Systems, methods, and devices for determining endpoints of a rest period using motion data

Abstract

Systems, methods, and devices for determining a temporal duration of a rest period using motion data are described herein. In one exemplary embodiment, one or more data filters are applied to received motion data to generate one or more data sets of the motion data. The motion data represents an amount of activity experienced by an individual over the course of a period of time, such as one day. An iterative process is performed to identify a starting point and an ending point of a rest period using the generated data set(s). After the starting and ending points are identified, a temporal difference between the starting and ending points is calculated, and a total temporal duration of the rest period is determined.

What is claimed is: 
     
         1 . A method for identifying a rest period of an individual, the method comprising:
 setting a threshold activity level for the rest period to an initial threshold activity level, wherein the threshold activity level is expressed in terms of an activity count during each epoch of a given temporal duration;   receiving motion data, via a wearable motion tracker having at least one motion sensor, representative of an amount of activity in terms of the activity count during each epoch of the given temporal duration experienced by the individual over a course of a given time interval greater than the given temporal duration;   applying at least a first data filter and a second data filter, different from the first data filter, to the motion data to generate at least a first data set and a second data set, different from the first data set, representative of the motion data;   determining a minimum point within the first data set;   determining a minimum time within the given time interval associated with the minimum point;   determining a first boundary of the rest period, wherein the first boundary comprises one of an upper or a lower boundary, via the steps of:
 determining a first intersection point where the first data set intersects with the threshold activity level, wherein the first intersection point occurs (i) after the minimum time with respect to the upper boundary or (ii) before the minimum time with respect to the lower boundary; 
 determining a first time within the given time interval associated with the first intersection point, the first time corresponding to a first approximation of the first boundary of the rest period; 
 determining a second intersection point where the second data set intersects with the threshold activity level based on the first time being used as a starting point for determining the second intersection point; 
 determining a second time within the given time interval associated with the second intersection point, the second time corresponding to a second approximation of the first boundary of the rest period; 
 determining a first position on the motion data corresponding to the second time; 
 determining a first raw data intersection point where the motion data intersects with the threshold activity level based on the second time being used as a starting point for determining the first raw data intersection point, wherein the first raw data intersection point occurs within the motion data (i) after the first time with respect to the upper boundary or (ii) before the first time with respect to the lower boundary; 
 determining a first raw data time within the given time interval associated with the first raw data intersection point, wherein the first raw data time occurs (i) after the first time with respect to the upper boundary or (ii) before the first time with respect to the lower boundary; and 
 setting the first raw data time as the first boundary of the rest period; 
   determining a second boundary of the rest period, wherein the second boundary comprises the lower or the upper boundary, opposite the first boundary, via the steps of:
 determining a second raw data intersection point where the motion data intersects with the threshold activity level; 
 determining a second raw data time within the given time interval associated with the second raw data intersection point, and 
 setting the second raw data time as the second boundary of the rest period; and 
   determining an amount of time of the rest period, identified via the first boundary of the rest period and the second boundary of the rest period, by calculating a temporal difference between the first raw data time and the second raw data time.   
     
     
         2 . The method of  claim 1 , wherein:
 the motion data comprises a plurality of data points; and   the plurality of data points is representative of the amount of activity in terms of the activity count detected during each epoch.   
     
     
         3 . The method of  claim 2 , wherein a plurality of samples is obtained during each epoch, and the amount of activity during each epoch is an aggregate of the plurality of samples during a respective epoch. 
     
     
         4 . The method of  claim 1 , wherein the first data filter comprises a moving Gaussian window of a first length of time and the second data filter comprises a moving Gaussian window of a second length of time, shorter than the first length of time. 
     
