U.S. patent application number 14/588363 was filed with the patent office on 2015-07-02 for methods and systems for data collection, analysis, formulation and reporting of user-specific feedback.
The applicant listed for this patent is Sensoria Inc.. Invention is credited to Mario ESPOSITO, Maurizio MACAGNO, Mauro M. RIZZI, Davide Giancarlo VIGANO'.
Application Number | 20150182843 14/588363 |
Document ID | / |
Family ID | 53480659 |
Filed Date | 2015-07-02 |
United States Patent
Application |
20150182843 |
Kind Code |
A1 |
ESPOSITO; Mario ; et
al. |
July 2, 2015 |
METHODS AND SYSTEMS FOR DATA COLLECTION, ANALYSIS, FORMULATION AND
REPORTING OF USER-SPECIFIC FEEDBACK
Abstract
Methods and systems provide user-specific feedback and
recommendations to a user based on a user profile, contextual
and/or biometric data, which may be collected using a sensing
system, such as a wearable sensor device(s). A host system
formulates and provides reports to the user relating to the user's
sensed activities or conditions, and may also provide
recommendations relating to performance and goal achievement,
injury prevention, what/if analysis, gait analysis, shoe purchase
suggestions, and the like. Data analysis algorithms may be used to
identify similar patterns across large collections of user profiles
and biometric data and apply data generated across a larger
population to formulate recommendations for specific users.
Inventors: |
ESPOSITO; Mario; (Redmond,
WA) ; MACAGNO; Maurizio; (Redmond, WA) ;
VIGANO'; Davide Giancarlo; (Redmond, WA) ; RIZZI;
Mauro M.; (Redmond, WA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Sensoria Inc. |
Redmond |
WA |
US |
|
|
Family ID: |
53480659 |
Appl. No.: |
14/588363 |
Filed: |
December 31, 2014 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61923161 |
Jan 2, 2014 |
|
|
|
Current U.S.
Class: |
700/91 |
Current CPC
Class: |
G06K 9/00342 20130101;
G09B 23/28 20130101; G09B 19/003 20130101 |
International
Class: |
A63B 71/06 20060101
A63B071/06 |
Claims
1. A method for formulating user-specific feedback, comprising: a.
providing an electronic user interface for populating user
information; b. providing an electronic user interface for
selecting a user-desired objective selected from the group of
physical training objectives, physical rehabilitation objectives;
and medical condition objectives; c. providing a host system
capable of: receiving the user information, the user-desired
physical training objective, and user biometric information sensed
by a user-worn or user-carried sensing system; analyzing the user
information, user-desired physical training or medical condition
improvement objective and user biometric information; and
formulating user-specific feedback for guiding a user to achieve
the user-desired objective; and d. providing a user interface for
communicating the user-specific feedback to the user.
2. The method of claim 1, wherein the user-worn or user-carried
sensing system is selected from the group consisting of: fitness
and activity trackers, heart rate devices, pedometers, medical
devices, and garment-worn sensing systems.
3. The method of claim 1, wherein the user-worn or user-carried
sensing system comprises at least one force or pressure sensor.
4. The method of claim 3, wherein the user-worn or user-carried
sensing system additionally comprises at least one of the following
sensing components: an accelerometer, a gyroscope, an orientation
sensing component, a location sensing component, and a temperature
sensor.
5. The method of claim 1, wherein the host system is additionally
capable of receiving additional information derived from a source
other than the user, and the additional information is analyzed,
with the user information, user-desired physical training or
medical condition improvement objective and user biometric
information to formulate user-specific feedback for guiding the
user to achieve the user-desired objective.
6. The method of claim 1, wherein the host system provides
pre-determined and/or programmable fitness/activity programs for
selection by a user that guide a user through a series of
activities.
7. The method of claim 1, wherein the host system provides a menu
of training programs selectable by the user and formulates the
user-specific feedback taking into account a selected training
program.
8. The method of claim 8, wherein the host system provides a menu
of virtual trainers or virtual caretakers selectable by the user
and formulates the user-specific feedback through a selected
virtual trainer or virtual caretaker.
9. The method of claim 1, additionally comprising providing a user
interface for selection of at least one aspect of user feedback
content, delivery format and delivery frequency.
10. The method of claim 1, wherein the host system additionally
programs at least one of the user feedback content, delivery
format, and delivery frequency on a dynamic basis.
11. A method for formulating user-specific feedback, comprising: a.
providing a user interface for populating user information; b.
providing a user interface for selecting at least one user-desired
objective selected from the group of physical training objectives,
physical rehabilitation objectives; and medical condition
objectives; c. providing a user interface for selecting a virtual
coach or virtual caretaker for delivery of feedback to the user; d.
providing a host system capable of: receiving the user information,
the at least one user-desired objective, and user biometric
information sensed by a user-worn or user-carried sensing system;
analyzing the user information, the at least one user-desired
objective and user biometric information; and formulating
user-specific feedback for guiding a user to achieve the
user-desired objective; and e. providing a user interface for
communicating the user-specific feedback to the user through the
virtual coach or virtual caretaker.
12. The method of claim 11, wherein the virtual coach or virtual
caretaker has name, voice and feedback style.
13. A method for formulating user-specific feedback, comprising: a.
providing an electronic user interface for creating a user account
and populating user information in the user account; b. providing
an electronic user interface for selecting user-desired
condition(s) or activity reporting parameters and user condition or
activity notification boundaries in the user account; c. providing
a host system capable of: receiving the user information, receiving
the user-desired condition or activity reporting parameters and the
user condition or activity notification boundaries; receiving user
data sensed by a user-worn or user-carried sensing system;
analyzing the user information, the user-desired condition or
activity reporting parameters and user condition or activity
notification boundaries; and formulating user-specific feedback
comprising reports for delivery to the user relating to sensed
conditions and activities and notifications when sensed conditions
or activities are outside notification boundaries; and d. providing
a user feedback interface for communicating the user-specific
feedback to the user.
14. The method of claim 13, wherein the user information includes
user authorization for providing access to the user information and
information sensed by a user-worn or user-carried sensing system by
an authorized third party.
15. The method of claim 13, wherein the third party is a
user-selected coach or caretaker.
16. The method of claim 13, additionally comprising providing a
third party electronic interface with the user account in the host
system and providing access to user information, information sensed
by a user-worn or user-carried sensing system, and user-desired
condition(s) or activity reporting parameters and boundaries to the
authorized third party.
17. The method of claim 16, wherein a third party electronic
interface additionally allows the authorized third party to
contribute to and communicate user-specific feedback to the user
through the user feedback interface.
18. The method of claim 13, additionally comprising formulating
user-specific feedback for providing user-specific recommendations
for changing a condition or activity when sensed conditions or
activities are outside notification boundaries.
19. A method for monitoring footwear usage, comprising: providing a
user interface for entry of user information relating to footwear
worn at specific times and/or during specific activities;
collecting data from a user-worn or user-carried sensing device and
calculating a cumulative distance travelled while specific footwear
is worn; and reporting the cumulative distance travelled while
specific footwear is worn to the user.
20. The method of claim 19, additionally comprising making a
recommendation to the user for replacing footwear when the
cumulative distance travelled while specific footwear is worn
exceeds a predetermined distance.
21. The method of claim 20, additionally comprising making a
recommendation to the user for replacing footwear with a purchase
of specific footwear based on comparison of user information, user
activity information and data collected from a larger population
that share attributes with the user.
22. The method of claim 21, additionally comprising providing a
user interface for purchasing new footwear.
