U.S. patent application number 17/143459 was filed with the patent office on 2021-10-07 for systems, apparatus, and methods related to modeling, monitoring, and/or managing metabolism.
The applicant listed for this patent is Massachusetts Institute of Technology. Invention is credited to Lawrence M. Candell, Christopher Ferraiolo, Gary A. Shaw, Andrew M. Siegel, George Zogbi.
Application Number | 20210307642 17/143459 |
Document ID | / |
Family ID | 1000005652752 |
Filed Date | 2021-10-07 |
United States Patent
Application |
20210307642 |
Kind Code |
A1 |
Candell; Lawrence M. ; et
al. |
October 7, 2021 |
SYSTEMS, APPARATUS, AND METHODS RELATED TO MODELING, MONITORING,
AND/OR MANAGING METABOLISM
Abstract
Systems, apparatus, and methods related to modeling, monitoring,
and/or managing metabolism of a subject include measuring a
respiratory quotient (RQ) level in a subject and/or optimizing and
executing a nonlinear feedback model to model energy substrate
utilization in the subject based on at least one of a macronutrient
composition and caloric value of food consumed by the subject, an
intensity and duration of activity by the subject, a rate and
maximum capacity of glycogen storage in the subject, a rate and
maximum capacity of de novo lipogenesis in the subject, a quality
and duration of sleep by the subject, and/or an RQ level in the
subject.
Inventors: |
Candell; Lawrence M.;
(Arlington, MA) ; Ferraiolo; Christopher;
(Newburyport, MA) ; Shaw; Gary A.; (Westford,
MA) ; Siegel; Andrew M.; (Arlington, MA) ;
Zogbi; George; (Arlington, MA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Massachusetts Institute of Technology |
Cambridge |
MA |
US |
|
|
Family ID: |
1000005652752 |
Appl. No.: |
17/143459 |
Filed: |
January 7, 2021 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
15221313 |
Jul 27, 2016 |
10925513 |
|
|
17143459 |
|
|
|
|
62197324 |
Jul 27, 2015 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
A61B 5/0871 20130101;
A61B 2560/0247 20130101; A61B 5/0816 20130101; A61B 5/0836
20130101; A61B 2560/0223 20130101; A61B 5/4866 20130101; G09B
19/0092 20130101; A61B 5/083 20130101; A61B 5/7275 20130101; A61B
5/0833 20130101; A61B 5/082 20130101; A61B 5/097 20130101; G09B
19/00 20130101 |
International
Class: |
A61B 5/083 20060101
A61B005/083; A61B 5/08 20060101 A61B005/08; A61B 5/097 20060101
A61B005/097; A61B 5/00 20060101 A61B005/00; G09B 19/00 20060101
G09B019/00 |
Goverment Interests
GOVERNMENT SUPPORT STATEMENT
[0002] This invention was made with Government support under
Contract No. FA8721-05-C-0002 awarded by the U.S. Air Force. The
Government has certain rights in the invention.
Claims
1-11. (canceled)
12. A computer-facilitated method for modeling metabolism in a
subject and/or managing body weight of the subject, the method
comprising: determining from data related to the subject obtained
via at least one input device: a macronutrient composition and
caloric value of food consumed by the subject; an intensity and
duration of activity by the subject; a rate and maximum capacity of
glycogen storage in the subject; and a rate and maximum capacity of
de novo lipogenesis in the subject; and optimizing, via at least
one processor, a nonlinear feedback model to model energy substrate
utilization in the subject based on the macronutrient composition
and caloric value of food consumed by the subject, the intensity
and duration of activity by the subject, the rate and maximum
capacity of glycogen storage in the subject, and the rate and
maximum capacity of de novo lipogenesis in the subject.
13. The computer-facilitated method of claim 12, further
comprising: obtaining metabolic data for the energy substrate
utilization in the subject, the metabolic data including
respiratory quotient (RQ) data acquired from the subject.
14. The computer-facilitated method of claim 13, further
comprising: controlling, via the at least one processor, operation
of the optimized nonlinear feedback model based on the metabolic
data to determine a target value of one or more energy substrate
utilization variables that at least one of maintains and increases
energy substrate utilization in the subject.
15. The computer-facilitated method of claim 14, wherein the one or
more energy substrate utilization variables comprise at least one
of: the macronutrient composition and caloric value of food
consumed by the subject; and the intensity and duration of activity
by the subject.
16. The computer-facilitated method of claim 12, further
comprising: determining from data related to the subject obtained
via the at least one input device a quality and duration of sleep
by the subject, and wherein optimizing, via the at least one
processor, the nonlinear feedback model is based further on the
quality and duration of sleep by the subject.
17. The computer-facilitated method of claim 12, further
comprising: obtaining at least one initial physiological parameter
associated with the subject, the at least one initial physiological
parameter including an initial body weight of the subject; and
18. The computer-facilitated method of claim 17, further
comprising: controlling, via the at least one processor, operation
of the optimized nonlinear feedback model based on the at least one
initial physiological parameter to determine a target value of one
or more energy substrate utilization variables that at least one of
maintains and alters the body weight of the subject
19. The computer-facilitated method of claim 18, wherein the one or
more energy substrate utilization variables comprise at least one
of: the macronutrient composition and caloric value of food
consumed by the subject; and the intensity and duration of activity
by the subject.
20. The computer-facilitated method of claim 17, wherein the at
least one initial physiological parameter further comprises at
least one of height, age, gender, body mass index (BMI), body fat
percentage, waist circumference, hip circumference, and chest
circumference.
21. A system for optimizing a nonlinear feedback model of energy
substrate utilization in a subject, the system comprising: at least
one input device for obtaining data related to the subject; at
least one memory device for storing the data related to the subject
and processor-executable instructions; and at least one processor
in communication with the at least one input device and the at
least one memory device, wherein upon execution of the
processor-executable instructions, the at least one processor:
determines from the data related to the subject: a macronutrient
composition and caloric value of food consumed by the subject; an
intensity and duration of activity by the subject; a rate and
maximum capacity of glycogen storage in the subject; and a rate and
maximum capacity of de novo lipogenesis in the subject; and
optimizes the nonlinear feedback model to model energy substrate
utilization in the subject based on the macronutrient composition
and caloric value of food consumed by the subject, the intensity
and duration of activity by the subject, the rate and maximum
capacity of glycogen storage in the subject, and the rate and
maximum capacity of de novo lipogenesis in the subject.
22. The system of claim 21, wherein the at least one processor,
upon execution of the processor-executable instructions, optimizes
the nonlinear feedback model to model energy substrate utilization
in the subject based on a quality and duration of sleep by the
subject, and the one or more energy substrate utilization variables
comprise at least one of the macronutrient composition and caloric
value of food consumed by the subject or the intensity and duration
of activity by the subject.
23. A system for managing body weight of a subject, the system
comprising: at least one input device for obtaining data related to
the subject; at least one memory device for storing the data
related to the subject and processor-executable instructions; and
at least one processor in communication with the at least one input
device and the at least one memory device, wherein upon execution
of the processor-executable instructions, the at least one
processor: determines from the data related to the subject at least
one initial physiological parameter associated with the subject,
the at least one initial physiological parameter including an
initial body weight of the subject; and controls operation of a
nonlinear feedback model to determine, based on the at least one
initial physiological parameter, a target value of one or more
energy substrate utilization variables that at least one of
maintains and alters the body weight of the subject, wherein: the
nonlinear feedback model is optimized to model energy substrate
utilization in the subject based on at least one of: a
macronutrient composition and caloric value of food consumed by the
subject; an intensity and duration of activity by the subject; a
rate and maximum capacity of glycogen storage in the subject; a
rate and maximum capacity of de novo lipogenesis in the subject;
and a quality and duration of sleep by the subject; and the one or
more energy substrate utilization variables comprise at least one
of: the macronutrient composition and caloric value of food
consumed by the subject; and the intensity and duration of activity
by the subject.
24. The system of claim 23, wherein the at least one initial
physiological parameter further comprises at least one of height,
age, gender, body mass index (BMI), body fat percentage, waist
circumference, hip circumference, and chest circumference.
25. The system of claim 23, wherein the nonlinear feedback model is
optimized further based on a quality and duration of sleep by the
subject.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application is a divisional of U.S. application Ser.
No. 15/221,313, filed on Jul. 27, 2016, which in turn claims the
priority benefit, under 35 U.S.C. 119(e), of U.S. Provisional
Patent Application 62/197,324, filed on Jul. 27, 2015. Each of
these applications is incorporated herein by reference in its
entirety.
TECHNICAL FIELD
[0003] The present disclosure relates generally to biofeedback
systems, apparatus, and methods. More specifically, the present
disclosure relates to systems, apparatus, and methods related to
modeling, monitoring, and/or managing a metabolic state of a
subject.
BACKGROUND
[0004] Obesity in humans has steadily increased worldwide. This
trend is not just confined to adults-currently, over one-third of
U.S. adults are obese-but also evident in children and adolescents.
Comorbidities associated with obesity include type 2 diabetes,
cancer (e.g., pancreatic and prostate cancers), cardiovascular
disease, asthma, gallbladder disease, osteoarthritis, and chronic
back pain. The most prevalent comorbidity associated with obesity
is type 2 diabetes, which is estimated to cost the United States in
excess of $300 billion per year.
[0005] Diet plans for preventing or reducing obesity provide often
conflicting and sometimes even diametrically opposed advice for
which macronutrients promote weight loss. More recently, the
"calorie-in/calorie-out" (CICO) model of metabolism has increased
in popularity. The use of the calorie to attribute energy to food
sources has been in place since the mid-1800s. FIG. 1 is a diagram
illustrating the "calorie-in/calorie-out" (CICO) model 100, which
posits that an imbalance between a subject's energy needs (calories
out 102, e.g., 1500-3000 Cal/day) and the amount of calories that
the subject consumes (calories in 104, e.g., 1000-3000 Cal/day)
results in either weight gain (virtually unlimited calorie storage
106) or weight loss (calorie burning 108). That is, if the subject
consumes calories in excess of the subject's energy needs, the
excess calories will not be burned or excreted, but rather stored
as glycogen or as adipose tissue, thereby contributing to obesity
and comorbidities.
[0006] Many diet plans now rely on tracking food calories as a
simplistic means to manage weight. Diet plans that are structured
around the CICO model promote overt control of calorie intake
and/or an exercise regimen to create an energy imbalance.
[0007] While the CICO model relies on the first law of
thermodynamics (i.e., energy cannot be created or destroyed), the
underlying assumption of the CICO model is that all food calories
are alike. As such, calorie counting and other diet plans that are
structured around the CICO model are not only tedious to implement
and prone to error, but often ignore the impact that different
macronutrient mixes and exercise intensities have on an
individual's homeostatic control system. As a consequence, diet
plans focused solely on energy balance (the CICO model) have failed
to reduce the worldwide trend of increasing obesity and
comorbidities despite their popularity.
SUMMARY
[0008] The inventors have recognized and appreciated that existing
metabolic models, particularly the CICO model, fail to capture the
physiological complexity of metabolism and therefore do not provide
sufficient guidance to reverse obesity trends in individuals or
populations. Instead, the inventors herein provide a metabolic
state model based on avoiding elevated blood glucose levels through
proper selection of dietary macronutrients and exercise sufficient
to enable the body's homeostatic system to achieve and maintain a
healthy body weight according to some embodiments.
[0009] In part, the inventors have recognized and appreciated that
the kinds of foods consumed and activities performed by a subject,
in concert with their unique genetic makeup, directly influence the
subject's metabolic state-fat/carbohydrate/protein burning,
fat/carbohydrate/protein storing, or neutral-and thereby account
greatly for weight gain and weight loss. According to some
embodiments, the inventors have developed personal respiratory
quotient (RQ) measurement devices and methods for providing
on-demand feedback indicative of a subject's real-time metabolic
state and guiding the subject's macronutrient intake and/or
activity levels to promote a desired metabolic state and/or achieve
and maintain a healthy body weight.
[0010] In one embodiment, a system for managing metabolism of a
subject includes at least one input device for obtaining data
related to the subject, at least one memory device for storing the
data related to the subject and processor-executable instructions,
and at least one processor in communication with the at least one
input device and the at least one memory device. Upon execution of
the processor-executable instructions, the at least one processor
determines, from the data related to the subject, metabolic data
characterizing energy substrate utilization in the subject, the
metabolic data including RQ data acquired from the subject, and
controls operation of a nonlinear feedback model to determine,
based on the metabolic data, a target value of one or more energy
substrate utilization variables, at least one of which maintains
and alters energy substrate utilization in the subject. The
nonlinear feedback model is optimized to model energy substrate
utilization in the subject based on at least one of a macronutrient
composition and caloric value of food consumed by the subject, an
intensity and duration of activity by the subject, a rate and
maximum capacity of glycogen storage in the subject, a rate and
maximum capacity of de novo lipogenesis in the subject, and a
quality and duration of sleep by the subject. The one or more
energy substrate utilization variables include at least one of the
macronutrient composition and caloric value of food consumed by the
subject and the intensity and duration of activity by the
subject.
[0011] In one embodiment, a system for optimizing a nonlinear
feedback model of energy substrate utilization in a subject
includes at least one input device for obtaining data related to
the subject, at least one memory device for storing the data
related to the subject and processor-executable instructions, and
at least one processor in communication with the at least one input
device and the at least one memory device. Upon execution of the
processor-executable instructions, the at least one processor
determines from the data related to the subject a macronutrient
composition and caloric value of food consumed by the subject, an
intensity and duration of activity by the subject, a rate and
maximum capacity of glycogen storage in the subject, and a rate and
maximum capacity of de novo lipogenesis in the subject. The at
least one processor further optimizes the nonlinear feedback model
to model energy substrate utilization in the subject based on the
macronutrient composition and caloric value of food consumed by the
subject, the intensity and duration of activity by the subject, the
rate and maximum capacity of glycogen storage in the subject, and
the rate and maximum capacity of de novo lipogenesis in the
subject.
[0012] In one embodiment, a system for managing body weight of a
subject includes at least one input device for obtaining data
related to the subject, at least one memory device for storing the
data related to the subject and processor-executable instructions,
and at least one processor in communication with the at least one
input device and the at least one memory device. Upon execution of
the processor-executable instructions, the at least one processor
determines from the data related to the subject at least one
initial physiological parameter associated with the subject, the at
least one initial physiological parameter including an initial body
weight of the subject, and controls operation of a nonlinear
feedback model to determine, based on the at least one initial
physiological parameter, a target value of one or more energy
substrate utilization variables that at least one of maintains and
alters the body weight of the subject. The nonlinear feedback model
is optimized to model energy substrate utilization in the subject
based on at least one of a macronutrient composition and caloric
value of food consumed by the subject, an intensity and duration of
activity by the subject, a rate and maximum capacity of glycogen
storage in the subject, a rate and maximum capacity of de novo
lipogenesis in the subject, and a quality and duration of sleep by
the subject. The one or more energy substrate utilization variables
comprise at least one of the macronutrient composition and caloric
value of food consumed by the subject and the intensity and
duration of activity by the subject.
[0013] In one embodiment, an apparatus for measuring an RQ level in
a subject includes a first input port for receiving respired air
from the subject, a measurement chamber for receiving the respired
air from the first input port, the measurement chamber being in
fluid communication with the first input port, a first sensor
located in the measurement chamber, the first sensor for measuring
a series of oxygen levels in the measurement chamber, a second
sensor located in the measurement chamber, the second sensor for
measuring a series of carbon dioxide levels in the measurement
chamber at a temporal rate sufficient to ascertain a respiration
rate of the subject, at least one output interface, at least one
memory for storing processor-executable instructions, the series of
oxygen level measurements, and the series of carbon dioxide level
measurements, and at least one processor coupled to the first
sensor, the second sensor, the at least one output interface, and
the at least one memory. Upon execution of the processor-executable
instructions, the at least one processor obtains a first portion of
the series of carbon dioxide level measurements, determines a first
respiration rate based on the first portion of the series of carbon
dioxide level measurements. The at least one processor further
iterates steps of obtaining a subsequent portion of the series of
carbon dioxide level measurements, determining a subsequent
respiration rate based on the subsequent portion of the series of
carbon dioxide level measurements, and comparing the subsequent
respiration rate to at least one prior respiration rate until a
stable breathing pattern is identified. The at least one processor
then obtains a stable breathing pattern portion of the series of
oxygen level measurements and a stable breathing pattern portion of
the series of carbon dioxide level measurements, determines an
average minimum oxygen level for a respiration cycle from the
stable breathing pattern portion of the series of oxygen level
measurements, determines an average maximum carbon dioxide level
for a respiration cycle from the stable breathing pattern portion
of the series of carbon dioxide level measurements, calculates the
RQ level from the average minimum oxygen level and the average
maximum carbon dioxide level, and at least one of displays and
transmits, via the at least one output interface, the calculated RQ
level.
[0014] In an embodiment, the at least one processor controls an
ambient air calibration process including the steps of comparing at
least one measurement to at least one expected value for ambient
air. If the at least one measurement is from the first sensor and
sufficiently different from the at least one expected value for
ambient air, the at least one processor may perform a span
calibration process on the first sensor to determine and apply a
gain correction to subsequent measurements from the first sensor.
If the at least one measurement is from the second sensor and
sufficiently different from the at least one expected value for
ambient air, the at least one processor may perform a zero-point
calibration process on the second sensor to determine and apply an
offset correction to subsequent measurements from the second
sensor.
[0015] In an embodiment, the at least one expected value for
ambient air at 760 mm Hg is at least one of about 19.5% (v/v) to
about 23.5% (v/v) oxygen and about 250 ppm to about 5,000 ppm
carbon dioxide. The at least one expected value for ambient air at
760 mm Hg may be at least one of about 20.9% (v/v) oxygen and about
400 ppm carbon dioxide. The ambient air calibration process may be
performed each time the first sensor and the second sensor are
initiated. The ambient air calibration may also employ measured
relative humidity to adjust the percent oxygen to account for water
vapor content. The ambient air calibration process further may
include storing in the at least one memory at least one of the gain
correction applied to subsequent measurements from the first sensor
and the offset correction applied to subsequent measurements from
the second sensor, such that a history of ambient air measurement
drift is maintained for reference.