     
         5 . The method of  claim 1 , wherein applying at least the first data filter and the second data filter, different from the first data filter further comprises:
 applying a third data filter, different from the first and second data filters, to the motion data to generate a third data set representative of the motion data;   applying a fourth data filter, different from the first, second and third data filters, to the motion data to generate a fourth data set representative of the motion data; and   applying a fifth data filter, different from the first, second, third and fourth data filters, to the motion data to generate a fifth data set representative of the motion data,   wherein determining the first boundary of the rest period further comprises:
 determining respective third, fourth, and fifth intersection points where the third, fourth and fifth data sets intersect with the threshold activity level based on a respective second, third and fourth time being used as a starting point for determining the respective third, fourth and fifth intersection points; 
 determining respective third, fourth and fifth times within the given time interval associated with the respective third, fourth and fifth intersection points, the respective third, fourth and fifth times corresponding to a respective third, fourth and fifth approximation of the first boundary of the rest period; 
 determining the first raw data intersection point where the motion data intersects with the threshold activity level based on the fifth time being used as a starting point for determining the first raw data intersection point, wherein the first raw data upper intersection point occurs within the motion data (i) after the first time with respect to the upper boundary or (ii) before the first time with respect to the lower boundary. 
   
     
     
         6 . The method of  claim 1 , further comprising:
 applying a moving Gaussian window to the motion data to generate an additional data set of the motion data, the moving Gaussian window having a length of 360 minutes;   determining at least two local minima of the additional data set;   determining that the minimum point is located between at least two of the local minima; and   determining that the minimum point is associated with the rest period.   
     
     
         7 . The method of  claim 1 , further comprising:
 repeating, for at least two different threshold activity levels, until a new determined rest period duration is equal to a previous determined rest period duration, the following:   modifying the threshold activity level to a modified threshold activity level greater than a previous threshold activity level;   determining a new rest period duration corresponding to the amount of time of the rest period based on the modified threshold activity level and the motion data;   comparing the new rest period duration with a previous determined rest period duration; and   responsive to the new determined rest period duration being equal to a previous determined rest period duration, selecting a lesser of two most recent modified threshold activity levels for use as an optimal threshold activity level in place of the initial threshold activity level.   
     
     
         8 . A system comprising:
 a wearable motion tracker comprising at least one motion sensor; and   a motion analysis device, wherein the motion analysis device comprises:
 communications circuitry that receives motion data from the wearable motion tracker, wherein the motion data is representative of an amount of activity in terms of an activity count during each epoch of a given temporal duration experienced by the wearable motion tracker over a course of a given time interval greater than the given temporal duration; 
 memory that stores the motion data; and 
 at least one processor configured to:
 set a threshold activity level for a rest period to an initial threshold activity level, wherein the threshold activity level is expressed in terms of the activity count during each epoch of the given temporal duration; 
 apply at least a first data filter and a second data filter, different from the first data filter, to the motion data to generate at least a first data set and a second data set different from the first data set, representative of the motion data; 
 determine a minimum point within the first data set; 
 determine a minimum time within the given time interval associated with the minimum point; 
 determine a first boundary of the rest period, wherein the first boundary comprises one of an upper or a lower boundary, via the steps of:
 determining a first intersection point where the first data set intersects with the threshold activity level, wherein the first intersection point occurs (i) after the minimum time with respect to the upper boundary or (ii) before the minimum time with respect to the lower boundary; 
 determining a first time within the given time interval associated with the first intersection point, the first time corresponding to a first approximation of the first boundary of the rest period; 
 determining a second intersection point where the second data set intersects with the threshold activity level based on the first time being used as a starting point for determining the second intersection point; 
 determining a second time within the given time interval associated with the second intersection point, the second time corresponding to a second approximation of the first boundary of the rest period; 
 determining a first position on the motion data corresponding to the second time; 
 determining a first raw data intersection point where the motion data intersects with the threshold activity level based on the second time being used as a starting point for determining the first raw data intersection point, wherein the first raw data intersection point occurs within the motion data (i) after the first time with respect to the upper boundary or (ii) before the first time with respect to the lower boundary; 
 determining a first raw data time within the given time interval associated with the first raw data intersection point, wherein the first raw data time occurs (i) after the first time with respect to the upper boundary or (ii) before the first time with respect to the lower boundary; and 
 assigning the first raw data time as being the first boundary of the rest period; 
 
 determine a second boundary of the rest period, wherein the second boundary comprises the lower or the upper boundary, opposite the first boundary, via the steps of:
 determining a second raw data intersection point where the motion data intersects with the threshold activity level; 
 determining a second raw data time within the given time interval associated with the second raw data intersection point, and 
 assigning the second raw data time as the second boundary of the rest period; and 
 
 determine an amount of time of the rest period, identified via the first boundary of the rest period and the second boundary of the rest period, by calculating a temporal difference between the first raw data time and the second raw data time. 
 