Description
REFERENCE TO EARLIER FILED PROVISIONAL PATENT APPLICATION
[0001] This application claims priority from U.S. Provisional
Patent Application No. 61/923,161, filed Jan. 2, 2014. The
disclosure of the previous application is incorporated by reference
herein in its entirety.
FIELD
[0002] This disclosure relates to data collected from one or more
sensors, including wearable sensors, methods and systems for using
sensor-collected data in combination with data from other sources,
analysis of the data, and formulation of user-specific feedback,
such as recommendations based on the user's profile, goals, needs,
contextual and/or biometric data. In some aspects, the present
disclosure relates to methods and systems for comparing the user's
data set to data collected across large collections of users,
and/or to contextual and/or biometric data categorized as similar,
and providing user-specific feedback relating to the user's
relationship to collections of similar users and/or to similar sets
of contextual and/or biometric data. In some embodiments, the
systems and methods allow a user to select feedback and reporting
parameters and to select performance and achievement metrics
forming the basis for providing user-specific recommendations
related to health, wellness, fitness, or for other purposes.
BACKGROUND
[0003] Various types of sensing systems for monitoring various
physiological parameters have been incorporated in bands,
wrist-worn devices, portable electronic devices, medical devices,
shoes, insoles, socks and other types of garments for various
applications, including recreational, fitness, sporting, military,
diagnostic and medical and health and wellness applications. The
use of sensing systems for fitness applications to monitor and
analyze activities such as running, walking, energy expenditure,
and the like, is now common. Medical applications for sensing
pressure, posture, gait, temperature and the like for purposes of
monitoring neuropathic and other degenerative conditions with the
goal of alerting an individual and/or medical service providers to
sensed parameters that may indicate the worsening of a condition,
lack of healing, and the like, have been proposed. Footwear-related
sensing systems directed to providing sensory data for patients
suffering from neuropathy, for gait analysis, rehabilitation
assessment, shoe research, design and fitting, orthotic design and
fitting, and the like, have been proposed.
[0004] Sensing devices and footwear having sensors incorporated for
monitoring pressure and other body parameters are well known.
Health and fitness data collection devices are currently available
for collecting a variety of biometric (e.g., physiological) data.
Heart rate monitors, pedometers, fitness tracking devices, location
tracking devices, blood glucose monitors, heart rhythm (e.g, ECG)
monitors, etc., are in common usage. Consumers of these types of
health and fitness data collection devices typically collect and
store multiple disconnected data sets, such as data sets relating
to activity type, activity time, distance travelled, average or
instantaneous speed, steps climbed, temperature, heart rate,
calories burned, perspiration, blood glucose levels, etc. In the
usual scenario, some or all of the data may be represented
graphically in a dashboard format on the device or on an associated
remote display device that reports the data to the user but
provides little or no user-specific analysis or feedback. Such
devices generally do not provide user-specific guidance or make
recommendations as to how to achieve a user's goals, how to
exercise or train more effectively, or the like. Methods and
systems described herein are directed to integrating data
contributed by a user with data collected by one or more sensor
systems and providing useful, user-specific feedback.
SUMMARY
[0005] The present disclosure relates to methods and systems for
collecting data from one or more sources or sensing systems and
interfacing with a host system having analytical capabilities and
tools to provide useful data analysis and, optionally, feedback to
a user or to a coach or care provider or other desired (or
permitted) person or group. In some embodiments, the methods and
systems may provide integrated sensing, data collection, data
analysis and data reporting functions. In some embodiments, a user
registers with a host system and provides user profile information
and data to the host system through a user interface, typically
provided on a user's electronic device. A user may also provide
authorization through a user interface for data collected by one or
more device, such as a wearable or portable sensor(s)--e.g., a
fitness band, wrist-worn sensing system, heart rate sensor,
wearable sensor such as a wearable pressure sensor, or the like--to
be communicated to the host system. The host system may collect
(and synchronize, if necessary) data from additional sensors or
data sources, analyze the collected data sets, produce results and
conclusions, and report to the user or a third party designated by
the user (e.g., a coach or care provider) through a user interface
or a third party interface.
[0006] In some embodiments, the sensing system itself may have data
processing, recording and/or data transfer capabilities. Thus, in
some embodiments, a sensing system such as a wrist-worn band or
watch-like device, an anklet, or another wearable or mobile device,
may have data processing, recording and/or transfer capabilities.
Such sensing systems may additionally have display capabilities, as
well audio, visual and/or tactile messaging capabilities. In some
embodiments, an auxiliary device, such as a dedicated electronic
device (DED), may be provided in communication with a sensing
system, and the DED may provide data processing, recording and/or
transfer capabilities and may additionally have display
capabilities, as well audio, visual and/or tactile messaging
capabilities. In some embodiments, a sensing system and/or DED may
communicate with and transfer data to one or more external
computing and/or display system(s), such as a smartphone, computer,
tablet computer, dedicated computing device, medical records system
or the like, using wired and/or wireless data communication means
and protocols. A sensing system, DED and/or an external computing
and/or display system may, in turn, communicate with a centralized
host computing system (located, e.g., in the Cloud), where further
data processing, analysis and formulation of user-specific feedback
takes place. Both substantially real-time feedback and delayed
feedback, including data displays, notifications, alerts, reports,
comparisons, recommendations, and the like, may be provided to the
user, caretaker and/or clinician according to user, caretaker
and/or clinician preferences.
[0007] In one aspect, sensing systems (including one or more
auxiliary or DED devices, and/or an external computing and/or
display device) collect and analyze biometric data from body sites
and provide intermittent or continuous monitoring and reporting of
conditions and/or activity parameters and, in addition, combine the
collected biometric data with additional data for purposes of
analyzing and reporting activity parameters and providing feedback
to the user, or to a coach in fitness applications, or to a
caretaker or medical professional in other applications for
purposes of reducing the incidence and/or severity of injuries,
improving a user's performance (e.g., gait), providing information
to caretakers, improving compliance with a prescribed regimen, and
accelerating the pace and quality of wound healing. In yet other
aspects, sensors, interfaces, systems and materials described
herein for collection and analysis of biometric and/or
biomechanical data may be used for a variety of sports-related,
military, fitness, medical, diagnostic and therapeutic
purposes.
[0008] Exemplary sensing systems incorporating wearable force or
pressure sensors and DEDs, as well as data handling and processing
routines and user interfaces, are described in U.S. Patent
Publication US2013/0192071 A1 (WO 2013/116242) and PCT
International Patent Application No. PCT/US14/049263, the
disclosures of which are incorporated herein in their entireties.
Other types of sensing systems, including many types of fitness and
activity trackers, heart rate devices, pedometers, various types of
medical condition sensors, and the like, may contribute data for
processing in methods and systems described herein. Many exemplary
systems will be provided with reference to fitness applications; it
will be appreciated that the disclosed methods and systems are
suited for use in many different environments and in a variety of
applications.
[0009] In some embodiments, sensing systems incorporating various
types of sensors, leads, traces and terminals may be mounted to
and/or incorporated in or associated with garments, such as socks,
shirts, underwear, leggings, footies, gloves, caps, sleeves, body
bands and brassieres, and other substrates, such insoles, shoes,
boots, belts, straps, bandages, wraps, wrapping bands, wound
dressings, sheets, pads, cushions, sporting equipment, and the
like. In some embodiments, a sensing system is implemented in a
wearable garment, such as a sock, a shirt, a glove, or the like and
the sensing system includes one or more force and/or pressure
sensors. In some embodiments, a DED may incorporate an
accelerometer, a gyroscope, an orientation sensing component, a
location sensing component, a temperature sensor, display
capability, visual, audio and/or tactile indicating capabilities,
and the like. Methods and systems for data collection, analysis and
formulation of user-specific feedback are particularly well suited
for use with sensing systems incorporating force or pressure
sensors in combination with a DED incorporating one or more of the
following: an accelerometer, a gyroscope, an orientation sensing
component, a location sensing component, or a temperature
sensor.
