 2016 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G06F 16/9535 20190101;
A23L 33/30 20160801; A61P 25/00 20180101; A61P 25/18 20180101; G16H
20/60 20180101 |
International
Class: |
G16H 20/60 20060101
G16H020/60; G06F 17/30 20060101 G06F017/30 |
Claims
1-87. (canceled)
88. A system for recommending foods to a user based on health data,
comprising: a database that stores user health data for each user
within a community of users, including vitals, genotypical and
phenotypical data, user food preference data and foods data that
includes macronutrient and micronutrient data for foods that may be
recommended to a user; a memory that stores program instructions,
including program instructions that are capable of implementing (i)
a classifier that classifies user health data into predetermined
diet types and micronutrient recommendations, (ii) a filtering
engine to filter the food data to determine available foods for a
user based on the food data, the diet type, and the user food
preference data; and (iii) a ranking engine that ranks available
meals for a user based on the micronutrient recommendations and
food data; and a processor, coupled to the database and the memory,
that when executing the program instructions causes the decision
tree logic to classify a user by diet type and nutrient
recommendations, causes the filtering engine to determine available
foods for the user and causes the ranking engine to rank and
translate the micronutrient recommendations and the available foods
for a user into specific food recommendations for the user.
89. The system according to claim 88, wherein the foods are
prepared meals.
90. The system according to claim 88, wherein the user health data
includes activity levels for at least some users.
91. The system according to claim 88, wherein the health data
further includes user goals that are used by at least one of the
filtering engine and the ranking engine to select foods for the
user.
92. The system according to claim 88, wherein the genotypical data
includes data on a user's FTO rs9939609.
93. The system according to claim 92, wherein the genotypical data
further data on a user's FTO rs1121980.
94. The system according to claim 88, wherein the meal data
includes a name, and quantity information corresponding to
carbohydrates, protein, fat, calories, and micronutrients.
95. A method for recommending meals to a user based on health data,
comprising: maintaining a database of users that stores (i) for at
least some users, a diet type vector for each user comprising
macronutrient and micronutrient ranges determined based on decision
logic from the user's health data, including vitals, genotypical
and phenotypical data, (ii) user food preference data, and (iii)
food data including macronutrient and micronutrient data
corresponding to foods that may be recommended to a user; at the
request of a requesting user, filtering the food data based on the
user's diet type vector and the user food preference data for the
requesting user to determine a set of available foods; when the
requesting user's diet type vector matches foods included in the
food data, including the matching foods in a list of available
foods for the requesting user as part of the filtering; excluding a
food from the list of available foods for the requesting user if
the food does not match the requesting user's preference data as
part of the filtering; and presenting to the requesting user the
list of available foods that match the user's diet type.
96. The method according to claim 95, further comprising ranking
the available foods for a user based on at least some of the
micronutrients in the user's diet type vector and the food data for
the foods on the available foods list.
97. The method according to claim 96, wherein at least some of the
users update their corresponding health data over time and the
vector for the updating user is updated based on the new health
data.
98. The method according to claim 96, wherein the diet type vector
for at least some users includes macronutrient ranges corresponding
to carbohydrates, protein and fat determined individually for each
of the at least some users based on that user's genotype data,
phenotype data and vitals maintained in the database.
99. The method according to claim 98, wherein the diet type vector
for at least some users includes at least one micronutrient range
determined using diagnostic logic based on each of the at least
some user's metabolic health as determined by the user's genotype
data, phenotype data and vitals.
100. The method according to claim 96, wherein the diet type vector
for at least some users includes at least one micronutrient range
determined using diagnostic logic based on each of the at least
some user's heart health as determined by the user's genotype data,
phenotype data and vitals.
101. The method according to claim 96, wherein the diet type vector
for at least some users includes at least one micronutrient range
determined using diagnostic logic based on each of the at least
some user's inflammation as determined by the user's genotype data,
phenotype data and vitals.
102. A computer program product including computer program
instructions stored on a media for causing a computer to
recommending meals to a user based on health data, comprising:
maintaining logic for causing the computer to maintain a database
of users that has (i) for at least some users, a diet type vector
for each user comprising macronutrient and micronutrient ranges
determined based on decision logic from such user's health data,
including vitals, genotypical and phenotypical data, (ii) user food
preference data, and (iii) meal data including macronutrient and
micronutrient data corresponding to each known meal; comparing
logic for causing the computer to compare, at the request of a
requesting user, the diet type vector for the requesting user to
the meal data and to compare the user food preference data for the
requesting user to the meal data; when the requesting user's diet
type vector matches a known meal included in the meal data,
including the known meal in a list of available meals for the
requesting user; excluding a known meal from the list of available
meals for the requesting user if the known meal does not match the
requesting user's preference data; and presenting to the requesting
user the list of available meals.
103. The computer program product according to claim 102, wherein
at least some of the users update their corresponding health data
over time and the vector for the updating user is updated based on
the new health data.
104. The computer program product according to claim 103, wherein
the diet type vector for at least some users includes macronutrient
ranges corresponding to carbohydrates, protein and fat determined
individually for each of the at least some users based on that
user's genotype data, phenotype data and vitals maintained in the
database.
105. The computer program product according to claim 103, wherein
the diet type vector for at least some users includes at least one
micronutrient range determined using diagnostic logic based on each
of the at least some user's metabolic health as determined by the
user's genotype data, phenotype data and vitals.
106. The computer program product according to claim 103, wherein
the diet type vector for at least some users includes at least one
micronutrient range determined using diagnostic logic based on each
of the at least some user's heart health as determined by the
user's genotype data, phenotype data and vitals.
107. The computer program product according to claim 103, wherein
the diet type vector for at least some users includes at least one
micronutrient range determined using diagnostic logic based on each
of the at least some user's inflammation as determined by the
user's genotype data, phenotype data and vitals.
Description
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application claims priority to U.S. Provisional Patent
Application No. 62/412,114, filed Oct. 24, 2016, the disclosure of
which is hereby incorporated by reference in its entirety for all
purposes.
FIELD OF THE INVENTION
[0002] The disclosed embodiments relate generally to health
diagnostic systems and methods, and in particular, to recommending
meals, recipes, foods and/or supplements based on a person's
vitals, genotypic and phenotypic data.
BACKGROUND OF THE INVENTION
[0003] Diet is a major factor in the health of individuals. Advice
on what to eat is prevalent today. The advice tends to be general
in nature and focuses on which foods to avoid, such as processed
foods or saturated and trans fats, or which foods to eat more of,
such as fruits, vegetables and whole grains. Certain health
conditions or diseases have also led to recommendations to avoid or
to eat certain foods. However, within other subgroups of people,
such as people with diabetes, there have not been individualized
nutritional recommendations. Moreover, even where nutrition
recommendations have been identified for individual disease states,
genotypes, or phenotypes, comprehensive dietary recommendations
have not been developed that consider the interplay between
multiple features of an individual.
[0004] U.S. Patent Application Publication No. 2012/0130732, for
example, describes methods and systems for providing personalized
nutrition and exercise advice to a subject. However, the methods
consider subject features individually, rather than together. For
example, as illustrated in FIG. 3, identification of low serum
ferritin levels in an individual result in a monotonous static
recommendation to eat red meat and liver, take iron supplements,
swim, and exercise less often. This advice does not consider,
however, how the interplay of other features of the subject affect
the recommendations provided. This publication also does not
comprehensively evaluate a user's genetics, phenotypical and other
information about a user to produce diet types for macronutrient
recommendations or combine macronutrient and micronutrient needs of
a user into daily, weekly, or other frequent meal, food, or
supplement recommendations that exhibit variety and that are ranked
for a user and that also may output recipes, or supplement
regimens.
[0005] U.S. Patent Application Publication No. 2012/0295256, for
example, describes methods and systems for providing weight
management advice by considering features associated with weight
management. However, the methods only consider recommendations
related to weight management, without considering other health
considerations.
[0006] U.S. Patent Application Publication No. 2013/0280681, for
example, describes methods and systems for providing food selection
recommendations based on a user's dietary history. However, the
methods do not consider the biological differences between
individuals that inform healthy eating.
[0007] Recent studies suggest that healthy individuals have greater
metabolic adaptability which facilitates phenotypic flexibility to
changing environmental conditions, including stressors (e.g.,
physical activity). van Ommen B. et al., Genes Nutr., 9(5):423
(2014). For example, impaired phenotypic flexibility has been
reported in overweight participants who may have reduced ability to
metabolize stored lipids for energy synthesis and in response,
slowly adapt to excess dietary fat intake, compared with lean
participants. Blaak E. et al., J Clin Endocrinol Metab, 91:1462-69
(2006). Further, the lack, or excess, of consumption of certain
dietary components, are known to impair phenotypic flexibility and
may ultimately affect optimal health. van Ommen B. et al., Supra.
The assessment of phenotypic flexibility involves the perturbation
of homeostasis and subsequent evaluation of specific
nutrition-related biomarkers. Challenge tests with various
combinations of macronutrients have been used to temporarily
disturb homeostasis (Stroeve 2015; Kardinaaal 2015; van Amelsvort
1989). However, these tests are inconvenient, typically requiring
an individual to visit a testing center to perform a lengthy
test.
[0008] Comprehensive analyses of an individual's genotypic and
phenotypic characteristics are not performed for the purpose of
recommending personalized meals or foods. As such, there remains a
need for specific techniques to analyze information for individuals
and help individuals to determine what they should eat. There
remains a further need for the nutritional recommendations to
reflect comprehensively a person's individuality and goals.
SUMMARY OF THE INVENTION
[0009] Various embodiments of systems and methods within the scope
of the appended claims each have several aspects, no single one of
which is solely responsible for the attributes described herein.
Without limiting the scope of the appended claims, after
considering this disclosure, and particularly after considering the
section entitled "Detailed Description," one will understand how
the aspects of various embodiments are used to enable specific
personalized nutrition systems and methods.
[0010] The disclosed systems and methods use data from individual
users, including their vitals data, such as waist circumference,
blood pressure and age; genotypical data including data on a user's
DNA and genetic variations such as particular single nucleotide
polymorphisms (SNPs), and phenotypical data relating to markers
obtained from blood samples from the individual. By focusing on
these and other types of data associated with a person's body,
rules and logic may be applied to classify individuals into
specific diet types that specify at the macronutrient and
micronutrient level a personalized diet and also what foods should
be eaten by the person. Moreover, with the addition of personal
goals as well as food preferences, a list of available meals,
recipes, hero foods, snacks or supplements can be selected,
customized, prioritized, and delivered for each user within a
community of users that is tailored to the well-being of each user
and that delivers a variety of healthy, different, and interesting
food recommendations on a daily, weekly, monthly, or other frequent
basis and that introduces healthy meal variation for each user over
time. In this manner, a user or each user in a population of users
is provided a variety of different prepared meals that may be
delivered to the user, recipes that may be prepared by the user,
food recommendations or supplement recommendations, all in order to
help the user on a daily, weekly, monthly, or other frequent basis
achieve a desired state of wellbeing or one or more health goals
through healthy and personalized consumption.
[0011] According to some embodiments, a system for recommending
foods to a user based on health data, comprises a database, a
memory and a processor. The database stores user health data for
each user within a community of users, including vitals,
genotypical and phenotypical data, user food preference data and
foods data that includes macronutrient and micronutrient data for
foods that may be recommended to a user. The memory stores program
instructions, including program instructions that are capable of
implementing (i) decision tree logic that classifies user health
data into predetermined diet types and micronutrient
recommendations, (ii) a filtering engine to filter the food data to
determine available foods for a user based on the user's diet type
and the user's food preference data; and (iii) a ranking engine
that ranks available meals for the user based on the micronutrient
recommendations and the food data. The processor is coupled to the
database and the memory and, when executing the program
instructions, causes the decision tree logic to classify the user
by diet type and nutrient recommendations, causes the filtering
engine to determine available foods for the user and causes the
ranking engine to rank and translate the micronutrient
recommendations and the food data for the available foods for the
user into specific food recommendations for the user.
[0012] According to some embodiments, the recommended foods are
prepared meals. According to some embodiments, the recommended
foods may be one or more of the following: prepared meals, recipes,
snacks, hero foods, which are foods high in certain nutrients of
value to users, or nutritional supplements. The health data in some
embodiments may include activity levels for at least some users.
The health data may further include in some embodiments user goals
such as weight loss or endurance that are used by the filtering
engine or the ranking engine to select foods for the user. The food
data may also include calorie information used by the filtering
engine or the ranking engine to select foods for the user. The
system may also makes lifestyle recommendations to the user to
improve the user's wellbeing based on the health data.
[0013] The vitals used by the system in some embodiments include
waist circumference and blood pressure and may further include age,
gender, height, weight, activity level and other information about
a user.
[0014] The genotypical data in some embodiments includes genetic
variants including single nucleotide polymorphisms that are
correlated with one or more of the following: body fat, blood
pressure, heart health and inflammation among other data. The
phenotypical data in some embodiments includes information on some
or all of the following: the user's insulin sensitivity,
cholesterol, triglicerides, and nutrient and mineral levels, among
other data. The user's food preference data in some embodiments
includes information on foods that the user will not eat or the
user's food religion, such as vegan or kosher.
[0015] In some embodiments, a method for recommending foods to a
user based on health data, includes maintaining a database of users
that stores (i) for at least some users, a diet type vector for
each user comprising macronutrient and micronutrient ranges
determined based on decision logic from the user's health data,
including vitals, genotypical and phenotypical data, (ii) user food
preference data, and (iii) food data including macronutrient and
micronutrient data corresponding to foods that may be recommended
to a user. At the request of a requesting user, the method includes
filtering the food data based on the user's diet type vector and
the user food preference data to determine a set of available foods
for the user. A food is excluded from the list of available foods
for the requesting user if the food does not match the requesting
user's preference data. In some embodiments, the method includes
presenting to the requesting user the list of available foods
matching the user's diet type. The list of matching foods may also
be ranked based on the micronutrients in the user's diet type
vector and the food data corresponding to the matching foods. Many
other factors may also be used to influence the ranking.
[0016] In one aspect, the disclosure provides a multi-nutrient
challenge beverage for measuring the metabolic adaptability of a
user, including: a) from 44 to 57 grams total fats; b) 75.+-.15
grams total carbohydrates; and c) 20.+-.3 grams total protein.
[0017] In some embodiments of the multi-nutrient challenge beverage
described above, the fat content of the beverage comprises from 10%
to 20% of the total weight of the beverage.
[0018] In some embodiments of the multi-nutrient challenge
beverages described above, the fat content of the beverage is
primarily from an edible vegetable oil.
[0019] In some embodiments of the multi-nutrient challenge
beverages described above, the edible vegetable oil is palm
oil.
[0020] In some embodiments of the multi-nutrient challenge
beverages described above, the carbohydrate content of the beverage
comprises from 10% to 30% of the total weight of the beverage.
[0021] In some embodiments of the multi-nutrient challenge
beverages described above, the carbohydrate content of the beverage
is primarily from monosaccharide sugar.
[0022] In some embodiments of the multi-nutrient challenge
beverages described above, the monosaccharide sugar is
dextrose.
[0023] In some embodiments of the multi-nutrient challenge
beverages described above, the protein content of the beverage
comprises from 2.5% to 10% of the total weight of the beverage.
[0024] In some embodiments of the multi-nutrient challenge
beverages described above, the protein content of the beverage is
primarily from a milk protein isolate.
[0025] In some embodiments of the multi-nutrient challenge
beverages described above, the beverage further including one of
more of a tastant, an emulsifier, a thickening agent, and a
preservative.
[0026] In one aspect, the disclosure provides a method for
measuring the metabolic adaptability of a user, including: (A)
obtaining data on a user's blood insulin levels, blood glucose
levels, and blood triglyceride levels prior to consumption of a
multi-nutrient challenge beverage, after a first period of time
following consumption of the multi-nutrient challenge beverage, and
after a second period of time following consumption of the
multi-nutrient challenge beverage; and (B) inputting the obtained
data into a metabolic adaptability classifier, wherein the first
period of time and second period of time following consumption of
the multi-nutrient challenge beverage are each no longer than 120
minutes, and wherein the challenge beverage is a challenge beverage
as described above.
[0027] In some embodiments of the method for measuring the
metabolic adaptability of a user described above, the data obtained
on the user's blood insulin levels, blood glucose levels, and blood
triglyceride levels is derived from a dried blood sample collected
by the user.
BRIEF DESCRIPTION OF THE FIGURES
[0028] So that the present disclosure can be understood in greater
detail, a more particular description may be had by reference to
the features of various embodiments, some of which are illustrated
in the appended drawings. The appended drawings, however, merely
illustrate the more pertinent features of the present disclosure
and are therefore not to be considered limiting, for the
description may admit to other effective features.
[0029] FIG. 1 is a block diagram illustrating an implementation of
a personalized food and nutrition recommendation system, in
accordance with some embodiments.
[0030] FIG. 2A is a flow chart illustrating a method of processing
user vitals, genotypical and phenotypical data to determine a diet
type for a user in accordance with some embodiments.
[0031] FIG. 2B is a flow chart illustrating a method of processing
user diet type determined based on a user's vitals, genotypic and
phenotypic data and information on available meals, recipes, foods
and/or supplements to determine available meals, recipes, foods or
supplements for a user in accordance with some embodiments.
[0032] FIG. 2C is a flow chart illustrating a method of ranking
available meals, recipes, foods and/or supplements for a user based
on a user's diet type and vitals, genotypic and phenotypic data in
accordance with some embodiments.
[0033] FIG. 3 is a list of phenotypic data that is used in
accordance with some embodiments for processing a user's diet
type.