[0016] In an embodiment, the apparatus further includes a second
input port for receiving a carbon dioxide cartridge for calibrating
at least one of the first sensor and the second sensor, the second
input port being in fluid communication with the measurement
chamber such that the measurement chamber receives carbon dioxide
released from the carbon dioxide cartridge.
[0017] In an embodiment, the at least one processor controls a full
calibration process, the full calibration process including
allowing the measurement chamber to fill with ambient air, setting
a span value for the first sensor to an expected value for the
ambient air, setting a zero-point value for the second sensor to an
expected value for the ambient air, coupling the second input port
with the carbon dioxide cartridge for releasing carbon dioxide from
the carbon dioxide cartridge, releasing the carbon dioxide into the
measurement chamber, setting a zero-point value for the first
sensor to zero once the carbon dioxide displaces the ambient air in
the measurement chamber, and iteratively measuring oxygen levels
with the first sensor in the measurement chamber as the ambient air
displaces the carbon dioxide in the measurement chamber until a
predetermined oxygen level is measured, the predetermined oxygen
level indicating the span value for the second sensor. The
predetermined oxygen level may be about 16.7%, and the span value
for the second sensor may be about 20%.
[0018] In an embodiment, the apparatus further includes a component
timed to prevent fluid communication between the first input port
and the measurement chamber during an initial portion of each
respiration cycle and allow fluid communication between the first
input port and the measurement chamber during an end-tidal portion
of each respiration cycle. The timed component may be at least one
of a mechanical shutter, a vacuum pump, and a purge fan. The first
input port may be compatible with at least one of a mouthpiece and
a sample tube for coupling the input port to at least one of the
mouth and a nostril of the subject. The at least one of the
mouthpiece and the sample tube may be at least one of disposable
and for hands-free use. The mouthpiece may be a repurposed sport
bottle cap. The apparatus further may include a heating element for
at least one of preventing and reducing condensation on the first
sensor and the second sensor.
[0019] In one embodiment, a kit for measuring an RQ level in a
subject includes a carbon dioxide cartridge, at least one of a
mouthpiece and a sample tube for coupling to at least one of the
mouth and a nostril of the subject, and a device for measuring the
RQ level in a subject. The device includes a first input port for
receiving, via the at least one of the mouthpiece and the sample
tube, respired air from the subject, a measurement chamber for
receiving the respired air from the first input port, the
measurement chamber being in fluid communication with the first
input port, a first sensor located in the measurement chamber, the
first sensor for measuring a series of oxygen levels in the
measurement chamber, a second sensor located in the measurement
chamber, the second sensor for measuring a series of carbon dioxide
levels in the measurement chamber at a temporal rate sufficient to
ascertain a respiration rate of the subject, a second input port
for receiving the carbon dioxide cartridge for calibrating at least
one of the first sensor and the second sensor, the second input
port being in fluid communication with the measurement chamber such
that the measurement chamber receives carbon dioxide released from
the carbon dioxide cartridge, at least one output interface, at
least one memory for storing processor-executable instructions, the
series of oxygen level measurements, and the series of carbon
dioxide level measurements, at least one processor coupled to the
first sensor, the second sensor, the at least one output interface,
and the at least one memory. Upon execution of the
processor-executable instructions, the at least one processor
obtains a first portion of the series of carbon dioxide level
measurements, determines a first respiration rate based on the
first portion of the series of carbon dioxide level measurements,
iterates steps of obtaining a subsequent portion of the series of
carbon dioxide level measurements, determining a subsequent
respiration rate based on the subsequent portion of the series of
carbon dioxide level measurements, and comparing the subsequent
respiration rate to at least one prior respiration rate until a
stable breathing pattern is identified, obtains a stable breathing
pattern portion of the series of oxygen level measurements and a
stable breathing pattern portion of the series of carbon dioxide
level measurements, determines an average minimum oxygen level for
a respiration cycle from the stable breathing pattern portion of
the series of oxygen level measurements, determines an average
maximum carbon dioxide level for a respiration cycle from the
stable breathing pattern portion of the series of carbon dioxide
level measurements, calculates the RQ level from the average
minimum oxygen level and the average maximum carbon dioxide level,
and at least one of displays and transmits, via the at least one
output interface, the calculated RQ level.
[0020] In one embodiment, a computer-facilitated method for
measuring an RQ level in a subject includes receiving, in a
measurement chamber via an input port in fluid communication with
the measurement chamber, respired air from the subject, obtaining,
via a first sensor located in the measurement chamber, a first
portion of a series of oxygen level measurements, obtaining, via a
second sensor located in the measurement chamber, a first portion
of a series of carbon dioxide level measurements, and determining,
via at least one processor, a first respiration rate based on the
first portion of the series of carbon dioxide level measurements.
The method further includes iterating steps of obtaining a
subsequent portion of the series of carbon dioxide level
measurements, determining, via the at least one processor, a
subsequent respiration rate based on the subsequent portion of the
series of carbon dioxide level measurements, and comparing, via the
at least one processor, the subsequent respiration rate to at least
one prior respiration rate until a stable breathing pattern is
identified. The method then includes obtaining a stable breathing
pattern portion of the series of oxygen level measurements and a
stable breathing pattern portion of the series of carbon dioxide
level measurements, determining, via the at least one processor, an
average minimum oxygen level for a respiration cycle from the
stable breathing pattern portion of the series of oxygen level
measurements, determining, via the at least one processor, an
average maximum carbon dioxide level for a respiration cycle from
the stable breathing pattern portion of the series of carbon
dioxide level measurements, calculating, via the at least one
processor, the RQ level from the average minimum oxygen level and
the average maximum carbon dioxide level, and at least one of
displaying and transmitting, via at least one output interface, the
calculated RQ level.
[0021] In one embodiment, a computer-facilitated method for
managing metabolism of a subject includes determining from data
related to the subject obtained via at least one input device a
macronutrient composition and caloric value of food consumed by the
subject, an intensity and duration of activity by the subject, a
rate and maximum capacity of glycogen storage in the subject, and a
rate and maximum capacity of de novo lipogenesis in the subject.
The method also includes optimizing, via at least one processor, a
nonlinear feedback model to model energy substrate utilization in
the subject based on the macronutrient composition and caloric
value of food consumed by the subject, the intensity and duration
of activity by the subject, the rate and maximum capacity of
glycogen storage in the subject, and the rate and maximum capacity
of de novo lipogenesis in the subject. The method further includes
obtaining metabolic data for the energy substrate utilization in
the subject, the metabolic data including RQ data acquired from the
subject, and controlling, via the at least one processor, operation
of the optimized nonlinear feedback model based on the metabolic
data to determine a target value of one or more energy substrate
utilization variables that at least one of maintains and increases
energy substrate utilization in the subject. The one or more energy
substrate utilization variables include at least one of the
macronutrient composition and caloric value of food consumed by the
subject and the intensity and duration of activity by the
subject.
[0022] In one embodiment, a computer-facilitated method for
managing metabolism of a subject includes determining from data
related to the subject obtained via at least one input device at
least one of a macronutrient composition and caloric value of food
consumed by the subject, an intensity and duration of activity by
the subject, a rate and maximum capacity of glycogen storage in the
subject, a rate and maximum capacity of de novo lipogenesis in the
subject, and a quality and duration of sleep by the subject. The
method also includes optimizing, via at least one processor, a
nonlinear feedback model to model energy substrate utilization in
the subject based on at least one of the macronutrient composition
and caloric value of food consumed by the subject, the intensity
and duration of activity by the subject, the rate and maximum
capacity of glycogen storage in the subject, the rate and maximum
capacity of de novo lipogenesis in the subject, and the quality and
duration of sleep by the subject. The method further includes
obtaining metabolic data for the energy substrate utilization in
the subject, the metabolic data including RQ data acquired from the
subject, and controlling, via the at least one processor, operation
of the optimized nonlinear feedback model based on the metabolic
data to determine a target value of one or more energy substrate
utilization variables that at least one of maintains and increases
energy substrate utilization in the subject. The one or more energy
substrate utilization variables include at least one of the
macronutrient composition and caloric value of food consumed by the
subject and the intensity and duration of activity by the
subject.
[0023] In one embodiment, a computer-facilitated method for
modeling metabolism in a subject includes determining from data
related to the subject obtained via at least one input device a
macronutrient composition and caloric value of food consumed by the
subject, an intensity and duration of activity by the subject, a
rate and maximum capacity of glycogen storage in the subject, and a
rate and maximum capacity of de novo lipogenesis in the subject.
The method also includes optimizing, via at least one processor, a
nonlinear feedback model to model energy substrate utilization in
the subject based on the macronutrient composition and caloric
value of food consumed by the subject, the intensity and duration
of activity by the subject, the rate and maximum capacity of
glycogen storage in the subject, and the rate and maximum capacity
of de novo lipogenesis in the subject.
[0024] In one embodiment, a computer-facilitated method for
managing body weight of a subject includes determining from data
related to the subject obtained via at least one input device a
macronutrient composition and caloric value of food consumed by the
subject, an intensity and duration of activity by the subject, a
rate and maximum capacity of glycogen storage in the subject, and a
rate and maximum capacity of de novo lipogenesis in the subject.
The method also includes optimizing, via at least one processor, a
nonlinear feedback model to model energy substrate utilization in
the subject based on the macronutrient composition and caloric
value of food consumed by the subject, the intensity and duration
of activity by the subject, the rate and maximum capacity of
glycogen storage in the subject, and the rate and maximum capacity
of de novo lipogenesis in the subject. The method further includes
obtaining at least one initial physiological parameter associated
with the subject, the at least one initial physiological parameter
including an initial body weight of the subject, and controlling,
via the at least one processor, operation of the optimized
nonlinear feedback model based on the at least one initial
physiological parameter to determine a target value of one or more
energy substrate utilization variables that at least one of
maintains and alters the body weight of the subject. The one or
more energy substrate utilization variables include at least one of
the macronutrient composition and caloric value of food consumed by
the subject and the intensity and duration of activity by the
subject.
[0025] In one embodiment, a computer-facilitated method for
managing body weight of a subject includes determining from data
related to the subject obtained via at least one input device at
least one of a macronutrient composition and caloric value of food
consumed by the subject, an intensity and duration of activity by
the subject, a rate and maximum capacity of glycogen storage in the
subject, a rate and maximum capacity of de novo lipogenesis in the
subject, and a quality and duration of sleep by the subject. The
method also includes optimizing, via at least one processor, a
nonlinear feedback model to model energy substrate utilization in
the subject based on at least one of the macronutrient composition
and caloric value of food consumed by the subject, the intensity
and duration of activity by the subject, the rate and maximum
capacity of glycogen storage in the subject, the rate and maximum
capacity of de novo lipogenesis in the subject, and the quality and
duration of sleep by the subject. The method further includes
obtaining at least one initial physiological parameter associated
with the subject, the at least one initial physiological parameter
including an initial body weight of the subject, and controlling,
via the at least one processor, operation of the optimized
nonlinear feedback model based on the at least one initial
physiological parameter to determine a target value of one or more
energy substrate utilization variables that at least one of
maintains and alters the body weight of the subject. The one or
more energy substrate utilization variables comprise at least one
of the macronutrient composition and caloric value of food consumed
by the subject and the intensity and duration of activity by the
subject.
[0026] In an embodiment, the at least one initial physiological
parameter further includes at least one of height, age, gender,
body mass index (BMI), body fat percentage, waist circumference,
hip circumference, and chest circumference.
[0027] In an embodiment, the nonlinear feedback model is optimized
further based on a quality and duration of sleep by the
subject.
[0028] It should be appreciated that all combinations of the
foregoing concepts and additional concepts discussed in greater
detail below (provided such concepts are not mutually inconsistent)
are contemplated as being part of the inventive subject matter
disclosed herein. In particular, all combinations of claimed
subject matter appearing at the end of this disclosure are
contemplated as being part of the inventive subject matter
disclosed herein. It should also be appreciated that terminology
explicitly employed herein that also may appear in any disclosure
incorporated by reference should be accorded a meaning most
consistent with the particular concepts disclosed herein.
[0029] Other systems, processes, and features will become apparent
to those skilled in the art upon examination of the following
drawings and detailed description. It is intended that all such
additional systems, processes, and features be included within this
description, be within the scope of the present invention, and be
protected by the accompanying claims.
BRIEF DESCRIPTION OF THE DRAWINGS
[0030] The skilled artisan will understand that the drawings
primarily are for illustrative purposes and are not intended to
limit the scope of the inventive subject matter described herein.
The drawings are not necessarily to scale; in some instances,
various aspects of the inventive subject matter disclosed herein
may be shown exaggerated or enlarged in the drawings to facilitate
an understanding of different features. In the drawings, like
reference characters generally refer to like features (e.g.,
functionally similar and/or structurally similar elements).
[0031] FIG. 1 is a diagram illustrating the
"calorie-in/calorie-out" (CICO) model of metabolism.
[0032] FIG. 2 is a diagram illustrating a blood glucose control
model in accordance with some embodiments.
[0033] FIG. 3 is a control block diagram illustrating the body's
blood glucose control system in accordance with some
embodiments.
[0034] FIG. 4 is a graph illustrating the relationship between
respiratory quotient (RQ) values and the combustion of fats and
carbohydrates in accordance with some embodiments.
[0035] FIG. 5 is a diagram illustrating the relationship between RQ
values and weight gain/loss in accordance with some
embodiments.
[0036] FIG. 6 is an overall block diagram for an executable
baseline metabolic model illustrating the relationship between food
consumption, insulin levels, body fat, blood glucose, etc., in
accordance with some embodiments.
[0037] FIG. 7 is a block diagram of the internal composition of the
glycogen storage module in FIG. 6 in accordance with some
embodiments.
[0038] FIG. 8A is a plot illustrating metabolic energy sources and
glycogen transfer for a baseline energy-balanced case in accordance
with some embodiments. FIG. 8B is a plot illustrating daily energy
substrate and glycogen stores in accordance with some
embodiments.
[0039] FIG. 9 is a plot illustrating stable adipose tissue stores
for an energy-balanced state in accordance with some
embodiments.
[0040] FIG. 10A is a plot illustrating metabolic energy sources and
glycogen transfer for the case in FIG. 8A but with high glycemic
index in accordance with some embodiments. FIG. 10B is a plot
illustrating daily energy substrate and glycogen stores for the
case in FIG. 8B but with high glycemic index in accordance with
some embodiments.
[0041] FIGS. 11A and 11B are side-by-side comparison of days 7 and
8 for two different glycemic index cases in accordance with some
embodiments.
[0042] FIG. 12A is a plot illustrating change in fat storage
resulting from increasing glycemic index of dietary carbohydrates
in accordance with some embodiments. FIG. 12B is a plot
illustrating change in fat storage resulting from increased dietary
intake of 300 kCal/day with no compensating exercise in accordance
with some embodiments.
[0043] FIG. 13A is a plot of change in fat storage resulting from
increased dietary intake of 300 kCal/day with fat burning exercise
regimen of 225 kCal/day in accordance with some embodiments. FIG.
13B is a plot of change in fat storage resulting from increased
dietary intake of 300 kCal/day with a carbohydrate burning exercise
regimen of 225 kCal/day in accordance with some embodiments.
[0044] FIGS. 14A and 14B are plots illustrating the results in
carbohydrate intake and daily RQ average from a published
carbohydrate overfeeding experiment.
[0045] FIGS. 15A-15F are plots comparing published results from the
carbohydrate overfeeding experiment with results produced by a
metabolic state model simulation in accordance with some
embodiments.
[0046] FIG. 16 is a cross-sectional view of an RQ device that
includes provision for measuring flow rate and hence energy
expenditure, in accordance with some embodiments.
[0047] FIGS. 17A-17G are perspective views of components of the RQ
device in FIG. 6 in accordance with some embodiments.
[0048] FIG. 18A is a perspective view of an RQ device, also with
provision for a flow-rate sensor, and FIG. 18B is a cross-sectional
view of the RQ device in FIG. 18A in accordance with some
embodiments.
[0049] FIG. 19 is another perspective view of the RQ device in FIG.
18A in accordance with some embodiments.
[0050] FIG. 20 is a process flow chart illustrating breath sampling
in accordance with some embodiments.
[0051] FIG. 21 is a process flow chart illustrating RQ device
calibration in accordance with some embodiments.
[0052] FIG. 22 is a process flow chart illustrating RQ measurement
in accordance with some embodiments.
[0053] FIG. 23 is a graph illustrating RQ measurements from a
subject over time in accordance with some embodiments.
[0054] FIG. 24 is a graph illustrating (hypothetical) energy
substrate utilization as it relates to speed/intensity of running
in accordance with some embodiments.
[0055] FIG. 25 is a graph illustrating energy utilization from
carbohydrates in two subjects while increasing speed on a treadmill
in accordance with some embodiments.
[0056] FIG. 26A is an image illustrating another RQ sensor
prototype in accordance with some embodiments. FIG. 26B is a series
of representative plots illustrating the breath-by-breath data and
RQ obtained with the sensor shown in FIG. 26A in accordance with
some embodiments.
[0057] FIG. 27 is an image illustrating components of a passive
side-stream sampling system in accordance with some
embodiments.
[0058] FIG. 28A is an image and FIG. 28B is a perspective wireframe
view of a mixing chamber for the passive side-stream system of FIG.
27 in accordance with some embodiments.
[0059] FIG. 29A is a representative plot illustrating O2 and CO2
gas concentration curves produced by a passive side-stream sampling
sensor in accordance with some embodiments.
[0060] FIG. 29B is a representative plot illustrating respiratory
quotient, and FIG. 29C is a plot of the corresponding breath volume
rate and energy expenditure estimates associated with the data from
the passive side-stream sampling sensor in accordance with some
embodiments.