[0010] In one embodiment, an authentication routine and/or user
identification system matches a DED and one or more associated
sensing system(s) (e.g., a collection of sensor(s) associated with
an underlying substrate) with a user, caretaker and/or clinician,
and may link user information or data from other sources to a
software- and/or firmware-implemented system residing on the DED or
on one or more external computing system(s). In some embodiments,
user information and data is linked to a wearable device or to a
mobile device such as a smart phone, and the wearable device or the
mobile device, or both, communicate with a centralized host
computing system or facility where data is stored, processed and
analyzed, and where output, feedback, communications, instructions,
information, and the like may be formulated for delivery back to
the user, caretaker and/or clinician.
[0011] Calibration routines may be provided to ensure that the
sensing system(s) and any intermediate devices are properly
configured to work optimally for each specific user and with the
data processing and feedback system selected. Configuration and
setup routines may be provided to guide the user (or authorized
third party) to input user information or data to facilitate data
collection, and various protocols, routines, data analysis and/or
display characteristics, and the like, may be selected by the user
(or authorized third party) to provide data collection and analysis
targeted to specific users. Notification and alarm systems may be
provided, and selectively enabled, to provide messages, warnings,
alarms, notifications and the like to the user, and/or to
authorized third parties, substantially in real-time, based on
sensed data. Notification and alarm limits and boundaries may be
set by the user, or by an authorized third party, or may be
determined and set by the host system.
[0012] Many specific examples described herein relate to fitness
applications. In some aspects, methods and systems described herein
relate to a platform and host system that allows end users to
select or design a training program and provides end users with
motivation, recommendations and valuable data (real-time and not)
to improve their training experience. The platform comprises a host
computing system that may communicate with multiple components and
services that range from mobile devices (smartphones and tablets)
to services on the cloud and web sites that leverage the user's
data in combination with biometric and/or contextual data, using
multilevel data processing and, optionally, artificial intelligence
processing methods, to create ad-hoc customized feedback to
specific users with the overall goal of physical condition and/or
performance improvement and user engagement. The platform may also
provide pre-determined and/or programmable fitness/activity
programs that guide users through a series of activities. For
example, a user may configure the platform to allow users to set
their own training regimen and the platform formulates feedback to
coach the user in accordance with that regimen. Alternatively, the
platform may additionally or alternatively provide a menu of
training programs or regimen selectable by users, and the platform
will then formulate feedback to coach the user in accordance with
the selected program and may recommend the use of additional or
different programs as the user progresses. In some embodiments, the
platform may additionally or alternatively provide a menu of
virtual trainers or caretakers selectable by the user and
formulates feedback in accordance with the style or personality of
the selected trainer or caretaker.
[0013] In some aspects, the host system leverages an electronic
delivery system by using smartphones and mobile electronics like
smart watches and accessories (e.g., the wearable internet of
things (IOT)) to communicate in real time to users by delivering
user-specific feedback and notifications much like a personal
trainer or caretaker would do. For example, the host system may
provide feedback for communicating to users, in real time, when
they are within or outside boundaries set with respect to various
parameters in a user's configurations or recommended by selected
programs. In a running training program, for example, the host
system may provide user-specific feedback based on the user's
profile, real time biometric data collected, and a selected
training program or reporting parameters and boundaries relating to
the user's gait when running (e.g., ball/heel strike of the foot),
the pace and cadence, the heart rate training zone, etc., for the
purpose of motivating and coaching the user during the workout and
providing recommendations for correcting the user's form when
necessary. The system may use different modes to deliver feedback
and notifications to end users, from audio cues and audio (e.g.,
spoken) recommendations, to text formed messages and automated
phone calls. The delivery mode, timing and frequency of reporting
and feedback formulated by the host system is preferably dynamic
over the course of the user's training program, rather than
static.
[0014] In some aspects, the system provides user selection from
different levels and styles of feedback (e.g., coaching), as well
as user-specific feedback. Thus, the system may provide basic
activity notifications (e.g., simple status information and
biometric data reporting relating to a current activity) to
real-time coaching tips and full training programs tailored for the
end user, as well as reports relating to the user's workout form,
compliance with the training program(s), progression, and the like.
User feedback may be provided in substantially real time, or
delayed, and may be delivered to the user in accordance with user
preferences and smart algorithms that predict the most effective
notification timing, frequency and delivery mode. In other aspects,
the system may be customized to provide real-time (and/or delayed)
observation and participation in the user's activities by
authorized third parties such as coaches, caregivers, and the like,
remotely via a mobile application or web site/service or the like.
In other aspects, the host system may assist the end user to
initiate operations on his/her behalf, such as communications with
others.
BRIEF DESCRIPTION OF THE DRAWINGS
[0015] FIG. 1 shows a highly schematic diagram illustrating a host
system interfacing with a plurality of data modules to provide
user-specific feedback.
[0016] FIG. 2 shows a schematic diagram illustrating an exemplary
automated training data module.
[0017] FIG. 3 shows a schematic diagram illustrating an exemplary
location awareness data module.
[0018] FIG. 4 shows a schematic diagram illustrating an exemplary
external integration data module.
[0019] FIG. 5 shows a schematic diagram illustrating an exemplary
environmental awareness data module.
[0020] FIG. 6 shows a schematic diagram illustrating an exemplary
notifications data module.
[0021] FIG. 7 shows a schematic diagram illustrating an exemplary
notifications delivery data module.
[0022] FIG. 8 shows a schematic diagram illustrating an exemplar
predictions data module.
[0023] FIG. 9 shows a schematic diagram illustrating an exemplary
self-learning data module.
[0024] FIG. 10 shows a schematic diagram illustrating an exemplary
emotional feedback data module.
[0025] FIG. 11 shows a schematic diagram illustrating an exemplary
motivational feedback data module.
[0026] FIG. 12A shows an exemplary virtual coach enable user
interface and an exemplary virtual coach selection.
[0027] FIG. 12B shows an exemplary user interface for a virtual
coaching program illustrating exemplary user selection choices for
running form feedback.
[0028] FIG. 12C shows a portion of another exemplary user interface
illustrating exemplary user selection choices for performance
feedback.
[0029] FIG. 12D shows yet another exemplary user interface
illustrating exemplary user selection choices for selectable
feedback options.
[0030] FIG. 13 shows a schematic diagram illustrating exemplary
user subpopulation clusters.
[0031] It will be understood that the appended drawings are not
necessarily to scale, and that they illustrate specific embodiments
of some aspects of systems and components described herein. The
present invention is not limited to the aspects illustrated in the
drawings or disclosed specifically herein.
DETAILED DESCRIPTION
[0032] In many embodiments, systems and methods described herein
leverage data collected from three main data sources. A first
category of data is user profile data, which may be provided
directly or indirectly by a user, and may include information about
the user, for example: gender, age, weight and height, foot size,
foot conformation, shoes owned and worn during monitored
activities, activity preferences and goals, reporting and data
presentation and delivery preferences, coaching preferences, etc. A
second category of data is biometric data, which refers to data
that may be collected by one or more sensing systems or data
collection device(s) connected with one or more sensors or sensing
systems that detects user biometric data in real-time including,
for example, heartbeat, pressure points, pressure exerted at
identified body locations, temperature, moisture, blood gas
compositions, body motion, position and location, body part (arm,
leg, foot, shoulder, torso, etc.) biomechanical motion and
position, etc. A third category of data is contextual data, which
refers to data that may be collected externally of the user and
communicated to and stored by a hardware component such as a
smartphone, a computer, a dedicated data collection device or host
system and/or software services. Contextual data may relate to
information such as GPS location, outside temperature, weather
conditions, terrain, user preferences, external user accounts
(e.g., calendar, social networks, etc.) Each category of data may
be communicated to and stored by a host system via any suitable
data collection and/or communication device, such as a smartphone,
an electronic sensing device, a computer or the like using any
suitable data communication system.