[0034] FIG. 4 is a list of genotypic data that is used in
accordance with some embodiments for processing a user's diet
type.
[0035] FIG. 5 depicts a mapping of combinations of macronutrient
recommendations into diet types in accordance with some
embodiments.
[0036] FIG. 6 depicts an illustrative set of ranges for seven
individualized diet types into which to categorize users based on
their vitals, genotype and phenotype in accordance with some
embodiments.
[0037] FIG. 7 depicts an illustrative collection of food groups and
serving sizes for seven different diet types in accordance with
some embodiments.
[0038] FIG. 8 is a list of micronutrients and in some cases foods
that are used in accordance with some embodiments for determining
micronutrient recommendations and meal or food ranking in
accordance with some embodiments.
[0039] FIG. 9 depicts a method of interacting with a user over a
network connection related to delivering meal, recipe, food and
supplement related information based on the user's vitals, genotype
and phenotype and other information provided by the user in
accordance with some embodiments.
[0040] FIG. 10 depicts an illustrative classifier that produces
macronutrient and micronutrient recommendations based on vitals,
genotypic and/or phenotypic data for a user in accordance with some
embodiments.
[0041] FIGS. 11A and 11B depict an illustrative classifier for
determining a carbohydrate recommendation based on vitals,
genotypic and/or phenotypic data in accordance with some
embodiments.
[0042] FIGS. 12A, 12B, and 12C depict an illustrative classifier
for determining a fats recommendation based on vitals, genotypic
and/or phenotypic data in accordance with some embodiments.
[0043] FIG. 13 depicts an illustrative classifier for determining a
protein recommendation based on vitals, genotypic and/or phenotypic
data in accordance with some embodiments.
[0044] FIGS. 14A, 14B, 14C, 14D, and 14E depict a list of hero
foods that are recommended to users in some embodiments.
[0045] FIG. 15 is a block diagram illustrating an implementation of
a personalized food and nutrition recommendation method, in
accordance with some embodiments.
[0046] FIG. 16 depicts an illustrative classifier for determining
monounsaturated fatty acid and fiber recommendations based on
vitals, genotypic and/or phenotypic data in accordance with some
embodiments.
[0047] FIG. 17 depicts an illustrative classifier for determining
dietary protein flexibility recommendations based on vitals,
genotypic and/or phenotypic data in accordance with some
embodiments.
[0048] FIG. 18 depicts an illustrative classifier for determining
dietary carbohydrate flexibility recommendations based on vitals,
genotypic and/or phenotypic data in accordance with some
embodiments.
[0049] FIG. 19 depicts an illustrative classifier for determining
dietary fat flexibility recommendations based on vitals, genotypic
and/or phenotypic in accordance with some embodiments.
[0050] FIGS. 20 depicts an illustrative classifier for determining
carbohydrate micronutrient recommendations based on vitals,
genotypic and/or phenotypic data in accordance with some
embodiments.
[0051] FIG. 19 illustrates insulin levels in subjects before and
after consuming a multi-nutrient challenge beverage, as measured
using capillary blood samples spotted on a substrate (insulin ADX)
and venous blood collected in a catheter (insulin venous).
[0052] FIG. 20 illustrates a linear regression comparing insulin
levels measured using capillary blood samples spotted on a
substrate (insulin ADX) with venous blood collected in a catheter
(insulin venous) before and after consuming a first multi-nutrient
challenge beverage.
[0053] FIG. 21 illustrates a linear regression comparing insulin
levels measured using capillary blood samples spotted on a
substrate (insulin ADX) with venous blood collected in a catheter
(insulin venous) before and after consuming a second multi-nutrient
challenge beverage.
[0054] FIG. 22 illustrates glucose levels in subjects before and
after consuming a multi-nutrient challenge beverage, as measured
using capillary blood samples spotted on a substrate (insulin ADX)
and venous blood collected in a catheter (insulin venous).
[0055] FIG. 23 illustrates a linear regression comparing glucose
levels measured using capillary blood samples spotted on a
substrate (insulin ADX) with venous blood collected in a catheter
(insulin venous) before and after consuming a first multi-nutrient
challenge beverage.
[0056] FIG. 24 illustrates a linear regression comparing glucose
levels measured using capillary blood samples spotted on a
substrate (insulin ADX) with venous blood collected in a catheter
(insulin venous) before and after consuming a second multi-nutrient
challenge beverage.
[0057] FIG. 25 illustrates triglyceride levels in subjects before
and after consuming a multi-nutrient challenge beverage, as
measured using capillary blood samples spotted on a substrate
(insulin ADX) and venous blood collected in a catheter (insulin
venous).
[0058] FIG. 26 illustrates a linear regression comparing
triglyceride levels measured using capillary blood samples spotted
on a substrate (insulin ADX) with venous blood collected in a
catheter (insulin venous) before and after consuming a first
multi-nutrient challenge beverage.
[0059] FIG. 27 illustrates a linear regression comparing
triglyceride levels measured using capillary blood samples spotted
on a substrate (insulin ADX) with venous blood collected in a
catheter (insulin venous) before and after consuming a second
multi-nutrient challenge beverage.
[0060] FIGS. 28A, 28B, 28C, 28D, 28E, 28F, 28G, and 28H are a flow
chart illustrating a method of providing food recommendations based
on the features of a user in accordance with some embodiments.
[0061] FIG. 29 depicts an illustrative method of collecting data
from users and about meals and available ingredients and
classifying the users into diet types and the meals according to
their data in order to match users with a variety of different,
heathy meal options on a daily, weekly, monthly or other frequency
basis that are individualized for the user and that may be
delivered to the user, in accordance with some embodiments.
[0062] In accordance with common practice the various features
illustrated in the drawings may not be drawn to scale. Accordingly,
the dimensions of the various features may be arbitrarily expanded
or reduced for clarity. In addition, some of the drawings may not
depict all of the components of a given system, method or device.
Finally, like reference numerals may be used to denote like
features throughout the specification and figures.
DETAILED DESCRIPTION
Overview
[0063] The various implementations described herein include
systems, methods and/or devices used to enable individualized meal
and food recommendations to a user based on that user's health
vitals, such as height, weight, blood pressure, age, waist
circumference; the user's genotype and in particular genetic
markers, such as SNPs, and phenotype data as determined by blood
tests.
[0064] The disclosed systems and methods use data from individual
users, including their vitals data, such as waist circumference,
blood pressure and age; genotypical data including data on a user's
DNA and genetic variations such as particular single nucleotide
polymorphisms (SNPs), and phenotypical data relating to markers
obtained from blood samples from the individual. By focusing on all
three types of data associated with a person's body, rules and
logic may be applied to classify individuals into specific diet
types that specify at the macronutrient and micronutrient level a
personalized diet and also what foods should be eaten by the
person. Moreover, with the addition of personal goals as well as
food preferences, a list of available meals, recipes, hero foods,
snacks or supplements can be selected, customized and prioritized
for each user. In this manner, a user is provided prepared meals
that may be delivered to the user, recipes that may be prepared by
the user, food recommendations or supplement recommendations in
order to help the user achieve a desired state of wellbeing or one
or more health goals.
Systems of the Invention
[0065] FIG. 1 depicts a block diagram of a system 100 according to
some embodiments of the invention. The system implements
personalized nutrition analysis for a user and facilitates
identifying meals, recipes, and foods or supplements (collectively
foods) for users and may further facilitate selling and delivering
meals and other foods to users. Referring to FIG. 1, the system 100
includes a plurality of users at user devices 101 that communicate
with a server, such as a web server interface 104, typically via a
network. The network may include the Internet, local area networks,
wide area networks, wired networks, optical networks, wireless
networks, telephone networks, cellular networks, email networks and
any other type of network or bus connection that allows the
exchange of data typically, though not limited to, through the
Internet Protocol. The user devices 101 may be mobile devices, such
as mobile phones, tablets, or laptop computers, for example.
Alternatively, the devices 101 may be desktop or other computers or
devices. The user devices 101 enable a plurality of users to
interact with the web interface server 104 to provide information
about the user to the web server 104 and to receive information
back from the web server interface 104. Generally, the user devices
101 includes a processor, memory, a screen, and input devices such
as a touchscreen, keyboard, keys, a mouse, or a microphone. The
user interacts with the user device 101 and the web server
interface 104 to exchange information between the system 100 and
the user 101.
[0066] The system 100 also may include devices 102 associated with
health service providers and devices 103 associated with meal,
recipe or supplement providers. The devices 102 and 103 are similar
to the user devices described above. The system 100 further
includes a user health database 105, a meal and recipe database 106
and a meals processing engine 107.
[0067] The user devices 101 may be used by users to provide health
information about themselves to the system 100. In particular, in
some embodiments, the user may log into the web server interface
104 and upon authentication provide to the system 100 information
about the user's vitals, such as the information shown in FIG. 1.
The user may further provide genotype and phenotype information,
for example, of the types shown in FIGS. 3 and 4. The user may in
some embodiments also provide information about the user's goals,
such as general wellbeing, weight loss, increase of muscle mass
and/or improving endurance. The user may also in some embodiments
provide information about the food preferences, for example food
religion (e.g., vegan, kosher, gluten free), or a list of foods
that the user prefers or does not like. This information may be
elicited through a browser interface with questions or lists of
questions with dropdown predetermined choices according to some
embodiments.
[0068] The devices 102 may be used by health service providers to
provide vitals, genotype or phenotype information regarding the
user to the system 100. In general, the user and/or healthcare
providers may enter or upload data via the web server.
Alternatively, the user and or health service providers may upload
the data for particular users directly to a database associated
with the system 100, such as the database 105. The database 105 may
be centralized or distributed and accessible by the system 100.
[0069] In general, the webserver 104 and devices 101 and 102 are
used for inputting data about each user's vitals, genotype and
phenotype. The web server interface 104 may serve a browser page
that authenticates users and/or health service providers and allows
them to enter relevant data into particular fields. Alternatively,
the web server interface may facilitate uploading files to the
database 105 or otherwise facilitating access to the database 105
to provide relevant information about users to the system. The web
server interface 104 may further include parsing and filtering
functionality that receives data on the vitals, genotypes and
phenotypes of users and converts the data into a recommendation
context with data populating fields that will be used by the system
100 for nutritional analysis according to some embodiments
described herein. Similarly, goals and food preference information
may be filtered and stored in the database 105.
[0070] Additional devices that interact with the system 100 may be
coupled to the system, including in some embodiments devices 103.
Devices 103 may be associated with meal, recipe or health
supplements providers (hereinafter meal providers). The meal
providers may provide meals, recipes or supplement information to
the system to be stored in the meal and recipe database 106. The
devices 103 may provide meal related information to the meal and
recipe database 106 via the web server interface through browser
entry, through uploading data via the web server interface 104 or
via the meals processing engine 107.
[0071] The devices 101 may further include activity trackers
associated with a user that provide additional information about
users to the system 100. For example, in some embodiments, activity
trackers may provide daily information about how many calories a
user has burned, how much sleep a user has gotten, how many steps a
user has taken, heart rate information, distance walked or run. In
some embodiments, other information about the user's activities may
be provided such as the type of activity done by the user and the
duration, such as swimming for one hour. The user's device may
automatically upload activity information or may upload it in
response to synchronization operations initiated by the user.
Additionally, the user may provide activity level, sleep and other
data about the user to the system 100 via a webpage served by the
web server interface 104 by uploading or linking a file with
activity data.
[0072] The meals processing engine 107 receives data from the web
server interface 104 or the devices, such as devices 103 regarding
meals, recipes or other foods or supplements and converts the data
into a format usable by the system 100 and then stores the data in
the database 106. In general, the information regarding meals and
recipes includes in some embodiments the number of calories
associated with the meal and macronutrient information, such as the
calories from protein, fat and carbohydrates. The meal information
in some embodiments includes the number of grams of fat, protein
and carbohydrates. In some embodiments, the meal and food
information includes amounts associated with micronutrients, such
as vitamins, or dietary fibers, or types of fats such as saturated,
monounsaturated, or polyunsaturated fats. The data associated with
foods, meals and/or recipes in terms of macronutrients and
micronutrients may be directly provided to the database 106 or may
be converted by a conversion process in the web server interface
104 or meals processing engine 107 in some embodiments into
actionable macronutrient and micronutrient information. Similarly,
hero foods, snacks or supplements may be described to the system in
terms of micronutrient and other macronutrient information by the
same processes describe above.
[0073] The web server interface 104 may maintain a user profile for
each user. The user profile may include, for example, some or all
of the following information: [0074] User Number, User id,
Password, biometric data [0075] User location or shipping address,
billing address or credit card information [0076] User email
address or telephone number at which to receive messages [0077]
User meal delivery data (daily, weekdays, # meals per week,
monthly, breakfast, lunch, dinner, snack, supplement) [0078] User
offer preferences (offer user recommended meals every day by
messages, weekly, monthly, other frequency) [0079] User activity
tracker information [0080] User organization affiliation [0081]
User diet type classification [0082] User goals and food
preferences [0083] User coaching preferences
[0084] The system 100 in some embodiments processes the information
received from users and providers to produce recommendations for
meals, recipes and supplements. The web interface server 104, for
example, includes information on each user in the user profile. The
user profile may specify, for example that a user is to be given a
meal recommendation for each meal three times a day. Alternatively,
the user profile may specify only one meal a day or five meals a
week. The profile may also call for delivery of the meals or
alternatively recipe recommendations according to some embodiments.
Additional details of how the system may be configured for users is
discussed below.
[0085] The system 100 determines foods for users, including in some
embodiments prepared meals, recipes, snacks, hero foods,
supplements or some or all of the foregoing. In some embodiments,
the determination is made in real time on request by a user. In
some embodiments, the system 100 determines meals for users at some
frequency determined by a user selecting from available options.
When the web server interface 104 determines that the system is
ready to identify recommended meals for a user the recommendation
process starts. This process uses the decision tree engine 108 to
produce macronutrient 109 and micronutrient 110 classifications for
each user, which result in each user being classified in one of
several possible diet types. Each diet type specifies ranges for
protein, fats and carbohydrates as shown in FIG. 6. The ranges may
be specified in grams or as percentages of calories.
[0086] The macronutrient 109 recommendations and the meal and
recipe database 106 are inputs to a user specific filtering engine
115. The filtering engine 115 filters meal data based on the user's
macronutrient classifications or diet type. The filtering engine
may also filter the meals and recipes based on the user's goals, or
food religion or food preferences. For example, if the user does
not like fish, meals with fish will be excluded by the filter.
Similarly, users whose food religion is vegan will have meals and
recipes that include meat filtered out. When goals such as weight
loss are factored in, certain meals may be filtered out based on
calories or macronutrient factors, including those specific to the
user. The result of the filtering engine 115 is a set of available
meals, recipes or supplements for the user, sometimes referred to
as the available meals 128.
[0087] The meal ranker engine 130 receives the available meals as
well the user's macronutrient 109 classifications or diet type, and
micronutrient 110 classifications. The meal ranker engine may also
receive the following information from the databases 105 and 106:
[0088] Data on calories, macronutrients and micronutrients for each
meal, recipe, food or supplement [0089] Data on diet type,
macronutrient and micronutrient recommendations for each user
[0090] Goals and user preference information
[0091] The meal ranker algorithm outputs recommendations for one or
more users. The meal ranker algorithm may rank meals, recipes,
supplements, hero foods, snacks or other information. The meal
ranker algorithm may take into account other user meals in a day or
supplements that the user regularly takes. It may also take into
account the activity level of the user, in addition to
macronutrient and micronutrients.
[0092] FIG. 2 depicts a method 200 of determining a diet type and a
micronutrient recommendation for a user based on vitals,
genotypical and phenotypical data. Referring to FIG. 2, according
to the method user vitals, genotypical and phenotypical data are
stored for a user in 202. The vitals data includes information
specific to the user, including, for example, the following
information: age, sex, waist circumference (size or
high/medium/low), and blood pressure measurements. The phenotypical
data is based on blood work done on the user. The phenotypical
information may include the data set forth in FIG. 3. In some
embodiments, the user is given a challenge beverage and samples of
the user's blood are taken at different times before and after
drinking the challenge beverage. The challenge beverage is
described in more detail in the Challenge Beverage section. In
general, the phenotypical data provides information about the
user's metabolic health, insulin sensitivity, heart health,
micronutrient levels, cholesterol and triglyceride levels and
inflammation. The genotypical markers in some embodiments are those
indicated in FIG. 4. In some embodiments, the genotypical markers
are single nucleotide polymorphisms (SNPs) that have a bearing on,
for example, gluten sensitivity, endurance performance, blood
pressure and sodium, insulin sensitivity, heart health, and
inflammation. More, fewer or different SNPs may be used as compared
to the ones identified in FIG. 4. The vitals, phenotypical and
genotypical data may be uploaded to the system by a user or health
care provider. Once the data is uploaded, for example into database
105, then in 204 individual data elements may be stored as part of
a recommendation context for the user. Diagnostic measurements,
which may be combinations of data elements from the vitals,
genotypical and phenotypical data, may also be determined and
stored in connection with a user as part of the recommendation
context for the user. In general, the recommendation context
includes actionable data related to a user's genotype, phonotype
and vitals that are to be used to determine the user's diet type,
macronutrient and micronutrient recommendations, which in turn form
the basis of meal, recipe, food and supplement recommendations.
[0093] In 206, decision tree logic is used on the recommendation
context, including the vitals, genotype and phenotype information.
The decision tree logic classifies the user according to specific
rules specified herein that result in diet type, macronutrient and
micronutrient recommendations. The diet type, macronutrient and
micronutrient classifications are based not just on one piece of
information. Rather, they are based on combinations of genotypical,
phenotypical and vitals information. In some embodiments, the diet
type, macronutrient and micronutrient classifications may also
factor in the user's goals and activity levels.