   
     
     
         9 . The system of  claim 8 , wherein:
 the motion data comprises a plurality of data points; and   the plurality of data points is representative of the amount of activity in terms of the activity count detected during each epoch.   
     
     
         10 . The system of  claim 9 , wherein a plurality of samples is obtained during each epoch, and the amount of activity during each epoch is an aggregate of the plurality of samples during a respective epoch. 
     
     
         11 . The system of  claim 8 , wherein the first data filter comprises a moving Gaussian window of a first length of time and the second data filter comprises a moving Gaussian window of a second length of time, shorter than the first length of time. 
     
     
         12 . The system of  claim 8 , wherein that at least one processor is further configured to:
 apply a third data filter, different from the first and second data filters, to the motion data to generate a third data set representative of the motion data;   apply a fourth data filter, different from the first, second and third data filters, to the motion data to generate a fourth data set representative of the motion data;   apply a fifth data filter, different from the first, second, third and fourth data filters, to the motion data to generate a fifth data set representative of the motion data; and   determine respective third, fourth, and fifth intersection points where the third, fourth and fifth data sets intersect with the threshold activity level based on a respective second, third and fourth time being used as a starting point for determining the respective third, fourth and fifth intersection points;   determine respective third, fourth and fifth times within the given time interval associated with the respective third, fourth and fifth intersection points, the respective third, fourth and fifth times corresponding to a respective third, fourth and fifth approximation of the first boundary of the rest period; and   determine the first raw data intersection point where the motion data intersects with the threshold activity level based on the fifth time being used as a starting point for determining the first raw data intersection point, wherein the first raw data upper intersection point occurs within the motion data (i) after the first time with respect to the upper boundary or (ii) before the first time with respect to the lower boundary.   
     
     
         13 . The system of  claim 12 , wherein:
 the first data filter comprises a first moving Gaussian window having a first length of 100 minutes;   the second data filter comprises a second moving Gaussian window having a second length of 80 minutes;   the third data filter comprises a third moving Gaussian window having a third length of 60 minutes;   the fourth data filter comprises a fourth moving Gaussian window having a fourth length of 40 minutes; and   the fifth data filter comprises a fifth moving Gaussian window having a fifth length of 20 minutes.   
     
     
         14 . The system of  claim 8 , wherein the at least one processor is further configured to:
 apply a moving Gaussian window to the motion data to generate an additional data set of the motion data, the moving Gaussian window having a length of 360 minutes;   determine at least two local minima of the additional data set;   determine that the minimum point is located between at least two of the local minima; and   determine that the minimum point is associated with the rest period.   
     
     
         15 . The system of  claim 8 , wherein the at least one processor is further configured to:
 repeat, for at least two different threshold activity levels, until a new determined rest period duration is equal to a previous determined rest period duration, the following:   modify the threshold activity level to a modified threshold activity level greater than a previous threshold activity level;   determine a new rest period duration corresponding to the amount of time of the rest period based on the modified threshold activity level and the motion data;   compare the new rest period duration with a previous determined rest period duration; and   responsive to the new determined rest period duration being equal to a previous determined rest period duration, select a lesser of two most recent modified threshold activity levels for use as an optimal threshold activity level in place of the initial threshold activity level.   
     
