Cyclical Behavior Modification

• US 20100332443A1
• Filed: 06/07/2010
• Published: 12/30/2010
• Est. Priority Date: 06/25/2009
• Status: Active Grant

• Current Assignee: Merck & Co., Inc., Proactive Life, Inc.
• Original Assignee: George Mason Intellectual Properties, Inc.
• Inventors: Gartenberg, Daniel
• Granted Patent: US 8,468,115 B2
• Time in Patent Office: Days
• US Class Current: 706/54
• CPC Class Codes: G16H 20/30 relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising
• CPC Class Codes: G16H 20/60 relating to nutrition control, e.g. diets
• CPC Class Codes: G16H 20/70 relating to mental therapies, e.g. psychological therapy or autogenous training
• CPC Class Codes: G16H 40/67 for remote operation

Abstract

Embodiments treat cyclical behaviors based on behavioral data describing physiological factors affecting the behaviors. According to various embodiments, a processor or probe produce, from the behavioral data, factor data concerning the factors. Using a processor or recommendation engine that analyzes the factor data, treatment data is recommended based on an estimate of how the at least one factor affects the at least one cyclical behavior. A processor or feedback engine determines, from the treatment data, behavioral feedback configured to produce new behavioral data and to harness the effects of the factors to improve the cyclical behaviors.

Claims

1. A non-transitory computer readable medium containing instructions that, when executed by or more processors, cause the one or more processors to perform an iterative process for treating at least one cyclical behavior based on behavioral data, the behavioral data describing physiological factors, the physiological factors correlating with the at least one cyclical behavior, the process comprising:

a) collecting behavioral data from a user;

b) extracting at least two diverse factor data points that describes at least one physiological factor from the behavioral data, the factor data points including;

i) score data that describes how the physiological factors affect the at least one cyclical behavior;

ii) comparison data that describes how at least two quantitatively similar factor data points affect the at least one physiological factor; and

iii) proxy data that describes a variable that correlates the relationship between the at least two quantitatively similar factor data points;

c) recommending, from the at least two quantitatively similar factor data points, treatment data based on an estimate of how the at least two quantitatively similar factor data points affect the at least one cyclical behavior;

d) formulating, from the treatment data, behavioral feedback configured to produce new behavioral data; and

e) issuing the behavioral feedback.

2. The medium of claim 1, further including determining if one of the factor data points is quantitatively similar to another of the factor data points.

3. The medium of claim 1, wherein the determining if one of the factor data points is quantitatively similar to another of the factor data points includes:

a) calculating, on a factor-wise basis, at least one set of distances between the comparison data of one of the factor data points and the comparison data of the another of the factor data points; and

b) deciding that the one of the factor data points is quantitatively similar to another of the factor data points if the distances are within at least one threshold.

4. The medium of claim 3, wherein the at least one threshold has a value that varies proportionately with the degree of dependence between score data and the comparison data.

5. The medium of claim 2, wherein recommending treatment data includes:

a) estimating how at least one of the factor data points affects the at least one cyclical behavior by identifying, from among quantitatively similar factor data points, the at least one of the factor data points whose score data has a desirable value; and

b) recommending the treatment data based on the proxy data of the identified at least one of the factor data point.

6. The medium of claim 5, wherein identifying at least one treatment data includes:

a) dividing the at least one sampling window into at least one block;

b) arranging any quantitatively similar factor data points into the blocks based on the proxy data of the quantitatively similar factor data points, such that at least one block contains at least one of the quantitatively similar factor data points;

c) combining score data of at least one of the quantitatively similar factor data points in each of the blocks that contains factor data points to produce a combined score for each block;

d) identifying the block that contains the desirable combined score as at least one a local extreme; and

e) identifying as the treatment data at least one of the factor data points within the block identified as the at least one local extreme.

7. The medium of claim 6, wherein:

a) the at least one cyclical behavior includes at least two cyclical behaviors; and

b) the score data of at least one of the quantitatively similar factor data points in each block includes a weighted combination of score data produced by combining at least one score associated with each of the at least two behaviors and weighted according to user preference data indicating the relative importance of the at least two cyclical behaviors.

8. The medium of claim 2, wherein determining behavioral feedback includes determining the behavioral feedback based on the treatment data, such that, after several iterations of implementing behavioral feedback, score data associated with at least one response converges towards the score data of at least one of the factor data points identified in the treatment data.

9. The medium of claim 2, wherein estimating treatment data includes:

a) when the factor data indicates that a priority behavior condition exists, identifying at least one factor data point, from among the at least one quantitatively similar factor data points, whose proxy data indicates that the at least one factor data point would best improve the priority behavior condition; and

b) estimating the treatment data based on the proxy data of the at least one factor data points.

10. The medium of claim 1, wherein the behavioral data includes at least one of the following:

a) at least one survey response;

b) at least one invasive physiological measure;

c) at least one non-invasive physiological measure;

d) at least one vigilance score concerning a user'"'"'s performance of at least one vigilance task;

e) at least one cognition score concerning a user'"'"'s performance of at least one cognitive task;

or

f) any combination thereof.

11. The medium of claim 1, wherein the at least one cyclical behavior includes at least one of the following:

a) sleep inertia;

b) the homeostatic sleep need;

c) the circadian sleep need;

d) dream recall;

e) reduction in sleep onset;

f) best time and duration to nap;

g) the best time to exercise;

h) the best type of exercise;

i) a healthier diet;

j) increased productivity

k) improved mood;

l) the best time to take a medication;

m) the best dosage of medication;

or

n) any combination thereof.

12. The medium of claim 1, wherein the behavioral feedback includes at least one of the following:

a) an alarm set at a wake-time;

b) at least one indication of when to go to sleep;

c) suggestion of amount of sleep;

d) when to exercise;

e) how much to exercise;

f) sleep instructions;

g) workout instructions;

h) medication instructions

i) diet instructions;

j) a vigilance task;

or

k) any combination thereof.

13. The medium of claim 11, wherein the behavioral data includes at least one of the following:

a) information relating to bedtime;

b) wake time;

c) at least one sleep quality survey;

d) at least one sleep efficiency survey;

e) at least one sleepiness survey;

f) severity of depression;

g) at least one habit survey;

h) at least one medication use survey;

i) invasively collected sample;

j) non-invasively collected sample;

k) a user'"'"'s fitness level;

l) a user'"'"'s diet habits;

m) a user'"'"'s exercise habits;

or

n) any combination thereof.

14. The medium of claim 1, wherein the at least one cyclical behavior includes a behavior that co-varies with depression severity.

15. The medium of claim 14, further including identifying, from the behavioral data, at least one cyclical behavior that co-varies with depression severity.

16. The medium of claim 1, wherein the at least one cyclical behavior includes a behavior that co-varies with a fitness behavior.

17. The medium of claim 16, further including identifying at least one user-selected fitness behavior.

18. The medium of claim 11, wherein the at least one physiological factor includes at least one of the following:

a) a homeostatic component;

b) a circadian component;

c) a sleep stage upon waking component;

d) choronobiology;

e) neurochemistry;

f) physical cycles (need a better word for this;

i.e. periods, food needs, menopause, etc.;

or

g) any combination thereof.

19. The medium of claim 1, wherein:

a) each iteration is performed within a sampling window; and

b) the size of the at least one sampling window is modified based upon changes in behavioral data.

20. The medium of claim 1, wherein the at least one cyclical behavior includes a behavior that co-varies with sleep interia.