[0033] The schematic diagram shown in FIG. 1 illustrates a host
system communicating with users (through various means, including
communication with sensing devices, electronic devices,
software-enabled systems, apps, or the like) to collect and store
user input. The host system additionally communicates with sensor
systems and collects and stores sensor input (through various
means, including communication with various types of sensing
systems and/or electronic devices, software-enabled systems, apps,
or the like). User and sensor input is communicated to, received by
and stored in the host system, and user feedback is formulated in
and delivered to the user from the host system.
[0034] The host system additionally incorporates or has access to a
plurality of logic modules which are described in greater detail
below and which function to collect and organize data from multiple
sources, perform data processing, apply intelligence, and provide
content and instructions to the host system for user feedback. The
logic modules, in some embodiments, function as machine expert
components that apply artificial intelligence logic rules that
leverage databases created through dedicated tools.
[0035] An interface providing user registration prompts and user
profile prompts may be provided and displayed to users, and users
may be prompted to enter personal data in a variety of fields. In
one embodiment, the user downloads and opens an app and enters user
information, and user information is stored in a data storage
facility provided in or accessible by a host system. In athletic
training (or other) applications involving running, walking or leg
movement, user profile data may include information such as user
name, gender, date of birth, weight, height, shoe size, stride
length, foot information such as arch type, tendency to pronote or
supinate, foot rotation characteristics (i.e., inward rotation,
slight or more pronounced outward rotation), static and dynamic leg
axes (i.e., bow legs, straight legs, knock knees). Explanatory
diagrams and menus may be provided to assist the user to accurately
characterize his/her foot, leg and movement attributes.
[0036] In other types of athletic training applications involving
other types of movement, e.g., golf, baseball, boxing, gaming,
etc., involving movement of other body areas, user profile
information may include information such as user name, gender, date
of birth, weight, height, equipment specifics and types, arm,
shoulder or torso information and characteristics, and the like. In
medical care applications, user profile data may include
information such as user name, gender, date of birth, weight,
height, medical condition, monitored biometric parameters and
boundaries, medical equipment, medications, medical care
professionals, and the like. It will be appreciated that many
different types of user profile information, user attributes and
equipment types and attributes may be useful and may be collected
in a user profile interface.
[0037] In fitness applications, additional information indicating
specific running shoe models owned and worn by the user may be
provided. In one embodiment, for example, the user profile
interface is pre-populated with different running shoe brands and
the user selects brands owned and sizes, frequency and type of
usage, and the like. The user may have an option to rate attributes
of the specific shoes, including comfort, price, etc. Additional,
more detailed information relating to the user's experience of
specific footwear may be collected, such as the user's subjective
shoe fit opinions (size fit, width fit, arch style or fit, support
and comfort). Collection of this data from users across various
populations allows other aspect of the host system to characterize
specific footwear in accordance with the feedback provided and make
footwear recommendations tailored to specific users based on their
foot type, shoe attributes and shoe experience across a population.
Similar features and interface options may be provided for other
types of fitness training, medical monitoring and health and
wellness applications. Detailed information may be collected, for
example, relating to a user's wearable accessories (gloves,
clothing, head gear, etc.), sports and fitness equipment (golf
clubs, bats, boxing gloves, etc.), and subjective user experience
data may also be collected. Collection of this data from users
across various populations allows the host system to characterize
specific user wearables and equipment and make recommendations for
wearables and equipment tailored to specific users based on their
user profile data.
[0038] Sensor input may include biometric and/or biomechanical data
from a variety of sensors, including wearable sensors (pressure and
force sensors, heart rate sensors, accelerometers, gyroscopes,
medical monitoring sensors for monitoring blood glucose,
temperature, heart rhythm, ECG, EEG, blood pressure, blood oxygen,
sleep quantity and quality, body weight and body mass index, etc.)
used by the user. Accelerometers, gyroscopes and other types of
location, position and motion sensors may additionally provide data
relating to user position, location, motion, and the like.
[0039] Sensor system(s), typically worn or carried by a user to
monitor biometric properties during an activity, or to monitor
biometric (e.g., physiological) properties in everyday life, are
typically registered to and provide data (continuously or
intermittently) to the host system. The sensor system itself,
and/or the host system, may provide data monitoring and feedback to
the user (or to a third party permitted by the user, such as a
coach or caretaker) on a continuous or intermittent basis. In an
exemplary embodiment, a sensor system comprising one or more
sensors located to sense force or pressure on the bottom of a
user's foot may be used in combination with a DED or one or more
external sources that collects locational data, positional data
(e.g., using an accelerometer and/or gyroscope) heart rate data,
etc.
[0040] In some embodiments, one or more of the following data
fields may be sensed and/or determined and reported to the user
through a user interface: heart rate, calories burned, distance,
pace, speed, ascent, descent, altitude, foot landing, foot contact,
number of steps, cadence, vertical speed, heart rate zone, calories
(fat), power, ambient temperature, ambient pressure, body
temperature, blood pressure, blood oxygen level and blood glucose
level. In other types of sensor systems, additional or different
data fields relating to biometric, activity, health and/or wellness
information of other types, of interest to users for other purposes
may be displayed to the user. In some embodiments, one or more of
the data fields listed above (optionally including additional data
fields) may be displayed to the user during (and following) an
activity. In some embodiments, the user may set boundaries and
request notifications when sensed values for selected parameters
exceed the predetermined boundaries. Various metrics may be
calculated and reported to the user, in real time or subsequently,
using a variety of media and reporting formats.
[0041] In one aspect, methods and systems described herein may be
used in association with one or more biometric (e.g.,
physiological) sensors and may function as a "Virtual Coach" or
"Virtual Caregiver." Relevant feedback, such as real time reporting
of biometric and other parameters, alarms and notifications when
predetermined thresholds or boundaries are violated,
recommendations for performing actions or modifying behaviors in
support of user goals and/or wellbeing is formulated by the host
system and communicated to the user and, optionally, to another
permitted party authorized by the user, such as a coach, a
caretaker or another permitted party, through an electronic device
interface. In another aspect, systems and associated methodologies
are disclosed for collecting data from the user using one or a
variety of sources; comparing user data to data collected from a
larger population and, particularly, to data collected from members
of a larger population that share attributes with the user or have
similar attributes or goals; providing user specific
recommendations based on the data collected from the user in
combination with similar data collected from other users having
some commonality to the current user; and performing user-defined
smart tasks or providing user-specific feedback based on such
data.
[0042] In some embodiments, the user selects an activity (e.g.,
running, walking, cycling, skiing, etc.) prior to initiating data
collection. In some embodiments, the system interfaces with sensors
and has the capability to determine the type of activity that the
user initiates (i.e., running, jogging, standing, sitting, walking,
biking, skiing, etc.) without requiring user input. In case of a
runner, the system may combine "how fast", "how far" data with "how
well" they are running Data analysis using systems and methods
described herein may provide the user with activity performance
metrics, coaching input, and/or injury prevention guidance in
real-time as the user engages in activity, or at a subsequent time.