DETAILED DESCRIPTION
[0061] The number of people afflicted by obesity, diabetes and
other metabolic disorders has grown dramatically and alarmingly in
the past decades, representing a major cost to society and
adversely affecting the quality of life of many. The present
application describes a system model for human metabolism developed
to gain insight into the relationship between these metabolic
disorders, their underlying causes, and to quantify the
relationship between metabolic health and food. Furthermore,
systems, apparatus, and methods related to modeling, monitoring,
and/or managing a metabolic state of a subject are disclosed in the
present application.
[0062] Existing metabolic models like the "calorie-in/calorie-out"
(CICO) model fail to capture the physiological complexity of
metabolism and therefore do not provide sufficient guidance to
reverse obesity trends in individuals or populations. Thus, a new
metabolic state model is provided for avoiding elevated blood
glucose levels through proper selection of dietary macronutrients
and exercise sufficient to enable the body's homeostatic system to
achieve and maintain a healthy body weight. According to some
embodiments, a personal respiratory quotient (RQ) measurement
device and/or method is described for providing on-demand feedback
indicative of a subject's real-time metabolic state and/or guiding
the subject's macronutrient intake and/or activity levels to
promote a desired metabolic state and/or achieve and maintain a
healthy body weight.
"Calorie-In/Calorie-Out" (CICO) Models
[0063] The use of calorie counting as a means to promote and
maintain weight loss has gained popularity over the past several
decades. While the core concepts driving the CICO model are based
on the conservation of energy principle, many individuals who have
followed a diet plan that is based on the CICO model have met with
failure to achieve and/or maintain their weight goals. The CICO
model does not provide any insight into the metabolic complexity
associated with consumption of food and conversion into useful
metabolic energy or stored energy.
[0064] Specifically, the CICO model does not make a distinction
regarding the macronutrient make-up of the food calories or the
homeostatic mechanisms that incite hunger or satiety in response to
food and activity levels. For example, both one liter of soda and a
salad provide about 450 kilocalories, but a body's response to
these two foods is different in terms of how the calories are
processed, how quickly the calories become available to meet energy
needs, how much of a sense of satiety the calories promote, and how
much the calories perturb the blood glucose levels from
nominal.
[0065] Moreover, recent studies indicate that the CICO model cannot
explain the prevalence of increased adiposity despite reduced
energy intake (see Mozaffarian et al., NEJM, 364 (25): 2392-2404
(2011); Ludwig et al., JAMA, 311 (21): 2167-2168 (2014)). For
example, one longitudinal study that included more than 120,000
subjects found that the kinds of food consumed affected the study
participant's weight gain over a span of four years (id.). Another
study proposed that diets that focus solely on calorie intake, and
specifically those that focus on the restriction of calories
consumed, may increase underlying metabolic dysfunction and, as a
result, increase hunger (id.).
[0066] While adherence to a calorie restricted diet can be
effective in producing weight loss, there are numerous reasons why
calorie restriction usually fails to achieve and/or maintain
long-term weight loss goals, including the following: [0067] 1)
tracking food calories is time consuming and tedious and
consequently may be prone to error; [0068] 2) restricting food
calories without regard to macronutrient content can lead to
nutritional imbalances that trigger hormones and neural responses
that result in overconsumption and/or an inability to adhere to the
calorie-restricted regimen; [0069] 3) genetic differences may
affect a subject's basal metabolic rates as well as a subject's
response to exercise and/or macronutrients, thus rendering
formulaic prescriptive diets to be, at best, approximations; [0070]
4) metabolic needs of a subject change with weight loss and
associated changes in the ratio of lean muscle mass to adipose
tissue, thereby requiring an individualized adaptive calorie
restriction formula to achieve a constant rate of weight loss or
maintenance of a reduced body weight; and [0071] 5) a scale--the
most common method for tracking weight loss--is effective in
revealing long-term body weight trends but has a high variance from
measurement to measurement due to varying states of the subject
including, but not limited to, levels of hydration, glycogen
stores, and contents of the stomach, intestines, and colon.
Aspects of an Optimized Metabolic State Model
[0072] According to some embodiments, a system model is premised on
the observation that the body places a very high priority on
maintaining blood glucose levels in a narrow range. For an average
size individual, normal blood sugar levels, on the order of 100
mg/dl, translate to a control objective of no more than 5 g of
glucose circulating in the body's entire blood volume. The fact
that this level stays fairly constant, regardless of whether food
is plentiful (hundreds of grams of glucose coming in to the
bloodstream per day) or non-existent, is evidence that the body
places a high priority on maintaining normal blood glucose levels
employing mechanisms to monitor blood sugar levels, methods to
signal that it is out of range, and means of actively controlling
and correcting blood glucose levels in response to these
signals.
[0073] At the cellular level, these processing can be quite
complex, however, even with no knowledge of the detailed metabolic
pathways and processes that enable these high level functions, the
engineering discipline of control systems theory may be applied to
develop a high-level system model that captures the dynamic
behavior of metabolic systems in response to food and levels of
physical activity. By definition, a system model is designed to
"hide" underlying complexity which is accomplished by incorporating
only the essential functions necessary to predict the system
behavior for the situations of interest. According to some
embodiments, a control system model enables assessment of the
body's ability to successfully control blood glucose, but more
importantly, it also enables predictions of limitations of this
control, and identification of situations in which the system is
unable to successfully maintain control.
[0074] Body weight is another quantity that the body seems to have
the ability to tightly control under some conditions, as evidenced
by the fact that many people maintain nearly constant body weight
over long periods of time without having to consciously count and
balance their calories consumed and their calories expended. Rather
than focusing on blood glucose, a system model may be developed
that describes the control of body weight in response to foods and
exercise. However, body weight control may be a lower priority than
blood glucose control and, as such, metabolic behavior is largely
explained by the priority of blood glucose control. In fact, a
principal mechanism the body employs to control blood sugar is to
suspend any constraint on body weight as a means of achieving tight
control of blood glucose.
[0075] The present application develops the control system approach
for modeling metabolism from a notional construct to an executable
numerical model.
[0076] The body's demonstrated ability to tightly control blood
sugar implies that the process is well suited to analysis and
modeling techniques developed for feedback control systems. FIG. 2
is a diagram illustrating a blood glucose control model 200 that
considers the inputs of food into the body and exercise by the body
202, the sensing of blood glucose level 204 by a blood glucose
sensing system 206, which generates a glucose control signal 208,
and a set of metabolic control functions 210 that respond to the
control signal and further inputs in accordance with some
embodiments. As illustrated in FIG. 2, in a feedback control
system, the value of the output parameter that is being controlled,
in this case blood glucose level 204, is measured and fed back into
the controller, along with other external inputs, and the
controller continuously adjust parameters in the system to achieve
the desired output.
[0077] Food and activity level represent the body's metabolic
interface to the outside world. Food intake, along with existing
energy stores, provides the fuel needed to sustain metabolic
processes. As much as 70% of the metabolic energy required over the
course of a day is devoted to sustaining autonomic processes such
as breathing, circulating blood, digestion. In addition to these
internal processes, individuals also require energy to move about
and perform daily activities.
[0078] The foods can place tremendous demands on the blood sugar
control system, since an influx of food may release large amounts
of glucose into the bloodstream. In contrast, intense exercise may
deplete large quantities of blood glucose, which must be
replenished quickly to sustain critical processes such as brain
function. In either case, in the context of a blood glucose control
model, the key parameter associated with food intake and activity
level, is the rate at which glucose calories appear or are depleted
from the blood stream.
[0079] Nutrients can have a direct effect on the flow of glucose
into the blood stream. The three main macronutrients (fat,
carbohydrates, and proteins) each have different effects on the
dynamics of glucose level in the blood. Fat is neutral to the rate
of appearance of glucose, since digestion of fat does not result in
any glucose being produced. In contrast, the calories absorbed from
carbohydrates are all destined to be released into circulation in
the form of glucose. The rate of glucose appearance in the blood
depends heavily on the timing of consumption, the quantity of
carbohydrates consumed, and the glycemic index (a measure the speed
with which food is turned into circulating glucose) associated with
the particular carbohydrate. Protein can also affect blood sugar,
since it can be converted into glucose through the process of
gluconeogenesis.
[0080] For typical modern diets, the glucose load into the
bloodstream is largely driven by digestion of carbohydrates. For
example, drinking a sugary beverage might deliver 50 grams of
glucose into the bloodstream at rates as high as about 6-7
grams/minute. If this were not compensated for in some way (see
actuators below), blood sugar levels would rise to over 1000 mg/dl
from this one drink. The fact that high glycemic index
carbohydrates can release glucose into the blood stream quickly
sets demand on the speed of response of the control system to
mitigate the effect.
[0081] Understanding the timing, quantity, and intensity of
physical activity, whether it is part of daily living or formal
exercise is essential to correctly model the ability to control
blood glucose levels. The demands that physical activity places on
the blood glucose control system are directly related to the rate
of depletion of glucose they induce. Low intensity exercise that
burns fat will have negligible effects on glucose control; however,
high intensity exercise utilizes glucose, and will have a direct
effect on the level of glucose in the blood.
[0082] Exercise can use glucose at rates of over 1000 kcal/hour,
which is on the order of 250 g of glucose per hour. Some of this
fuel comes directly from glycogen stores but still has important
impacts on the function of the control system.
[0083] During eating or exercise, the blood glucose sensing
function of the body will detect changes in the glucose level and
will respond by signaling the body to act in a way to offset this
change. Understanding the details of how insulin and its
counterpart, glucagon, broadcast their control signals to the body
is not essential to correctly modeling the functional behavior of
the control loop. It is sufficient to know that subsystems in the
body respond to increasing and decreasing blood glucose levels by
increasing and decreasing the level of the insulin and glucagon
control signals.
[0084] Adopting control system terminology, in accordance with some
embodiments, the mechanisms by which the body responds to the
glucose control signals may be referred to as "actuators."
According to some embodiments, four mechanisms may impact the
quantity of glucose in circulation. In each of these functions, it
is important to consider how much capacity they have to offset the
rates of appearance and disappearance of glucose in the
bloodstream. These capacity limits represent a nonlinear element of
the control system and the effect of the nonlinearity is to produce
complex dynamic responses despite a relatively small number of
functional elements in the overall system. These nonlinear effects
are critical to the understanding of type 2 diabetes and obesity.
TABLE 1 below summarizes the four mechanisms, assesses the rate of
glucose control that they can achieve, and catalogs the limit
(capacity) of their control authority. For each of these four
actuators, it is noteworthy that there is a direct connection
between the presence of insulin and the rise in the level of the
actuation, though this confirmation of insulin's integral role in
affecting glucose control is not required for this model to be
valid.
TABLE-US-00001 TABLE 1 Known Glucose Response Control Actuation
Control Capacity To Mechanism Method Rate Limit Insulin Increase
Burn Glucose Moderate Max Yes Metabolic to lower level (50 g/hr)
metabolic Rate rate (50 g/hr) Store Glu- Move glucose Very high
Glycogen Yes cose as out of blood (250 g/hr) stores Glycogen (~400
g total) Reduce Fat Increase glucose High 0% fat use Yes Burning
burning for fuel (150 g/hr) (50 g/hr) Convert Remove glucose
Moderate DNL rate Yes Glucose to by conversion (30 g/hr) in Liver
Fat (30 g/hr)
[0085] "DNL" is used in TABLE 1 and throughout the present
application as an acronym for de novo lipogenesis, which is
described further herein.
[0086] The exact behavior of each of these mechanisms, as well as
their capacity limitations, will vary across individuals and may
also change with time for a given individual and are also a
function of the recent history of food intake and exercise.
[0087] In response to a signal that blood sugar is high, the body
can react by increasing the basal metabolic rate to consume more
glucose from the bloodstream. Even when an individual is not
intentionally physically active, they are still burning many
calories for basal functions, as much as 70% of the total calories
per day. These calories will draw from the fuel available in the
bloodstream, and will serve to reduce the level of glucose in the
blood. If, in response to the control signal, the metabolic rate
increases, the rate of glucose consumption will increase producing
the desired effect in response to the control signal. The increase
in metabolic energy level may be achieved through a combination of
increased temperature, body motion, or other metabolic function.
For the purposes of the model, it does not matter what explains the
increase in rate only that it occurs and is subject to rate and
capacity limits. The metabolic rate cannot be arbitrarily increased
in response to the signal, so its control authority to remove
glucose is limited to on the order of 50 g/hour (basal rates of
.about.2500 kCal/day).
[0088] The body has the ability to respond to the control signal by
storing glucose in a place other than the bloodstream. The body
stores glucose in the form of glycogen in the liver and in skeletal
muscle distributed throughout the body. While the exact behavior of
the two storage forms are different, from the perspective a control
system model, moving glucose in and out of storage represents a
powerful tool for managing the rates of glucose appearance and
disappearance in the blood. It has the capability to act very
quickly (perhaps 250 g/hour), though the total storage capacity is
typically about 400 grams. Once the storage becomes full, it can no
longer sink excess glucose at any rate. Likewise, if glycogen
stores are empty, they cannot be the source of glucose to
compensate for declining blood glucose levels.
[0089] The body can meet its metabolic fuel needs by using any
available nutrients. When using fat as a nutrient, it has no effect
on blood glucose level. When using glucose, it directly depletes
the glucose level in the blood. Therefore, shifting the fuel mix
towards glucose is another control mechanism the body can use to
respond to the control signal indication of elevated blood glucose.
By adjusting the fuel mix, the body can quickly make large swings
in the rate of glucose disappearing. Of course, it cannot drive the
mix beyond the extremes of 0% or 100%, which represent a hard limit
on the employment of this actuator and in fact since the
brain/blood barrier prevents fat from entering the brain, some
amount of glucose is necessary to sustain brain function and
explains in part why blood glucose control is a priority.
[0090] The body modulates the level of fat burning by converting
fats from free fatty acids (which can be used as fuel) to
triglycerides (which cannot). Understanding this specific mechanism
is not necessary to implement the control model.
[0091] Finally, the body has the capacity to convert glucose into
fat, a process known as de novo lipogenesis. This process removes
glucose from the blood thereby lowering blood glucose. There is
some debate as to what rate de novo lipogenesis can occur at, and
under what conditions it occurs. It appears to activate as a last
resort to blood glucose control when the control authority of the
three other mechanisms described above become saturated.
[0092] According to some embodiments, a metabolic state model is
optimized based on avoiding elevated blood glucose levels through
proper selection of dietary macronutrients and exercise sufficient
to enable the body's homeostatic system to achieve and maintain a
healthy body weight.
[0093] These inputs, the sensing function, control signal, and the
actuators described above may be combined in a way that shows their
connectivity and allows explanation of their interactive function.
FIG. 3 is a control block diagram illustrating the body's blood
glucose control system in accordance with some embodiments. In FIG.
3, food calories 302 from carbohydrates 304 and fats 306 are input
into the model. Carbohydrates 304, in the form of glucose, in the
circulation are depicted flowing from left to right, and dietary
fat 306 is depicted as flowing from left to right into fat storage
or adipose tissue 308. For fat, model 300 is extremely simple.
Since the body has essentially unlimited capacity to store fat, the
disposition of dietary fat can be modeled by initially placing all
consumed fat into this vast storage reservoir. According to some
embodiments, the fat storage may be treated as unlimited and/or the
fat storage may be allowed to contribute to metabolic needs. For
dietary carbohydrates, storage options are limited, and
carbohydrates are depicted flowing directly into the bloodstream,
at a rate determined by the glycemic index, raising the blood
glucose level 310. Excess carbohydrates in the circulation may be
temporarily stored as glycogen 312 in the liver and the skeletal
muscles. Excess carbohydrates in the circulation also may be used
to fulfill metabolic needs 314. For example, metabolic needs may be
met with energy from carbohydrates 316 and/or energy from fat 318.
Further excess carbohydrates in the circulation may result in de
novo lipogenesis 320.
[0094] According to some embodiments, the metabolic state model
acts to maintain a constant blood glucose level 310. For example, a
healthy adult human regulates blood glucose levels within a narrow
band of between about 80 to 110 mg/dl.times.50 dl (i.e., about 5 g
or about 20 calories). This blood glucose level is roughly constant
whether food is plentiful or scarce. If carbohydrates are not
temporarily stored as glycogen 312, used for metabolic needs 314,
or converted into fat 320, .beta.-cells in the pancreas 322 will
produce the hormone insulin 324 to help control blood glucose.
[0095] According to some embodiments, the metabolic state model
identifies the critical metabolic state control loops or actuators,
depicted in FIG. 3 as independent valves 326, 328, 330, and 332.
The master regulator for each valve is insulin. The pancreas 322
monitors the level of circulating glucose 310, and emits insulin
324 as a control signal to four actuators for lowering blood
glucose. Each of the actuators is shown in the figure as a valve
that responds to the control signal. The glycogen storage valve 326
shuttles glucose into the limited stores of glycogen 312, the
metabolism valve 328 controls the metabolic rate 314, the fat
burning valve 330 turns off to maximize the use of glucose, and the
lipogenesis valve 332 ramps up de novo lipogenesis 320.
[0096] According to some embodiments, these metabolic state control
mechanisms include: [0097] a) burning more glucose for energy
through intentional increase in activity (e.g., exercise) or
unintentional homeostatically-induced responses (e.g., fidgeting);
[0098] b) storing more glucose as glycogen in the liver and
skeletal muscles (facilitated by depleting glycogen storage, e.g.,
through exercise); [0099] c) blocking the use of dietary fat as an
energy source (i.e., store the dietary fat instead so that
metabolic needs require more glucose); and [0100] d) converting
more glucose into fat (i.e., de novo lipogenesis).
[0101] According to some embodiments, any or all of these responses
may be rate and capacity limited. These rates and capacities may be
adjusted depending on factors including, but not limited to, the
species, gender, age, and genetic factors associated with each
individual subject. In some embodiments, metabolic needs are capped
at, for example, about 1500 calories per day depending on the
individual subject. The glycogen storage capacity in an adult human
is approximately 15 g/kg of body weight. Thus, in some embodiments,
glycogen storage is capped at, for example, about 1400 calories
depending on the individual subject. In some embodiments, de novo
lipogenesis is capped at, for example, about 300 calories per day
depending on the individual subject.