     
         16 . A wearable motion tracker, comprising:
 at least one motion sensor that captures motion data representative of an amount of activity detected during each epoch of a given temporal duration experienced by the at least one motion sensor within a time period greater than the given temporal duration;   memory that stores the motion data; and   at least one processor configured to:
 set a threshold activity level for a rest period to an initial threshold activity level, wherein the threshold activity level is expressed in terms of the activity count during each epoch of the given temporal duration; 
 apply at least a first data filter and a second data filter, different from the first data filter, to the motion data to generate at least a first data set and a second data set different from the first data set, representative of the motion data; 
 determine a minimum point within the first data set; 
 determine a minimum time within the given time interval associated with the minimum point; 
 determine a first boundary of the rest period, wherein the first boundary comprises one of an upper or a lower boundary, via the steps of:
 determining a first intersection point where the first data set intersects with the threshold activity level, wherein the first intersection point occurs (i) after the minimum time with respect to the upper boundary or (ii) before the minimum time with respect to the lower boundary; 
 determining a first time within the given time interval associated with the first intersection point, the first time corresponding to a first approximation of the first boundary of the rest period; 
 determining a second intersection point where the second data set intersects with the threshold activity level based on the first time being used as a starting point for determining the second intersection point; 
 determining a second time within the given time interval associated with the second intersection point, the second time corresponding to a second approximation of the first boundary of the rest period; 
 determining a first position on the motion data corresponding to the second time; 
 determining a first raw data intersection point where the motion data intersects with the threshold activity level based on the second time being used as a starting point for determining the first raw data intersection point, wherein the first raw data intersection point occurs within the motion data (i) after the first time with respect to the upper boundary or (ii) before the first time with respect to the lower boundary; 
 determining a first raw data time within the given time interval associated with the first raw data intersection point, wherein the first raw data time occurs (i) after the first time with respect to the upper boundary or (ii) before the first time with respect to the lower boundary; and 
 assigning the first raw data time as being the first boundary of the rest period; 
 
 determine a second boundary of the rest period, wherein the second boundary comprises the lower or the upper boundary, opposite the first boundary, via the steps of:
 determining a second raw data intersection point where the motion data intersects with the threshold activity level; 
 determining a second raw data time within the given time interval associated with the second raw data intersection point, and 
 assigning the second raw data time as the second boundary of the rest period; and 
 
 determine an amount of time of the rest period, identified via the first boundary of the rest period and the second boundary of the rest period, by calculating a temporal difference between the first raw data time and the second raw data time. 
   
     
     
         17 . The wearable motion tracker of  claim 16 , wherein:
 the motion data comprises a plurality of data points; and   the plurality of data points is representative of the amount of activity in terms of the activity count detected during each epoch.   
     
     
         18 . The wearable motion tracker of  claim 17 , wherein a plurality of samples is obtained during each epoch, and the amount of activity during each epoch is an aggregate of the plurality of samples during a respective epoch. 
     
     
         19 . The wearable motion tracker of  claim 16 , wherein the first data filter comprises a moving Gaussian window of a first length of time and the second data filter comprises a moving Gaussian window of a second length of time, shorter than the first length of time. 
     
     
         20 . The wearable motion tracker of  claim 16 , wherein the at least one processor is further configured to:
 apply a third data filter, different from the first and second data filters, to the motion data to generate a third data set representative of the motion data;   apply a fourth data filter, different from the first, second and third data filters, to the motion data to generate a fourth data set representative of the motion data;   apply a fifth data filter, different from the first, second, third and fourth data filters, to the motion data to generate a fifth data set representative of the motion data;   determine respective third, fourth, and fifth intersection points where the third, fourth and fifth data sets intersect with the threshold activity level based on a respective second, third and fourth time being used as a starting point for determining the respective third, fourth and fifth intersection points;   determine respective third, fourth and fifth times within the given time interval associated with the respective third, fourth and fifth intersection points, the respective third, fourth and fifth times corresponding to a respective third, fourth and fifth approximation of the first boundary of the rest period; and   determine the first raw data intersection point where the motion data intersects with the threshold activity level based on the fifth time being used as a starting point for determining the first raw data intersection point, wherein the first raw data upper intersection point occurs within the motion data (i) after the first time with respect to the upper boundary or (ii) before the first time with respect to the lower boundary.