Methods and systems of the present invention may also be used to
monitor and guide injury rehabilitation, as well as to monitor and
guide users who have activity limitations and/or deficits. In an
elderly patient or injury rehabilitation scenario, for example, a
host system providing a virtual caregiver option to a user
integrates data from multiple sensors, generates a gait analysis on
the fly (standing or walking) and compares it with similar data
sets to evaluate the user's gait and posture and predict the chance
of that user falling, or to make recommendations that may improve
the user's gait or posture, or reduce the user's chances of
falling. Notifications and/or alarms may be programmed to alert
either or both the user and a caregiver, for example, of the higher
risk of the patient falling and recommend the use of a walker or
wheel chair. The system may also be able to track and report user
weight changes over periods of time.
[0043] In one aspect, methods and systems described herein provide
a platform for collecting and storing user information and data;
collecting and storing data from user-worn and/or user enabled
sensor systems; collecting and storing data from sources external
to users; analyzing and integrating the data collected from various
sources; formulating feedback (in the form of information, data,
notifications, recommendations, and the like) relevant to the user
based on user-input preferences or instructions and/or based on
information gleaned from larger populations; and delivering the
user-specific feedback to the user, optionally with additional
content. The platform is typically operated by a third party host
operator and typically resides on a host system, such as on a
host-operated computing system located at a facility remote from
the user (e.g., in the Cloud). The host platform may communicate
(directly or indirectly through one or more intermediate devices)
with user-worn and/or user enabled sensor systems for data
collection, and with external sources for collection of external
data using various types of communications protocols; similarly,
the host platform may communicate (directly or indirectly) with the
user through one or more intermediate devices (e.g., an electronic
device having data storage, processing and/or display capabilities
such as a smartphone, computer, tablet, smart watch, or other smart
device).
[0044] The host system incorporates a plurality of discrete logic
modules, each module comprising and/or accessing one or more
database(s) and machine learning algorithms and applying logic
rules, optionally artificial intelligence logic rules, to formulate
user-specific feedback. The exemplary logic modules illustrated in
FIG. 1 including "Self Learning," "Predictions," "Auto Training
Programs," "Motivation," "Environmental Awareness," "Emotions,"
"Notifications," "Location Awareness," "Notification Delivery," and
"External Integrations." Additional or fewer modules may be
incorporated in any host system, depending on the user population,
the type of sensor input(s), the desired user feedback, and the
like. An exemplary structure for each of the individual logic
modules shown in FIG. 1 is illustrated in FIGS. 2-11. It will be
appreciated that these module configurations are exemplary, and
that additional and different information, data, content and logic
may be applied in these or additional modules.
[0045] FIG. 2 shows an exemplary Auto Training Programs module.
This module may provide access to a library of automated and
programmed or programmable user sessions, such as fitness training
sessions, rehabilitation sessions, or the like. In some situations,
multiple programmed or programmable sessions may be presented as
feedback to the user for user selection; in some situations, the
host system may select a programmed or programmable session for
presentation to a specific user based on user profile data or other
data relating to the user. In another aspect, an auto training
program module may provide additional information, such as
recommended time of day for a specific user to perform an activity,
recommended activities, recommended locations, and the like.
[0046] FIG. 3 shows an exemplary Location Awareness module. A host
system environmental awareness module may track a user's location
and environment using locational data (such as GPS data) collected
from a user worn or carried sensing or electronic device that
interfaces with the host system. A host system location awareness
module may include geofencing capabilities (e.g., the capability to
identify a geographical perimeter based on its latitude and
longitude location) that allow a user or a device operating within
the system to set boundaries and provide alerts or notifications to
a user when he or she enters and/or exits a geofenced boundary.
Geofenced boundaries may additionally trigger the application and
presentation of Smart Tasks (described in more detail below).
[0047] FIG. 4 shows an exemplary integrations module, which may
interface with external sources of data as well as external user
accounts, networks, or the like. In the exemplary integration
module shown, data from other sensing devices, such as a JAWBONE
device, a FITBIT device (or another activity monitoring device), a
sleep monitoring device, or other sources of sensed information are
communicated to the module. An integration module may also include
integrations with one or more of the following: an electronic user
calendar (e.g., for scheduling or proposing workout or activity
schedules that are compatible with the user's calendar and free
time), an electronic user music account or media account (e.g.,
NETFLIX, AMAZON, Pandora, iTunes, etc., for automatically
retrieving music or entertainment compatible with the user's
workout or activity, or to recommend entertainment or products
compatible with the user's lifestyle and life patterns), a user's
social or professional networks (e.g., FACEBOOK, TWITTER, LINKED
IN, etc.). This type of integration module may allow the host
system to make recommendations based on a variety of information
derived externally of the sensing system. Thus, reference to a
user's calendar through an integrations module may allow the host
system to make recommendations to the user relating to scheduling
activities (e.g., runs, walks, events, etc.), may allow the
platform to select or make recommendations to the user relating to
entertainment while engaging in activities (e.g., music, movie and
other media programming, etc.), and to discern the user's mood and
make recommendations based on the content of external user accounts
and connections.
[0048] FIG. 5 shows an exemplary environmental awareness module.
Environmental awareness may additionally involve data relating to
current and/or future predicted weather conditions, current terrain
types or terrain types (e.g., paved or non-paved road or path,
topography, etc.) experienced along a user's projected activity
path, other environmental factors such as traffic lights, traffic
patterns, traffic conditions, and the like that a user may
experience along a projected activity path. Providing awareness of
and recommendations relating to terrain type may involve alerting a
user to terrain along a specified route, recommending routes or
alternative routes based on terrain type and user limitations, and
the like.
[0049] Environmental awareness may additionally involve data
relating to current and/or future predicted proximity to friends,
sights, business, or the like, along a current or projected
activity path of a user. In some embodiments, an environmental
awareness module may provide one or more of the following features:
Cross road/Traffic Light Detection--e.g., a way to warn users of
crossroads, traffic lights and other environmental features that
may pose a danger to the user and perform a smart-task to stop or
pause or lower the volume of music (if playing) to ensure user is
aware of potential dangers; automatically create a music selection
based on the local area; inform users of friend and workout
partners that are in the current area; and inform users about
entering and exiting a specific area (e.g., entering a playground
area and warn the user to be careful of children at play).
[0050] FIG. 6 illustrates an exemplary Notification module
structure. Using this module, the notification type--e.g., audio
cues (non-verbal), audio messages (verbal), push notifications,
texts and automated phone calls may be selected (by the user or the
host system) as the delivery mode for various notifications.
Notifications of different types and having different content may
be delivered at different times using different delivery modes. The
timing, frequency and priority of notifications is important, and
may be handled in a delivery module or in a Notification Delivery
Logic module, as described below. Real time and/or offline
(non-real time) notifications may be provided, as selected by the
user or as determined by the platform. Notifications of many types
may be provided, including (without limitation) notifications
regarding compliance to a selected plan or regimen,
activity-related parameters and reports; technique-related
recommendations for improved performance, injury prevention, etc.,
positive reinforcement for encouraging activity, etc. Off-line
notifications may include activity reports provided to the user in
a programmed or programmable format.
[0051] Different types of audio cues (non-vocal sounds, e.g.,
bing/bang sounds) may be used to notify users when they are doing
something correctly or incorrectly. The audio cues may be different
when the user is performing an activity correctly (bing) or when a
sensed parameter is within a boundary than when a sensed parameter
is outside a predetermined boundary (bang). This type of
notification aids the user during a monitoring period. Audio
messages, such as spoken cues (generated, e.g., from a
text-to-speech engine), may be used to notify users of the status
of their activity or to report real time sensed values (e.g.,
current speed, pace, distance, etc.), or to apply changes to the
activity to comply with the parameters (e.g., to increase or
decrease a Heart Rate Zone). Audio messages may be provided to
motivate the user to improve sensed values and overall
performance.