[0102] Also shown in this model is the notion of glucose overflow
334. When blood glucose exceeds a subject's homeostatic mechanisms
by reaching a level at which the excess glucose cannot be disposed
of through a combination of combustion, conversion or storage, then
other mechanisms manifest themselves in response. The glucose will
overflow into the surrounding tissues (e.g., accumulating in
tissues, joints, retinas, etc.) and the kidneys will pass glucose
into urine, which is characteristic of type 2 diabetes. These are
not considered as actuation methods above, since they represent
disease conditions.
[0103] In some embodiments, the metabolic state model provides
insights into and feedback regarding the relationship between
chronically elevated blood glucose levels and obesity.
[0104] Diabetes is defined as the impaired ability to control blood
glucose, so this model has direct applicability to understanding
its behavior. For type 1 diabetes, the impairment is in the sensing
system. Despite the elevation of blood glucose, no signal, or a
very weak one, is produced to signal the actuation mechanisms.
Whether the missing signal is caused by an inability to sense
glucose in the blood or an inability to produce insulin to
broadcast the message, the actuators are never enabled despite
having ample capacity to control blood sugar. When insulin is
manually added to the blood stream, the full set of actuation
mechanisms will properly function to control blood glucose levels,
though maintaining the precise level of insulin required to balance
incoming glucose with the setting of the actuators is very
difficult to achieve without constant monitoring of the blood
glucose level. This model may be employed to better understand the
dynamics of the insulin control system, in much the same way as an
artificial pancreas might function.
[0105] In some embodiments, this model is also particularly useful
in providing insights into type 2 diabetes. In this disease state,
the body is also unable to control blood glucose levels; however,
it is not due to a malfunction of the sensing portion of the
control system. Typical insulin levels with type 2 diabetes are
much higher than normal, yet the blood glucose control system does
not adequately respond to the sensor signal. The condition is
referred to as insulin resistance, suggesting that the actuation
mechanisms need higher signal levels than previously required to
reach the proper control level. However, there is not a clear
understanding or consensus regarding the root cause of this
apparent loss in insulin sensitivity.
[0106] In some embodiments, this model may offer an alternative
explanation of the precursors and root cause of type 2 diabetes. If
all of the four actuation mechanisms described above are
functioning at their full capacity, then the presence of higher
levels of insulin will have minimal effect on the rate at which
glucose is cleared from the blood, and all of the symptoms of type
2 diabetes will be present, despite high levels of insulin. Blood
glucose control will fail if the control authority available in the
actuators is less than the rate of glucose appearance in the blood
stream. This interpretation of type 2 diabetes implicates the
macronutrient content we ingest as the root cause diabetes
symptoms. Any individual whose actuation mechanisms cannot keep up
with the rate that carbohydrates are being absorbed will exhibit
diabetes symptoms. Changes to diet that add carbohydrates and that
move toward high glycemic index foods will challenge the available
control authority of a greater fraction of our population. This
explanation of type 2 diabetes does not require postulating or
explaining the origins and mechanism of insulin resistance.
Conversely, the model suggests that if carbohydrate rates of
appearance are reduced to be compatible with the individual's
control authority to process carbohydrates, then the symptoms of
type 2 diabetes may quickly disappear.
[0107] The genesis of the metabolic model was to functionally model
the mechanisms involved in responding to changes in blood glucose.
However, the insight the model provides may extend beyond
understanding just the control of blood glucose, by quantifying how
the blood glucose control settings affect body weight.
[0108] Contrary to the view that obesity is the cause of insulin
resistance and the onset of type 2 diabetes, this model suggests an
alternative sequence of events according to some embodiments. For
example, the model predicts that when the body is struggling to
reduce high blood glucose levels, it will shut off the burning of
fat and in some situations convert excess glucose into fat in an
attempt to keep up with the rate of glucose input. If, as a result
of dietary habits, an individual is constantly releasing high
levels of glucose into the blood stream, the response of these
control mechanisms will manifest as an increase in stored fat.
Consequently, for an overweight individual seeking to reduce body
weight through a low fat (hence high carbohydrate) diet, the
metabolic settings the body is choosing in an effort to control
glucose prevent the fat burning needed to reduce body weight.
[0109] In some embodiments, the model may be used to predict an
individual's response to a given diet and how the response varies
with number of calories, macronutrient mix, glycemic index,
duration of exercise, and/or intensity of exercise. Tuned to an
individual, the model may be used to determine what dietary intake
and nutrient mix will ensure that blood glucose control does not
override the body's signals related to weight control, enabling the
individual to burn dietary and stored fat, rather than
carbohydrates, to meet metabolic energy needs.
[0110] In some embodiments, described further herein, a fully
numerical simulation of the control system of FIG. 3 may be used to
produce quantitative assessments of how diet and exercise affect
diabetes and obesity. In further embodiments, also described
further herein, an apparatus comprising a metabolic sensor may
enable an individual to easily customize this model to their own
metabolic limits and behavior, as well as obtain on-demand feedback
on how hard the body is working to keep blood glucose in the normal
range.
[0111] The metabolic state of a body (e.g., a fat burning or a de
novo lipogenesis state) can be accurately tracked by measurement of
the respiratory exchange ratio (RER). As used herein in the
specification and in the claims, RER is interchangeable with
respiratory quotient (RQ). Both RER and RQ are measurements of the
ratio of carbon dioxide (CO.sub.2) production and oxygen (O.sub.2)
consumption.
[0112] FIG. 4 is a graph illustrating the relationship between
respiratory quotient (RQ) values and the combustion of fats and
carbohydrates in accordance with some embodiments. RQ measurements
may inform whether a subject is primarily in a carbohydrate-burning
state (RQ.apprxeq.1.0), in a mixture of carbohydrate and
fat-burning state (e.g., RQ.apprxeq.0.85), or primarily in a
fat-burning state (RQ.apprxeq.0.7). Note that RQ values between 1.0
and 0.7 represent a mixture of fat, carbohydrate, and protein
burning states. During rest and low activity levels, RQ values in
excess of 1.0 indicate that a subject is turning carbohydrates into
fat (i.e., de novo lipogenesis state).
[0113] FIG. 5 is a diagram 500 illustrating the relationship
between RQ values and weight gain/loss in accordance with some
embodiments. RQ measurements may inform whether a subject is in a
state that contributes to weight gain 502 (RQ.gtoreq.1.0),
maintains weight 504 (e.g., RQ matched to the food quotient,
nominally 0.85), or contributes to weight loss 506, with RQ near
the low end of the scale (RQ.apprxeq.0.7) indicating a high
percentage of metabolic energy derived from fat.
[0114] RQ is a measurement of the ratio of the volume of carbon
dioxide produced per unit time to the volume of oxygen consumed per
unit time and may be calculated using equation (1) below:
R .times. Q = V .times. CO 2 V .times. O 2 ( 1 ) ##EQU00001##
[0115] RQ can be used to represent when a subject is in a
carbohydrate burning state or in a fat burning state. For example,
the carbohydrate burning state may be represented as chemical
equation (2) below, which yields the RQ value of 1.0 according to
equation (3) below:
C 6 .times. H 1 .times. 2 .times. O 6 + 6 .times. O 2 .fwdarw. 6
.times. CO 2 + 6 .times. H 2 .times. O ( 2 ) RQ = V .times. CO 2 V
.times. O 2 = 6 6 = 1.0 ( 3 ) ##EQU00002##
[0116] The fat burning state may be represented as chemical
equation (4) below, which yields the RQ value of 0.7 according to
equation (5) below:
C 1 .times. 8 .times. H 3 .times. 6 .times. O 2 + O 2 .fwdarw. 18
.times. CO 2 + 18 .times. H 2 .times. O ( 4 ) RQ = V .times. CO 2 V
.times. O 2 = 1 .times. 8 2 .times. 7 = 0.7 ( 5 ) ##EQU00003##
[0117] According to some embodiments, the incorporation of RQ
measurements in a metabolic state model provides advantages over
food calorie tracking and restriction by directly assessing the
precise metabolic state of a subject on demand. Unlike
calorie-based weight management, coupled with body weight tracking,
RQ provides immediate feedback to a subject as to whether or not
diet and exercise activities, which may be modulated by genetic
factors unique to each individual, are putting the subject in a
homeostatic zone where body weight and glucose are both
successfully controlled. Specifically, when a subject's glucose
system has been overwhelmed for long periods of time (indicated by
blocking of fat burning, and conversion of excess glucose into fat
via de novo lipogenesis), the subject's body manifests these
metabolic responses in the form of weight gain. RQ measurements may
be used to avoid overconsumption of carbohydrates, enabling a
subject to maintain normal blood glucose levels thereby allowing
weight-related homeostasis mechanisms to properly function and, as
a consequence, lose or maintain weight. In addition to weight
control, RQ measurements also may be useful in early detection,
prevention, and/or management of type 2 diabetes and/or other
metabolic diseases. Additional uses of RQ measurements include, but
are not limited to, supporting athletic training and assessing
athletic endurance.
[0118] The invention will be further described in the following
examples, which are not intended to limit the scope of the
claims.
EXAMPLES
An Executable and Optimized Metabolic State Model
[0119] An executable state model optimized for metabolism is
disclosed according to some embodiments.
[0120] Inputs into the executable metabolic state model may
include, but are not limited to, one or more RQ values, a
macronutrient composition and caloric value of food consumed by the
subject (e.g., carbohydrates versus fats), an intensity and
duration of activity by the subject (e.g., carbohydrate-burning
activity versus fat-burning activity), a rate and maximum capacity
of glycogen storage in the subject, a rate and maximum capacity of
de novo lipogenesis in the subject, and/or a quality and duration
of sleep by the subject. In some embodiments, an executable
metabolic state model is run on an RQ device and/or communicatively
coupled with an RQ device. Additional variables may include, but
are not limited to, information associated with measures of blood
glucose, glycogen, average carbohydrate intake, basal carbohydrate
metabolism, body fat, average fat intake, basal fat metabolism,
body weight, heart rate, respiration rate, average insulin level,
and/or insulin level in the subject.
[0121] Outputs from the model may include, but are not limited to,
raw measurement values and/or feedback in the form of suggested
nutritional and/or exercise modifications or goals relating to
weight management, disease management, and/or athletic
training.
[0122] According to some embodiments, the numerical model reveals
the time response behavior of metabolism to both food and activity
levels and, consequently, enables investigation of the impact of
dietary macronutrients and exercise on the selection of energy
substrate and the control of body weight. An executable model may
require quantifying how each of the four control methods described
respond to the presence of the control signal (e.g., an "insulin"
control signal, which takes on positive values in response to high
blood glucose levels and negative values in response to low blood
glucose levels, with the negative values representative of glucagon
signaling), and how the magnitude of the response becomes saturated
as it reaches its control limit. In some embodiments, a numerical
description of how the insulin control signal itself is generated
in response to changes in blood glucose level is developed.
[0123] With simple descriptions of these component pieces, hardware
and/or software tools, such as Simulink.RTM. (available from
MathWorks.RTM., Natick, Mass.), which is built on the idea of
signal flow graphs, may be used to simulate the model behavior when
all of these pieces interact. FIG. 6 is an overall block diagram
600 for a baseline metabolic model in accordance with some
embodiments. The simulation allows for inputs of food and exercise,
a model of each of the four mechanisms to process carbohydrates,
fat stores behavior, an insulin secretion model, and monitors of
key parameters.
[0124] Many variations and extensions to the baseline executable
model of FIG. 6 are possible. For example, Simulink.RTM. includes
an extensive library of functional blocks that may be specialized
and interconnected to represent the information flow associated
with the processes of interest. In FIG. 6, four functional block
types are depicted in accordance with some embodiments. One
functional block type is a pulse generator 602, which is used in
the metabolic model to generate periodic feeding or exercise
schedules. For example, pulse generator 602a represents
carbohydrate burning exercise, and pulse generator 602b represents
fat burning exercise. A second functional block type is a summing
junction 604. A third functional block type is an integrator 606
with upper and lower saturation limits. A fourth functional block
type is a first order transfer function 608, with time constant
1/a, which is used in the metabolic model to generate daily
averages. Viewing scopes 610 are used in the metabolic model to
plot time evolution of one or more user-selected parameters,
including basal carbohydrate metabolism, basal fat metabolism, body
fat, glycogen level, total blood glucose, and average insulin.
[0125] In FIG. 6, five modules are depicted in accordance with some
embodiments. Each module may be a multi-input and/or multi-output
user-defined subsystem, which further includes a number of
interconnected functional blocks. Module 612 executes food
consumption function(s), module 614 executes de novo lipogenesis
function(s), module 616 executes glycogen storage function(s),
module 618 executes metabolism function(s), and module 620 executes
pancreas function(s). Each module is simply a means of grouping
together functional blocks that together implement a common
function. More or fewer modules, even no modules, may be used in
accordance with some embodiments.
[0126] The food consumption module 612 as well as two exercise
blocks 602a, 602b together provide the dietary intake and activity
levels over the duration of the simulation in accordance with some
embodiments. Within module 612, the time for each meal, calorie,
and macronutrient composition is specified. The carbohydrate
burning exercise block 602a is used to represent high intensity,
carbohydrate burning activities, while the fat burning exercise
block 602b is used to represent lower intensity, fat burning
exercise, such as walking. The blocks may be employed together to
represent any combination of fat plus carbohydrate burning
exercise. Employing the pulse generators 602 to represent exercise
allows the duration and frequency of the exercise to be specified
as well as the phasing or time of day for the start of exercise.
The exercise model can be made as complex and variable from
day-to-day as desired, but in the example of FIG. 6, is modeled as
a fixed duration at the same time each day.
[0127] The de novo lipogenesis module 614 in FIG. 6 implements de
novo lipogenesis based on the insulin levels and available glycogen
storage capacity in accordance with some embodiments. When the
insulin levels are high, implying high blood glucose, and there is
little or no remaining glycogen storage capacity available, de novo
lipogenesis is initiated to begin converting circulating blood
glucose to fat for storage. De novo lipogenesis fat is added to any
dietary fat and the combined signal is reduced by any fat burning
exercise and heightened metabolic rate before being added to the
body fat stores, which can be viewed on the body fat viewing
scope.
[0128] The dietary carbohydrates take a slightly more complicated
path through the model in FIG. 6, passing through three summing
junctions where the dietary intake can be reduced by de novo
lipogenesis activity, glycogen storage, and metabolic rate, before
passing through an integrator that tracks the blood glucose level,
subsequently passing this signal to the pancreas module 620 where
the insulin control signal is generated based on the blood glucose
levels.
[0129] The glycogen storage module 616 controls the glycogen
storage and retrieval process which is modeled as a whole-body
effect, with no distinction between glycogen storage in liver
versus muscle tissue. This distinction may be incorporated in the
module if desired.
[0130] FIG. 7 is a block diagram of the internal composition of the
glycogen storage module 616 in FIG. 6, which is described in detail
to provide a representative example of the internal structure of a
functional module according to some embodiments. In FIG. 7, some
new functional block types are depicted, including a signal input
port 702 for a user defined module, which is used in the glycogen
storage module 616 to input carbohydrate exercise rate 702a and
insulin level 702b. Another new functional block type is a gain
block 704, which is used in the glycogen storage module 616 to
multiply an incoming signal by a user-defined gain term K. Yet
another new functional block type is a signal output port 706 for a
user defined module, which is used in the glycogen storage module
616 to output glycogen transfer from blood 706a and glycogen level
706b.
[0131] If insulin level 702b is positive, insgain/glycap block 704
is scaled by the fraction of remaining glycogen storage capacity
(i.e., (glycap-glylevel)/glycap), with the result that insulin
impact on glycogen storage varies from a maximum of insgain, when
glycogen stores are fully depleted, to zero, when glycogen storage
is at capacity.
[0132] If insulin level 702b is negative (i.e., glucagon), insulin
impact on glycogen retrieval varies from insgain to zero in
proportion to glylevel/glycap. The function of these blocks
comprising the glycogen storage module 616 and their ranges are
summarized in TABLE 2 below.
TABLE-US-00002 TABLE 2 Element Units Range Default Input: Insulin
Level -- Defined in insulin .+-.30 module Input: Carb Exercise Rate
Kcal/hr Defined in carb 0 exercise module Gain2: insgain/glycap
kcal.sup.-1 Set in module 50/2300 mask parameters Switch:
controlled by insulin kcal .+-.insgain -- polarity Store 1:
Integrator Output kcal 0 to glycap 0.5*glycap Store 1: Saturation
indicator -- 0 unsaturated, -- .+-.1 when saturated Output:
Glycogen Level -- 0 to 1 0.5 fraction Output: Glycogen Transfer
kcal/hr insgain* 50*30 (insulin level)
[0133] Turning attention again to the overall metabolic model block
diagram in FIG. 6, the metabolism module 618 monitors the glycogen
storage level and the insulin level in order to adjust the fraction
of metabolic energy derived from carbohydrates versus fats, with
the ability to completely shut off fat as a source of energy and
rely exclusively on carbohydrates (glucose). The metabolism module
618 may include a user-settable basal metabolic rate with diurnal
variation and also as the capability to increase the basal
metabolic rate over a user-specified range in response to high
levels of insulin.
[0134] For anyone familiar with the cellular level processes for
supplying metabolic energy demands, storing excess carbohydrates
and fats and releasing the hormones necessary to control these
processes, the elements comprising the block diagram in FIG. 6 may
seem too sparse and/or simple to provide useful insight into these
complex metabolic processes. However, while the model is simple by
design, the output of the model can be quite complex and varied in
response to dietary and exercise inputs. For example, the model may
be shown to capture the salient metabolic responses identified in a
comprehensively documented fourteen-day study of carbohydrate
overfeeding with reasonable accuracy as described further
herein.