[0094] The decision tree logic presents a specific implementation
of determining diet types, macronutrient and micronutrient
recommendations. The decision trees operate based on input from
vitals, genotypical and phenotypical information for each user and
are a particular application of rules that classify users into at
least one of several diet types and recommended micronutrient
levels. The diet types then become the basis for meal and recipe
recommendations.
[0095] In 208 the system may optionally transmit the personalized
diet type, macronutrient and micronutrient information to the user.
The information may be part of a recommendation to supplement the
user's diet with particular hero foods or particular vitamin
supplements or part of a narrative or set of coaching instructions
for the user. In 210, the macronutrient and micronutrient
information is stored for the user.
[0096] In 212 the diet type is determined for the user and may be
stored in the database 105 in association with the user. The diet
type may be determined in 212 directly from macronutrient
information. Alternatively diet type may be determined based on
mapping one or more macronutrient recommendations or one or more
macronutrient and micronutrient recommendations to a set of
predetermined diet types for the system. For example, the
macronutrient recommendation may be broken down into eight
combinations: Fats (f and F), Carbohydrates (c and C), and Protein
(p and P). The upper case letter designation refers to an increased
level as compared to the lower level. The table below shows an
example of mapping sets of macronutrient recommendations to five
diet types or diet type vectors.
TABLE-US-00001 TABLE 1 Description of illustrative diet types. Diet
Type F/C/P Description Balanced Harvester FCP High carb, medium
fat, medium protein FCp Grain Seeker+ fCP High carb, low fat,
medium protein Grain Seeker fCp High carb, low fat, low protein
Protein Seeker fcP Low carb, low fat, high protein Hunter FcP Low
carb, medium fat, high protein
[0097] FIG. 5 shows another mapping of diet types based on
macronutrient recommendations according to some embodiments. Here,
there are twelve potential combinations of macro nutrient
recommendations: Fats (f and F), Carbohydrates (c and C), and
Proteins (p, P+ and P++). The diet types each reflect different
levels of macronutrients that are personalized for the user based
on vitals, genotype and phenotype data. FIG. 6 shows a table 600
that provides illustrative ranges for the seven diet types, or diet
type vectors, shown in FIG. 5, according to some embodiments.
Referring to the table 600, each diet type is shown with a
recommended daily calorie intake of 2000 calories. The number of
calories may be customized for each person based on sex, age,
activity level and other factors or may be considered on a meal by
meal basis. The table also includes recommended percentage ranges
for each diet type or diet type vector that correspond in some
embodiments to macronutrient recommendations. The macronutrient
recommendations are shown as elements 605. Table elements 610 show
illustrative values for calories associated with carbohydrates, fat
and protein for each diet type for an exemplary meal falling within
the ranges of the diet type. For each diet type, recommended meals
falls within the macronutrient ranges 605 for each user. Table
elements 615 show illustrative values in grams of carbohydrates,
fat and protein for each diet type for an exemplary meal falling
within the ranges of the diet type.
[0098] There may be different biological diet types for different
groups of users or all of the diet types may be the same across the
user population of a particular system 100. The diet types may
range in number, but in some embodiments there are between six and
nine biological diet types. There may be more or fewer depending on
the design of the system or the overall vitals, phenotypical and
genotypical variation found within the entire user community or
groups of users defined by geography, organizations, families or
other factors if desired.
[0099] After the diet types are determined for each user, the diet
type information may be transmitted to the user in 214. The diet
types in some embodiments may contain informative labels for the
user to comprehend the type of diet that is recommended for the
user. For example, diet type labels may include "balanced
harvester, grain seeker, protein seeker, hunter, and other terms
that are associated with macronutrient attributes of the diet type.
In 216, the system may optionally transmit narratives describing
ranges and the types of foods, snacks and meals that the user
should eat. The narratives may include additional information about
goals, micronutrient intake, supplements and other information
related to the user's nutritional needs.
[0100] FIG. 2B depicts a method of determining available meals,
recipes or foods for a user based on a user's diet type and other
information. In 220, the system 100 collects and stores information
from the user, such as on goals, weight loss, fitness, well-being,
increasing muscle mass or improving endurance. In some embodiments,
the information on goals may be collected from the user by serving
a webpage with a drop down menu of choices for the user to select.
The goals set forth herein are illustrative only and may include
any goals that have a bearing on the number of calories or types of
meals, foods or supplements that a user with those goals might want
to eat. The goals are stored in the database 105 associated with
the system 100.
[0101] In 222, the system 100 collects and stores user activity
data, such as one or more user's daily exercise or activity levels
in the database 105. This data collection may be done by
synchronizing a remote activity level tracker device or database
associated with the user with the database 105 to transfer data to
the database 105 on a user's activity levels. Alternatively, a user
may upload a general description of the user's regular activity,
daily activity, weekly activities, monthly activities or one time
activities. The user may be prompted to enter this data or may be
given a web page with drop down menus to use to describe regular or
one time activities. The system may determine recommended meals or
foods for users in some embodiments based on activity levels in a
particular day. Alternatively, the activity levels may be used to
determine calories burned by the user over periods of time and then
used in meal recommendations to the user.
[0102] In 224, the system 100 collects and stores food intake
information associated with the user in some embodiments. The food
intake information may include: (i) information the user identifies
to the system, for example in some embodiments, in response to a
web page that the system provides to the user asking for food
intake information; or (ii) information on meals or recipes that
the user has purchased and consumed through the system. In either
case, the user may identify for the system foods and supplements
that the user has eaten or plans to eat in order to get meal or
recipe recommendations for breakfast, lunch or dinner in a given
day; to get snack, supplement or other food recommendations over
the course of several days or a week based on what the user is
expected to eat during that time period. The food intake
information for one or more users may be stored in the database
105.
[0103] In 226, the system 100 collects and stores food preference
information for each user. The food preference information may
include in some embodiments: (i) a list of foods that the user is
allergic to; (ii) a list of foods that the user does not like to
eat; or (iii) a list of foods that the user likes to eat; (iv) the
user's food religion (kosher, vegan, pescatarian and similar). Food
preferences for one or more user are stored in the database 105.
The food preferences may be provided by each user in response to a
web pages soliciting this information with selectable choices. This
information may also be uploaded by a user or a health or other
service provider to the database 105.
[0104] In 228, the system receives information on meals, recipes
and/or hero foods that are available for recommendation to the user
and stores the information in the meals and recipe database 106.
This information may be provided in some embodiments by
administrators of the system 100 to the database meals and recipe
database 106. Alternatively, meals, recipe and other food and
supplement information may be provided by health service providers
102, meal or recipe providers 103 or users 101. The information
such as recipes or available foods or meals in the database 106 may
be designated to be specific to a user or specific to a group of
users, for example a family, those users in a geographic area, or
those users who work at a particular organization. Alternatively,
some meals, foods, recipes or supplements may be designated in the
database 106 to be available to all users or many groups of
users.
[0105] In general, the meals and recipe information for each meal
or recipe includes information on the calories of the meal or
recipe and macronutrient information, such as calories from fat,
carbs and protein or grams of fat, carbs and protein. The
information may also include information of the type shown in FIG.
7 for each meal or recipe. The meal and recipe information may also
include information on micronutrients, such as the volume, weight,
or RDA percentage of one or more micronutrients. The meal
processing engine 107 may provide macronutrient and micronutrient
information based on the contents of the meal, recipe, food or
supplement and known averages for the types of food in the recipe
or meal or the types of nutrients in the food or supplements being
described. Alternatively, the macronutrient and micronutrient
information for the meal, recipe, food or supplement may be input
by a meal or recipe provider or an administrator of the system.
Meals or foods may also be stored with a breakfast, lunch, dinner,
snack, hero food, supplement or other similar designation to
facilitate specific recommendations to the user. Meals or recipes
may be designated in more than one category in some
embodiments.
[0106] In 230, meals, recipes, foods and/or supplements in the
database 106 that are associated with the user may be filtered in
order to determine available meals, recipes, foods or supplements
for the user. One or more filters may be selected an applied for
each user. For example, in some embodiments the available meals and
recipes are filtered based on the user's biological diet type 116.
This filtering is based on, for example, macronutrient
recommendations and meals that do not fit within macronutrient
ranges are filtered out.
[0107] In some embodiments, in 117 a user's food preferences are
used to filter the available meals, recipes, foods or supplements.
When a user's food preferences indicate that the user cannot eat
fish, for example, then meals or recipes with fish will be filtered
out. Similarly, other meals with one or more ingredients that are
not allowed or desired for a user are filtered out in some
embodiments.
[0108] In some embodiments, a user may provide other criteria in
118 that is used to filter meals. For example, a user might have a
goal of not exceeding 500 calories at dinner. This criteria may be
used to filter available dinners that have fewer than 500 calories.
Similarly, a user may specify a criteria that the user is searching
for one or more dinner meals or recipe. This criteria may be used
to filter out breakfast or lunch recipes.
[0109] After any user (or user group) specific filtering 115 is
applied to the available meals, recipes, foods and/or supplements,
the available meals, recipes, foods and/or supplements 120 are
generated and stored in connection with the user. These are
available meals, recipes, food and/or supplements for a user based
on each user's preferences, biological diet type and other criteria
in some embodiments.
[0110] FIG. 2C depicts a method of generating meal, recipe, food or
supplement recommendations for a user according to some
embodiments. The method of 2C may be applied to selecting meals or
recipes. Similarly, the method of 2C may be applied to selecting
snacks, such as hero foods or other snacks with an ingredient list
or supplements. Available meals, recipes, foods or supplements
stored in 232 may be retrieved in 240 in connection with a
particular user in order to make one or more recommendations to the
user. In 242, the system 100 retrieves macronutrient and
micronutrient recommendations for the user, diet type information
associated with the user, and other meal ranking parameters. One or
more of the following meal ranking parameters may be used in some
embodiments: [0111] Meal type--breakfast, lunch, dinner or snack;
[0112] User activity level; [0113] User goals; [0114] User food
intake; [0115] User group or organization; [0116] Cost of meals or
recipes; [0117] Availability of ingredients for meals or recipes;
[0118] Micronutrients; [0119] Macronutrients; [0120] Calories;
[0121] Available meals associated with other users and the ability
to share ingredients among a user group for which meals are being
prepared; [0122] Past user meal selections; and [0123] Meal variety
in view of past meal selections;
[0124] The meal ranking parameters in some instances are specific
to users, user groups or geographies where users are located. In
other instances, the meal ranking parameters may be specific to the
meal preparer, or to the specific meals or recipes or
ingredients.
[0125] In 244 a meal ranker algorithm is applied. In some
embodiments, the meal ranker algorithm ranks meals based on the
user's micronutrient recommendations and the ability of the meal to
provide those micronutrients. This is performed in some embodiments
by applying for at least some micronutrients recommend for the
user, the following equation:
((Micronutrient amount in the meal-Micronutrient recommendation for
the user)/(Micronutrient recommendation for the user+Micronutrient
amount in the meal)) 2
[0126] Each micronutrient subject to the calculation is then summed
together for each meal. The highest ranked meal has the lowest
micronutrient score. The meals are ranked from first to last based
on the lowest to highest micronutrient score. The top X meals or
recipes are then transmitted or recommended to the user in 244. The
value of X may be any number that is designed to give the user some
choices without flooding the user with too many choices. When
snacks supplements or hero foods are being ranked or recommended,
those may be transmitted in 246 to the user. The foods, such as
prepared meals, recipes, hero foods or supplements, are ranked
and/or recommended for the user and may also be stored for the
user. FIG. 8 depicts a list of micronutrients (or basic foods) that
may be given values specific to a user and used to score each meal,
recipe or snack in the meal ranker algorithm and that also may be
given values in each meal, recipe, snack or supplement in the
database 106.
[0127] The user may be given a web page to specify what
recommendations the user is looking for in order to drive the
method of FIG. 2C. For example, the user may be seeking a dinner
recipe or to order meals for the next week. The user may specify
that the user wants the top 10 recommended meals and/or recipes in
some embodiments. The user may specify that the user wants only
dinner recipes or breakfast, lunch and/or dinner meals and recipes
to choose from. Similarly the user may specify snacks or
supplements. The meal ranker algorithm will select from the
available meals, recipes, foods and supplements and make
recommendations according to the methods described herein after
ranking.
[0128] Other techniques for ranking factor in cost, calories, and
goals. Still other techniques may take into account meals (and
ingredients) being made available to other users based on their
respective diet types so that there are economies of scale for the
food preparation process when there are a plurality of users for
which meals are being prepared. Still other techniques may store
selections of the user in response to past meal recommendations.
This may be used to determine both what the user likes because of
the user choices as well as what the user does not like because the
user does not selected certain recommended meals. In some
embodiments, different hueristic equations may be used to optimize
selections for users. In some embodiments, the other ranking
parameters may be given a score between 0 and 1 (or more than that)
and then added to the micronutrient summation. Meal ranking is then
performed for each meal based on its overall score with the low
score representing a higher rank. There are many ways to rank
meals, recipes, foods and/or supplements based on macronutrient and
micronutrient content and macronutrient and micronutrient
recommendations for the user and other meal ranking parameters and
it will be understood by those having ordinary skill in the system
may prioritize and score meals in a variety of ways all of which
are within the scope of the invention.
Decision Tree Engine
[0129] According to some embodiments of the invention, a user's
diet type and recommended meals and foods are based on an
individualized determination of each user's macronutrient and
micronutrient needs. Referring to FIG. 10, these needs are
determined by receiving vitals 1002, phenotype 1004 and genotype
1006 data from each user.
Vital Information
[0130] In general, the vitals data may include data such as shown
below:
TABLE-US-00002 UserId # or alphanumeric Height # Weight # Sex M/F
Waist Circumference High/Low or High/Med/Low or >33/<33 Blood
Pressure High/Low or #/# Activity Level High/Med/Low or
Calories/day or other measure
In some embodiments, body mass index (BMI) can also be used.
Phenotypic Information
[0131] In addition to the vitals information, the system also
utilizes measurements of phenotypic and genotypic biomarkers to
assess a number of physiological factors such as metabolic health
and endurance, insulin response, etc., as is more fully described
below. The phenotype and genotype data in some embodiments is as
shown in FIGS. 3 and 4 respectively.
[0132] The phenotype data generally includes information obtained
from blood testing on the user. In some embodiments, the user's
blood is sampled after fasting and at future times after ingestion
of a challenge beverage as described in more detail below. The
challenge beverage is designed to provide carbohydrates, fats and
proteins to the user and then measure the user's response at
intervals. The blood samples provide some insight into the user's
ability to process sugars, fats and proteins based on the changes
in biomarkers present in the blood over time. The blood samples
also may include information about cholesterol, vitamin and/or
mineral levels, triglicerides, hormone levels and other
information.
[0133] Accordingly, the user takes a blood sample at a fasting
state, drinks the challenge beverage and then takes blood samples
at a number of different time points, usually from one to three
time intervals, with a fasting level, a measure at 30 minutes and
another at two hours finding use in many situations, although other
time periods can be done, including, but not limited to, thirty
minutes, one hour, two hours and three hours. The blood levels of
one or more of the following phenotypic biomarkers are then assayed
and input into the system, with from one, 5, 10, 15, 20, 25 or all
28 being tested in some embodiments.
[0134] In one embodiment, glucose levels are measured as a marker
of metabolic health and insulin sensitivity as it relates to
metabolic health. Accordingly, glucose can be measured at t=0
(fasting), t=30 minutes (glucose_t30) and t=120 minutes
(glucose_t120).
[0135] In one embodiment, C-peptide biomarkers are used as a
measure of metabolic health and insulin sensitivity as it relates
to metabolic health. The connecting peptide, or C-peptide, is a
short 31-amino-acid polypeptide that connects insulin's A-chain to
its B-chain in the proinsulin molecule and is a marker for how much
insulin a user is making. Accordingly, the C-peptide levels can be
measured at t=0, t=30 minutes (C-peptide_t30) and t=120 minutes
(C-peptide_t120).
[0136] In one embodiment, the blood level of carotenoids in the
plasma are tested for all time points as an indication of
carotenoid intake. In one embodiment, a disposition index is
measured as this is an indicator of beta cell function and thus can
be used to assess metabolic health and insulin sensitivity. In one
embodiment, a hepatic insulin index is done on each time point,
which measures hepatic glucose production (HGP) and calculates
indices of hepatic insulin resistance as an indicator of metabolic
health, insulin sensitivity.
[0137] In one embodiment, several different cholesterol levels are
determined at all time points, including HDL, LDL, total
cholesterol and using a ratio of total cholesterol: HDL
cholesterol.
[0138] In one embodiment, total cholesterol is measured at all time
points. In one embodiment, HDL cholesterol levels are measured at
all time points, which is an indicator of heart health. In one
embodiment, LDL cholesterol levels are measured at all time points
as well.
[0139] In one embodiment, high sensitivity C-reactive protein is
measured at all time points as a biomarker for inflammation. The
cut points of low risk (<1.0 mg/L), average risk (1.0 to 3.0
mg/L), and high risk (>3.0 mg/L) may be used.
[0140] In one embodiment, a magnesium category test is measured at
all time points which is a marker for blood pressure, inflammation
and insulin sensitivity.
[0141] In one embodiment, an Omega-3 index is done at all time
points, which can be used for recommendations regarding the intake
of omega 3 for heart health.
[0142] In one embodiment, a potassium category test is done at all
time points, which is relevant to blood pressure and heart
health.