[0052] Device push notifications, which generally comprise messages
that are pushed by the host system to the user's device, may appear
as text messages on a device. How such messages appear may be also
dependent on the type of platform (Operating System) that the
device is running E-mail messages may also originate at the host
system and be communicated to the user's email inbox. These
messages may include consolidated reports including varying detail
relating to the user's past activities, the system's analysis of
the user's activities and performance, suggestions and
recommendations for future activities, improvements, and the like.
In another scenario, automated phone calls may originate with the
host system and be automatically placed to users informing them of
particular events, activity recommendations, warnings or alerts,
such as change of weather or the like.
[0053] FIG. 7 illustrates an exemplary Notification Delivery Logic
module structure, including notification timing, frequency and
priority, as well as delivery mode or mechanism. These elements may
be selected by the user and/or determined by the platform. The
format, content and timing of notification deliveries may vary so
that the notifications remain fresh and of interest to the user.
This module may also provide priority routing of conflicting
messages or resolve conflicting message timing issues.
[0054] FIG. 8 illustrates an exemplary Prediction module structure.
Based on historical data collected and stored corresponding to each
user, this module of the host system may make predictions and
recommendations to users such as the best day and weather
conditions for particular types of activities, best time of day for
a workout, best shoes to wear for a particular workout, route or
terrain type, best "pal" for workout sharing and data comparison,
etc.
[0055] FIG. 9 illustrates an exemplary Self learning module that
includes both historical user data leveraging and social learning
components. The social learning component may involve grouping, or
clustering of participating users having similar attributes
(described below). Based on grouping users by attributes, "Foot
Pals" having similar activity profiles and other selected
attributes may be identified for each user, and recommendations
relating to activity tips, training tips, shoe recommendations, and
the like. Historical user data may be analyzed by the host system
module to provide user recommendations relating, for example, to
workout patterns and regimen, activity types, locations, frequency,
solo vs. group participation, and suggestions concerning footwear,
terrain, equipment, speed, frequency, distance, and the like.
[0056] A host system "emotion" module, as schematically illustrated
in FIG. 10, may direct additional feedback content to a user, such
as jokes and facts, based on the user's location, activity, mood
detection, or the like. Additional content such as music,
playlists, interesting facts or stories, or the like, may be
selected for and/or recommended to a user based on mood detection,
user activity and performance metrics, geographical location, and
the like. A host system "motivation" module may additionally be
provided, as schematically illustrated in FIG. 11, to formulate and
track user performance-based awards, points, etc., and to prompt
the user to participate in activities.
[0057] Methods and systems for formulating and providing
user-specific feedback based on user profile information and sensed
data may include one or more of the modules described above, and
may include additional functionality. In some embodiments, host
systems having at least two, or at least three, or at least four,
or at least five, or at least six, or at least seven, or at least
eight of the functional modules described above and shown in FIGS.
2-11 are provided. Formulation of user-specific feedback may be
based on any combination of user profile input, user activities,
user goals, real time sensed (biometric) data, real time contextual
data, coordinated analysis of multiple real time sensed and
contextual data sets, and coordinated analysis of databases
containing historical sensed and contextual, as well as user data.
Various settings for controlling and adjusting the type, content,
frequency and delivery mode for feedback may be provided and may be
accessible to a user directly on a sensing device and/or on an
intermediate mobile electronic device. In many embodiments, various
control settings may be provided in a user's mobile application. In
some embodiments, various control settings may additionally or
alternatively be communicated from the host system to the user's
sensing device or mobile electronic device.
[0058] Methods and systems described herein may be used to provide
user-specific feedback delivered through a "Virtual Coach" or
"Virtual Caregiver" format, as mentioned above. Different levels of
coaching and caretaking may be delivered to individual users,
ranging from basic activity notification (e.g., simple status of
the current activity and simple metrics relating to current
activity) to real time coaching or caretaking tips and full
training or care programs tailored specifically for the end user.
The Virtual Coach and Virtual Caretaker features may provide users
with a series of real time communications relating to the user's
current condition, workout parameters and form, compliance with a
rehabilitation or training regimen, progress toward a goal, or the
like. These features may be used to communicate to a user when
he/she is within or outside a boundary set in these configurations
for the purpose of engaging and motivating the user during activity
and providing an opportunity for the user to modify his/her
behavior. These settings may be defined in a configuration setting
in a mobile device or a mobile application that monitors user's
biometric data.
[0059] A Virtual Coach or Virtual Caretaker feature provided by the
host system may allow the user to select from among a variety of
pre-set fitness/activity/health and wellness programs that guide
users through a series or activities. In some embodiments, the host
system may present a user with a selection of specific individuals
or avatars representing virtual coaches or caretakers, each having
a unique name, voice and feedback style. Users may select a
specific virtual coach or caretaker, and feedback is then delivered
through the selected virtual coach or caretaker. FIG. 12A
illustrates a sample user interface showing a virtual coach feature
enabled and one exemplary virtual coach, "Mara." In some
embodiments, the host system may select or assign a specific
virtual coach or virtual caretaker to a specific user based on user
profile information and user data. In some embodiments, the user
may select a specific virtual coach or caretaker for delivery of
feedback, and the user may also select, or program, many aspects of
the feedback content, delivery format, and delivery frequency.
Sample activity feedback selections, content and user interfaces
are illustrated in FIGS. 12B, 12C and 12D.
[0060] The host system as described herein may also be linked to
and used by professional coaches and medical care providers, such
as physicians, nurses, physical therapists and other care providers
to facilitate remote monitoring of the activities and/or
condition(s) of users and to communicate feedback and
recommendations to users and/or to the third party provider(s). In
one scenario, for example, a professional athletic coach or
trainer, or a professional caregiver or medical professional, may
be linked to the host system and a user's account on the host
system. In some applications, one or more (authorized) third
parties (e.g., coach or caretaker or training partner or colleague)
may monitor a user's condition or performance in substantially real
time as biometric data is acquired by a sensing system worn or
carried by the user and communicated to the host system. In some
applications, one or more (authorized) third parties may monitor a
user's condition or performance over a period of time to detect
trends and/or gauge current performance levels or conditions and
provide feedback to the user.
[0061] In some applications, an additional interface with the host
system may be provided for third party entry of configurations for
specific users, such as data collection content and frequency,
alert and notification boundaries, and the like. In some
applications, the third party coach or caretaker may provide
content such as user-specific feedback in the format of training
protocols, activity and performance modifications and
recommendations, motivational messages, and the like, to the user
through the host system. Thus, in some embodiments, the host system
serves as an interface and analytical reporting system, collecting
and providing real time and/or historical data relating to a user's
condition, activities and/or performance to one or more
user-permitted third party participants, and relaying
communications between a monitored end user utilizing a sensing
system and a coach or caregiver monitoring the status and/or
activities of the monitored end user.
[0062] In some embodiments, the host system and user interfaces may
be customized for use by a particular user or company, and user
interfaces, reporting and notification formats, and the like may be
customized by or co-branded with a third party for presentation to
a collection of users. In this setting, the host system typically
allows the third party to provide content, protocols, activity and
performance modifications and recommendations, reporting and
notification formats and delivery modes, configurations, settings,
and the like, while the host system may serve as an interface and
reporting system providing feedback to users.