[0135] The performance of the model and the function of the modules
in FIG. 6 may be illustrated with a simple baseline reference case
and modification of the food and exercise inputs.
[0136] For example, in a baseline case, the basal metabolic rate in
the metabolism module 618 is set to 1800 kCal/day, with a diurnal
variation of .+-.300 kcal, and the feeding schedule consists of
breakfast, lunch, and dinner with no between-meal snacks. The
macronutrient mix and calorie equivalent of the meals is shown
below in TABLE 3.
TABLE-US-00003 TABLE 3 Parameter kcal Carb % .tau..sub.FAT (hrs)
.tau..sub.GLY (hrs) Breakfast 550 56 0.5 0.2 Lunch 500 56 0.5 0.2
Dinner 775 56 0.5 0.2 Total or Average 1825 56 0.5 0.2
[0137] Note that for this baseline example, daily dietary intake is
just 25 kcal higher than the basal metabolic rate and all of the
meals are comprised of 56% carbohydrates and 44% fats with the
absorption time constants for the carbohydrates and fats, different
from one another but constant over all meals. These parameters
represent a nearly unrealizable uniformity for actual meals but
serve to illustrate the salient features of the model. For
simplicity and clarity of illustration, the meal schedule and
composition is held fixed for each simulation day.
[0138] In the reference baseline, there is no exercise or physical
activity scheduled other than the diurnal basal metabolic rate of
1800 kCal/day. The available fat stores at the beginning of the
simulation are initialized to 25,000 kcal which, at a nominal 3500
kcal/lb corresponds to approximately 7 lb of adipose tissue. The
glycogen storage capacity is initialized to 2300 kcal, the
equivalent of 575 g of carbohydrate stored in the liver and
muscles, with glycogen stores half full at the beginning of the
simulation.
[0139] According to some embodiments, the metabolic model is run
for a simulation time period of two weeks, or 336 hours, and
completes execution in less than about one second. The seven view
scope icons depicted in FIG. 6 represent viewers in which the
parameter identified can be viewed. Additional scopes may be added
to display and compare multiple parameters in one plot.
[0140] FIG. 8A is a plot illustrating metabolic energy sources and
glycogen transfer for the baseline energy-balanced case in
accordance with some embodiments. In particular, FIG. 8A depicts
the change in glycogen stores and the energy derived from carbs and
fats in units of kcal/hr. The impact of the three meals is evident
in the daily spikes in blood glucose and metabolic activity. There
is also movement of carbohydrates in and out of glycogen storage
although the amounts are small. Integration of the curves results
in the daily averages shown in FIG. 8B.
[0141] FIG. 8B is a plot illustrating daily energy substrate and
glycogen stores in accordance with some embodiments. In particular,
FIG. 8B presents this information in the form of a daily average of
carbohydrate and fat energy sources along with the glycogen storage
all in kCal/day. After the initial transient associated with
initializing the model, the average daily energy expenditure
approaches a steady state matched to the dietary intake, with small
transients related to the three daily meals, carbohydrates moving
in and out of storage, and diurnal variation in the basal metabolic
rate.
[0142] The implication is that there is an energy balance which, as
FIG. 9 illustrates, results in a stable value of stored fat over
the simulated two week time period. FIG. 9 is a plot illustrating
stable adipose tissue stores for energy balanced state in
accordance with some embodiments.
[0143] In the baseline case, the dietary intake is closely matched
to the resting metabolic rate and the glycemic index of the diet is
low, thereby minimizing blood glucose spikes and resulting in a
stable weight and good control of blood glucose. In the next
example, the impact on metabolism is assessed from simply
increasing the glycemic index of the diet, from 0.2 to 0.6, without
changing either the total dietary calories or macronutrient mix.
The increase in glycemic index by a factor of three results in
carbohydrates entering the blood stream three times more quickly,
spiking the blood sugar, generating a strong insulin response and
more forcefully activating the glucose control mechanisms.
[0144] FIG. 10A is a plot illustrating metabolic energy sources and
glycogen transfer for the case in FIG. 8A but with high glycemic
index in accordance with some embodiments. Relative to FIG. 8A,
blood glucose excursions are larger as are the glycogen transfer
rates and carbohydrate metabolism. Relative to the low glycemic
index baseline case, FIG. 10A clearly exhibits more rapid temporal
variation consistent with a higher glycemic index in FIG. 8A, and
larger excursion from nominal or average levels for all of the
signals shown. FIG. 10B is a plot illustrating daily energy
substrate and glycogen stores for the case in FIG. 8B but with high
glycemic index in accordance with some embodiments. Relative to
FIG. 8B, the peak carbohydrate metabolism is increased, the average
fat metabolism is reduced and glycogen storage variability is
increased.
[0145] To better compare the two cases and understand the impact of
the high glycemic index, FIGS. 11A and 11B are side-by-side
comparison of days 7 and 8 for the two different glycemic index
cases in accordance with some embodiments. Referring to FIG. 11A,
corresponding to the low glycemic index case, the glycogen
transfers, which are in units of kcal, are relatively small, on the
order of 40 kcal (10 g) or less. The blood glucose level rises
after each meal from the baseline set point of 30 kcal to a peak of
around 40 kcal. There is little evidence of hypoglycemia. Note that
while the dinner meal has the highest fat content, the fat
metabolism, in units of kcal/hr, exhibits the lowest peak for the
dinner meal because priority has been given to metabolizing blood
glucose from the carbohydrate content of the meal and moving some
of the excess blood glucose into glycogen storage.
[0146] In FIG. 11B, corresponding to the high glycemic index case,
the blood glucose excursions are larger, peaking at 66 kcal with
lows less than 13 kcal, indicative of hypoglycemia, and the
glycogen transfer excursions are also larger with 110-kcal storage
peaks and 30-kcal retrieval peaks. The fat metabolism is suppressed
relative to the low glycemic case since as the as the blood sugar
spikes due to the higher glycemic index, carbohydrate metabolism
peaks and reliance on fat to meet metabolic energy needs is
suppressed. At the same time more of the dietary carbohydrate
intake is being moved into glycogen storage. These actions are all
driven by higher levels of circulating insulin. At the completion
of carbohydrate digestion blood glucose rates fall precipitously
since there is no longer a source of glucose but the circulating
blood insulin level does not return to normal immediately leading
to episodes of hypoglycemia. In a real situation rather than a
simulation, these hypoglycemic events might lead to additional
carbohydrate snacking in response to the low blood glucose.
However, in the simulation, the feeding schedule is prescribed
(which one might think of as a dieter with an "iron will" or a
clinical study in which diet is strictly controlled).
[0147] FIG. 12A is a plot illustrating change in fat storage
resulting from increasing glycemic index of dietary carbohydrates
while preserving the baseline dietary caloric intake in accordance
with some embodiments. Total increase in stored fat over the
two-week simulation is small (1000-kcal equivalent to a weight gain
of about one-quarter pound) but is indicative of an energy
imbalance induced solely by a change in glycemic index. Since the
metabolic response to the blood sugar spikes is to increase
reliance on circulating glucose to meet energy needs while reducing
reliance on circulating fat, the net result, as shown in FIG. 12A,
is that body weight is increased as dietary fat moves into storage
during episodes of high blood glucose. As a response to the
subsequent episodes of hypoglycemia, glycogen, rather than the
dietary fat, is moved out of storage to counteract the resulting
low blood glucose.
[0148] Since the baseline model exhibited an energy balance, with a
basal metabolic rate of 1800 kCal/day and a food intake of 1825
kCal/day, one might wonder why simply increasing the glycemic index
would result in weight gain when the system was previously in
energy balance. The answer to the apparent mystery of where the
extra energy came from to produce weight gain has to do with one of
the four control mechanisms identified in TABLE 1. In particular,
as blood glucose levels rise, carbohydrate metabolism increases in
order to dispose of glucose at a higher rate. The increased
metabolism may take the form of higher body temperature, higher
agitation, but, apart from actively increasing exercise, there is a
limit to the total metabolic rate that can be achieved, which in
the baseline model considered here is set to 100 kcal/hr. In the
baseline energy-balanced example with low glycemic index, the
maximum metabolic rate was never approached. However, with the
higher glycemic index, the maximum metabolic rate needed to control
the blood glucose is higher than the allowable maximum which leads
to higher rates of glycogen storage and lower rates of fat
metabolism resulting in higher levels of fat storage. During the
episodes of hypoglycemia, the metabolism adapts to lower values and
begins recruiting stored glycogen, not the previously stored fat,
to raise blood glucose levels. The fat that was stored during the
blood glucose spikes remains in storage since it will not help with
blood glucose control. The overall cycle of the adaptive metabolism
rate hitting a maximum followed by a nadir, lowers the average
metabolism over the course of the three meals and the reduced
average energy demands account for the increase in body fat.
[0149] The previous examples did not include any explicit exercise
periods. To further demonstrate the efficacy of the model and the
insights it affords, the model is returned to the reference diet
defined in TABLE 3, but 100 kcal are added to each meal of the
three daily meals, increasing the daily caloric intake from 1825
kcal to 2125 kcal. The glycogen storage is also initialized to full
capacity at the start of the simulation, which may be
representative of a sedentary individual. FIG. 12B is a plot
illustrating change in fat storage resulting from increased dietary
intake of 300 kCal/day with no compensating exercise in accordance
with some embodiments. As shown in FIG. 12B, without the
introduction of exercise, the increased calorie intake of 300
kCal/day results in a fat gain of 4090 kcal (about 1 lb) during the
course of the two-week simulation.
[0150] A misconception sometimes promoted by fitness trainers is
that "if you want to lose weight, you should preferentially engage
in low intensity exercise, such as walking, since it is fat
burning." Without question, low intensity exercise favors fat over
carbohydrate oxidation. However, in terms of optimal weight loss,
this advice ignores the complex interactions of the metabolic
subsystems and the priority placed on blood glucose control. In
other words, without a quantitative model to guide understanding,
it is easy even for the experts to be misled regarding the impact
of diet and exercise choices. To illustrate, the simulation is run
again, this time adding exercise to offset the 300 kCal/day
increase in dietary intake. All of the exercise is first added as
low-intensity fat burning, and then the simulation is repeated with
all of the exercise represented as high intensity carbohydrate
burning exercise. For the purpose of comparison, the exercise is
begun each day at 6 am and is 1.5 hours in duration with a total
calorie expenditure of 150 kcal/hr. The 150 kcal/hr is not
representative of high intensity exercise, but the same time
duration and calorie rate for the carbohydrate burning exercise are
used to keep as many variables the same for the sake of comparison
and to focus attention on the impact of fat burning versus
carbohydrate burning. In terms of justifying this assumption, one
could imagine high intensity exercise with rest periods
interspersed in order to achieve the same average kcal rate as the
low intensity exercise over the 1.5 hour duration.
[0151] FIG. 13A is a plot of change in fat storage resulting from
increased dietary intake of 300 kCal/day with fat burning exercise
regimen of 225 kCal/day in accordance with some embodiments. As
shown in FIG. 13A, the fat burning exercise reduced the weight gain
from 4090 kcal with no exercise over the two week simulation, down
to 1390 kcal with fat burning exercise. Note however that the
positive slope on the fat accumulation curve implies that
additional gains will occur in subsequent weeks under this feeding
and exercise regimen.
[0152] In comparison, FIG. 13B is a plot of change in fat storage
resulting from increased dietary intake of 300 kCal/day with a
carbohydrate burning exercise regimen of 225 kCal/day in accordance
with some embodiments. As indicated in FIG. 13B, carbohydrate
burning exercise of the same calorie value and duration as the fat
burning exercise results in a peak fat gain of 1080 kcal on day 6,
the weight gain then begins to fall off slightly to a stable 1020
kcal by the end of the second week.
[0153] The fact that the same level of energy expenditure leads to
different body fat outcomes for fat burning versus carbohydrate
burning exercise may not seem surprising, but it does seem
counter-intuitive that fat burning exercise would be the less
effective exercise to achieve weight loss. This and the previous
example clearly reveal the inadequacy of a simple
calories-in/calories-out model of weight loss.
[0154] While the executable metabolic model of FIG. 6 behaves as
expected for the contrived examples described in the previous
sections, a more rigorous evaluation of the model fidelity may
require comparison with carefully controlled studies in which the
dietary intake, activity levels, and metabolic state are monitored
over one or more diurnal cycles. A challenge in validating the
metabolic model in this way is that few published studies provide a
complete description the metabolic state of the subjects at the
beginning of the study or report macronutrient composition of meals
and feeding and exercise schedules in sufficient detail. For
example, there is seldom knowledge of the state of an individual's
glycogen stores at the beginning or end of an experiment, and the
previous examples have shown that the state of glycogen stores has
a significant impact on how an individual responds to a high
carbohydrate diet.
[0155] From a modeling perspective, TABLE 4 below summarizes the
desired information for proper initialization of a metabolic model,
according to some embodiments, and identifies the corresponding
parameters reported in three of the more comprehensive studies
uncovered in the literature and incorporated by reference herein:
Acheson et al., Am. J. Clin. Nutrition, 48:240-47 (1988); McDevitt
et al., Am. J. Clin. Nutrition, 74 (6):737-46 (2001); and Jebb et
al., Am. J. Clin. Nutrition, 58:455-62 (1993).
TABLE-US-00004 TABLE 4 Info Acheson McDevitt Jebb Documented
Desired (1988) (2001) (1993) Duration Days to 3 + 7 + 2 4 7 + 12
(days) observe trends Number of More is better 3 13 5 subjects
Subject Gender, age, Young male Middle- Healthy, physiology BMI,
fitness athletes aged women ages 19-42 Body Measured Weight,
Weight, Weight, composition at least daily BMI, BMI, BMI, % body
fat DXA density Fat and Measured Daily 4-day Daily carbohydrate
daily aggregate balance Subject Measured N/A N/A 1 hour BMR daily
per day Initial Known state Depleted N/A N/A glycogen at start
stores Energy Measured at Daily 4-day Daily expenditure least
diurnally average aggregate average Exercise Intensity and 3 per
day 4 per day Ergometer duration 3 .times. 40 min Meal Time and 3
per day 4 per day N/A schedule duration Meal Macronutrients, Daily
mix, 4-day Macro- composition GI no GI aggregate nutrients, kcal
Hydration Daily intake N/A N/A N/A RQ Minimum Daily N/A Daily
hourly average average DNL RQ and .sup.2H.sub.2O Daily by
.sup.2H.sub.2O trace to N/A rates RQ > 1 fat stores Protein
Quantified Quantified N/A Urinalysis metabolism through Blood
plasma urinalysis CGM, 7 .times. Triglycerides, N/A analysis
insulin, leptin, triglycerides, insulin, triglycerides insulin,
glucose, throughout day glucose, leptin at lipoproteins 96
hours
[0156] For example, in Acheson (1988), a comprehensive fourteen-day
metabolic energy balance study included a nearly complete
description of the parameters necessary to establish a ground truth
reference for validating the efficacy of the metabolic state model
according to some embodiments. Briefly, three healthy young men,
ages 21-22, weights 62-72 kg, heights 174-180 cm, and body fats
11-14%, with no family history of diabetes or obesity participated.
During the first three days, the subjects consumed a restricted
diet, high in fat and low in carbohydrates, and followed an
exercise program to deplete their glycogen stores prior to tracking
the energy balance over a period of ten days. Halfway through this
energy restrictive period, the subjects were admitted into a whole
room indirect calorimetry chamber in which respiratory exchange
measurements were to be continued for ten days. After 36 hours in
the chamber, their diet was changed to a high-carbohydrate, low-fat
diet that was ingested for the following seven days. During the
last two days, while still in the chamber, the subjects received
limited amounts of a high protein diet essentially devoid of
carbohydrates. The subjects then left the respiration chamber but
continued to consume the high-fat, low-carbohydrate diet for a
further two days.
[0157] The restricted high-fat, low-carbohydrate diet consumed on
days 1-3, 13, and 14 provided 1600 kcal comprising 15% protein, 75%
fat, and 10% carbohydrate. During the overfeeding period (days 4-10
inclusive), the high-carbohydrate, low-fat diet provided 3600 kcal,
comprising 11% protein, 3% fat, and 86% carbohydrate. Energy intake
was increased progressively each day while the composition was kept
constant with the intent to provide 1500 kcal in excess of the
previous day's energy expenditure, which was measured in the
respiration chamber. By day 10 the energy intake had thus increased
to 5000 kcal. For the next (and last) two days in the respiration
chamber, the subjects consumed a high-protein, low-calorie diet
(protein sparing modified fat, 600 kcal). The diet was then changed
to the same restricted high-fat, low-carbohydrate diet eaten on
days 1-3 for the last 2 days of the experiment spent outside the
chamber.
[0158] FIGS. 14A and 14B are plots illustrating the results in
carbohydrate intake and daily RQ average of this carbohydrate
overfeeding experiment. Note that from the onset of carbohydrate
overfeeding starting on day 4, it took more than 24 hours for the
daily average RQ to exceed 1.0 from a glycogen-depleted starting
point of 0.75, and 4 days for the carbohydrate over feeding to
effectively saturate the glycogen storage capacity and raise daily
average RQ to its maximum observed value of 1.16.
[0159] FIGS. 15A-15F are plots comparing published results from the
carbohydrate overfeeding experiment with results produced by a
metabolic state model simulation in accordance with some
embodiments. The metabolic model was initialized with the data
corresponding to the dietary intake, activity, and average
physiology of the three subjects in the carbohydrate overfeeding
experiment. The solid curves in FIGS. 15A-15F represent the
composite values for the three individuals involved in the study as
reported in the published paper. With the exception of FIG. 15C,
the dotted curves represent the values predicted by the metabolic
state model given the feeding and exercise schedules during the
study. In some embodiments, the model may include a protein module;
however, in this case, the caloric values of the protein
macronutrients, shown in FIG. 15A, were lumped with the fat input.
Unlike the study, which took fourteen days to complete, the
simulation of the study completed in less than one second.