[0143] In one embodiment, the ratio of two essential amino acids
ARA/AA and EPA is measured at all time points. The AA/EPA ratio is
an indication of levels of cellular inflammation, with a ratio of
1.5 to 3 indicating low inflammation, 3 to 6 indicating moderate
inflammation, 7 to 15 is elevated inflammation and >15
indicating high inflammation.
[0144] In one embodiment, sodium levels are measured at all time
points as an indicator of blood pressure and heart health and for
intake recommendations.
[0145] In one embodiment, the blood level of triglycerides are
measured at a fasting state (t=0), and then at 30 minutes and 120
minutes, as an indication of heart health, blood lipids, metabolic
health and metabolic syndrome.
[0146] In one embodiment, vitamin A levels are measured at all time
points for intake recommendations.
[0147] In one embodiment, vitamin B6 levels are measured at all
time points as an indicator of blood pressure and heart health and
for intake recommendations.
[0148] In one embodiment, vitamin C levels are measured at all time
points as an indicator of blood pressure and for intake
recommendations.
[0149] In one embodiment, vitamin D levels are measured at all time
points for intake recommendations.
[0150] In one embodiment, vitamin B6 levels are measured at all
time points for intake recommendations.
[0151] In one embodiment, zinc levels are measured at all time
points for intake recommendations.
Genotypic Data
[0152] In general, the genotype data is taken from DNA analysis on
the user. Certain single nucleotide polymorphisms (SNPs) or genetic
markers may be selected based on their correlation with health and
dietary intake and are depicted in FIG. 4. In general, one or more
of the following 34 genotypic biomarkers are tested, with from at
least about 5, 10, 15 20, 25, 30 or all 34 finding use in many
embodiments.
[0153] As will be appreciated by those in the art, any number of
standard SNP detection techniques can be used, including, but not
limited to, hybridization methods, enzyme based methods and nucleic
acid sequencing methods. Hybridization methods include, but are not
limited to, dynamic allele-specific hybridization (DASH) genotyping
which takes advantage of the differences in the melting temperature
in DNA that results from the instability of mismatched base pairs;
this is frequently done as in known in the art using molecular
beacon technologies or SNP microarray technologies. Enzymatic
methods include enzyme based amplification technologies, where the
amplification only occurs and/or doesn't occur based on the
presence or absence of the SNP, such as polymerase chain reaction
(PCR), oligonucleotide ligation assays (OLA), primer extension
methods, etc. Nucleic acid sequencing methods utilize a number of
different technologies, including single molecule sequencing
(Pacific Biosciences), sequencing by synthesis (Illumina),
pyrosequencing (454), ion semiconductor (Ion Torrent), and
sequencing by ligation (SOLiD).
[0154] In some embodiments, the user's blood is tested for the
presence of the angiotensin I-converting enzyme insertion/deletion
(ACE I/D) polymorphism ACE rs1799752, the presence of which is
associated with human physical performance including endurance, see
Ma et al., PLOS, The Association of Sport Performance with ACE and
ACTN3 Genetic Polymorphisms: A Systematic Review and Meta-Analysis.
PLoS ONE8(1): e54685, hereby incorporated by reference in its
entirety.
[0155] In some embodiments, the user's blood is tested for the
presence of the angiotensin I-converting enzyme insertion/deletion
(ACE I/D) polymorphism ACE rs4646994, the presence of which is
associated with blood pressure and sodium recommendations. The most
influential dietary factor for the renin-angiotensin system (RAS)
is sodium. Interactions between the ACE I/D polymorphism, sodium
intake and the RAS system determine blood pressure and therefore
influence risk for hypertension.
[0156] In some embodiments, the user's blood is tested for the
presence of the ADAMT69 risk allele rs4607103, the presence of
which is associated with insulin sensitivity, insulin secretion and
fiber recommendations.
[0157] In some embodiments, the user's blood is tested for the
presence of the ADRB3 rs4994, the presence of which is associated
with human physical performance including endurance,
[0158] In some embodiments, the user's blood is tested for the
presence of the AGT rs5051 SNP, the presence of which is associated
with blood pressure and sodium recommendations.
[0159] In some embodiments, the user's blood is tested for the
presence of the AGT rs699 SNP, the presence of which is associated
with blood pressure and sodium recommendations.
[0160] In some embodiments, the user's blood is tested for the
presence of the APOA5-A4-C3-A1 rs964184, the presence of which is
associated with macro fat recommendations, diet type, blood
pressure, insulin sensitivity (specifically fat consumption).
[0161] Cholesteryl ester transfer protein (CETP) is an important
regulator of plasma HDL-C. Several genetic mutations in the CETP
gene were found to be associated with HDL-C levels. Accordingly, in
some embodiments, the user's blood is tested for the CETP rs1532624
allele, the presence of which is an indicator of heart health based
on LDL and a total cholesterol diagnosis.
[0162] In some embodiments, the user's blood is tested for the CETP
rs1532624 allele, the presence of which is an indicator or useful
for classifying the carbohydrate diet types and insulin sensitivity
low carb tree.
[0163] In some embodiments, the user's blood is tested for CYP1A2
rs762551, with the rs762551(A) allele being associated as a "fast
metabolizer" and the (C) allele is by comparison a slower
metabolizer of certain substrates (including caffeine).
[0164] The FADS1 gene codes for the fatty acid delta-5 desaturase,
a key enzyme in the metabolism of long-chain polyunsaturated
omega-3 and omega-6 fatty acids. In some embodiments, the user's
blood is tested for one or both of FADS1 rs174546 or rs174548, as
variants in the fatty acid desaturase 1 (FADS1) gene are also
associated with altered polyunsaturated fatty acids (PUFAs) such as
omega-3, and the presence of these SNPs is used as an indicator of
heart health, blood pressure for the epa dha recommendation (omega
3), for intake omega-3.
[0165] The FTO gene encodes the fat mass and obesity-associated
protein (also known as alpha-ketoglutarate-dependent dioxygenase
FTO). In some embodiments, the user's blood is tested for the FTO
rs11221980 SNP, the presence of which is used for diet type
classification (carbs and fats), and as a marker for insulin
sensitivity for fat consumption, insulin sensitivity for low
carbohydrates, and weight maintenance for energy balance.
[0166] In some embodiments, the user's blood is tested for the FTO
rs9939609 SNP, the presence of which is used for diet type
classification (carbohydrates, proteins and fats), and as a marker
for blood pressure relating to fat.
[0167] In some embodiments, SNPs associated with group-specific
component (vitamin D binding protein) GC gene area tested as they
have been linked by several studies to vitamin D serum
concentrations. The allele associated with lower vitamin D, and
thus the potential for vitamin D insufficiency, is rs2282679(C).
Thus in some embodiments, the user's blood is tested for the GC
rs2282679 SNP, the presence of which is related to the
recommendation for vitamin D levels as well as for
inflammation.
[0168] In some embodiments, the user's blood is tested for the
presence of the GC rs4588 SNP, the presence of which is related to
the recommendation for vitamin D levels as well as for
inflammation.
[0169] In some embodiments, the user's blood is tested for the
presence of the GC rs7041 SNP, the presence of which is related to
the recommendation for vitamin D levels as well as for
inflammation.
[0170] The T-allele of GCKR (glucokinase regulatory protein (GCKR)
gene) SNP rs780094 is associated with increased triglycerides.
Accordingly, in some embodiments, the user's blood is tested for
the presence of the GCJR rs7800094 SNP, the presence of which is
related to insulin sensitivity for fasting glucose levels.
[0171] HLA-DQ is a gene family for a .alpha..beta. heterodimer cell
surface receptor. In some embodiments, a user's blood is tested for
an HLA-DQ SNP, as a number of these are related to celiac disease
and gluten sensitivity. In some embodiments, the SNP is the
HLA-DQ2.2 rs2395182 SNP. In some embodiments, the SNP is the
HLA-DQ2.2 rs4713586 SNP. In some embodiments, the SNP is the
HLA-DQ2.2 rs7775228 SNP. In some embodiments, the SNP is the
HLA-DQ2.5 rs2187668. In some embodiments, the SNP is the HLA-DQ7
rs4639334 SNP.
[0172] The rs4402960 SNP in the insulin like growth factor 2 mRNA
binding protein (IGF2BP2 r54402960) are associated with type-2
diabetes risk and is thus used as a biomarker for the fat diet type
and insulin sensitivity for fat consumption. In some embodiments,
the user's blood is tested for the presence of the IGF2BP2
rs4402960 SNP.
[0173] The IL6 rs1800795 SNP is a SNP in the promoter of the IL-6
gene that is associated with inflammation. In some embodiments, the
user's blood is tested for the presence of the IL6 rs1800795
SNP.
[0174] The MCM6 gene encodes the protein DNA replication licensing
factor MCM6, one of the highly conserved minichromosome maintenance
complex proteins that are essential for the initiation of
eukaryotic genome replication. The MCM6 rs4988235 SNP is associated
with lactose intolerance and lactose sensitivity. In some
embodiments, the user's blood is tested for the presence of the
MCM6 rs4988235 SNP.
[0175] The MTHFR gene encodes the vitamin-dependent enzyme,
methylenetetrahydrofolate reductase, involved in folate metabolism
and thus associated with blood pressure in terms of riboflavin. The
MTHFR rs1801133 SNP Homozygous rs1801133(T;T) individuals have
.about.30% of the expected MTHFR enzyme activity, and
rs1801133(C;T) heterozygotes have .about.65% activity, compared to
the most common genotype, rs1801133(C;C). In some embodiments, the
user's blood is tested for the presence of the MTHFR rs1801133
SNP.
[0176] The nitrous oxide synthase gene NOS3 gene variant rs1799983
is strongly associated with coronary artery disease; a large study
found that homozygosity for rs1799983(T;T) increases risk of
ischemic heart disease and can be used as a biomarker for blood
pressure for cocoa flavanols and resveratrol recommendations. In
some embodiments, the user's blood is tested for the presence of
the NOS3 gene variant rs1799983.
[0177] The PPARG rs1801282 associates with type 2 diabetes and
interact with physical activity, as diet type (fats), insulin
sensitivity for fat consumption In some embodiments, the user's
blood is tested for the presence of PPARG rs1801282 (Pro12Ala).
[0178] In some embodiments, the user's blood is tested for the
presence of the R577X rs1815739 SNP. This SNP, in the ACTN3 gene,
encodes a premature stop codon in a muscle protein called
alpha-actinin-3. The polymorphism alters position 577 of the
alpha-actinin-3 protein. In publications the (C;C) genotype is
often called RR, whereas the (T;T) genotype is often called XX. The
(T;T) is under-represented in elite strength athletes, consistent
with previous reports indicating that alpha-actinin-3 deficiency
appears to impair muscle performance and is accordingly a marker
for muscle performance.
[0179] In some embodiments, the user's blood is tested for the
presence of the TCF7L2 (Transcription Factor 7 Like 2) rs7903146
SNP as this is one of two SNPs within the TCF7L2 gene that have
been reported to be associated with type-2 diabetes, It is used as
a biomarker for diet types relating to carbohydrates and fats,
blood pressure for fat, insulin sensitivity for low carbohydrates,
and weight maintenance for energy balance.
[0180] The TNF rs1800629 SNP in the tumor necrosis factor-alpha
gene, rs1800629, is also known as the TNF-308 SNP. Occasionally the
rs1800629(A) allele is referred to as 308.2 or TNF2, with the more
common (G) allele being 308.1 or TNF1. The (A) allele is associated
with higher levels of TNF expression. This SNP has been linked to a
wide variety of conditions including inflammation. Accordingly, in
some embodiments, the user's blood is tested for this SNP.
[0181] In some embodiments, the user's blood is tested for the
presence of the VDR rs1544410, also known as the BsmI polymorphism,
is a SNP in the Vitamin D receptor (VDR) and is used as a marker
for Vitamin D.
[0182] The decision tree Engine 108 receive the vitals, genotype
and phenotype data for each user and convert this data into
macronutrient and micronutrient recommendations. The
recommendations are essentially vectors that correlate relevant
macronutrients or micronutrients with a level or range for each
user. In the case of macronutrients, the user's vector includes
values as shown for Carbohydrates, Fats and Protein. An
illustrative decision tree for carbohydrates is shown in FIGS. 11A
and B. An illustrative decision tree for Fats is shown in FIG. 12.
An illustrative decision tree for Protein is shown in FIG. 13. In
general, the decision trees receive the inputs of vitals, genotype
and genotype data, and through the application of rules and logic,
the decision trees produce the user's macronutrient recommendation
vector. The range of values produced and included in the user's
macronutrient recommendation vector may be as shown in FIGS. 5 and
10. Alternatively, values, value ranges thresholds may be applied.
As shown in FIG. 5, the macronutrient recommendations may be mapped
into diet types. Alternatively, the decision tree or decision logic
may directly output diet types from input values. The macronutrient
recommendations and diet types for each user in some embodiments
are based on vitals, phenotype and genotype data for each user.
[0183] The micronutrient recommendations for each user are
similarly based on the vitals, phenotype and genotype data for each
user. However, certain micronutrient recommendations may be based
on less than all three data types. A list of micronutrients and/or
foods, levels for all or some of which may be determined for each
user are shown in FIG. 8. Meals, recipes, foods, snacks and
supplements that are stored in the database 106 also may include
information on levels of micronutrients such as those in the list
of FIG. 8. Both for the macronutrient recommendations and the
micronutrient recommendations, the decision logic may include
determining intermediate values that are used in determining
multiple macronutrient or micronutrient recommendations. Some
examples of intermediate values include
[0184] The decision tree engine may implemented in program
instructions that implement decision tree logic that are stored in
memory of a computer and then are executed by a processor within
the computer to process the inputs and produce macronutrient,
micronutrient and diet types based on the vitals, genotypical and
phenotypical data for each user. The decision trees may be static.
Alternatively, the decision tree logic may be updated over time.
The relevant vitals, phenotypical or genotypical data for each user
that is used in the recommendations may also change over time in
some embodiments. The changes in decision tree logic may be driven
by new scientific information about food and the impact of genotype
or phenotype on health in some embodiments. In some embodiments,
the decision tree logic be updated based on feedback from results
of users of the system as the vitals and phenotypical data of users
change over time based on their meals. activity levels and
aging.
[0185] In general, each of the methods and processes shown and
described herein may be implemented on a server or other computer
and the web server interface, decision tree engine, filtering
engine and meal ranker engine may implemented by a server or other
network connected computer. These computers may one computer or may
be centralized or distributed and may share data with each other
and other network elements shown in FIG. 1 via the Internet, local
area networks, wide area networks or other networks. The processes
in some embodiments are implements as program instructions that may
be stored as software or firmware in the memory of a device or
other computer and executed by a processor. In general, for each of
the devices, servers and engines shown herein, the device includes
a memory, a processor, input/output units, and networking units.
The processor executes program instructions to perform the
processes shown and described herein, including database queries,
web interfaces, meals processing, health decision trees, filtering,
meal ranking and other user interactions to ensure user
registration, meal and food recommendations and in other instances
payment and arranging for delivery of meals or other food.
[0186] The databases include stored data regarding users, which may
be stored in an encrypted and secure manner. Additional information
that is collected or generated during the processes shown and
described herein may be stored in the databases. In general, the
databased are network connected and may store or provide
information in response to queries to any of the network elements
in order to facilitate the processes shown and described
herein.
[0187] FIG. 15 illustrates methods and systems for personalized
food and nutrition recommendation system 1500, in accordance with
some embodiments. Information about the user 1502 is collected,
e.g., one or more of genotypic information 1506, phenotypic
information 1508 which, in some embodiments includes metabolic
adaptability information determined, for example, through analysis
of the user's blood following consumption of a multi-nutrient
challenge beverage as described herein, food preferences 1510
(e.g., food likes, dislikes, food religions, or other dietary
preferences), anthropometrics 1512 (e.g., physical measurements of
the individual), goals 1514 (e.g., weight loss, muscle building, or
increases in energy), dietary patters 1516 (e.g., eating habits or
food logs), and activity patterns 1518 (e.g., typical physical
activities, exercise logs, or measured caloric outputs). In some
embodiments, information about the user is collected multiple
times, e.g., before initial classification and one or more times
after adapting a particular diet. In some embodiments, information
collected after implementation of a food habit is used to track
changes in the user and/or adjust classification of the user based
on changes accompanying the adapted food habits. For example, a
user initially identified as having elevated blood pressure may be
initially classified as requiring a diet low in fats. However, upon
re-testing after implementing a low fat diet, it may be found that
the user's blood pressure has been reduced. This information can be
used to reclassify the user as no longer requiring a diet low in
fats, e.g., in combination with other risk factors.
[0188] The information about the user is applied to one or more
food recommendation classifiers, e.g., one or more of diet type
classifier 1520, micronutrient recommendation classifier 1522,
caloric recommendation classifier 1524, hero food classifier 1525,
and a supplement recommendation classifier 1552, to provide one or
more food classifications and/or recommendations for the user,
e.g., one or more of a diet type 1526, a micronutrient
recommendation profile 1528, a source recommendation profile 1530,
a caloric recommendation 1532, a hero food recommendation 1533, and
a supplement recommendation classifier.
[0189] In one embodiment, a method for recommending foods to a user
includes obtaining genotypic data about the user comprising a
plurality of first features X={x.sub.1 . . . , x.sub.m} (e.g., one
or more of the genotypes described above with respect to FIG. 1
and/or identified in FIG. 4), wherein each respective feature xi in
the plurality of first features X is a status of a locus in a
plurality of loci and obtaining phenotypic data about the user
comprising a plurality of second features Y={y.sub.1 . . . y.sub.n}
(e.g., one or more of the phenotypes described above with respect
to FIG. 1 and/or identified in FIG. 3), wherein each respective
feature y.sub.i in the plurality of second features Y is a status
of a phenotype in a plurality of phenotypes.
