Data assessment and visualization platform for use in a network environment including computing devices

Abstract

Generally described, one or more aspects of the present application relate to enabling determination of a predicted effect of a target action on sensor data generated on a computing device. More specifically, the present disclosure provides a system that can analyze sensor data generated on a computing device before and after a target action, generate analytics data indicating the effect of the target action on the sensor data generated on the computing device, and store the analytics data in association with the one or more characteristics associated with the computing device one or more characteristics of the target action. Subsequently, the system can analyze a given set of characteristics associated with a different computing device and characteristics and sensor data generated on said different computing device, and output, based on the previously generated analytics data, an indication of the predicted effect of a given action or set of actions on the sensor data to be generated on said different computing device.

What is claimed is: 
     
         1 . A system for providing analytics relating to accelerometer data generated on computing devices, the system comprising:
 a data repository comprising computer hardware and storing analytics data associated with one or more computing devices; and   a data analysis server comprising computer hardware, wherein the data analysis service is configured to at least:
 receive a first request to analyze historical accelerometer data collected from a plurality of computing devices, wherein the first request includes (i) first characteristics data associated with the plurality of computing devices, and (ii) second characteristics data associated with a plurality of target actions, wherein at least a first subset of the historical accelerometer data is associated with a first type of target actions and a second subset of the historical accelerometer data is associated with a second type of target actions different from the first type of target actions; and 
 identify a set of temporal windows based at least in part on the historical accelerometer data associated with a first computing device of the plurality of computing devices and a first target action associated with a reference point, wherein identifying the set of temporal windows comprises one or both of (i) identifying two or more pre-reference temporal windows preceding the reference point based at least in part on the historical accelerometer data collected prior to the reference point exhibiting a first threshold level of change, or (ii) two or more post-reference temporal windows following the reference point based at least in part on the historical accelerometer data collected subsequent to the reference point exhibiting a second threshold level of change different from the first threshold level of change; 
 generate outcome data based at least in part on a comparison of the historical accelerometer data in two or more temporal windows of the set of temporal windows, wherein the outcome data indicates a level of effectiveness of the first target action for the historical accelerometer data collected from the first computing device; and 
 store, in the data repository, the outcome data in association with (a) at least a portion of the first characteristics data associated with the first computing device and (b) at least a portion of the second characteristics data associated with the first target action, such that the outcome data is retrievable in a characteristics-specific manner, 
   wherein the data analysis server is further configured to:
 receive, from a second computing device, a second request to analyze additional accelerometer data collected from a third computing device and not included in the historical accelerometer data, wherein the second request includes third characteristics data associated with the third computing device not included in the plurality of computing devices; 
 retrieve, from the data repository, first analytics data based at least in part on the third characteristics data and a second target action of the plurality of target actions, wherein the first analytics data is a subset of the outcome data stored in the data repository that corresponds to at least a portion of the third characteristics data and the second target action; 
 determine, based at least in part on the first analytics data and the additional accelerometer data collected from the third computing device, (a) a predicted effectiveness value associated with the second target action for the additional accelerometer data and (b) an indication of predicted accelerometer data values to be collected from the third computing device subsequent to the second target action; 
 retrieve, from the data repository, second analytics data based at least in part on the third characteristics data and a third target action of the plurality of target actions different from the second target action, wherein the second analytics data is a subset of the outcome data stored in the data repository that corresponds to at least a portion of the third characteristics data and the third target action; 
 determine, based at least in part on the second analytics data and the additional accelerometer data collected from the third computing device, (c) a predicted effectiveness value associated with the third target action for the additional accelerometer data and (d) an indication of predicted accelerometer data values to be collected from the third computing device subsequent to the third target action; 
 generate one or both of (i) a first comparison of (a) the predicted effectiveness value associated with the second target action for the additional accelerometer data and (c) the predicted effectiveness value associated with the third target action for the additional accelerometer data, or (ii) a second comparison of (b) the indication of predicted accelerometer data values to be collected from the third computing device subsequent to the second target action and (d) the indication of predicted sensor data values to be collected from the third computing device subsequent to the third target action; and 
 output, via a user interface associated with the second computing device, a visualization of one or both of the first comparison or the second comparison. 
   
     
     
         2 . The system of  claim 1 , wherein the first target action is identical to the second target action. 
     