[0160] A number of the measured parameters from the study are
compared to the simulation output, including total calorie intake
in FIG. 15A, daily energy expenditure in FIG. 15B, total energy
expenditure in FIG. 15C, unhydrated glycogen storage in FIG. 15D,
change in fat stores in FIG. 15E, and RQ in FIG. 15F. The degree of
correlation between the published data and the simulated output
provides compelling evidence that, without resorting to detailed
modeling of cellular metabolic pathways and processes, the impact
of dietary macronutrients and exercise can be captured by this
system-level feedback control model.
[0161] Realizing that blood glucose control is a critical function
that takes precedence over other functions of the homeostatic
control system leads to a somewhat different perspective regarding
the causes and treatments for excess body weight and obesity. TABLE
5 below compares some of the implications of the metabolic state
model to commonly held views.
TABLE-US-00005 TABLE 5 Metric Conventional View Model View Obesity
Contributor to Mechanism by which T2-diabetes the body attempts to
avoid becoming diabetic High insulin Caused by insensitivity Caused
by overwhelming levels to Insulin (Insulin the 4 glucose control
Resistance) actuators Diet for Requires overt Requires active
avoidance weight loss control of calorie of carbohydrate
intake/expenditure "overconsumption" Exercise for Low intensity is
best Choose exercise to counteract weight loss because it's fat
burning macro-nutrient imbalance Feedback for On-demand body weight
On-demand metabolic state weight loss measurements (scale)
measurements (RQ, insulin, etc.) Excess body Root causes are Root
causes are historical weight decreased activity changes in
macronutrient population and larger food mix that subvert
homeostatic trend portions control loops
[0162] As noted above, the human body has only four actions to take
to control blood glucose levels and two of these--shutting off fat
burning and turning glucose into fat--lead to weight gain.
Consequently, weight gain leading eventually to obesity is not
necessarily the cause of diabetes, but rather the body's best
effort to avoid becoming diabetic by clearing excess glucose from
the blood stream. Eventually, when these four processes are
overwhelmed, the diagnosis is diabetes, but it is not caused by
obesity.
[0163] Similarly, high insulin levels are often attributed to
insulin insensitivity whereas the model clearly shows that the
glucose control options are rate limited, and as the pancreas
increases the control signal (insulin), eventually the rate limits
of the processes are reached. Over time, the burden of persistently
high blood glucose levels may degrade the effectiveness of the
control mechanisms, but the model suggests that the degradation of
the response to insulin (insulin resistance) is not the
precipitating event but a consequence of long-term overburdening of
the glucose control mechanisms.
[0164] The fact that body weight may fluctuate by a kilogram or
more during the course of a day and that body weight fluctuations
are caused by a number of processes not related to fat gain or
loss, means that body weight is only a reliable indicator of
progress over time periods of many days or weeks. As a consequence,
making the association between dietary and exercise choices and
their impact on weight loss goals on any given day, whether good or
bad, is challenging to discern by tracking body weight, since the
impact may not be discernable for days.
Apparatus for Measuring RQ
[0165] Given the differences in the way individuals respond to
dietary macronutrients and exercise, the challenge is provide
feedback, on demand, regarding an individual's metabolic state. On
a long-term basis, there are methods for quantifying metabolic
health through blood draws and analysis, and many individuals
obtain snapshots of their metabolic health during an annual
physical. However, to control weight and assess daily choices
regarding macronutrient intake and exercise, it would be beneficial
to be able to measure blood glucose levels, insulin levels, and
disposition of macronutrients on a daily or hourly basis. While a
number of blood analysis measurements can now be made by means of
finger pricks and reagent strips, including blood glucose and
ketone levels, these measurements are invasive, require consumables
which can be costly over time and, most importantly, don't directly
reveal insulin levels or energy substrate mix.
[0166] However, referring to the block diagram of FIG. 3, there is
a simple, non-invasive method of measuring individual energy
substrate mix and inferring the state of three of the four
metabolic control valves 326, 328, 330, and 332 in the diagram.
While not a direct measure of insulin levels, knowledge of the
state of these valves conveys information about metabolic health
and normalcy of insulin levels. According to some embodiments, this
may be determined non-invasively from an analysis of exhaled
breath. The method is based on a simple breath analysis to measure
a quantity known as the respiratory exchange ratio (RER) which, in
the absence of short-term breathing and anaerobic exercise
artifacts, is representative of the respiratory quotient (RQ) at
the cellular level.
[0167] The burning of fat and carbohydrates to supply metabolic
energy needs is essentially a combustion process and, as such,
consumes oxygen (O.sub.2) and produces combustion by-products, in
particular carbon dioxide (CO.sub.2). Furthermore, the ratio of
CO.sub.2 produced to O.sub.2 consumed is different for fat burning
than for carbohydrate burning. At the cellular level, the ratio of
CO.sub.2 produced to O.sub.2 consumed is called the respiratory
quotient (RQ) and is equal to 1.0 for carbohydrate combustion and
0.7 for fat combustion. If carbohydrates are being used exclusively
to supply all of the metabolic energy needs of the whole body, the
RQ will be about 1.0 and if fats are being used predominantly, the
RQ will be closer to 0.7. As indicated in the discussion of the
metabolic model, in general, metabolic energy needs are met by a
combination of carbohydrate and fat burning, in which case the
value of RQ provides an indication of the percent of metabolic
energy derived from carbs versus the percent derived from fats.
[0168] The O.sub.2 needed for combusting carbs and fats is drawn
from inhaled air, and similarly, the CO.sub.2 produced by the
processes is expelled in exhaled air. Consequently, under a fairly
broad set of conditions, measuring the gas concentration of an
individual's exhaled breath, to determine the O.sub.2 consumed and
the CO.sub.2 produced, provides all the information needed to
determine the percent of metabolic energy derived from carbs versus
fats. Protein can also be used to supply metabolic energy, but it
is a minor player in supplying metabolic energy needs and, in any
case, has an RQ in the mid-range between carbs and fats, nominally
about 0.8 depending upon the details of the protein.
[0169] Clearly the ability to measure and track one's RQ would
provide valuable information about the impact of specific food and
exercise choices on metabolism, and thus provides a means to assess
how dietary macronutrient choices and levels of activity drive the
body's choice of fuels. The implication is that if a low-cost
personal sensor can be developed, it can be used to provide
information to the user on-demand regarding the current state of
his metabolic system. Armed with such a sensor, one could, if
desired, track the impact of dietary and exercise choices on, for
example, an hour-by-hour basis including during exercise
itself.
[0170] Apparatus for measuring and/or tracking RQ are disclosed
according to some embodiments. An RQ device may provide real-time
assessment of a subject's metabolic state. Because of the real-time
measurement capacity of a subject's RQ device, RQ measurements may
be used to effect immediate changes in a subject's behaviors that
affect the subject's metabolic state, such as, for example,
modification of nutritional intake and/or activity level. According
to some embodiments, an RQ device includes and/or is
communicatively coupled with a computer-implemented metabolic state
model.
[0171] In some embodiments, such as those illustrated in FIGS.
16-19, systems and apparatus for measuring RQ use low-cost
commercial electro-optical sensors which, because of their small
size, weight, and power, have time constants longer than a breath.
Consequently, the sensor design emphasizes the capture and holding
of end-tidal portions of the exhaled breath sample in the
measurement chamber during the inhale cycle in order to allow the
slow sensors sufficient time to settle to an accurate measurement.
The challenges to implementing this direct breath sampling sensor
include: [0172] 1) avoiding condensation on the sensor, which is
achieved in at least some embodiments by heating the sensor to the
expected temperature of the exhaled breath; [0173] 2) preventing
ambient air from entering the measurement chamber during
inhalation, which is achieved in at least some embodiments by using
multiple large valves to ensure low respiratory burden; [0174] 3)
correcting for changes in measurement chamber conditions, which is
achieved in at least some embodiments by providing sufficiently
fast and accurate temperature and humidity sensors; [0175] 4)
correcting for the inability of at least some sensors (too slow) to
measure temporal variations in O.sub.2 and CO.sub.2 concentrations
over each breath profile, which is achieved in at least some
embodiments by incorporating an adaptive energy expenditure
estimation algorithm.
[0176] FIG. 16 is a cross-sectional view of a portable RQ device
1600 in accordance with some embodiments. In FIG. 16, the RQ device
1600 includes a mouthpiece socket 1602, a flow rate aperture 1604,
and an intake flow valve 1606 for receiving expired breath from a
subject. As shown in FIG. 16, an input port on the device (i.e.,
the mouthpiece socket 1602) may be designed to be compatible with a
sports cap on a disposable bottle, thereby providing for an
inexpensive source of mouthpieces for the same or multiple users.
The RQ device in FIG. 16 also defines a measurement chamber with
sensors 1608 positioned therein. Printed circuit boards (PCBs)
1610, a display 1612, a purge fan 1614, and a battery 1616 are also
included in accordance with some embodiments.
[0177] FIGS. 17A-17G are perspective views of components of the RQ
device in FIG. 16 in accordance with some embodiments. FIGS.
17A-17C show more detailed views of the body of the device. FIG.
17D shows the circuit boards, FIG. 17E shows a flow sensor, FIG.
17F shows an O.sub.2 sensor, and FIG. 17G shows a CO.sub.2
sensor.
[0178] FIG. 18A is a perspective view and FIG. 18B is a
cross-sectional view of a second RQ device in accordance with some
embodiments. The RQ device FIG. 19 is another perspective view of
the RQ device in FIG. 18A in accordance with some embodiments.
[0179] An RQ device may have at least one input port. In some
embodiments, a subject breathes in and out through a mouthpiece
and/or cap that is attached to an input port of an RQ device. In
other embodiments, a subject breathes in and out through one or
more nasal tubes attached to an input port. The sampling of the
subject's breath may be hands-free to allow for continuous sampling
of exhaled breath without interfering with conversation,
consumption of food and drink, or other activities and/or to
facilitate the use of the device during movement or exercise. For
example, a subject may be fitted with nasal tubes and/or a
mouthpiece that connects to an input port and, optionally, is
secured around the subject's head, thus allowing for hands-free
sampling. Other hands-free configurations may include securing
sampling tubes around a subject's torso, arm, neck, or lower body.
Hands-free sampling is enabled, at least in part, because, unlike
energy expenditure (Calorie) measurements, measurement of RQ does
not require measuring of volume rates.
[0180] In some embodiments, an RQ device is fitted with one or more
valves that control flow of air through the device during subject
exhalation (e.g., an exhaust valve) and/or inhalation (e.g., an
intake flow valve). A valve may help enforce unidirectional flow of
air during the subject's breathing. Unidirectional airflow, in tum,
may support more accurate measurements. A valve also may aid in
capturing a subject's end tidal air, as explained further below. In
some embodiments, at least one valve is provided for exhalation.
Two, three, four, five, six, seven, eight, nine, ten, fifteen, or
twenty valves may be provided for exhalation. In some embodiments,
at least one valve is provided for inhalation. Two, three, four,
five, six, seven, eight, nine, ten, fifteen, or twenty valves may
be provided for inhalation. Different types of valves may be used.
For example, the intake flow valve in FIG. 16 is a flapper valve.
Flapper valves are also used as the intake and exhaust valves in
FIGS. 18A and 18B in accordance with some embodiments.
[0181] In some embodiments, an RQ device includes a fan to remove
or at least partially remove gases from the device and/or prevent
or at least partially reduce condensation in the device. A fan may
also contribute to keeping the measurement chamber closed during
the initial period of exhalation and/or open only during the
end-tidal portion of the breath. For example, the RQ device in FIG.
6 shows a purge fan. In some embodiments, an RQ device includes a
shutter to enable the capture and measurement of a subject's
expired end tidal air fraction, as explained further below.
[0182] According to some embodiments, an RQ device includes at
least one electronic component for controlling and/or communicating
information about the device including, but not limited to,
measurements made with the device. The electronic components may be
electrically coupled using one or more printed circuit boards as
shown in FIGS. 16 and 17D. Additional electronic components for
controlling and/or communicating information about the device may
be external to the RQ device but communicatively coupled (with
wires or wirelessly) with the RQ device as shown in FIG. 18A. An RQ
device may be further configured to house an internal battery, as
shown in FIG. 16, and/or to be coupled with an external power
source, as shown in FIG. 18A. An internal battery may be
rechargeable (with wires or wirelessly) or disposable. In further
embodiments, an RQ device may include a user interface. The user
interface may include an input component and/or an output component
(e.g., visual, aural, and/or haptic signals or displays). For
example, the RQ device in FIG. 16 includes a visual display screen.
Alternatively or in addition, an RQ device may be coupled (with
wires or wirelessly) with an external input device and/or output
device (e.g., a smartphone, tablet, laptop, or other computing
device).
[0183] In some embodiments, an RQ device includes one or more
sensors or detectors. The one or more sensors may include a flow
sensor and/or a gas sensor.
[0184] A flow sensor is a component for sensing a rate of fluid
flow and may include, but is not limited to, a microsensor that
measures the transfer of heat caused by the moving fluid (e.g., a
thermal mass flow meter), a laser-based interferometer, a
photoacoustic Doppler sensor, and/or a mechanical flapper valve or
vane that is pushed by the fluid to drive, for example, a rotary
potentiometer.
[0185] An RQ device may sample and/or measure a subject's end tidal
breath, maximum expired CO.sub.2, and/or minimum expired O.sub.2.
According to some embodiments, an O.sub.2 sensor and/or a CO.sub.2
sensor is arranged to capture the end tidal concentrations of
O.sub.2 and/or CO.sub.2, respectively.
[0186] A gas sensor is a component for sensing and/or measuring one
or more gas types and may include, but is not limited to, an
electrochemical gas sensor, an infrared point sensor, an infrared
imaging sensor, a semiconductor, a spectrophotometer, a tunable
diode laser spectrometer, photoacoustic spectrometer, and/or a
holographic gas sensor. In some embodiments, a gas sensor is an
all-optical gas sensor for detecting and measuring a gas
concentration using the characteristic optical/spectral absorption
of the gas species. A gas sensor may be a slow gas analysis
sensor.
[0187] According to some embodiment, optical gas sensing technology
reduces the need for expendables and/or frequent replacement of
consumable sensor elements such as electrochemical fuel cell
sensors. For example, commercial sensors may use an electrochemical
cell for O.sub.2 detection, the anode of which is consumed during
measurements such that the sensor requires frequent re-calibration
and eventual replacement (typically after 6-12 months of use).
Thus, an all-optical gas sensor may be used to minimize size,
weight, power, and/or use of consumables.
[0188] In some embodiments, an RQ device includes a CO.sub.2
all-optical gas sensor and/or an O.sub.2 all-optical gas sensor.
For example, an O.sub.2 sensor may be a fluorescence quenching
sensor, as shown in FIG. 17F, and/or a CO.sub.2 sensor may be a
nondispersive infrared (NDIR) sensor, as shown in FIG. 17G. In some
embodiments, the housing/packaging of a commercially-available
sensor (e.g., the O.sub.2 sensor in FIG. 17F) may be modified by
removing an apical portion of the housing and/or modulating a
thickness of a ruthenium component of the sensor.
[0189] One challenge of using these low-power, low-cost sensors for
the measurement of O.sub.2 and/or CO.sub.2 is that the sensor time
constants may be long compared to the respiration periods and the
time constants may be unequal in some embodiments. In order to
compensate for the long and/or unequal time constants associated
with the sensors, the measurement chamber of the device may be
designed to be relatively small and relatively close to the
subject's mouth. Such embodiments account for the effect of dead
space air (i.e., the volume of air that is inhaled but does not
take part in gas exchange). The dead space air associated with the
mouth, esophagus, and upper lungs is exhaled first and subsequently
pushed through the measurement chamber by remaining exhalant until,
finally, the end tidal breath is captured in the measurement
chamber where it resides during the next inhalation cycle.
[0190] In some embodiments, as a consequence, these slow gas
analysis sensors are exposed to end tidal air during most of a
breathing cycle and therefore asymptotically approach the true end
tidal concentration of O.sub.2 and CO.sub.2 with only brief
exposure to dead space air during the next exhalation. This
technique of trapping only end tidal air in the measurement chamber
may be accomplished through the use of low-cost, mechanical valves,
as discussed above, that consume minimal or no additional power in
accordance with some embodiments. The influence of a subject's dead
space air may be reduced further by the use of a mechanical
shutter, a vacuum pump, and/or a fan timed to keep the measurement
chamber closed during the initial period of exhalation, and open
only during the end tidal portion of the subject's breath.
[0191] Another challenge associated with the unequal time constants
of the O.sub.2 and CO.sub.2 sensors is that at any instant of time,
one sensor may have settled to, for example, about 90% of the final
value while the slower sensor may only have settled to, for
example, about 50% of the final value. Therefore, taking a ratio of
the CO.sub.2 concentration to the O.sub.2 concentration measured at
the same time may produce large errors. In order to contend with
the unequal time constants of the sensors, in some embodiments, the
measurements are not time-synchronized. Instead, a maximum value of
CO.sub.2 during a breath is measured, a minimum value of O.sub.2
during a breath is measured, and these measurement are used to
estimate the RQ value.
[0192] Following the acquisition of the minimum expired O.sub.2
(i.e., end tidal O.sub.2 sample) and the maximum expired CO.sub.2
(i.e., end tidal CO.sub.2 sample) the average RQ is computed over
valid breaths as follows:
R .times. Q A .times. V .times. G = .SIGMA. .function. [ CO 2
.times. e .times. x .times. p N 2 .times. e .times. x .times. p -
CO 2 .times. i .times. n N 2 .times. i .times. n ] .SIGMA.