[0190] The method then includes assigning a respective diet type
D.sub.j in a plurality of diet types D={D.sub.1 . . . , D.sub.q}
(e.g., assigning one of diet types 1-7 as described above with
respect to FIGS. 1 and 5) to the user by inputting a first
sub-plurality X.sub.1 of the plurality of first features X and a
first sub-plurality Y.sub.1 of the plurality of second features Y
into a diet type classification model (e.g., diet type classifier
1520 in FIG. 15, health decision tree engine 108 in FIG. 1, and/or
and illustrative macronutrient classification models in FIGS.
11-13).
[0191] The method also includes assigning a micronutrient
recommendation profile R.sub.j={r(z.sub.i) . . . , r(z.sub.s)}
comprising a recommendation r(z.sub.i) for each respective
micronutrient z.sub.i in a plurality of micronutrients Z={z.sub.1 .
. . , z.sub.s} (e.g., one or more of the micronutrients identified
in FIG. 8 and/or described above with reference to FIG. 1) to the
user by inputting a second sub-plurality X.sub.2 of the plurality
of first features X and a second sub-plurality Y.sub.2 of the
plurality of second features Y into a micronutrient classification
model (e.g., one or more illustrative micronutrient classification
model in FIGS. 16-23).
[0192] Finally, the method includes ranking one or more foods in a
plurality of foods L={N.sub.1 . . . , N.sub.t} (e.g., foods 1534
such as meals in a master library of meals or menu of selected
meals, for example, a weekly menu of meals), wherein each
respective food N.sub.i in the plurality of foods has a
corresponding nutrition profile P.sub.Ni={D.sub.ki, P(z.sub.ki)}
comprising an assigned diet type D.sub.k in the plurality of diet
types D and an assigned micronutrient profile
P(z.sub.kk)={v(z.sub.1) . . . v(z.sub.s)}, wherein the
micronutrient profile P(z.sub.k) includes a respective value
v(z.sub.i) for each micronutrient Z.sub.i in the plurality of
micronutrients Z, by comparing the diet type D.sub.j and
micronutrient recommendation profile R.sub.j assigned to the user
to the nutrition profiles P.sub.N of foods N in the plurality of
foods L (e.g., via one or more of user specific filtering engine
115 as described with respect to FIG. 1, meal ranker engine 125 as
described with respect to FIG. 1, and food selection classifier
1536 described with respect to FIG. 15).
[0193] In some embodiments, assigning a respective diet type
D.sub.j includes assigning macronutrient recommendations for fat,
carbohydrate, and protein intake to the user and then matching the
assigned macronutrient recommendations to a diet type D (e.g., one
of the seven diet types described above with reference to FIG.
5).
[0194] For example, in some embodiments, the method includes
assigning a macronutrient fat intake recommendation F.sub.j to the
user by inputting a third sub-plurality X.sub.3 of the plurality of
first features X and a third sub-plurality Y.sub.3 of the plurality
of second features Y into a fat recommendation classification model
(e.g., the fat recommendation classifier described above with
reference to FIG. 12). In some embodiments, the user is assigned
either a low fat dietary recommendation (f) or a regular fat
dietary recommendation (F). In other embodiments, the fat
macronutrient dietary recommendation is one of more than two
classes of recommendations, e.g., one of three, four, five, or more
classes of recommendations.
[0195] In some embodiments, the method also includes assigning a
macronutrient carbohydrate intake recommendation C.sub.j to the
user by inputting a fourth sub-plurality X.sub.4 of the plurality
of first features X and a fourth sub-plurality Y.sub.4 of the
plurality of second features Y into a carbohydrate recommendation
classification model (e.g., the carbohydrate recommendation
classifier described above with reference to FIG. 11). In some
embodiments, the user is assigned either a low carbohydrate dietary
recommendation (c) or a regular fat dietary recommendation (C). In
other embodiments, the carbohydrate macronutrient dietary
recommendation is one of more than two classes of recommendations,
e.g., one of three, four, five, or more classes of
recommendations.
[0196] In some embodiments, the method also includes assigning a
macronutrient protein intake recommendation P.sub.j to the user by
inputting a fifth sub-plurality X.sub.5 of the plurality of first
features X and a fifth sub-plurality Y.sub.5 of the plurality of
second features Y into a carbohydrate recommendation classification
model (e.g., the protein recommendation classifier described above
with reference to FIG. 11). In some embodiments, the user is
assigned either a low protein dietary recommendation (p) or a
regular protein dietary recommendation (P). In some embodiments,
the user is assigned either a low protein dietary recommendation
(p), a regular protein dietary recommendation (P), or a high
protein dietary recommendation (P+). In some embodiments, the user
is assigned either a low protein dietary recommendation (p), a
regular protein dietary recommendation (P), a high protein dietary
recommendation (P+), or an extra high protein dietary
recommendation (P++). In other embodiments, the carbohydrate
macronutrient dietary recommendation is one of more than four
classes of recommendations, e.g., one of five, six, seven, or more
classes of recommendations.
[0197] In some embodiments, the method includes comparing the
assigned macronutrient fat intake recommendation Fj, macronutrient
carbohydrate intake recommendation C.sub.j, and macronutrient
protein intake recommendation P.sub.j to the plurality of diet
types D={D.sub.1 . . . , D.sub.q}. In some embodiments, every
combination of fat, carbohydrate, and protein dietary
recommendations defines a different diet type. In other
embodiments, certain combinations of fat, carbohydrate, and protein
dietary recommendations are classified in a same diet type (for
example, in the diet type classifications described above with
respect to FIG. 5, FCP+ and FCP++ combinations both correspond to
Diet Type 2). In yet other embodiments, one or more combination of
fat, carbohydrate, and protein dietary recommendations is
associated with more than one diet type, for example, based on one
or more additional factors (e.g., a particular genotypic marker,
phenotypic marker, metabolic adaptability feature, food preference,
food religion, anthropometric feature, user goal, dietary pattern,
or activity pattern).
[0198] In some embodiments, the food classifications and/or
recommendations assigned to the user are used to provide ranked
food recommendations 1548 using food selection classifier 1536. In
some embodiments, the user's food classifications and/or
recommendations, along with list of foods 1534 (e.g., a list of all
meals in a menu database, or a sub-selection of meals, such as a
menu of meals to be prepared on a particular week) are input into
food selection classifier 1536, which optionally includes one or
more of diet type prioritization algorithm 1538, preference filter
1540, allergy and/or sensitivity filter 1542, source filter 1544,
and micronutrient ranking algorithm 1546. In various embodiments,
any or all of these components are used in any order to rank foods
for recommendation to a user.
[0199] In some embodiments, food selection classifier 1536 assigns
a numerical value to one or more of foods 1536. In some
embodiments, the numerical value for a particular food reflects
both a diet type suitability of the food for a user and a
micronutrient suitability of the food for a user. For example, in
some embodiments, the food is assigned a first number corresponding
to a diet type of the food and a second number corresponding to a
micronutrient profile of the food. For example, a food assigned to
a first Diet Type may be assigned a value of 1 and a food assigned
to a second Diet Type may be assigned a value of 5. Then a second
value is assigned to each food based on a similarity of the
micronutrients in the food to a micronutrient recommendation
profile of the user. In some embodiments, the two numbers are kept
separate, e.g., as an ordered pair of numbers (X, Y) or X. Y. In
other embodiments, the two numbers may be combined arithmetically,
e.g., by generating a sum of the two numbers. In this fashion, the
foods can then be ranked numerically to determine which foods are
best suited for the user.
[0200] In some embodiments, Diet type prioritization algorithm 1538
filters or ranks foods (e.g., meals) based on a comparison between
the diet type assigned to a user and a diet type assigned to the
food (e.g., meal). For example, in some embodiments, each food is
classified as belonging to one of the Diet Types (e.g., Diet Types
1-7, as described herein with reference to FIG. 5) and foods having
the same Diet Type designation as a user's Diet Type assignment are
prioritized over foods having different Diet Type designations as
the user's Diet Type assignment. In some embodiments, a food having
a Diet Type designation that is different from the user's Diet Type
assignment is filtered out (e.g., removed from a list of eligible
foods for the user).
[0201] In some embodiments, the food is assigned a Diet Type
designation based on the fat, carbohydrate, and protein contents of
the food. In some embodiments, the fat, carbohydrate, and protein
contents of the food are used to classify the food according to the
same fat, carbohydrate, and protein consumption recommendations
assigned to users. For example, a food with a carbohydrate content
below a threshold value (e.g., according to the percent of
carbohydrates by weight or calories in the food) is assigned a low
carbohydrate food designation (c) that corresponds to a low
carbohydrate dietary recommendation (c). Conversely, a food with a
carbohydrate content above a threshold value (e.g., according to
the percent of carbohydrates by weight or calories in the food) is
assigned a high carbohydrate food designation (C) that corresponds
to a low carbohydrate dietary recommendation (C). Likewise, the
food is assigned one of a plurality of fiber dietary
recommendations (e.g., for F) and protein dietary recommendations
(e.g., p or P; or p, P, or P+; or p, P, P+, or P++). The
combination of fat, carbohydrate, and protein classification of the
food is then mapped to a Diet Type (e.g., one of Diet Types 1-7, as
described herein with reference to FIG. 5).
[0202] In some embodiments, preference filter 1540 is applied to
deprioritize foods that does not comply with a user's preference
(e.g., vegetarian, dairy-free, gluten free, kosher, etc.). In some
embodiments, the system removes a food that does not comply with a
user's preference from a list of eligible foods for the user.
[0203] In some embodiments, allergy/sensitivity filter 1542 is
applied to deprioritize foods the user is allergic to and or is
sensitive. In some embodiments, the system removes a food the user
is allergic to or sensitive to from a list of eligible foods for
the user. For example, in some embodiments, food selection
classifier 1536 applies a sodium filter to deprioritize or remove
meals with a sodium content above a threshold level when the user
has been identified as having a salt sensitivity. In some
embodiments, food sensitivities are determined based on a user
feature 1504 (e.g., a genotype 1506, phenotype 1508, or metabolic
adaptability characteristic).
[0204] In some embodiments, source filter 1544 is applied to
deprioritize foods that do not comply with a source recommendation
for the user (e.g., a MUFA or Fiber source recommendation as
described herein with reference to FIG. 16). In some embodiments,
the system removes a food that does not comply with a source
recommendation for the user from a list of eligible foods for the
user.
[0205] In some embodiments, micronutrient ranking algorithm 1546 is
applied to prioritize foods with micronutrient profiles that most
closely match a micronutrient recommendation profile assigned to
the user (e.g., user micronutrient classifications 110 described
herein with reference to FIG. 1 and/or micronutrient recommendation
profile 1528 as described herein with reference to FIG. 15).
[0206] In some embodiments, food selection classifier 1536 adjusts
the ranking of one or more meals (e.g., deprioritizes) belonging to
a same meal family (e.g., meals having similar bases that vary, for
example, primarily by the identity of the protein) as a higher
ranked meal. For example, where a list of available meals includes
both beef over noodles and chicken over noodles, the lower ranked
meal will be deprioritized with in the ranking to avoid presenting
the user with highly similar meal choices.
[0207] In some embodiments, the systems and methods described
herein also include providing a caloric recommendation C.sub.j to
the user by inputting a sixth sub-plurality X.sub.6 of the
plurality of first features X and a sixth sub-plurality Y.sub.6 of
the plurality of second features Y into a caloric recommendation
classification model (e.g., caloric recommendation classifier 1524
illustrated in FIG. 15). In some embodiments, the caloric
recommendation classifier uses features of the user, e.g., one or
more of gender, age, height, weight, waist circumference, and
activity levels, to assign a caloric recommendation (e.g., caloric
recommendation 1532 illustrated in FIG. 15) to the user, for
example, a recommendation on how many calories to consume at a
single meal, an entire day, a week, etc.
[0208] In some embodiments, food selection classifier 1536 applies
caloric recommendation 1532 to prioritize foods (e.g., meals) that
closely match the user's caloric requirements. In some embodiments,
the system deprioritizes a food (e.g., a meal) that does not
conform with a user's caloric recommendation, e.g., a food with a
calorie content that exceeds a maximum calorie content determined
based on the user's caloric recommendation and/or a food with a
calorie content less than a minimum calorie content determined
based on the user's caloric recommendation. In some embodiments,
the system removes a food that does not conform to a user's caloric
recommendation from a list of eligible foods for the user.
[0209] In some embodiments, one or more ranked food recommendations
1548 are presented to the user, e.g., through a web-based user
interface. In some embodiments, the ranked food recommendations
correspond to meals that can be prepared and/or delivered to the
user. The user selects user food selections 1550 from ranked food
recommendations 1548, which are prepared and/or delivered to the
user in some embodiments (e.g., as food delivery 1556 illustrated
in FIG. 15).
[0210] In some embodiments, ranked food recommendations 1548
represent a sub-plurality of all available foods 1534, which most
closely fit food classifications and/or recommendations for the
user. In some embodiments, the user selects a number of meals to be
displayed, e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more meals for a
particular week. In some embodiments, the user specifies the number
and types of meals to be displayed, e.g., a certain number of
breakfasts, a certain number of lunches, and a certain number of
dinners. The system then selects the meals that best match the
user's food profile (e.g., classifications and/or recommendations)
and displays suggested meals to the user. In some embodiments, the
system also displays one or more alternative meals to the user that
the user may select in lieu of a suggested meal. In some
embodiments, the alternative meals are those ranked just below the
suggested meals by the food selection classifier.
[0211] In some embodiments, the system monitors and analyses user
food selections 1550 over one or more user selection events and
uses the information to refine food selection classifier 1536 for
the user. For example, where the user consistently chooses an
alternative meal containing chicken for a suggested meal containing
salmon, the system may update food selection classifier 1536 for
the user to more heavily weight meals containing chicken and/or
less heavily weight meals containing salmon. In some embodiments, a
learning classifier algorithm is implemented to refine the output
of food selection classifier 1536 for the individual.
[0212] In some embodiments, the system monitors and analyses user
food selections 1550 over one or more user selection events for a
plurality of users and uses the information to refine a master list
of meals (e.g., foods 1536), selection of meals for a particular
menu (e.g., selection of foods 1534 from a master list of foods),
and/or development of new meals to be added to a master list of
meals. For example, if the system identifies a pattern that users
select meals containing chicken more often than meals containing
beef, the system may refine an algorithm used to select potential
meals to offer chicken dishes more often and/or beef dishes less
often on a global scale (e.g., for all or a subset of users of the
system.)
[0213] In some embodiments, the methods and systems described
herein apply features 1504 of the user to a supplement
recommendation classifier (e.g., supplement recommendation
classifier 1552 illustrated in FIG. 15) to provide a supplement
recommendation (e.g., supplement recommendation 1554). In some
embodiments, the supplements recommended to a user are selected
from a predetermined list of supplements that address different
health needs, e.g., one or more of metabolic health, cholesterol
reduction, maintenance of polyunsaturated fat (e.g., omega-3 fatty
acids) levels, blood pressure control, cardiac health, and general
health (e.g., in a gender-specific or gender-neutral fashion).
[0214] In some embodiments, the supplement recommendation
classifier ranks potential supplement recommendations for a user
(e.g., based on a classifier that considers, for example, one or
more of the importance of the supplement to health and the user's
need for the particular supplement) and selects up to a
predetermined number of supplement recommendations to provide the
user (e.g., the top 2, 3, 4, 5, 6, 7, 8, 9, or more supplements).
For example, in one embodiment, the supplement recommendation
classifier may rank a first supplement over a second supplement
because the first supplement has been shown to greatly reduce
incidence of cardiac failure, while the second supplement has a
largely cosmetic effect, regardless of the user's relevant needs
for the two supplements. In another embodiment, the supplement
recommendation classifier may rank the second supplement, with the
largely cosmetic effect, higher than the first supplement,
associated with greatly reduced incidence of cardiac failure, if a
user has a much greater need for the second supplement than for the
first supplement.
[0215] In some embodiments, a metabolic supplement is recommended
to a user that would benefit from assistance with maintaining blood
glucose levels. In one embodiment, a metabolic supplement contains
one or more of green tea catechins and chromium picolinate, known
to contribute to maintenance of normal blood sugar.
[0216] In some embodiments, a phytosterol supplement is recommended
to a user that would benefit from assistance maintaining healthy
cholesterol levels because phytosterols have been shown to reduce
cholesterol levels.
[0217] In some embodiments, a cardiac health supplement is
recommended to a user that would benefit from assistance
maintaining a healthy cardiac system. In one embodiment, a cardiac
health supplement contains one or more of coenzyme Q10 and
grapeseed extract, both of which promote healthy blood vessels.
[0218] In some embodiments, an omega-3 fatty acid supplement is
recommended to a user that would benefit from assistance
maintaining healthy polyunsaturated fat levels. In one embodiment,
an omega-3 fatty acid supplement contains one or more of fish oil
and algal oil because EPA and DHA contribute to maintenance of
healthy omega-3 fatty acid levels.
[0219] In some embodiments, an omega-3 fatty acid supplement is
recommended to a user that would benefit from assistance lowering
their blood pressure. In one embodiment, an omega-3 fatty acid
supplement contains one or more of fish oil and algal oil because
EPA and DHA contribute to maintenance of normal blood pressure.