[0063] In some embodiments, the host system may additionally
provide one or more purchase and payment modules. In these
embodiments, user profile data may include payment authorization
data, and the host system may provide interfaces between users and
third party sales outlets. If the host system presents a particular
shoe recommendation to a user, for example, or recommends
particular sports or fitness equipment or accessories, the host
system may present sales outlets offering recommended products to
the user and facilitate a sales transaction. If the host system
makes a recommendation for matching a user with a particular coach
or trainer or caregiver, for example, the host system may present
subscription or payment options for retaining a particular coach or
trainer or caregiver and/or purchasing other types of goods or
services.
[0064] In some embodiments, systems and methods disclosed herein
integrate data collected over a population or sub-population of
users and groups users in clusters according to similarity or
affinity of one or more user data fields or characteristics. In one
embodiment, for example, the host system is configured to use any
number of individual data points, also referred to as attributes or
features, to create groups (clusters) of users that have an
affinity with respect to selected attributes. Users falling within
common clusters may be classified as "pals"--e.g., "foot pals,"
"ski pals," (bike) "riding pals," etc. Health and wellness-related
clusters may also be formulated, with similar groupings of users
having common characteristics, goals, and the like. Examination and
comparison of other data fields corresponding to members of the
cluster allows the system to formulate and provide recommendations
to a user based on comparison of the user's data with the attribute
data of reference individuals sharing similar attributes, and/or a
reference group determined by classification of user data available
in the system.
[0065] For example, a first classification might cluster a user
population or sub-population by gender. A next level, second
classification might add age, then additional attributes, including
height, weight, feet characteristics (e.g., size, pronation, etc.),
and so on. The final result may be an N-dimensional space where
each subject is classified in relation to other members of the
group or sub-group. Groups of affine subjects may be determined by
pivoting around fewer data attributes (M dimensions). The group
membership may be determined by calculating the "distance" of
subjects from the center of the cluster and may be visualized by
positioning individuals at the relevant distance from the center of
the cluster in relation to other individuals in the group.
Exemplary user subpopulation clusters are schematically illustrated
in FIG. 13.
[0066] In one specific example, let C be a data point representing
the center of a cluster where each dimension is the numerical
expression of the individual features, for example, men of age 35,
weight 180 lbs., height 6', foot size 10.5, over-pronating. Let X
be a data point corresponding to a specific subject--for example,
John, age 36, weight 160 lbs., height 6'2'', foot size 10.5,
over-pronating. If the distance D=Distance(X, C) is below a set
threshold, the subject will be considered part of the specific
cluster. Note that any suitable algorithm may be used to create
clusters, as long as it's computationally sustainable. This usually
translates in finding an efficient, yet accurate, distance
function.
[0067] In one embodiment, a variation of the "k-means clustering"
algorithm provides a suitable way to build the desired clusters.
The standard k-means algorithm assigns a set of k random additional
vectors (or points) to define cluster "centroids." For each
iteration, the following computation is carried out: each data
point is assigned to the cluster centroid closest to it; the
centroid is moved to the average position of all the data points
that belong to it; and, if any of the centroids moved in the last
step, perform another iteration, otherwise exit. This algorithm
allows identification of suitable clusters for various populations
and sub-populations.
[0068] In another embodiment, if we want to create clusters "on the
spot," based on a subset of features, or pivoting around specific
centers, k specific centers can be provided, and clusters may be
computed in a single iteration. Clustering of individuals in groups
and sub-groups may also be performed using different classification
algorithms and techniques, including hierarchical clustering,
agglomerative nesting, and k-NN (nearest neighbor) algorithms. Each
new user can be classified and grouped in any of the existing
clusters. The classification can be used immediately to provide
recommendations that pertain to the relative distance of subjects
in the group.
[0069] The type of cluster classification and the recommendations
available using the type of cluster classification previously
described is somewhat static, as the recommendation will always be
the same provided that the user stays in the same cluster. A more
dynamic set of recommendations can be provided to the user when
data based on their physical activity, or sensed data collected
during physical activity, is included in the analysis. In one
embodiment, for example, using data acquired from the foot pressure
from multiple users that share commonalities (i.e. foot pals), the
system can build recommendations that can be pushed to the runner
during the physical activity. The runner may be prompted, for
example, to modify their behavior throughout an activity (for
example, they may be prompted to start running slower or more
quickly, or to change their stride, cadence or the striking point.
In this case, the host system may observe changes in biometric user
data patterns and recommend changes in user behavior to improve the
user's performance, to prevent an injury, and/or keep the runner on
target to attain his/her goal(s).
[0070] As the user complies with a recommendation, the virtual
coach can run multiple simulations of type "what if" and "expected
outcome is". The best one that applies to the running subject can
be used by the system as the next recommendation that is going to
be delivered to them. When the user ignores a recommendation, a
motivational message can be used to win the confidence and build up
the compliance with the recommendation system.
[0071] To illustrate, let's consider the following scenario:
Subject A has the following user attributes; Weight: 190 lbs.,
Height: 5'7'', Build: Medium, Arch Type: Medium, Foot Width:
Normal, Current goal: run for 5 miles at 7.5 mph, no stops, Status:
active and running regularly, Goal status: Achieved and sustaining
Subject B has the following attributes: Weight: 195 lbs., Height:
5'6'', Build: slender, Arch Type: Medium, Foot Width: Normal,
Current goal: run for 4 miles at 4.5 mph, with 2 stops, Status:
working on it, very little success so far, Goal status: not
achieving. The classification compares Subject B's profile with
hundreds of others that are similar, but have already achieved the
goal or are currently seeing better results. When a matching
population (cluster) has been assembled, the system compares other
relationships between the best candidates and subject B's features,
using other attributes like body build, dedication, time in
achieving results and so on, as well as run patterns, based on the
biometric data. Subject A is a good candidate to use as a prototype
for the recommendation because Subject A has an affine profile
compared to subject B (same cluster), and Subject A has a better
achievement result.
[0072] A set of multiple hypotheses can be built and ranked; the
best applicable recommendation is provided to subject B (for
example, to follow similar patterns observed in subject A, such as
number of steps per minute, stride length, distance covered, etc.).
As the subject complies with the recommendations, the system can
compare the likelihood of positive impact against the goal, then
compares against the past of other subjects (e.g. subject A) in the
same pool, and create a baseline for subject B, to continue
evaluate the expected outcome vs. the actual outcome. A positive
observed outcome following a specific recommendation(s) is noted
and positively affects the system to make the same recommendation
for other subjects with similar characteristics. A negative
observed outcome following a specific recommendation is also noted
and negatively affects the system and reduces the likelihood of
giving that specific recommendation in future.
[0073] User-specific feedback and recommendations can be provided
in real-time (i.e. when the user is performing an activity) so that
the user can benefit immediately from the recommendations.
Appropriate means of real-time communication for the
recommendations may include: audible cues and/or messages
delivered, for example, via specific user interface(s) on
smart-phone/computer applications; audible messages via automated
phone call, text message or e-mail accounts, text messages,
e-mails, twitter accounts, etc. Recommendations may also be
provided to the user in a non-real time scheduling format (e.g.,
after the fact or as a result of a more complex, off-line
processing). Appropriate means of delivering offline communications
to users may include: end-of-day daily reports via email; digital
dashboards (web or smart-phone/computer apps); Social Media
services (e.g. Twitter, Facebook, etc.).
[0074] In addition to recommendations formulated using
classification of data and pattern matching algorithms, the host
system may also be capable of performing "smart tasks" based on
specific user directives, as a direct consequence of the occurrence
of a specific combination of real-time data (individual data points
or aggregated). In one embodiment, a smart task is characterized
by: a condition that is evaluated and an action that is executed
when the condition is verified. A smart task may be activated by
the user through a dedicated subsystem (e.g., digital dashboard) in
which the user can view, create, update and delete smart tasks.