     
         3 . The system of  claim 1 , wherein the first target action is different from the second target action. 
     
     
         4 . The system of  claim 1 , wherein the plurality of target action includes one or more of a surgical intervention, a therapy, a drug, or a method. 
     
     
         5 . The system of  claim 1 , wherein the first target action is associated with one or more of a medical practitioner, a medical intervention date, a medical device used, a medical device manufacturer, a drug infused, an amount of drug infused, a medical intervention duration, or a medical intervention time of day. 
     
     
         6 . A computer-implemented method for providing analytics relating to patients and medical interventions, the method comprising:
 receiving a first request to analyze historical sensor data collected from a plurality of computing devices, wherein the first request includes (i) first characteristics data associated with the plurality of computing devices, and (ii) second characteristics data associated with a plurality of target actions, wherein at least a first subset of the historical sensor data is associated with a first type of target actions and a second subset of the historical sensor data is associated with a second type of target actions different from the first type of target actions; and   identifying a set of temporal windows based at least in part on the historical sensor data associated with a first computing device of the plurality of computing devices and a first target action associated with a reference point, wherein identifying the set of temporal windows comprises one or both of (i) identifying two or more pre-reference temporal windows preceding the reference point based at least in part on the historical sensor data collected prior to the reference point exhibiting a first threshold level of change, or (ii) two or more post-reference temporal windows following the reference point based at least in part on the historical sensor data collected subsequent to the reference point exhibiting a second threshold level of change different from the first threshold level of change;   generating outcome data based at least in part on a comparison of the historical sensor data in two or more temporal windows of the set of temporal windows, wherein the outcome data indicates a level of effectiveness of the first target action for the historical sensor data collected from the first computing device; and   storing, in a data repository, the outcome data in association with (a) at least a portion of the first characteristics data associated with the first computing device and (b) at least a portion of the second characteristics data associated with the first target action, such that the outcome data is retrievable in a characteristics-specific manner;   receiving, from a second computing device, a second request to analyze additional sensor data collected from a third computing device and not included in the historical sensor data, wherein the second request includes third characteristics data associated with the third computing device not included in the plurality of computing devices;   retrieving, from the data repository, first analytics data based at least in part on the third characteristics data and a second target action of the plurality of target actions, wherein the first analytics data is a subset of the outcome data stored in the data repository that corresponds to at least a portion of the third characteristics data and the second target action;   determining, based at least in part on the first analytics data and the additional sensor data collected from the third computing device, (a) a predicted effectiveness value associated with the second target action for the additional sensor data and (b) an indication of predicted sensor data values to be collected from the third computing device subsequent to the second target action;   retrieving, from the data repository, second analytics data based at least in part on the third characteristics data and a third target action of the plurality of target actions different from the second target action, wherein the second analytics data is a subset of the outcome data stored in the data repository that corresponds to at least a portion of the third characteristics data and the third target action;   determining, based at least in part on the second analytics data and the additional sensor data collected from the third computing device, (c) a predicted effectiveness value associated with the third target action for the additional sensor data and (d) an indication of predicted sensor data values to be collected from the third computing device subsequent to the third target action;   generating one or both of (i) a first comparison of (a) the predicted effectiveness value associated with the second target action for the additional sensor data and (c) the predicted effectiveness value associated with the third target action for the additional sensor data, or (ii) a second comparison of (b) the indication of predicted sensor data values to be collected from the third computing device subsequent to the second target action and (d) the indication of predicted sensor data values to be collected from the third computing device subsequent to the third target action; and   outputting, via a user interface associated with the second computing device, a visualization of one or both of the first comparison or the second comparison.   
     
     
         7 . The computer-implemented method of  claim 6 , wherein the first target action is identical to the second target action. 
     
     
         8 . The computer-implemented method of  claim 6 , wherein the first target action is different from the second target action. 
     
     
         9 . The computer-implemented method of  claim 6 , wherein the sensor data includes one or more of step count, step size, distance traveled, or flights of stairs climbed. 
     
     
         10 . The computer-implemented method of  claim 6 , wherein the sensor data includes one or more of accelerometer data, gyroscope data, magnetometer data, or calorimeter data. 
     
     
         11 . The computer-implemented method of  claim 6 , wherein the plurality of target action includes one or more of a surgical intervention, a therapy, a drug, or a method. 
     