[0220] In some embodiments, recommended supplements are delivered
to the user (e.g., along with user food selections as part of food
delivery 1556). In some embodiments, as a consequence of providing
the user with one or more recommended supplements, the system
provides feedback to one or both of the food selection classifier
engine (e.g. meal ranker engine 125 as described herein with
reference to FIG. 1 and/or food selection classifier 1536 as
described herein with reference to FIG. 15) and hero food
recommendation engine, that the user has been provided a
supplement. In some embodiments, the food selection classifier
engine and/or hero food recommendation engine considers that the
user is taking supplements when making a future food
recommendation. For example, in some embodiments, in response to an
input that the user has or will be provided a fish oil supplement,
the food selection classifier deprioritizes foods (e.g., meals)
containing fish and/or foods (e.g., meals) high in omega-3 fatty
acids, because the user is receiving a large amount of omega-3
fatty acids from the fish oil supplements. In one embodiment, the
system will remove a food (e.g., a meal) containing fish and/or
high in omega-3 fatty acids, from a list of foods available to the
user while the user is receiving fish oil supplements. Likewise, in
some embodiments, a hero food recommendation engine (e.g., meal
ranker engine 125 in FIG. 1 and/or hero food recommendation
classifier engine 1525 in FIG. 15) deprioritizes and/or removes a
hero food recommendation high in omega-3 fatty acids while the user
is receiving fish oil supplements.
[0221] In some embodiments, the systems and methods described
herein also include providing a hero food recommendation H.sub.j to
the user by inputting a seventh sub-plurality X.sub.7 of the
plurality of first features X and a seventh sub-plurality Y.sub.7
of the plurality of second features Y into a hero food
recommendation classification model (e.g., a meal ranker engine 125
as described herein with respect to FIG. 1 and/or a hero food
recommendation classifier engine 1525 as described herein with
respect to FIG. 15). In some embodiments, the hero food
recommendation classifier uses features and/or Diet Type
assignments to recommend one or more hero foods (e.g., one or more
hero foods shown in FIG. 14) to the user.
[0222] It should be understood that the particular order in which
the operations in the methods and systems described above with
respect to FIG. 15 have been described is merely an example and is
not intended to indicate that the described order is the only order
in which the operations could be performed. One of ordinary skill
in the art would recognize various ways to reorder the operations
described herein. Additionally, it should be noted that details of
other processes described herein with respect to other methods
described herein are also applicable in an analogous manner to
methods and systems described above with respect to FIG. 15. For
example, information collection methods, the classifiers,
genotypes, phenotypes, vitals, communication networks, computer
infrastructures, etc. described above with reference to FIG. 15
optionally have one or more characteristics of the information
collection methods, classifiers, genotypes, phenotypes, vitals,
communication networks, computer infrastructures, etc. described
with reference to FIG. 1. For brevity, these details are not
repeated here.
[0223] In some embodiments, the methods described herein include
assigning one or more source recommendation to an individual. In
some embodiments, the source recommendations include a fiber source
recommendation, suggesting that the user eat foods higher in fiber
(e.g., a recommendation that the user consumes foods with a minimum
amount of fiber or in which a minimum percentage of carbohydrates
are fibers). In some embodiments, the source recommendations
include a monounsaturated fatty acid source recommendation,
suggesting that the user eat foods higher in monounsaturated fatty
acids (e.g., a recommendation that the user consumes foods with a
minimum amount of monounsaturated fatty acids or in which a minimum
percentage of fats are monounsaturated fatty acids).
[0224] FIG. 16 shows an illustrative classifier for providing
monounsaturated fatty acid (MUFA) and fiber source recommendations
(e.g., an exemplary source recommendation profile S.sub.j, as
illustrated in FIG. 15), in accordance with some embodiments. In
some embodiments, a classifier providing source recommendations is
implemented as part of a micronutrient recommendation classifier,
e.g., as illustrated in FIG. 15. In other embodiments, a classifier
providing source recommendations is implemented separate from a
micronutrient recommendation classifier.
[0225] In FIG. 16, user features (e.g., genotypes, phenotypes,
vitals, anthropometrics, and metabolic adaptability traits) that
lead to a MUFA or Fiber source recommendation are shown of the left
hand side of the table. The source recommendation assigned to the
user trait is represented by an `X` on the right side of the table.
For example, identifying the user as having elevated blood pressure
results in both a MUFA and a fiber recommendation, in accordance
with some embodiments. (*) Individuals with an increased waist
circumference (WC) plus the FTO risk variant will also get a fiber
recommendation because of their increased WC (e.g., independent of
their rs9939609 allele status). (**) Individuals with a low
disposition index with impaired fasting glucose (IFG), impaired
glucose tolerance (IGT), or IGT & IFG will also get a fiber
recommendation because of their IFG, IGT, or IGT & IFG.
[0226] In some embodiments, the methods described herein include
providing the user with information about their metabolic
flexibility associated with consuming one or more of fats,
carbohydrates, and protein. For example, FIG. 17 shows an
illustrative classifier for providing the user with information
about their metabolic flexibility associated with consuming
protein, in accordance with some embodiments. In FIG. 17, user
features (e.g., genotypes, phenotypes, vitals, anthropometrics, and
metabolic adaptability traits) that result in information about a
user's protein consumption flexibility are shown of the left hand
side of the table. The flexibility associated with the user's
feature is shown on the right side of the table. For example,
determining the user has elevated blood pressure identifies the
user as having flexibility to consume a diet rich in protein (e.g.,
in which 18-30% of the user's calories come from protein).
[0227] FIG. 18 shows an illustrative classifier for providing
micronutrient recommendations based on user features (e.g., as
described above with respect to health decision tree engine 108 in
FIG. 1 and/or micronutrient recommendation classifier 1522 in FIG.
15), in accordance with some embodiments. In FIG. 18, user features
(e.g., genotypes, phenotypes, vitals, anthropometrics, and
metabolic adaptability traits) determinative of a micronutrient
recommendation are displayed across the top of the table, while the
micronutrient is identified at the left of the table.
[0228] In some embodiments, a default micronutrient recommendation
is provided (e.g., one associated with a daily recommended intake
for the micronutrient) and the system modifies the micronutrient
recommendation when detecting a user feature associated with an
increased need for, or beneficial results of, consuming more or
less of the particular micronutrient. For example, as illustrated
in FIG. 17, base-line recommendations (DRI) for the micronutrient
are shown in the column next to the micronutrient. Modified
micronutrient recommendations for a user identified with a
particular feature are shown below the feature identified and
in-line with the micronutrient. For example, as illustrated in FIG.
20, a user identified as having elevated or high impaired glucose
tolerance (e.g., as identified using a challenge beverage test as
further described herein) is assigned one or more of the following
recommendations: that they consume 90 grams of whole grains, that 5
grams out of every 100 grams of carbohydrates they consume are
alpha-cyclodextrin, 8 grams out of every 100 grams of carbohydrates
they consume are arabinoxylan, 3.5 grams out of every 100 grams of
carbohydrates they consume are beta-glucans, and 14 grams of every
100 grams of carbohydrates they consume are resistant starch.
[0229] In some embodiments, the systems and methods provided herein
apply classifiers providing recommendations for one or more of the
micronutrients listed in FIG. 8. In some embodiments, a
micronutrient classifier is informed by studies linking improved
health to the administration of a micronutrient to subjects with a
specific feature (e.g., genotype, phenotype, metabolic flexibility,
anthropometric characteristic, etc.).
[0230] In one embodiment, the disclosure provides a method 2800 for
providing personalized food recommendations. The method includes
obtaining (2802) feature data about a user, for example, one or
more features as described herein with reference to FIG. 1 (e.g.,
via user health database 105), FIG. 2 (e.g., storing (202) user
vitals, genotypic, and phenotypic data), FIG. 3 (e.g., illustrative
phenotypes), FIG. 4 (e.g., illustrative genotypes), FIG. 10 (e.g.,
user vitals data 1002, user phenotypic data 1004, and user
genotypic data 1006), and FIG. 15 (e.g., feature data 1504). In
some embodiments, the user feature data includes one or more of a
plurality of genotypic markers X={x.sub.1 . . . x.sub.m} (2804) of
the user, a plurality of phenotypes Y={y.sub.1 . . . y.sub.n}
(2806) of the user, one or more metabolic adaptability
characteristics (2808), e.g., as identified using a multi-nutrient
challenge beverage, one or more food preference (2810), one or more
user goals (2814), one or more user dietary patterns (2816), and
one or more user activity patterns (2818).
[0231] In some embodiments, the method includes assigning (2820) a
respective diet type D.sub.j in a plurality of diet types
D={D.sub.1 . . . D.sub.q} (e.g., diet types 1-7 as described herein
with reference to FIG. 5) to the user by inputting user features,
including a first sub-plurality X.sub.1 of the plurality of first
features X and a first sub-plurality Y.sub.1 of the plurality of
second features Y, into a diet type classification model (e.g.,
health decision tree engine 108 as described herein with reference
to FIG. 1 and/or diet type classifier 1520 as described herein with
reference to FIG. 15).
[0232] In some embodiments, assigning a respective diet type
includes (2822): assigning a macronutrient fat intake
recommendation F.sub.j to the user by inputting a third
sub-plurality X.sub.3 of the plurality of first features X and a
third sub-plurality Y.sub.3 of the plurality of second features Y
into a fat recommendation classification model (e.g., as described
herein with reference to FIG. 12), assigning a macronutrient
carbohydrate intake recommendation C.sub.j to the user by inputting
a fourth sub-plurality X.sub.4 of the plurality of first features X
and a fourth sub-plurality Y.sub.4 of the plurality of second
features Y into a carbohydrate recommendation classification model
(e.g., as described herein with reference to FIG. 11), and
assigning a macronutrient protein intake recommendation P.sub.j to
the user by inputting a fifth sub-plurality X.sub.5 of the
plurality of first features X and a fifth sub-plurality Y.sub.5 of
the plurality of second features Y into a protein recommendation
classification model (e.g., as described herein with reference to
FIG. 13).
[0233] In some embodiments, the method includes assigning (2824) a
micronutrient recommendation profile R.sub.j={r(z.sub.i) . . . ,
r(z.sub.s)} including a recommendation r(z.sub.i) for each
respective micronutrient z.sub.i in a plurality of micronutrients
Z={z.sub.1 . . . , z.sub.s} to the user by inputting user features,
including a second sub-plurality X.sub.2 of the plurality of first
features X and a second sub-plurality Y.sub.2 of the plurality of
second features Y, into a micronutrient classification model (e.g.,
health decision tree engine 108 as described herein with reference
to FIG. 1 and/or micronutrient recommendation classifier 1520 as
described herein with reference to FIG. 15).
[0234] In some embodiments, the method includes assigning (2826)
one or more source recommendations S.sub.j to the user by inputting
user features, including a sub-plurality of first features X and a
sub-plurality of second features Y, into a source classification
model (e.g., micronutrient recommendation classifier 1520 as
described herein with reference to FIG. 15 or a classifier
implemented separately from micronutrient recommendation classifier
1520 and/or a an illustrative source classifier as described herein
with reference to FIG. 16). In some embodiments, a source
recommendation includes a recommendation for dietary fiber (e.g.,
as described herein with reference to FIGS. 15 and 16). In some
embodiments, a source recommendation includes a recommendation for
dietary monounsaturated fatty acids (e.g., as described herein with
reference to FIGS. 15 and 16).
[0235] In some embodiments, the method includes assigning (2832) a
caloric recommendation C.sub.j to the user by inputting user
features into a caloric recommendation classification model (e.g.,
caloric recommendation classifier 1525 as described herein with
reference to in FIG. 15). In some embodiments, the caloric
recommendation is based on a user daily activity level (2834). For
example, in some embodiments the user is presented with a
questionnaire asking about their physical activity levels during a
normal day (e.g., at work, school, and/or home). In some
embodiments, the caloric recommendation is based on a user exercise
level (2836). For example, in some embodiments, the user is
presented with a questionnaire asking about the physical activities
they routinely engage in (e.g., sports, weight-lifting,
cardiovascular exercising, and outdoor activities). For example,
the user is asked about one or more of what activities they
routinely participate in, how often they participate in the
activities, and how vigorously they participate in the activities.
In some embodiments, activity information is provided by an
electronic activity monitor. In some embodiments, the user's
reported daily physical activity levels and/or leisure activity
levels are weighted according to a model of the caloric output
and/or caloric requirement for each activity and then used to
arithmetically personalize a daily caloric requirement, e.g., as
based off of a starting caloric requirement for a male or female,
optionally considering other features of the individual (e.g., one
or more phenotype, metabolic adaptability characteristic, or
anthropometric measurement).
[0236] In some embodiments, the method includes assigning (2838)
one or more hero food recommendations H.sub.j (e.g., one or more
hero foods as described herein with reference to FIG. 14) to the
user by inputting user features, including a sub-plurality of first
features X, a sub-plurality of second features Y, and/or a dietary
type, into a hero food recommendation classification model (e.g.,
caloric recommendation classifier 1525 as described herein with
reference to in FIG. 15).
[0237] In some embodiments, the method includes assigning (2838)
one or more supplement recommendations V.sub.j to the user by
inputting user features, including a sub-plurality of first
features X and a sub-plurality of second features Y, into a
supplement recommendation classification model (e.g., supplement
recommendation classifier 1552 as described herein with reference
to in FIG. 15).
[0238] In some embodiments, the method includes recommending one or
more foods to the user by inputting (F) one or more of the user
features and/or recommendations into a food recommendation
classifier (e.g., meal ranker engine 125 as described herein with
reference to FIG. 1 and/or food selection classifier 1536 as
described herein with reference to FIG. 15). In some embodiments, a
plurality of foods (e.g., a plurality of meals) is input into the
classifier and the food recommendation classifier selects one or
more foods (e.g., meals) that best match the dietary needs of the
user based on the one or more user features and/or
recommendations.
[0239] For example, in one embodiment, the method includes ranking
(2842) one or more foods in a plurality of foods L={N.sub.1 . . . ,
N.sub.t} (e.g., a list of meals), where each respective food
N.sub.i in the plurality of foods has a corresponding nutrition
profile P.sub.Ni={D.sub.ki, P(z.sub.ki)} comprising an assigned
diet type D.sub.k in the plurality of diet types D and an assigned
micronutrient profile P(z.sub.k)={v(z.sub.1) . . . , v(z.sub.s)},
where the micronutrient profile P(z.sub.k) includes a respective
value v(z.sub.i) for each micronutrient z.sub.i in the plurality of
micronutrients Z.
[0240] In some embodiments, ranking one or more foods includes
deprioritizing (2844) a food N.sub.i that does not conform to a
user preference. For example, deprioritizing a meal containing
chicken for a user with a vegetarian preference. In some
embodiments, deprioritizing (2846) includes assigning the food a
lower rank in the ranking of the one or more foods in the plurality
of foods L. For example, assigning a meal containing beef a lower
ranking than a meal containing salmon for a user with a preference
for fish as a protein. In some embodiments, deprioritizing (2848)
includes removing the food from a list of eligible foods for the
user. For example, removing a dish containing pork as an option for
a user with a kosher food preference. In some embodiments,
different types of food preferences will result in different rules
for food prioritization. For example, in one embodiment, a
preference for a particular food religion will result in removing a
food from a list of foods available to the user, while a preference
for a particular protein source may just prioritize meals
containing that protein as compared to meals containing other
proteins.
[0241] In some embodiments, ranking one or more foods includes
prioritizing (2850) foods N by comparing the diet type D.sub.j
assigned to the user with the diet types D.sub.k assigned to each
food N.sub.i. In some embodiments, prioritizing (2852) includes
assigning a food N.sub.1 having a same diet type D.sub.k1 as the
diet type D.sub.j assigned to the user a higher rank in the ranking
of the one or more foods than a food N.sub.2 having a different
diet type D.sub.k2 as the diet type D.sub.j assigned to the user.
For example, ranking a meal having a high protein content higher
than a meal containing a low protein content for a user with a diet
type associated with a high protein requirement (e.g., associated
with a P+ or P++ dietary protein recommendation as described
herein). In some embodiments, prioritizing (2854) includes removing
a food N.sub.3 having a different diet type D.sub.k3 as the diet
type D.sub.j assigned to the user from a list of eligible foods for
the user, e.g., the plurality of foods. For example, removing a
meal having a high carbohydrate content and low protein content
from a list of available foods for a user with a diet type
associated with a high protein requirement (e.g., associated with a
P+ or P++ dietary protein recommendation as described herein) and a
low carbohydrate requirement (e.g., associated with a c dietary
carbohydrate recommendation as described herein).
[0242] In some embodiments, ranking one or more foods includes
deprioritizing (2856) a food N.sub.i that does not conform to a
user allergy and/or sensitivity. For example, deprioritizing a meal
high in caffeine for a user with a caffeine sensitivity. In some
embodiments, deprioritizing (2858) includes assigning the food a
lower rank in the ranking of the one or more foods in the plurality
of foods L. For example, assigning a meal containing a cream sauce
a lower ranking than a meal containing a tomato sauce for a user
with a lactose sensitivity. In some embodiments, deprioritizing
(2860) includes removing the food from a list of eligible foods for
the user. For example, removing a dish containing peanut butter as
an option for a user with a peanut allergy. In some embodiments,
different types of food sensitivities and allergies will result in
different rules for food prioritization. For example, in one
embodiment, a peanut allergy will result in removing a food from a
list of foods available to the user, while sensitivity for caffeine
may just result in deprioritizing meals containing caffeine.