.function. [ O 2 .times. i .times. n N 2 .times. i .times. n - O 2
.times. e .times. x .times. p N 2 .times. e .times. x .times. p ] (
6 ) ##EQU00004##
[0193] FIG. 20 provides a process flow chart illustrating a method
for breath sampling 2000 in accordance with some embodiments. In
step 2002, gas sensors are sampled at at least 1 Hz. In step 2004,
offset and gain correction is applied to raw sample data with
two-point (gain and offset) calibration constants for O.sub.2 and
CO.sub.2 sensors 2006. In step 2008, the fraction of expired
nitrogen is computed using the following equation:
%N.sub.2exp=1-%O.sub.2exp%CO.sub.2exp (7)
In step 2010, the fraction of oxygen consumed is computed according
to:
%O.sub.2in=20.9%-%O.sub.2exp (8)
[0194] In step 2012, the mean and stand deviations are computed.
Sample outliers are discarded. If the fraction of expired oxygen
has reached a minimum 2014 and fraction of expired carbon dioxide
has reached a maximum 2016, then an end tidal oxygen sample 2018
and an end tidal carbon dioxide sample 2020 are used to compute
average RQ over valid breaths in step 2022 according to equation
(6) above.
[0195] Over a period of extended use, all gas sensing technologies
are subject to drift. As a result, existing indirect calorimetric
sensors either require frequent re-calibration with a precision gas
mixture, or the use of a pre-calibrated expendable cartridge. An
important improvement in some embodiments of an RQ device is the
ability for a user to quickly and easily calibrate the sensor using
an inexpensive and readily available CO.sub.2 cartridge.
Calibration of an O.sub.2 sensor and/or CO.sub.2 sensor may be
achieved by performing a zero (offset term) and a span (gain term)
calibration on each sensor. Typically this process would be
performed by a trained technician in a laboratory using calibrated
gas mixtures; however, that is unnecessary according to at least
some embodiments. In some embodiments the zero gas concentration
condition for both sensors can be met by purging the sensor chamber
with a halocarbon gas obtained from a commercially-available
aerosol "gas duster" container.
[0196] First, all-optical gas sensors may require less frequent
full-scale calibration. In accordance with some embodiments, an RQ
device, when turned on, may automatically or manually undergo a
pre-use ambient air calibration that takes into consideration
average ambient air levels of O.sub.2 and CO.sub.2. A limited
ambient-air calibration may be performed each time a sensor is
powered up. Ambient air calibration may assume that, except in very
confined rooms with inadequate ventilation, dry ambient air O.sub.2
concentration is about 20.9%, which may be adjusted downward to
account for relative humidity, and CO.sub.2 concentration is about
0.04%. Ambient-air calibration allows for an adjustment or
correction of gain in an O.sub.2 sensor (most susceptible to gain
drift) and/or an adjustment or correction of offset in a CO.sub.2
sensor (most susceptible to offset drift). Thus, in some
embodiments, ambient-air calibration helps to maintain sensor
accuracy over time.
[0197] Historical ambient-air calibration measurements may be
stored in and/or accessible to an RQ device for subsequent
comparison to future ambient O.sub.2 and CO.sub.2 measurements in
order to determine an amount of drift in the settings. If
predetermined amount of drift in the settings is exceeded, a full
calibration cycle may be performed by the user.
[0198] A full calibration may be implemented, when necessary,
automatically or manually by, for example, using a CO.sub.2
cartridge in connection with the ambient-air calibration procedure.
First, the O.sub.2 gain and/or CO.sub.2 offset may be computed at
power up from ambient air. Subsequently, CO.sub.2 cartridge may be
introduced into the measurement chamber of an RQ device. In some
embodiments, an RQ device includes a port or chamber for receiving
CO.sub.2 into the measurement chamber from a CO.sub.2 cartridge. By
filling the measurement chamber with CO.sub.2, residual O.sub.2 is
purged or at least reduced. At that point, the zero offset for
O.sub.2 may be determined so that a full two-point calibration of
the O.sub.2 sensor is achieved. As the CO.sub.2 leaks from the
measurement chamber, it is replaced by ambient air, with a known or
calculable concentration of O.sub.2 (e.g., about 20.9%) and
CO.sub.2 (e.g., about 0.04%) in accordance with some embodiments.
Because the O.sub.2 sensor has undergone a two-point offset and
span calibration, it may serve as a calibrated reference, which
together with the known or calculable concentration of O.sub.2, may
be used to determine a span calibration for the CO.sub.2 sensor.
For example, when the O.sub.2 sensor reads about 16.72%, then about
80% of the CO.sub.2 has leaked out of the measurement chamber and
been replaced with ambient air, so the CO.sub.2 sensor should
measure a concentration of about 20%.
[0199] Although these sensors offer long term stability, another
challenge associated with using optical gas sensors is that the
optics are subject to contamination resulting in a loss of
calibration. Because contamination of optical surfaces on a sensor
may degrade accuracy of subsequent measurements and the humidity of
exhaled air is close to 100%, a sensor may be heated to a
temperature slightly higher than, for example, the subject's body
temperature, in order to exceed the dew point and thus prevent
condensation of moisture from exhaled breath on the sensor's
optical surfaces in accordance with some embodiments. In some
embodiments, a sensor is automatically or manually heated or
allowed to warm to a predefined temperature including, but not
limited to, 30.degree. C., 31.degree. C., 32.degree. C., 33.degree.
C., 34.degree. C., 35.degree. C., 36.degree. C., 37.degree. C.,
38.degree. C., 39.degree. C., 40.degree. C., 45.degree. C., or
50.degree. C.
[0200] FIG. 21 is a process flow chart illustrating a method for RQ
device calibration 2100 in accordance with some embodiments. A
history of ambient-air span and zero recalibrations 2102 is
maintained. In step 2104, drift history is reviewed to determine
whether full calibration is needed. If not, step 2106 is performed
to update the calibration history. In step 2108, ambient oxygen and
carbon dioxide levels are reviewed to determine whether they are
within specification. If they are, no calibration is required 2110.
If they are not within specification, oxygen span or carbon dioxide
zero-point is recalibrated in step 2112 and the calibration history
is updated 2106. However, if drift history indicates that full
calibration is needed, the sensor measurement chamber is allowed to
fill with ambient air in step 2114, oxygen span is set to 20.9% and
carbon dioxide zero-point is set to 0.04% in step 2116, a carbon
dioxide cartridge is attached to the measurement chamber valve to
purge the chamber in step 2118, oxygen is set to 0% in step 2120,
and calibration is continued as carbon dioxide leaks out over time
2122. Once the oxygen sensor reading is 16.7.+-.0.1% in step 2124,
the carbon dioxide span is set to 20% in step 2126, and full
calibration is completed 2128.
[0201] The evolution of the design of the RQ devices in FIGS. 16,
17A-17G, 18A-18B, and 19, including the challenges associated with
certain aspects of RQ measurement and the solutions provided, are
detailed below in TABLE 6.
TABLE-US-00006 TABLE 6 1st Gen Challenge Approach Problems
Innovation Breath sampling Capture single RQ variable due to
Average over end-tidal breath ventilation artifacts several breaths
on a single breath Multiple breath Require deep, slow Accentuates
Physically modify averaging breaths to accommodate ventilation
artifacts COTS sensor slow O.sub.2 sensor to gain about 10x speed
up Demarcate each Use CO.sub.2 waveform O.sub.2 min and CO.sub.2
Average O.sub.2 and breath sample and shape to reliably max not
necessarily CO.sub.2 separately rather average RQs demarcate
breaths phased in time than breath-by-breath RQs Inhaling through
Inhale through Abnormal breathing Incorporate flapper sensor
dilutes the nose and pattern creates valves to allow mouth sample
volume exhale through ventilation artifacts breathing exclusively
the mouth Number of breath Observe real-time Not a viable Compute
mean samples needed RQ plot until approach for a and variance and
steady state reached non-technical user stop when variance
sufficiently small Ventilation artifacts Observe real-time Not a
viable Discard outliers based from cough, laugh, RQ plot to
visually approach for a on history of yawn, etc. detect artifacts
non-technical user breath-by-breath averages Condensation Employ
metal The power and Heat only the of about 100% RH case and heat to
time to reach condensation sensitive breath fogs optics 38.degree.
C. to avoid 38.degree. C. is prohibitive components condensation
Thermal insulation Insulate measurement Foam tends to Employ
impermeable to speed and hold chamber with foam preferentially
absorb insulation materials 38.degree. C. temperature gas
components CO.sub.2 sensor Use known gas Not a viable approach
Employ CO.sub.2 sensor calibration drifts mixtures to recalibrate
for low-cost model that allows over time CO.sub.2 sensor personal
use calibration from ambient air O.sub.2 sensor calibration Use
known gas Not a viable Employ O.sub.2 sensor drifts over time
mixtures to recalibrate approach for low- model that allows O.sub.2
sensor cost personal use calibration from ambient air RQ can be
biased Make only Unreliable RQ Incorporate flow by lactate
buffering resting RQ measurements during rate sensor to gauge
measurements intense exercise exercise intensity from VO.sub.2
[0202] According to some embodiments, an RQ device is designed
and/or used to measure resting RQ. The underlying rational for
measuring resting RQ is that the energy substrate mix at rest
provides for a more accurate assessment of a subject's metabolic
state. One design challenge to acquiring the subject's resting RQ
is a potential for variations in the subject's breathing rate and
ventilation, which would bias the RQ measurement. Thus, in some
embodiments, an RQ device is designed to compute an RQ measurement
based on an average of several breaths.
[0203] An RQ device may measure CO.sub.2 concentration at a high
temporal rate (e.g., at least 5 times the breathing rate) to track
the CO.sub.2 variation over time. The CO.sub.2 concentration
measurements may be used to demarcate each breath and thereby
ascertain a resting respiration rate. The variability in
respiration rate, together with the measurement history of the
O.sub.2 and CO.sub.2 concentrations, may be used to determine when
a subject's breathing is stable, thereby producing a reliable
representation of the subject's resting RQ. In some embodiments, a
history of a subject's resting respiration rate and/or a
measurement of the subject's heart rate are used to distinguish
resting RQ measurements from RQ measurements made during periods of
activity or exercise.
[0204] FIG. 22 is a process flow chart illustrating a method of RQ
measurement 2200 in accordance with some embodiments. A daily
baseline of resting RQs (e.g., upon waking, before/after meals,
before/after exercise, and/or before bedtime) 2202 is maintained.
In step 2204, RQ measurement is initiated. In step 2206, the RQ
sensor is turned on. Once a ready indication is received in 2208,
indicating that the gas sensors have warmed to 38.degree. C. and/or
ambient-air auto-calibration has executed, ambient air calibration
correction history may be updated in step 2210. In step 2212, a
user breathes normally into the sensor through a mouthpiece and/or
nasal cannula. The sensor automatically detects new breath and
determines maximum carbon dioxide in each breath and minimum oxygen
in each breath in step 2214. Average carbon dioxide samples and
standard deviations are computed as are average oxygen samples and
standard deviations in step 2216. In step 2218, the standard
deviations are compared to stability limits. If they fall outside,
outliers are discarded in step 2220 and the process returns to step
2214. However, if they fall within, RQ is computed in step 2222,
optionally archived in step 2224 (e.g., for further optimization),
and RQ is displayed via an audio, graphic, or tactile indication
2226.
[0205] According to some embodiments, a subject may take RQ
readings at various times throughout a day as defined by actual
time and/or activity. For example, a subject may measure RQ before
and/or after fasting, consuming food and/or drink, exercising,
and/or sleeping. In some embodiments, a subject may measure RQ at
specific time intervals including, but not limited to, 1 hour, 2
hours, 3 hours, 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9
hours, 10 hours, 11 hours, or 12 hours before or after an
activity.
[0206] In some embodiments, a subject breathes into an RQ device,
through either a mouthpiece or a nasal cannula (e.g., in a
hands-free configuration). At least one sensor detects new breaths,
a maximum CO.sub.2 concentration per breath, and/or a minimum
O.sub.2 concentration per breath. In some embodiments, an RQ device
computes the average CO.sub.2 concentration and the average O.sub.2
concentration to determine a standard deviation (SD) of the
readings. If the SD of the readings is within predefined limits, an
RQ measurement may be output including, but not limited to,
displayed, stored, and/or transmitted. If the SD of the readings is
outside the predefined limits, the outlier readings may be
discarded. In that case, the subject may continue to breathe into
the RQ device until readings that fall within the acceptable SD
range are collected.
[0207] In some embodiments, RQ measurements provide a way to finely
monitor and tune a subject's metabolic state as a source of
individual biofeedback. RQ values may impart: instant knowledge of
the fraction of fat versus carbohydrate energy a subject is
utilizing; an indication of whether a subject is in a de novo
lipogenesis metabolic state (i.e., converting excess blood glucose
into fat); (3) an indication of carbohydrate stress (implying high
blood glucose levels); and/or (4) an understanding of glycogen
capacity. These personal biofeedback indicators make it possible to
track a subject's metabolic state and/or modify diet and exercise
accordingly to, for example, achieve/maintain a weight goal, manage
metabolic disease, and/or improve metabolic fitness and
endurance.
[0208] Unlike calorie counting, in some embodiments, RQ
measurements encourage diet and exercise adjustment by providing
on-demand feedback regarding the impact of macronutrient choices
and activity levels to avoid carbohydrate overloading that in tum
leads to storage, rather than metabolic combustion, of dietary fats
as well as conversion of glucose to stored fat via de novo
lipogenesis, both of which contribute eventually to obesity. In
some embodiments the personalized biofeedback capacity of the RQ
measurement device is used to manage or reduce the likelihood of
type 2 diabetes. RQ measurements may replace or reduce the
frequency of capillary blood glucose measurements required to
manage blood sugar. For example, resting RQ of 2:1.0 indicates high
blood glucose (hyperglycemia), whereas resting RQ of about 0.7
indicates a range of normal to low blood glucose
(hypoglycemia).
[0209] The use of routine RQ measurements also may support
individual diagnostics. In some embodiments, a database of "normal"
or historical levels of RQ measurements under various conditions
(e.g., fasting, glucose challenge, post-prandial, etc.) is
maintained. When newly acquired RQ readings consistently vary from
the normal or historical measurements, an indication of a loss of
metabolic homeostasis may be provided to the subject or a clinician
for early warning as to the onset of a disease state.
[0210] In some embodiments, RQ measurements may be used to monitor
metabolic response. FIG. 23 is a graph illustrating RQ measurements
from a subject over three days in accordance with some embodiments.
The subject is a healthy adult male human who had been training for
a half-marathon at the time the RQ readings were collected. In FIG.
23, on October 11, two days prior to running the half-marathon, the
subject's RQ values were measured at six separate times. First, the
subject's fasting RQ was measured in the morning (10:09 AM),
indicating that about 56% of the energy utilized by the subject was
obtained from carbohydrates. After eating a low-carbohydrate
breakfast (i.e., an omelet) (11:36 AM), still only 56% of the
energy utilized by the subject was obtained from carbohydrates.
Note that following a period of activity (i.e., thirty minutes of
walking and swimming for half a mile) (1:15 PM), the
carbohydrate-based energy utilized by the subject fell from about
56% to about 50%, indicating that the exercise affected the
subject's metabolism, increasing the proportion of energy supplied
from fat versus carbohydrates. At lunch the subject consumed a high
carbohydrate meal (i.e., two pieces of pizza and half a cookie)
(2:25 PM), after which the carbohydrate-based energy utilized by
the subject began to rise, first to about 55% and, several hours
later (4:40 PM), to about 95%. However, after further activity
(i.e., a brisk walk) (5:11 PM), the carbohydrate-based energy
utilized by the subject dropped to about 11%, indicating that the
subject's metabolism was dominated by the low-intensity fat-burning
exercise.
[0211] On October 12, the day before the half-marathon, the subject
consumed large quantities of carbohydrates (i.e., "carb loading").
An RQ reading taken later that evening (9:08 PM) indicated an RQ
greater than 1.0 implying that all of the energy utilized by the
subject was obtained from carbohydrates. Furthermore, the subject
was in a de novo lipogenesis metabolic state in which excessive
blood glucose levels were being converted to stored fat. However,
on October 13, three hours after completing the half-marathon and
before eating, the carbohydrate-based energy utilized by the
subject fell to 11%, indicating that the subject was again in a
high fat burning state due to the combination of no food intake and
depleted glycogen stores.
[0212] In some embodiments, RQ measurements may be used to predict
or provide a guideline or suggestion for performing an activity.
FIG. 24 is a graph illustrating representative metabolic fuel
utilization as it relates to speed/intensity in accordance with
some embodiments. In particular, FIG. 24 illustrates the utility of
measuring RQ for the assessment of optimal running speed in a
hypothetical marathon scenario. To first order, a runner bums about
the same calories per unit distance regardless of speed (i.e., at
high speed the distance is covered in less time but requires a
higher level of exertion than at lower speeds). In the absence of
nutritional supplements, running at maximum speed may deplete the
runner's circulating blood glucose and glycogen stores, a
phenomenon known as "bonking" or "hitting the wall", after which
speed and performance drop precipitously. As implied in FIG. 24 and
illustrated by TABLE 7, knowledge of RQ versus running speed,
enables a runner to select a pace to achieve an appropriate mix of
fuels (e.g., fats plus carbohydrates versus carbohydrates alone) in
order to delay or avoid glycogen depletion (i.e., running smart)
and thus achieve better overall performance than simply running at
maximum intensity (i.e., running fast).
TABLE-US-00007 TABLE 7 Parameter Run Fast Run Smart Calories
expended (per mile) 130 130 Initial glycogen reserves (Calories)
1500 1500 Initial speed (mph) 8.5 7.2 Initial pace (min:sec per
mile) 7:04 8:17 Distance to glycogen depletion (miles) 11.5 26.2
Peak RQ during 26.2 mi race 1.00 0.83 Average RQ over 26.2 mi race
0.80 0.83 Average speed over 26.2 mi race (mph) 6.1 7.2 Average
pace over 26.2 mi race (min:sec per mile) 9:50 8:17 Finish time
(hr:min) 4:16 3:37
[0213] In some embodiments, RQ measurements may provide a
predictive measure of a subject's endurance. FIG. 25 is a graph
illustrating energy utilization from carbohydrates in two subjects
moving on a treadmill at incrementally increasing speeds in
accordance with some embodiments. Both subjects are healthy adult
male humans with resting energy expenditures from carbohydrates of
about 41% and about 52%, respectively. RQ values were measured at
each speed, approximately every two minutes. The RQ readings
indicate that a subject with the lower resting RQ (i.e., a lower
percentage energy usage from carbohydrates) is able to maintain a
higher treadmill speed prior to reaching an all carbohydrate energy
consumption state. This anecdotal example also underscores the fact
that individual metabolisms are unique and thus benefit from
personalized, on-demand information about metabolic state, rather
than prescriptive, formulaic diets and exercise regimens.