     
         12 . The computer-implemented method of  claim 6 , wherein the first target action is associated with one or more of a medical practitioner, a medical intervention date, a medical device used, a medical device manufacturer, a drug infused, an amount of drug infused, a medical intervention duration, or a medical intervention time of day. 
     
     
         13 . A non-transitory computer-readable medium storing instructions that, when executed by a computing system, cause the computing system to perform operations comprising:
 receiving a first request to analyze historical sensor data collected from a plurality of computing devices, wherein the first request includes (i) first characteristics data associated with the plurality of computing devices, and (ii) second characteristics data associated with a plurality of target actions, wherein at least a first subset of the historical sensor data is associated with a first type of target actions and a second subset of the historical sensor data is associated with a second type of target actions different from the first type of target actions; and   identifying a set of temporal windows based at least in part on the historical sensor data associated with a first computing device of the plurality of computing devices and a first target action associated with a reference point, wherein identifying the set of temporal windows comprises one or both of (i) identifying two or more pre-reference temporal windows preceding the reference point based at least in part on the historical sensor data collected prior to the reference point exhibiting a first threshold level of change, or (ii) two or more post-reference temporal windows following the reference point based at least in part on the historical sensor data collected subsequent to the reference point exhibiting a second threshold level of change different from the first threshold level of change;   generating outcome data based at least in part on a comparison of the historical sensor data in two or more temporal windows of the set of temporal windows, wherein the outcome data indicates a level of effectiveness of the first target action for the historical sensor data collected from the first computing device; and   storing, in a data repository, the outcome data in association with (a) at least a portion of the first characteristics data associated with the first computing device and (b) at least a portion of the second characteristics data associated with the first target action, such that the outcome data is retrievable in a characteristics-specific manner;   receiving, from a second computing device, a second request to analyze additional sensor data collected from a third computing device and not included in the historical sensor data, wherein the second request includes third characteristics data associated with the third computing device not included in the plurality of computing devices;   retrieving, from the data repository, first analytics data based at least in part on the third characteristics data and a second target action of the plurality of target actions, wherein the first analytics data is a subset of the outcome data stored in the data repository that corresponds to at least a portion of the third characteristics data and the second target action;   determining, based at least in part on the first analytics data and the additional sensor data collected from the third computing device, (a) a predicted effectiveness value associated with the second target action for the additional sensor data and (b) an indication of predicted sensor data values to be collected from the third computing device subsequent to the second target action;   retrieving, from the data repository, second analytics data based at least in part on the third characteristics data and a third target action of the plurality of target actions different from the second target action, wherein the second analytics data is a subset of the outcome data stored in the data repository that corresponds to at least a portion of the third characteristics data and the third target action;   determining, based at least in part on the second analytics data and the additional sensor data collected from the third computing device, (c) a predicted effectiveness value associated with the third target action for the additional sensor data and (d) an indication of predicted sensor data values to be collected from the third computing device subsequent to the third target action;   generating one or both of (i) a first comparison of (a) the predicted effectiveness value associated with the second target action for the additional sensor data and (c) the predicted effectiveness value associated with the third target action for the additional sensor data, or (ii) a second comparison of (b) the indication of predicted sensor data values to be collected from the third computing device subsequent to the second target action and (d) the indication of predicted sensor data values to be collected from the third computing device subsequent to the third target action; and   outputting, via a user interface associated with the second computing device, a visualization of one or both of the first comparison or the second comparison.   
     
     
         14 . The non-transitory computer-readable medium of  claim 13 , wherein the first target action is identical to the second target action. 
     
     
         15 . The non-transitory computer-readable medium of  claim 13 , wherein the first target action is different from the second target action. 
     
     
         16 . The non-transitory computer-readable medium of  claim 13 , wherein the sensor data includes one or more of step count, step size, distance traveled, or flights of stairs climbed. 
     
     
         17 . The non-transitory computer-readable medium of  claim 13 , wherein the sensor data includes one or more of accelerometer data, gyroscope data, magnetometer data, or calorimeter data. 
     
     
         18 . The non-transitory computer-readable medium of  claim 13 , wherein the plurality of target action includes one or more of a surgical intervention, a therapy, a drug, or a method.