[0243] In some embodiments, ranking one or more foods includes
deprioritizing (2862) foods N by comparing the source
recommendation S.sub.j assigned to the user with the nutrition
profile P.sub.N of each food N.sub.i, e.g., deprioritizing a food
N.sub.i that does not conform to a user source recommendation. For
example, deprioritizing a meal low in fiber for a user with a fiber
source recommendation. In some embodiments, deprioritizing (2864)
includes assigning a food N.sub.1 that does not conform to a user
source recommendation a lower rank in the ranking of the one or
more foods than a food N.sub.2 that does conform to a user source
recommendation. For example, assigning a meal with high fiber
content above a meal having low fiber content for a user with a
fiber source recommendation. In some embodiments, deprioritizing
(2866) includes removing the food from a list of eligible foods for
the user. For example, removing a dish having a low fiber content
as an option for a user with fiber source recommendation. In some
embodiments, different types of source recommendations will result
in different rules for food prioritization. For example, in one
embodiment, a fiber source recommendation with result in the
removal of foods with low fiber content, while a monounsaturated
fatty acid source recommendation will result in the prioritization
of foods rich in monounsaturated fatty acids.
[0244] In some embodiments, ranking one or more foods includes
prioritizing (2868) foods N by comparing the micronutrient
recommendation profile R.sub.j assigned to the user with the
micronutrient profile P(z.sub.ki) assigned to each food N.sub.i. In
some embodiments, prioritizing (2870) includes assigning, within a
diet type D.sub.k, a food N.sub.1, having a micronutrient profile
P(z.sub.k1) that more closely matches the user's micronutrient
recommendation profile R.sub.j than the micronutrient profile
P(z.sub.k2) of a food N.sub.2 having the same diet type as food
N.sub.1, a higher ranking than food N.sub.2.
[0245] In some embodiments, ranking one or more foods includes
deprioritizing (2872) (e.g., further lowering a ranking of) a food
N.sub.1 having a lower ranking than a food N.sub.2 when food
N.sub.1 and food N.sub.2 belong to a same food family. For example,
where two meals are substantially identical other than for the
identity of the protein (e.g., a chicken dish and a beef dish
served over rice), if the chicken dish is ranked higher than the
beef dish, the beef dish is deprioritized with respect to other,
previously lower ranked dishes, in order to provide the user with
diverse food choices/recommendations.
[0246] In some embodiments, ranking one or more foods includes
deprioritizing (2874) a food N.sub.i by comparing a supplement
recommended to the user to the nutrition profile P.sub.N of each
food N. For example, where the method includes recommending and/or
delivering a nutrient supplement in addition to one or more foods,
the system will compensate for the nutrients by deprioritizing
foods rich in that nutrient. In some embodiments, deprioritizing
(2876) includes lowering the ranking of food N.sub.1 that is rich
in a nutrient present in the supplement recommended to the user.
For example, where the user is receiving a fish oil supplement, a
meal containing salmon is ranked below a meal containing chicken
because salmon is rich in omega-3 fatty acids. In some embodiments,
deprioritizing (2878) includes removing a food N.sub.1 that is rich
in a nutrient present in the supplement recommended to the user
from a list of eligible foods for the user. For example, where the
user is receiving a fish oil supplement, a meal containing salmon
is removed from a list of foods eligible to the user. In some
embodiments, different supplement recommendations will result in
different rules for food prioritization. For example, in one
embodiment, receiving a fish oil supplement will remove meals
containing salmon as an available food, while receiving a
multivitamin supplement will result in lowering a ranking of a food
rich in one of the vitamins in the supplement.
[0247] In some embodiments, ranking one or more foods includes
deprioritizing (2880) foods N by comparing a caloric recommendation
C.sub.j assigned to the user with the nutrition profile P.sub.N of
each food N.sub.i. For example, ranking a higher calorie meal above
a lower calorie meal for an extremely active user with a high
caloric recommendation. In some embodiments, deprioritizing (2882)
includes assigning a food N.sub.1 that does not conform to a user
caloric recommendation a lower rank in the ranking of the one or
more foods than a food N.sub.2 that does conform to a user caloric
recommendation. In some embodiments, deprioritizing (2884) includes
removing a food N.sub.1 that does not conform to a user caloric
recommendation from a list of eligible foods for the user.
[0248] In some embodiments, the method includes presenting (2886)
to the user a sub-plurality of ranked foods from the list of ranked
foods for selection of one or more foods to be prepared and/or
delivered to the user. For example, after ranking a group of 100
foods, the system displays the five foods ranked highest according
to the ranking classifier (e.g., meal ranker engine 125 as
described herein with respect to FIG. 1 and/or food selection
classifier 1536 as described herein with reference to FIG. 15). In
some embodiments, presenting (2888) includes providing (2888) at
least one primary food recommendation and at least one secondary
food recommendation that the user may substitute for the primary
food recommendation. For example, he system displays to the user
the highest ranked food according to the ranking classifier as the
default food for the user, but also displays the second highest
ranked food according to the ranking classifier as a substitute for
the default food.
[0249] In some embodiments, the method includes preparing and/or
delivering (2890) a food selected (G) for the user based on a
system recommendation (e.g., a food selected based on a
recommendation from a diet type classifier, a micronutrient
recommendation classifier, a source recommendation classifier, a
hero food recommendation classifier, a supplement recommendation
classifier, and/or a food selection classifier). In some
embodiments, the food is selected based on a diet type D.sub.j
assigned to the user (2892). In some embodiments, the food is
selected based on a micronutrient recommendation profile R.sub.j
assigned to the user (2894). In some embodiments, the food is
selected based on a source recommendation S.sub.j assigned to the
user (2898). In some embodiments, the food is a hero food selected
based on a hero food recommendation H.sub.j assigned to the user
(2898). In some embodiments, the food is a supplement selected
based on a supplement recommendation V.sub.j assigned to the user
(2902). In some embodiments, the food is selected based on a
ranking of foods from a list of foods available to the user (2904).
In some embodiments, the food is a prepared meal (2906). In some
embodiments, the food is selected by the user based on a ranking of
foods presented to the user (2908). In some embodiments, the food
is a prepared meal (2910).
[0250] In some embodiments, the method includes providing (2912)
the user with a food recommendation based (H) on a system
recommendation (e.g., a food selected based on a recommendation
from a diet type classifier, a micronutrient recommendation
classifier, a source recommendation classifier, a hero food
recommendation classifier, a supplement recommendation classifier,
and/or a food selection classifier). In some embodiments, the food
recommendation is based on a diet type D.sub.j assigned to the user
(2914). In some embodiments, the food recommendation is based on a
micronutrient recommendation profile R.sub.j assigned to the user
(2914). In some embodiments, the food recommendation is based on a
source recommendation S.sub.j assigned to the user (2916). In some
embodiments, the food recommendation is based on a hero food
recommendation H.sub.j assigned to the user (2918). In some
embodiments, the food recommendation is based on a supplement
recommendation V.sub.j assigned to the user (2920). In some
embodiments, the food recommendation is based on a caloric
recommendation C.sub.j assigned to the user.
[0251] FIG. 29 depicts an illustrative method of collecting data
from users and about meals and available ingredients and
classifying the users into diet types and the meals according to
their data in order to match users with a variety of different,
heathy meal options on a daily, weekly, monthly or other frequency
basis that are individualized for the user and that may be
delivered to the user. Referring to FIG. 29, there is a user
population 2902 associated with a system according to some
embodiments of the invention for making meal, food, recipe and
supplement recommendations to each user. In 2904, each user
provides information a DNA sample and a blood sample as described
in this application from which genotype and phenotype data may be
obtained. In addition other information including but not limited
to vitals, goals, and exercise is collected.
[0252] In 2906 the collected genotype, phenotype and other data
2905 is stored or otherwise made available on the system and for
each user, specific genotypical and phenotypical biomarkers are
selected for use in classifying a user according to a diet type. In
addition in 2906, certain data from the other data is selected to
be used in the classification of each user into a diet type. The
biomarkers selected may change over time. In 2908, each user is
classified into a diet type that is stored on the system for that
user along with data corresponding to the user's micronutrients
needs and other information that is useful for selecting meals for
the user such as calories, allergies and other information
described elsewhere herein. This information 2909 including diet
types, micronutrient needs and other information may be provided to
the meal ranking and recommendation algorithm.
[0253] In addition to a population of users of the system, a set of
meals and/or ingredients are available. The meals may include
foods, prepared meals, supplements or recipes. Data corresponding
to each meal, supplement or food is collected in 2912 and stored.
In 2914, data associated with each meal 2913 is received and
processed in order select a subset of data or to create new data
corresponding to the meal that will be used in meal selection for
the user.
[0254] In 2916, the system receives selected data associated with
the meal such as protein, carbohydrates, fats, micronutrient data,
calories and other detailed information as described elsewhere
herein and optionally codes the meals in a form that facilitates
correlating meals with diet types and ranking them. For example, a
meal might be coded 0, 5 or 10 and if there are six diet types, all
codes 0, 5 or 10 might be available for consumption by certain diet
types. However, for others only meal types 5 and 10 might be
available, while for still others only diet type 0 may be
available. In any event, the meals may be coded and the code used
along with a map correlating diet type with acceptable codes in a
meal ranking algorithm. The selected meal data, micronutrient data
and any selected codes 2017 may be provided to the meal ranking
process 2918.
[0255] In 2918, a meal ranking and recommendation is performed in
order to provide a healthy variety of food recommendations to a
user on a daily, weekly, monthly or other basis. The
recommendations, which may be a ranked subset from a large number
of choices compatible with a user's diet type and micronutrient
needs, may be of food, supplements, recipes, prepared meals, or
hero foods as described elsewhere herein, including in connection
with FIGS. 1 and 15, the meal ranker engine 125 and the element
1536. In 2920, the meals recommended for each user are presented to
the corresponding user through email, messaging or the user logging
in to the system and being presented with them there. The user
selects a meal or multiple meals, foods, recipes or supplements in
2922 for the day, week or month. The user is presented with a
healthy variety of meals that are each a match for the user's
genotype and phenotype and the user's selections may also be fed
back into the meal ranker 2918 as shown so that the user's
preferences are considered in the recommendation. In 2924, selected
meals, foods, or supplements may be delivered to the user.
Classifiers
[0256] In some embodiments, classifiers for determining nutritional
recommendations based on user vitals, genotypic and/or phenotypic
data can be developed or refined by training a decision rule using
data from one or more training sets and applying the trained
decision rule to data from users interested in receiving
nutritional recommendations. Information on pattern recognition and
prediction algorithms for use in data analysis algorithms for
constructing decision rules if found, for example, in National
Research Council; Panel on Discriminant Analysis Classification and
Clustering, Discriminant Analysis and Clustering, Washington, D.C.:
National Academy Press and Dudoit et al., 2002, "Comparison of
discrimination methods for the classification of tumors using gene
expression data." JASA 97; 77-87, the entire contents of which are
hereby incorporated by reference herein in their entirety for all
purposes.
[0257] In some embodiments, a classifier for determining
nutritional recommendations based on user vitals, genotypic, and/or
phenotypic data (e.g., for classifying a diet type, one or more
macronutrient recommendation, one or more micro-nutrient
recommendation, one or more source recommendation, or one or more
hero food recommendation, or one or more food ranking or
recommendation) may be built de novo by compiling existing clinical
study results, performing and/or integrating new clinical study
results, and/or observational theory. In some embodiments, one or
more classifiers are further refined after implementation based on
individual or population feedback.
[0258] For example, in an embodiment where the metabolic
adaptability of an individual (e.g., as determined using a
multi-nutrient challenge beverage) informs a diet type classifier,
the metabolic adaptability of the individual may be determined one
or more times following adaption of a particular diet type to track
changes in the individual's metabolic adaptability following
implementation of a particular diet. In this fashion, detrimental
changes to the user's metabolic adaptability when on a particular
diet can be identified and the diet type classifier can be refined
such that the individual is classified into a more suitable diet
type.
[0259] In some embodiments, a refined classifier is implemented in
a user-independent fashion, e.g., refinement of a particular
classifier based on data from a plurality of users leads to a
change in a diet type classifier used to assign diet types to all
users. In other embodiments, a refined classifier is implemented in
a user-specific fashion, e.g., refinement of a food selection
classifier based on observations that a particular user chooses
certain types of meals (e.g., meals containing quinoa, or does not
choose certain types of foods (e.g., meals including salmon as a
protein), leads to a change in the food selection classifier
implemented for that specific user, but not other users.
[0260] Relevant algorithms for decision rule include, but are not
limited to: discriminant analysis including linear, logistic, and
more flexible discrimination techniques (see, e.g., Gnanadesikan,
1977, Methods for Statistical Data Analysis of Multivariate
Observations, New York: Wiley 1977; tree-based algorithms such as
classification and regression trees (CART) and variants (see, e.g.,
Breiman, 1984, Classification and Regression Trees, Belmont,
Calif.: Wadsworth International Group; generalized additive models
(see, e.g., Tibshirani, 1990, Generalized Additive Models, London:
Chapman and Hall; neural networks (see, e.g., Neal, 1996, Bayesian
Learning for Neural Networks, New York: Springer-Verlag; and Insua,
1998, Feedforward neural networks for nonparametric regression In:
Practical Nonparametric and Semiparametric Bayesian Statistics, pp.
181-194, New York: Springer, the entire contents of each of which
are hereby incorporated by reference herein. Other suitable data
analysis algorithms for decision rules include, but are not limited
to, logistic regression, or a nonparametric algorithm that detects
differences in the distribution of feature values (e.g., a Wilcoxon
Signed Rank Test (unadjusted and adjusted)).
[0261] In some embodiments, the decision rule is based on multiple
measured values, e.g., two, three, four, five, ten, twenty, or more
measured values, corresponding to observables from multiple data
sets, e.g., two, three, four, five, ten, twenty, or more data sets.
In some embodiments, decision rules may also be built using a
classification tree algorithm. Other data analysis algorithms known
in the art include, but are not limited to, Classification and
Regression Tree (CART), Multiple Additive Regression Tree (MART),
Prediction Analysis for Microarrays (PAM), and Random Forest
analysis. Such algorithms classify complex spectra and/or other
information in order to distinguish subjects as normal or as having
a particular medical condition. Other examples of data analysis
algorithms include, but are not limited to, ANOVA and nonparametric
equivalents, linear discriminant analysis, logistic regression
analysis, nearest neighbor classifier analysis, neural networks,
principal component analysis, quadratic discriminant analysis,
regression classifiers and support vector machines. Such algorithms
may be used to construct a decision rule and/or increase the speed
and efficiency of the application of the decision rule and to avoid
investigator bias. For further review of algorithm classifiers, see
Duda, 2001, Pattern Classification, John Wiley & Sons, Inc.,
New York. pp. 396-408 and pp. 411-412, Hastie et al., 2001, The
Elements of Statistical Learning, Springer-Verlag, New York,
Chapter 9, and Breiman, 1999, "Random Forests-Random Features,"
Technical Report 567, Statistics Department, U. C. Berkeley,
September 1999, the entire contents of which are hereby
incorporated by reference herein in their entireties for all
purposes.
Challenge Beverage
[0262] A challenge food or beverage may be used to evaluate a
user's biological response to various foods and macronutrients.
Exogenous factors, including food and drink, constantly stress our
body's capacity to maintain physiological homeostasis. Our body's
ability to adequately react to these external challenges to
maintain homeostasis is termed "phenotypic flexibility." Phenotypic
flexibility is determined by a series of interconnected
physiological processes and molecular mechanisms. Challenge tests
that temporarily disturb homeostasis, including challenge tests
based on carbohydrates (oral glucose tolerance test, OGTT), lipids
(oral lipid tolerance test, OLTT), protein (oral protein tolerance
test, OPTT), and/or combinations thereof, have been used to test
these processes and access phenotypic flexibility.
[0263] Challenge tests based on individual macronutrients may not
be representative of an individual's diet. Furthermore, effects
elicited by single macronutrient challenges do not include all
process associated with phenotypic flexibility. A mixed
macronutrient challenge test is used to evaluate all processes
triggered by each individual challenge test at once and also to
trigger all physiological systems representative of phenotypic
flexibility. According to some embodiments, a challenge test
includes consuming a food that includes relative large quantities
of glucose, lipids, and protein. According to some embodiments, the
challenge beverage includes only glucose, lipids or protein in
large quantities, or a combination of them. In some embodiments,
the challenge food is a beverage or a solid food. According to some
embodiments, a challenge beverage includes or is made with the
following ingredients:
TABLE-US-00003 Ingredient Weight in mg Percentage by Weight Water
268.106 60.922 Organic Palm Oil - Olein 60.000 13.634 18C; fully
melted Dextrose; Non-GMO 83.380 18.947 MPI 90 23.350 5.306 Canola
lecithin - Non GMO 0.933 0.212 Natural Flavors 3.525v 0.801 gellan
gum 0.132 0.030 Trisodium Citrate 0.570 0.130 Sodium Hydroxide 10%
0.084 0.019 440.080 100.000
[0264] In some embodiments, the water is heated and mixed with the
other ingredients. The natural flavors may include vanilla in some
embodiments or cassia flavors or combinations of both. In some
embodiments, the natural flavors may be entirely different, or
encompass other flavors in combination with natural flavors
identified herein. The beverage in some embodiments is sterilized,
homogenized and packed. The sterilization in some embodiments is by
direct steam injection. The challenge beverage serving size in some
embodiments is approximately 415 mg. However, the overall portion
may be much smaller or larger depending on a range of factors,
including the size of the individual, the expected range of the
test results, the number of types of macronutrients present in the
challenge beverage and taste. There may be in some embodiments
multiple challenge beverage or food options for a single person to
take multiple tests. Alternatively, there may be in some
embodiments multiple challenge beverages available to choose from,
including different sizes or flavors based on the personal
preference of the user. In some embodiments, a blood test is done
prior to the consumption of a challenge beverage. Blood tests at
time intervals are done as described above after a user consumes a
challenge beverage.