[0214] FIG. 26A is an image illustrating another RQ sensor
prototype in accordance with some embodiments. The sensor employs
optical techniques to determine the concentration of oxygen and
carbon dioxide in expired breath and consequently does not require
expendables. The sensor measures the oxygen consumed and carbon
dioxide produced by averaging several breaths to account for
short-term ventilation effects. FIG. 26B is a series of
representative plots illustrating the breath-by-breath data and RQ
obtained with the sensor shown in FIG. 26A in accordance with some
embodiments. For purposes of illustration, the elapsed time for the
measurements shown in FIG. 26B is a little over three minutes, but
a shorter measurement, on the order of 30-60 s is adequate to
provide a stable estimate of RQ. Note from the plots in FIG. 26B
that while there is some variation in RQ the average is 0.93,
indicating that a majority (77%) of the metabolic energy during the
measurement is being derived from carbohydrate burning.
[0215] A key application envisioned for such as sensor is to
provide daily or even hourly feedback to the user regarding
dominant source of metabolic energy, essentially their metabolic
energy zone. In terms of weight loss by reduction of adipose
tissue, an RQ that hovers around 1.0 throughout the day implies
that there is little energy being derived from fat and what energy
is derived from fat is probably coming from circulating dietary fat
rather than the stored fat that is the objective of the weight loss
effort. Over-consumption of carbs, where overconsumption is
dependent on the individual metabolism and energy needs, may even
lead to an RQ>1, which is indicative of carbs being converted
and stored as fat for future energy needs. Thus a necessary, but
not sufficient, condition for weight loss through reduction of
stored fat is for the RQ to be below 1. The lower the RQ, the
greater percentage of energy derived from fat and, if the
individual is not overconsuming dietary fat, the energy will be
derived from stored fat producing weight loss.
[0216] By design, the metabolic model is intended to capture
salient metabolic performance with the least amount of complexity.
Consequently, there are many options for augmenting the model with
additional modules (thus raising complexity) in order to simulate
metabolic performance with greater detail or over a wider range of
dietary inputs and activity levels.
[0217] A personal RQ sensor, coupled with a metabolic modeled
tailored to the individual, provides a basis for developing more
effective regimens for weight loss, diabetes management, and
athletic training programs. Rather than a one-size fits all
formulaic diet for weight loss, the metabolic model and sensor
provide the quantitative tools to tailor macronutrient intake and
exercise activity to achieve weight loss and weight management
goals. A non-invasive personal RQ sensor has the potential to
detect and track episodes of high blood glucose reducing the
frequency of capillary blood testing and providing a higher density
of data to correlate against dietary intake and activity in order
to develop an individualized predictive model.
Passive Side-Stream Sampling
[0218] Existing breath-by-breath metabolic sensors employ
relatively fast and expensive gas sensors (e.g., with sample rate
time constants on the order of 150 ms or less to track variations
in O.sub.2 and CO.sub.2 gas concentrations over the course of a
breath) and/or active pumps to sample and deliver exhalant to the
measurement chamber at a constant volume rate.
[0219] According to some embodiments, a passive side-stream
sampling system may achieve accurate RQ and energy expenditure
estimates using slower, less expensive, smaller, lighter, and/or
lower power gas sensors and a low-cost and/or low-power
differential pressure sensor.
[0220] FIG. 27 is an image illustrating components of a passive
side-stream sampling system according to some embodiments. In FIG.
27, a sample tube 2700 includes a mouthpiece 2702 and a flow
restrictor or venture 2704. Two pressure tubes 2706 connected to a
differential pressure flow sensor are used to measure the
volumetric flow of inhaled and exhaled breath. Venture 2704 is
configured to create a back pressure which forces a small fraction
(e.g., a few percent) of the exhaled air into at least one
side-stream pick-off 2708 into a tube 2710 connected to a mixing
chamber 2712 where the O.sub.2 and CO.sub.2 gas concentrations are
measured. In some embodiments, the flow restrictor 2704 provides a
linear pressure/flow relationship, which then provides a stable
sidestream gas splitting fraction over the entire breath flow
range. In other embodiments, the flow restrictor 2704 provides a
more quadratic pressure/flow relationship, so the sidestream
fraction increases at higher breath flow rates as compared to lower
flow rates. In some embodiments, tube 2710 comprises a water vapor-
and temperature-permeable polymer membrane, such as Nafion.RTM.
tubing, configured to reduce and/or equilibrate both the
temperature and humidity of the exhaled breath sample with ambient
conditions. The rate of flow of the breath sample into mixing
chamber 2712 may be set proportional to the flow rate of the
exhaled air, thereby implementing proportional analog sampling of
each breath without the need for an active pump. In some
embodiments, several breaths may be required to displace all of the
air in mixing chamber 2712, so the chamber acts as a breath gas
integrator in which the O.sub.2 and CO.sub.2 gas concentrations are
averaged via mixing over several breaths and the rate of change is
commensurate with the time constants of slower gas sensors. A key
element of the mixing chamber design is a low-cracking pressure
(e.g., <about 1 mm H.sub.2O) exit or exhaust valve 2714 to
sustain breath sampling even at low pressures developed by the flow
restrictor 2704 encountered toward the end of an exhale when the
breath flow is low. The advantage of a sidestream flow consisting
of a fixed-fraction of the mainstream breath flow is that it allows
gas averaging over the entire breath flow composition range,
spanning from anatomical dead space gas through end-tidal alveolar
gas. This provides the same functionality of a Douglas bag sampling
system, but in a much more compact and portable form factor. In
some embodiments, a Universal Serial Bus (USB) cable 2716 or
another connector, cable, and/or protocol may be used for
connection, communication, and/or power supply between mixing
chamber 2712 and, for example, a processor, memory, and/or energy
storage device.
[0221] FIG. 28A is an image and FIG. 28B is a perspective wireframe
view of a breath sensing system which includes a mixing chamber in
accordance with some embodiments. In FIGS. 28A and 28B, the mixing
chamber includes an exhaust valve 2802 and replaceable battery
2804, which could also be a rechargeable battery. In some
embodiments, a breath sensing system may be configured for
connection to a battery, a wired or wireless connection to a
battery carried elsewhere on a user's body or located nearby, or a
connection to an external power supply (e.g., an AC plug). In some
embodiments, this electrical connection includes a USB,
FireWire.RTM., and/or similar connectorized cable.
[0222] FIG. 29A is a representative plot illustrating O2 and CO2
gas concentration curves produced by a passive side-stream sampling
sensor in accordance with some embodiments. FIG. 29B is a
representative plot illustrating respiratory quotient, and FIG. 29C
is a plot of the corresponding breath volume rate and energy
expenditure estimates associated with the data from the passive
side-stream sampling sensor in accordance with some
embodiments.
[0223] According to some embodiments, the benefits of employing
passive side-stream sampling with a miniature mixing chamber over
breath-by-breath approaches include, but are not limited to, the
following: [0224] utilization of smaller, lighter, less expensive,
and/or lower power gas sensors for greater system efficiency and/or
mobility; [0225] passive proportional gas concentration sampling of
each breath instead of an active pump for greater system
efficiency, mobility, and/or simplicity (especially with respect to
calibration); [0226] utilization of simple, low cost differential
pressure flow sensor and single exhaust valve for greater system
efficiency and/or simplicity; [0227] reduced valve leakage due to
low pressure of the side-stream sampling during inhalation for
greater system efficiency and/or accuracy; [0228] improved
accounting for dead space contributions to expired breath via
breath averaging in the mixing chamber for greater system
efficiency and/or accuracy; and [0229] more reliable energy
estimates based on matching the temporal dynamic of the gas
concentrations to the time constants of the sensors for greater
system efficiency and/or accuracy.
Conclusion
[0230] While various inventive embodiments have been described and
illustrated herein, those of ordinary skill in the art will readily
envision a variety of other means and/or structures for performing
the function and/or obtaining the results and/or one or more of the
advantages described herein, and each of such variations and/or
modifications is deemed to be within the scope of the inventive
embodiments described herein. More generally, those skilled in the
art will readily appreciate that all parameters, dimensions,
materials, and configurations described herein are meant to be
exemplary and that the actual parameters, dimensions, materials,
and/or configurations will depend upon the specific application or
applications for which the inventive teachings is/are used. Those
skilled in the art will recognize, or be able to ascertain using no
more than routine experimentation, many equivalents to the specific
inventive embodiments described herein. It is, therefore, to be
understood that the foregoing embodiments are presented by way of
example only and that, within the scope of the appended claims and
equivalents thereto, inventive embodiments may be practiced
otherwise than as specifically described and claimed. Inventive
embodiments of the present disclosure are directed to each
individual feature, system, article, material, kit, and/or method
described herein. In addition, any combination of two or more such
features, systems, articles, materials, kits, and/or methods, if
such features, systems, articles, materials, kits, and/or methods
are not mutually inconsistent, is included within the inventive
scope of the present disclosure.
[0231] The above-described embodiments can be implemented in any of
numerous ways. For example, embodiments of modeling, monitoring,
and managing metabolism disclosed herein may be implemented using
hardware, software or a combination thereof. When implemented in
software, the software code can be executed on any suitable
processor or collection of processors, whether provided in a single
computer or distributed among multiple computers.
[0232] Further, it should be appreciated that a computer may be
embodied in any of a number of forms, such as a rack-mounted
computer, a desktop computer, a laptop computer, or a tablet
computer. Additionally, a computer may be embedded in a device not
generally regarded as a computer but with suitable processing
capabilities, including a Personal Digital Assistant (PDA), a smart
phone or any other suitable portable or fixed electronic
device.
[0233] Also, a computer may have one or more input and output
devices. These devices can be used, among other things, to present
a user interface. Examples of output devices that can be used to
provide a user interface include printers or display screens for
visual presentation of output and speakers or other sound
generating devices for audible presentation of output.
[0234] Examples of input devices that can be used for a user
interface include keyboards, and pointing devices, such as mice,
touch pads, and digitizing tablets. As another example, a computer
may receive input information through speech recognition or in
other audible format.
[0235] Such computers may be interconnected by one or more networks
in any suitable form, including a local area network or a wide area
network, such as an enterprise network, and intelligent network
(IN) or the Internet. Such networks may be based on any suitable
technology and may operate according to any suitable protocol and
may include wireless networks, wired networks or fiber optic
networks.
[0236] The various methods or processes outlined herein may be
coded as software that is executable on one or more processors that
employ any one of a variety of operating systems or platforms.
Additionally, such software may be written using any of a number of
suitable programming languages and/or programming or scripting
tools, and also may be compiled as executable machine language code
or intermediate code that is executed on a framework or virtual
machine.
[0237] Also, various inventive concepts may be embodied as one or
more methods, of which an example has been provided. The acts
performed as part of the method may be ordered in any suitable way.
Accordingly, embodiments may be constructed in which acts are
performed in an order different than illustrated, which may include
performing some acts simultaneously, even though shown as
sequential acts in illustrative embodiments.
[0238] All publications, patent applications, patents, and other
references mentioned herein are incorporated by reference in their
entirety, including, but not limited to, the following:
[0239] 1. Acheson et al., Am. J. Clin. Nutrition, 48:240-47
(1988);
[0240] 2. Bouchard et al., Obesity Res., 2 (5):400-10 (1994);
[0241] 3. Cruickshank et al., J. Physiol., 80 (2):179-92
(1933);
[0242] 4. Ganz et al., Diabetology & Metabolic Syndrome, 6
(50): 1-8 (2014);
[0243] 5. Guh et al., BMC Public Health, 9 (88):1-20 (2009);
[0244] 6. Hargrove, J. Nutrition, 136 (12):2957-61 (2006);
[0245] 7. Jebb et al., Am. J. Clin. Nutrition, 58:455-62
(1993);
[0246] 8. Ladenheim, Drug Design, Dev. & Therapy, 1867-75
(2015);
[0247] 9. Ludwig et al., JAMA, 311 (21):2167-68 (2014);
[0248] 10. McDevitt et al., Am. J. Clin. Nutrition, 74 (6):737-46
(2001);
[0249] 11. Mozaffarian et al., NEJM, 364 (25):2392-2404 (2011);
and
[0250] 12. Weiss et al., N. Eng. J. Med., 350 (23):2362-74
(2004).
[0251] All definitions, as defined and used herein, should be
understood to control over dictionary definitions, definitions in
documents incorporated by reference, and/or ordinary meanings of
the defined terms.
[0252] The indefinite articles "a" and "an," as used herein in the
specification and in the claims, unless clearly indicated to the
contrary, should be understood to mean "at least one."
[0253] The phrase "and/or," as used herein in the specification and
in the claims, should be understood to mean "either or both" of the
elements so conjoined, i.e., elements that are conjunctively
present in some cases and disjunctively present in other cases.
Multiple elements listed with "and/or" should be construed in the
same fashion, i.e., "one or more" of the elements so conjoined.
Other elements may optionally be present other than the elements
specifically identified by the "and/or" clause, whether related or
unrelated to those elements specifically identified. Thus, as a
non-limiting example, a reference to "A and/or B", when used in
conjunction with open-ended language such as "comprising" can
refer, in one embodiment, to A only (optionally including elements
other than B); in another embodiment, to B only (optionally
including elements other than A); in yet another embodiment, to
both A and B (optionally including other elements); etc.
[0254] As used herein in the specification and in the claims, "or"
should be understood to have the same meaning as "and/or" as
defined above. For example, when separating items in a list, "or"
or "and/or" shall be interpreted as being inclusive, i.e., the
inclusion of at least one, but also including more than one, of a
number or list of elements, and, optionally, additional unlisted
items. Only terms clearly indicated to the contrary, such as "only
one of` or "exactly one of," or, when used in the claims,
"consisting of" will refer to the inclusion of exactly one element
of a number or list of elements. In general, the term "or" as used
herein shall only be interpreted as indicating exclusive
alternatives (i.e., "one or the other but not both") when preceded
by terms of exclusivity, such as "either," "one of," "only one of,"
or "exactly one of" "Consisting essentially of," when used in the
claims, shall have its ordinary meaning as used in the field of
patent law.
[0255] As used herein in the specification and in the claims, the
phrase "at least one," in reference to a list of one or more
elements, should be understood to mean at least one element
selected from any one or more of the elements in the list of
elements, but not necessarily including at least one of each and
every element specifically listed within the list of elements and
not excluding any combinations of elements in the list of elements.
This definition also allows that elements may optionally be present
other than the elements specifically identified within the list of
elements to which the phrase "at least one" refers, whether related
or unrelated to those elements specifically identified. Thus, as a
non-limiting example, "at least one of A and B" (or, equivalently,
"at least one of A or B," or, equivalently "at least one of A
and/or B") can refer, in one embodiment, to at least one,
optionally including more than one, A, with no B present (and
optionally including elements other than B); in another embodiment,
to at least one, optionally including more than one, B, with no A
present (and optionally including elements other than A); in yet
another embodiment, to at least one, optionally including more than
one, A, and at least one, optionally including more than one, B
(and optionally including other elements); etc.
[0256] In the claims, as well as in the specification above, all
transitional phrases such as "comprising," "including," "carrying,"
"having," "containing," "involving," "holding," "composed of," and
the like are to be understood to be open-ended, i.e., to mean
including but not limited to. Only the transitional phrases
"consisting of" and "consisting essentially of" shall be closed or
semi-closed transitional phrases, respectively, as set forth in the
United States Patent Office Manual of Patent Examining Procedures,
Section 2111.03.
[0257] As used herein in the specification and in the claims, the
terms "subject," "individual," "recipient," "host," "user," and the
like are used interchangeably to refer to a either a human or an
animal subject.
[0258] As used herein in the specification and in the claims, the
term "metabolism" refers to biochemical processes whereby nutrients
are converted to energy for use by a mammal. Metabolism can be
further defined as including "catabolic" and "anabolic" processes.
A catabolic process refers to biochemical events that occur in a
mammal that result in the breaking apart of molecules. An anabolic
process refers to biochemical events that occur in a mammal that
result in the creation of molecules.
[0259] As used herein in the specification and in the claims, the
term "metabolic disease" may include, but is not limited to, type 1
diabetes, type 2 diabetes, obesity, gout, lipid disorders,
hyperthyroidism, hypothyroidism, dyslipidemia, hypolipidemia,
galactosemia, phenylketonuria, metabolic syndrome, or
phenylketonuria. Diabetes is a metabolic disease that results in
excess sugar in the blood and urine. Type 1 diabetes is
characterized by a destruction of the pancreatic beta cells that
produce insulin. Type 2 diabetes is characterized by insulin
resistance, wherein the insulin produced by the pancreatic beta
cells is not efficiently utilized to convert glucose to other
energy forms.
[0260] As used herein in the specification and in the claims, the
term "sensor" refers to a transducer that detects an aspect of the
environment and provides a corresponding output. The phrases
"all-optical gas sensor" or "all-optical gas sensing" refer to a
sensor(s) that utilizes optics in the detection of specific
wavelengths (e.g., IR spectroscopy) for the detection of the
presence of certain gases. One example of an all optical gas sensor
is the CO.sub.2 nondispersive infrared (NDIR) optical gas
sensor.
[0261] As used herein in the specification and in the claims, the
term "hands-free" refers to an ability of a user to operate or
practice a system, apparatus, or method without, or with only
minimal use, of the user's hand(s).
* * * * *