[0075] Conditions can be very coarse (i.e. based on a large amount
of aggregated data), for example: total number of steps during a
workout, total number of miles during a run, total number of steps
taken in a week, total number of miles travelled in a week,
distance recorded during a run, speed recorded during a run,
average minute per mile, etc. These conditions will generally be
evaluated at regular intervals periodically, and will typically
fire at most once during the reference period (workout, run, week,
etc.) Conditions can also be very granular (i.e. based on specific
data patterns), for example: x % variation in stride; x % variation
in pronation; x % variation in eversion; progressive change in
speed; x lbs. delta weight detected (dehydration). These conditions
are generally evaluated at a higher frequency, and can fire several
times during the reference period (workout, run, week, etc.).
[0076] The host system may provide a preset menu of "smart tasks"
that the user can enable/disable and/or tune with actual parameters
(e.g. change the variation percentage in stride). The user may also
create new smart tasks by duplicating and modifying existing ones,
or by activating new smart tasks by selecting specific patterns in
their data. The user interface allows the user to be flexible with
their choices. In some embodiments, smart tasks can auto-update to
new parameters. For example, smart tasks tracking all-time records
(e.g., reaching a total of 10000 steps) may automatically update to
the next level of achievement once the user has beat them. Or,
smart tasks tracking performance records (e.g., run a time below a
6 minute mile) may automatically track against the best
performance. Smart tasks can be seen as powerful mechanisms that
may provide immediate feedback to the user for motivational and
self-improvement purposes and, alternatively, that may provide a
means to detect unusual patterns and provide corrective feedback,
as well as a means to communicate with others (e.g., acquaintances,
personal trainer(s), etc.) the results of specific activities or
the occurrence of specific conditions.
[0077] Actions can be any action that the host system supports.
Depending on the type of the task and the goal, they may include:
direct feedback to the user (motivational stimuli, corrective
feedback); indirect feedback to the user (e.g. power/booster songs
from their collection); feedback to friends (bragging, competitive
and motivational stimuli, social networking); end-of-day reporting,
digital dashboard; and alerting.
[0078] In some embodiments, methods and systems described herein
provide smart feedback relating to footwear wear patterns or
recommendations for replacement. In one specific example, when the
user identifies particular footwear worn during certain activities
and the host system is capable of calculating the distance
travelled by the user while wearing particular footwear and
participating in certain activities, the host system can determine
and monitor a user's mileage accumulated on specific footwear. The
host system may provide user-specific feedback relating to the
user's mileage accumulated on specific footwear to the user on a
continuous or intermittent basis through a user feedback interface.
The host system may alternatively or additionally alert the user
when the mileage accumulated using particular footwear has reached
a predetermined number of miles, and the host system may
additionally recommend replacement of the footwear based on
accumulated mileage. Shoe purchase recommendations may be provided
to the user based on user profile information, user activities,
user preferences, and external footwear database information. Shoe
purchase recommendations may also be provided based on data
collected from a larger population that share attributes with the
user or have similar user profile information.
[0079] In one specific example of recommendations made based on
clustering, consider the following scenario: "Provide the best
fitting shoe recommendation for a specific customer based on the
entire population data." The assumption is that people with similar
anatomical features will experience similar comfort or pain levels
in wearing a shoe. Therefore, individuals having similar individual
foot and body structural features provide the best basis for
footwear fit predictions.
[0080] Let's consider a user (or foot) profile with the following
features: Gender, Age, Weight, Height, Foot Size, Arch Type,
Pronation Type, Prevalent Activity type and intensity. Let's also
consider the following information provided by (some or all) users:
Brand (make, model) of shoes worn; subjective fitting information,
including Size Fit, Width Fit, Arch Support, Comfort, and Frequency
of Usage. The system will cluster the user population based on the
user (foot) profile data. The resulting clusters identify groups
having affinity (similar characteristics) across selected data
categories. Depending on the number of features we select in a
specific query, different groups can result (e.g. subjects [male,
age 40, over-pronating] vs. [male, age 40, over-pronating, size
10.5]). An additional classification may cluster the shoes, in
relation to users, based on the subjective fitting information
provided by each user. A ranked list of shoes may be assembled
based on the fitting information for each specific cluster of users
and used to provide user-specific feedback.
[0081] For example, let A, B, C be three clusters of users in our
population. Let S1, S2, . . . SN be a set of shoes that the
population has come to try/wear. For each cluster, the collection
of shoes SJ . . . SN may be ranked based on relative relevance of
such shoes for the sub-population of users in the cluster. For
example, S1 is recommended favorably by 5 users in cluster A, 2
users in cluster B, 0 users in cluster C. S2 is recommended
favorably by 3 users in cluster A, 2 users in cluster B, I 0 users
in cluster C. S.sub.3 is recommended favorably by I user in cluster
A, 5 users in cluster B, 2 users in cluster C. Also, S.sub.1 is
negatively recommended by I users in cluster A, I users in cluster
B, 3 users in cluster C. Assuming, for the sake of simplicity, that
a favorable recommendation counts as +1, while a negative
recommendation counts as -1. The resulting ranked list for cluster
A is (S1, S2, S3) (total rank. -4, 3, 1); the resulting ranked list
for cluster B is (S3, S2, S1) (total rank: 5, 2, I). The resulting
ranked list for cluster C is (S.sub.2, S.sub.3) (total rank: 10, 2,
-3). The calculation of relevance for the ranking algorithm is
generally more sophisticated, because the evaluation of the shoe is
more granular (using, for example, a rank of 1 to 5 for each of the
subjective fitting attributes assigned by each user on a shoe).
[0082] After the classification is performed, a user can receive
shoe recommendations simply by providing their foot profile. The
recommendation will be accurate as long as enough data points
(i.e., a sufficient data population) are available in the knowledge
base. Users may also be able to provide their own "feedback" data,
augmenting the overall knowledge base and altering the clusters and
classification for the ranking algorithms.
[0083] While the present invention has been described above with
reference to specific embodiments and the accompanying drawings in
which specific embodiments are shown and explained, it is to be
understood that persons skilled in the art may modify the
embodiments described herein without departing from the spirit and
broad scope of the invention. Accordingly, the descriptions
provided above are considered as being illustrative and exemplary
of specific structures, aspects and features within the broad scope
of the present invention and not as limiting the scope of the
invention. The various embodiments described herein may be combined
to provide further embodiments. The described devices, systems and
methods may omit some elements or acts, may add other elements or
acts, or may combine the elements or execute the acts in a
different order than that illustrated, to achieve various
advantages of the disclosure. These and other changes may be made
to the disclosure in light of the above detailed description. It
will also be understood that while the above description and the
appended claims refer to methods for accomplishing certain tasks
and providing certain feedback, the invention and the disclosure
also provides means and systems for implementing the described
methods using a host system, as described, interfacing with one or
more electronic devices.
[0084] In the present description, where used, the terms "about"
and "consisting essentially of" mean.+-.20% of the indicated range,
value, or structure, unless otherwise indicated. It should be
understood that the terms "a" and "an" as used herein refer to "one
or more" of the enumerated components. The use of the alternative
(e.g., "or") should be understood to mean either one, both, or any
combination thereof of the alternatives, unless otherwise expressly
indicated. As used herein, the terms "include" and "comprise" are
used synonymously, and those terms, and variants thereof, are
intended to be construed as non-limiting. In general, in the
following claims, the terms used should not be construed to limit
the disclosure to the specific embodiments disclosed in the
specification.
* * * * *