[0265] An example challenge beverage in some embodiments may
comprise:
TABLE-US-00004 Total Fat 61 g Saturated Fat 26 g Trans Fat 0 g
Polyunsaturated Fat 7 g Monounsaturated Fat 25 g Cholesterol 15 mg
Sodium 150 mg Total Carbohydrates 77 g Dietary Fiber 0 g Sugars 75
g Protein 20 g
[0266] In some embodiments, the disclosure provides a
multi-nutrient challenge beverage for measuring the metabolic
adaptability of a user containing fats, carbohydrates, and
proteins. In some embodiments, the multi-nutrient challenge
beverage contains from 44 to 66 grams total fats, 75.+-.15 grams
total carbohydrates, and 20.+-.3 grams total protein.
[0267] In some embodiments, the multi-nutrient challenge beverage
contains 60.+-.6 grams total fats. In other embodiments, the
multi-nutrient challenge beverage contains 50.+-.6, 51.+-.6,
52.+-.6, 53.+-.6, 54.+-.6, 55.+-.6, 56.+-.6, 57.+-.6, 58.+-.6, or
59.+-.6 grams totals fats. In some embodiments, the fat content of
the multi-nutrient challenge beverage comprises from 10% to 20% of
the total weight of the beverage. In other embodiments, the fat
content of the multi-nutrient challenge beverage comprises
10%.+-.2%, 11%.+-.2%, 12%.+-.2%, 13%.+-.2%, 14%.+-.2%, 15%.+-.2%,
16%.+-.2%, 17%.+-.2%, 18%.+-.2%, 19%.+-.2%, or 20%.+-.2% of the
total weight of the beverage.
[0268] In some embodiments, the fat content of the beverage is
primarily (e.g., at least 85%, 90%, 95%, or 99% of the fat content
is derived) from an edible vegetable oil. Vegetable oils are
primarily triglycerides extracted from plants. Non-limiting
examples of vegetable oils include, but are not limited to, palm
oil, coconut oil, corn oil, cottonseed oil, olive oil, peanut oil,
rapeseed oil (e.g., canola oil), safflower oil, sesame oil, soybean
oil, sunflower oil, and mixtures thereof. In one embodiment, the
edible vegetable oil is palm oil.
[0269] In some embodiments, the fat content of the beverage is
primarily (e.g., at least 85%, 90%, 95%, or 99% of the fat content
is derived) from edible nut oil. Nut oils are primarily
triglycerides extracted from nuts. Non-limiting examples of nut
oils include, but are not limited to, almond oil, beech nut oil,
brazil nut oil, cashew oil, hazelnut oil, macadamia nut oil,
mongongo nut oil, pecan oil, pine nut oil, pistachio nut oil,
walnut oil, pumpkin seed oil, and mixtures thereof.
[0270] In some embodiments, the multi-nutrient challenge beverage
contains 80.+-.15 grams total carbohydrates. In other embodiments,
the multi-nutrient challenge beverage contains 60.+-.5, 65.+-.5,
70.+-.5, 75.+-.5, 80.+-.5, 85.+-.5, or 90.+-.5, grams totals
carbohydrates. In some embodiments, the carbohydrate content of the
multi-nutrient challenge beverage comprises from 10% to 30% of the
total weight of the beverage. In other embodiments, the
carbohydrate content of the multi-nutrient challenge beverage
comprises 20%.+-.8%, 20%.+-.6%, 20%.+-.4%, 20%.+-.2%, about 18%,
about 19%, about 20%, about 21%, or about 22% of the total weight
of the beverage. In other embodiments, the carbohydrate content of
the multi-nutrient challenge beverage comprises 10%.+-.2%,
11%.+-.2%, 12%.+-.2%, 13%.+-.2%, 14%.+-.2%, 15%.+-.2%, 16%.+-.2%,
17%.+-.2%, 18%.+-.2%, 19%.+-.2%, 20%.+-.2%, 21%.+-.2%, 22%.+-.2%,
23%.+-.2%, 24%.+-.2%, 25%.+-.2%, 26%.+-.2%, 27%.+-.2%, 28%.+-.2%,
29%.+-.2%, or 30%.+-.2% of the total weight of the beverage.
[0271] In some embodiments, the carbohydrate content of the
beverage is primarily (e.g., at least 85%, 90%, 95%, or 99% of the
carbohydrate content is derived) from monosaccharide sugar.
Non-limiting examples of monosaccharide sugars include, but are not
limited to, pentose sugars (e.g., arabinose, lyxose, ribose,
xylose, ribulose, and xylulose), hexose sugars (e.g., allose,
altroses, glucose (dextrose), mannose, gulose, Idose, galactose,
talose, psicose, fructose, sorbose, and tagatose), heptose sugars
(e.g., sedoheptulose, mannoheptulose, and
L-glycero-D-manno-heptose). In one embodiments, the carbohydrate
content of the beverage is primarily (e.g., at least 85%, 90%, 95%,
or 99% of the carbohydrate content is derived) from glucose
(dextrose).
[0272] In some embodiments, the multi-nutrient challenge beverage
contains 20.+-.10 grams total protein. In some embodiments, the
multi-nutrient challenge beverage contains 10.+-.5, 15.+-.5,
20.+-.5, 25.+-.5, or 30.+-.5 grams total protein. In other
embodiments, the multi-nutrient challenge beverage contains
15.+-.2, 16.+-.2, 17.+-.2, 18.+-.2, 19.+-.2, 20.+-.2, 21.+-.2,
22.+-.2, 23.+-.2, 24.+-.2, or 25.+-.2 grams total protein. In some
embodiments, the protein content of the multi-nutrient challenge
beverage comprises from 2.5% to 10% of the total weight of the
beverage. In other embodiments, the protein content of the
multi-nutrient challenge beverage comprises 2%.+-.2%, 3%.+-.2%,
4%.+-.2%, 5%.+-.2%, 6%.+-.2%, 7%.+-.2%, 8%.+-.2%, 9%.+-.2%, or
10%.+-.2%, of the total weight of the beverage.
[0273] In some embodiments, the protein content of the beverage is
primarily (e.g., at least 85%, 90%, 95%, or 99% of the protein
content is derived) from protein isolated from an edible source,
e.g., from soy, whey, or milk. In one embodiment, the protein
content of the beverage is primarily (e.g., at least 85%, 90%, 95%,
or 99% of the protein content is derived) from a milk protein
isolate. Protein isolates, such as milk protein isolates, are used
as emulsifiers and stabilizers in dairy products such as yogurt,
ice cream and ice cream novelties, and liquid and powdered
nutritional formulations. They are also used as a protein source in
protein-enrichment applications such as powdered and ready-to-drink
beverages for sports nutrition, adult nutrition, and weight
management.
[0274] Other sources of edible protein include, without limitation,
milk protein (e.g., lactose-free skim milk or milk protein
isolate), soy milk, whey protein, caseinate, soy protein, egg
whites, gelatins, collagen and combinations thereof.
[0275] In some embodiments, a multi-nutrient challenge beverage
also contains one or more of a tastant (e.g., a flavoring agent),
an emulsifier, a thickening agent, and a preservative.
[0276] Non-limiting examples of tastants (e.g., flavoring agents)
include vanilla, cocoa, strawberry, and peanut butter.
[0277] Non-limiting examples of emulsifiers useful in a challenge
beverage include canola lecithin, propane-1,2-diol alginate,
konjac, polyoxyl 8 stearate, polyoxyethylene stearate, polysorbate
20, polysorbate 80, ammonium phosphatides, diphosphates, methyl
cellulose, hydroxypropyl cellulose, hydroxypropyl methyl cellulose,
ethyl methyl cellulose, carboxymethylcellulose, sodium carboxy
methyl cellulose, sodium caseinate, magnesium stearate, sorbitan
monostearate, sorbitan tristearate, sorbitan monolaurate, and
sorbitan monopalmitate. In one embodiment, canola lecithin is used
as an emulsifying agent in a challenge beverage described herein.
Typically, the emulsifier is present in the challenge beverage at
from about 0.01% to 2.0% by weight.
[0278] Non-limiting examples of thickening agents include gellan
gum, alginic acid, sodium alginate, potassium alginate, ammonium
alginate, calcium alginate, propane-1,2-diol alginate, agar,
carrageenan, processed eucheuma seaweed, locust bean gum (carob
gum), guar gum, tragacanth, acacia gum, xanthan gum, karaya gum,
tara gum, pectin, xanthan, starches and modified starches, and
mixtures thereof. In one embodiment, gellan gum is used as a
thickening agent in a challenge beverage described herein.
[0279] Non-limiting examples of preservatives include citrates,
e.g., sodium citrate and potassium citrate, benzoic acid,
benzoates, e.g., sodium, calcium, and potassium benzoate, sorbates,
e.g., sodium, calcium, and potassium sorbate, polyphosphates, e.g.,
sodium hexametaphosphate (SHMP), dimethyl dicarbonate, and mixtures
thereof. Also of use are antioxidants, such as ascorbic acid, EDTA,
BHA, BHT, TBHQ, EMIQ, dehydroacetic acid, ethoxyquin,
heptylparaben, and combinations thereof. In one embodiment, sodium
citrate is used as a preservative in a challenge beverage described
herein.
[0280] In some embodiments, other ingredients are added to a
challenge beverage composition including, but not limited to, one
or more flavanols, aeidulants, coloring agents, minerals, vitamins,
herbs, soluble fibers, non-caloric sweeteners, oils, carbonation
components, and the like.
[0281] In some embodiments, a method for measuring the metabolic
adaptability of a user is provided. The method includes obtaining
data on a user's blood insulin levels, blood glucose levels, and
blood triglyceride levels prior to consumption of a multi-nutrient
challenge beverage, after a first period of time following
consumption of the multi-nutrient challenge beverage, and after a
second period of time following consumption of the multi-nutrient
challenge beverage, and inputting the obtained data into a
metabolic adaptability classifier. In some embodiments, the first
period of time and second period of time following consumption of
the multi-nutrient challenge beverage are each no longer than 120
minutes. In some embodiments, the challenge beverage is a challenge
beverage described herein.
[0282] In some embodiments, the data obtained on the user's blood
insulin levels, blood glucose levels, and blood triglyceride levels
is derived from a dried blood sample collected by the user.
[0283] It will be understood that changes may be made to the
composition of the challenge beverage or food as discussed above
and that the above example is illustrative only.
EXAMPLE 1
Challenge Beverage Validation Study
[0284] In order to validate the use of a multi-nutrient challenge
beverage for determining metabolic adaptability of individuals, a
trial was established using two challenge beverages containing 75
grams of carbohydrates, 50-60 grams of fats, and 20 grams of
protein. Specifically, the study was designed to assess
postprandial lipid and glycemic responses and gastrointestinal
tolerance for the challenge beverages, assess the feasibility of
assessing postprandial responses in dried capillary blood samples,
and assess the feasibility of performing the test over a shorter
time frame, e.g., within two hours.
[0285] Briefly, 18 subjects between the ages of 30-60, having a
body mass index of from 18.5 to 30 kg/m.sup.2 and normal GI
function, were randomly administered either Challenge Beverage A
(75 g carbohydrates, 60 g fat, 20 g protein, 940 kcal) or Challenge
Beverage B (75 g carbohydrates, 50 g fat, 20 g protein, 860 kcal).
The subjects fasted for 10 to 14 hours prior to administration and
avoided vigorous physical activity (24 hours), alcohol consumption
(24 hours), and tobacco use (1 hour) before administration. An
intravenous catheter was inserted and venous and capillary blood
samples were taken ten minutes prior to administration. After
consumption of the assigned beverage, venous and capillary blood
samples were collected at 30, 60, 90, 120, 180, and 240
minutes.
[0286] The collected samples were then analyzed as outlined in
Table 2 to determine the following parameters: [0287] Change in
triglyceride concentrations from 0 to 120 min [the pre-consumption
measurement (t=-10 min) will be counted as time 0 for the
calculation]; [0288] Changes in glucose and insulin concentrations
from 0 to 30, 120, 180, and 240 min [the pre-consumption
measurement (t=-10 min) will be counted as time 0 for the
calculation]; [0289] Changes in triglyceride concentration from 0
to 180 and 240 min [the pre-consumption measurement (t=-10 min)
will be counted as time 0 for the calculation]; [0290] Triglyceride
area under the curve (AUC) from 0 to 60, 90, 120, 180, and 240 min
[the pre-consumption measurement (t=-10 min) will be counted as
time 0 for the calculation]; [0291] Glucose and insulin AUC from 0
to 60, 90, 120, 180, and 240 min [the pre-consumption measurement
(t=-10 min) will be counted as time 0 for the calculation]; [0292]
Peak values for TG, glucose, and insulin; and [0293] Composite
score and individual ratings (nausea, GI rumblings, abdominal pain,
bloating, flatulence, and diarrhea) using a GI tolerability
questionnaire.
[0294] FIG. 19 shows plots of the average insulin levels detected
in the venous catheter collected blood samples (Insulin Venous) and
the dried capillary blood samples (Insulin ADX) for both challenge
beverages. FIGS. 20 and 21 illustrate linear regressions comparing
the insulin levels detected in the venous samples and the capillary
samples for Challenge Beverage A (FIG. 20) and Challenge Beverage B
(FIG. 21). As shown in the figures, there was a strong correlation
between the insulin levels detected in the venous catheter
collected blood sample and the dried capillary blood sample for
both challenge beverages, evidencing that insulin sampling could be
performed using dried blood spot (DBS) technology. Further, the
measured insulin response following consumption of both challenge
beverages peaked around 120 minutes, evidencing that longer time
points were not necessary for sufficient measurement of an
individual's insulin response to food.
[0295] FIG. 22 shows plots of the average glucose levels detected
in the venous catheter collected blood samples (Glucose Venous) and
the dried capillary blood samples (Glucose ADX) for both challenge
beverages. FIGS. 23 and 24 illustrate linear regressions comparing
the glucose levels detected in the venous samples and the capillary
samples for Challenge Beverage A (FIG. 23) and Challenge Beverage B
(FIG. 24). As shown in the figures, there was a strong correlation
between the glucose levels detected in the venous catheter
collected blood sample and the dried capillary blood sample for
both challenge beverages, evidencing that glucose sampling could be
performed using dried blood spot (DBS) technology. Further, the
measured glucose response following consumption of both challenge
beverages peaked around 120 minutes, evidencing that longer time
points were not necessary for sufficient measurement of an
individual's glucose response to food.
[0296] FIG. 25 shows plots of the average triglyceride levels
detected in the venous catheter collected blood samples
(Triglycerides Venous) and the dried capillary blood samples
(Triglycerides ADX) for both challenge beverages. FIGS. 26 and 27
illustrate linear regressions comparing the triglyceride levels
detected in the venous samples and the capillary samples for
Challenge Beverage A (FIG. 26) and Challenge Beverage B (FIG. 27).
As shown in the figures, there was a strong correlation between the
triglyceride levels detected in the venous catheter collected blood
sample and the dried capillary blood sample for both challenge
beverages, evidencing that triglyceride sampling could be performed
using dried blood spot (DBS) technology. Further, the measured
triglyceride response following consumption of both challenge
beverages first peaked around 120 minutes, evidencing that longer
time points were not necessary for sufficient measurement of an
individual's triglyceride response to food.
[0297] Advantageously, the use of dried capillary blood samples, as
compared to venous liquid samples, requires minimal sample volumes,
facilitates non-invasive sampling, does not require special
training for collection, and provides stability of the sample at
room temperature. All of the benefits facilitate home sample
collection and delivery to a clinical laboratory by regular
mail.
[0298] It will be understood that, although the terms "first,"
"second," etc. may be used herein to describe various elements,
these elements should not be limited by these terms. These terms
are only used to distinguish one element from another. For example,
a first contact could be termed a second contact, and, similarly, a
second contact could be termed a first contact, which changing the
meaning of the description, so long as all occurrences of the
"first contact" are renamed consistently and all occurrences of the
second contact are renamed consistently. The first contact and the
second contact are both contacts, but they are not the same
contact.
[0299] The terminology used herein is for the purpose of describing
particular embodiments only and is not intended to be limiting of
the claims. As used in the description of the embodiments and the
appended claims, the singular forms "a", "an" and "the" are
intended to include the plural forms as well, unless the context
clearly indicates otherwise. It will also be understood that the
term "and/or" as used herein refers to and encompasses any and all
possible combinations of one or more of the associated listed
items. It will be further understood that the terms "comprises"
and/or "comprising," when used in this specification, specify the
presence of stated features, integers, steps, operations, elements,
and/or components, but do not preclude the presence or addition of
one or more other features, integers, steps, operations, elements,
components, and/or groups thereof.
[0300] As used herein, the term "if" may be construed to mean
"when" or "upon" or "in response to determining" or "in accordance
with a determination" or "in response to detecting," that a stated
condition precedent is true, depending on the context. Similarly,
the phrase "if it is determined [that a stated condition precedent
is true]" or "if [a stated condition precedent is true]" or "when
[a stated condition precedent is true]" may be construed to mean
"upon determining" or "in response to determining" or "in
accordance with a determination" or "upon detecting" or "in
response to detecting" that the stated condition precedent is true,
depending on the context.
[0301] The foregoing description, for purpose of explanation, has
been described with reference to specific implementations. However,
the illustrative discussions above are not intended to be
exhaustive or to limit the claims to the precise forms disclosed.
Many modifications and variations are possible in view of the above
teachings. The implementations were chosen and described in order
to best explain principles of operation and practical applications,
to thereby enable others skilled in the art.
* * * * *