Automated health data acquisition, processing and communication system and method

Abstract

To provide an automatic health data acquisition method, a processing method and a communication system for classifying the user activity.SOLUTION: Disclosed is a method in a self-learning activity identification system, which includes the steps of: transmitting sensor information 402, user profile information 404 and external information 406 to a classification engine 408; detecting a change in a heart rate, a respiration rate or other parameters of a user; classifying the detected change as the particular activity, providing the classified activity to a computing device associated with the user as the graphical representation of a timeline, a health score, and a social network; and determining whether or not to reclassify the activity information.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present application relates generally to automated health data acquisition, and more particularly to automatically classifying health-related activity information taking into account biological and / or behavioral tracking and / or external context. [Background technology]

[0002] Despite advances in various areas of technology, barriers remain to assessing a person's relative health status in a rapid, cost-effective, and timely manner. Assessing an individual's relative health status is important and has not been adequately addressed due to rising healthcare costs and the prevalence of diseases related to unhealthy lifestyles, such as diabetes and heart disease. Furthermore, in many parts of the world, access to physicians is limited. Even in the developed world, a doctor's time is considered a precious commodity, and there are often long waiting lists and doctor-specialist referral systems that must be navigated before an appointment can be made. As a result, accessing a medical professional to receive information about one's health can be very difficult.

[0003] In recent years, fitness tracking devices have been adopted to monitor an individual's health and activity. Devices for biological and physiological monitoring, such as blood pressure, heart rate, and body temperature, can communicate with computing devices via Bluetooth® or other wireless protocols. Behavioral and activity monitoring can also be provided via one or more devices, such as pedometers and other accelerometer-based devices, which can also wirelessly transmit signals representative of an individual's fitness activity to one or more computing devices. Moreover, hardware has been integrated and implemented into communication devices, such as smartphones, to track user activity and biological and physiological information. Unfortunately, such tracking devices operate alone and do not collectively contribute to providing an accurate and timely representation of a person's biological, physiological, and / or activity-based state.

[0004] More recently, mobile apps have been developed that integrate signals received from various fitness tracking devices. Exercise goals can be defined by users of such apps, and as the user exercises, the user can be notified of how well they are doing toward achieving those goals. For example, activity diaries (e.g., offered by the "Moves" app) can record walking, cycling, running, and other activities and provide users with reports on distance, duration, number of steps, and number of calories burned for each activity.

[0005] Despite recent advances in fitness tracking technology, fitness activities are still presented in misleading ways, requiring user input and manual intervention to correct errors. Summary of the Invention [Problem to be solved by the invention]

[0006] It is with respect to these and other considerations that the present application is presented. [Means for solving the problem]

[0007] In one or more implementations, the present application includes a system and method for classifying user activity. A passive tracking device, a processor configured to receive information from the tracking device, and a database accessible by the processor and storing tracking device information, user profile information, and external information are disclosed. The processor is configured to execute instructions that cause the processor to perform various steps, such as defining a first unit of activity having a first start time corresponding to detection of a user engaging in the activity and monitoring the tracking device information, the external information, or both. The processor is further configured to establish a first end time for the first unit of activity using the monitored information, thereby automatically classifying the first unit of activity. The classification of the first unit of activity is output to a display of the computing device, and the classification of the first unit of activity is stored in the database. Additionally, a user interface is provided that includes selectable options associated with the first unit of activity. Then, in response to at least one received selection of the selectable options, the classification is corrected by: combining the first activity unit with a second activity unit having a second start time and a second end time, whereby the corrected classification has a start time equal to the first start time and an end time equal to the second end time; further, the first activity unit is merged with the second activity unit having a second start time and a second end time, whereby the corrected classification has a start time equal to the first start time and an end time equal to the second end time; alternatively, the first activity unit is split into at least two activity units, each of the at least two activity units having a different respective start time and a different respective end time; the corrected classification is output to a display of the computing device, and the corrected classification is stored in a database.

[0008] In accordance with an implementation of the present application, a system and method are provided for restricting computer transmissions based on reclassification of a plurality of users, at least in response to masked numerical scores received over a data communications network and each representing health-related information associated with a respective one of the plurality of users. A processor is configured to receive a first masked numerical score from a computing device associated with each of the plurality of users, the processor being configured to execute instructions stored on a processor-readable medium. The instructions cause the processor to process the first masked numerical score to associate the user within one of the plurality of user bins, and further to receive a second masked numerical score from the computing device associated with each of the plurality of users, the second masked numerical score being different from the first masked numerical score. The second masked numerical score is then processed to associate the one user within a different one of the plurality of user bins. The processor is further configured to respond to processing the second masked numerical score with transmitting a message to a computing device associated with the one user indicating that the one user is associated with a different one of the plurality of user bands.

[0009] In one implementation, a system and method for classifying user activity is provided. A tracking device, a processor configured to receive information from the tracking device, and a database are provided. The database is accessible by the processor and stores: tracking device information; user profile information; and external information. The processor is configured to execute instructions that cause the processor to perform various steps. For example, the processor is configured to define a first activity unit having a first start time corresponding to detection of a user engaging in an activity. The processor is further configured to monitor the tracking device information, the external information, or both, and establish a first end time for the first activity unit using the monitored information. The processor is further configured to automatically classify the first activity unit and output the classification of the first activity unit to a display of the computing device. Furthermore, the classification of the first activity unit is stored in the database, and a user interface is provided that includes selectable options associated with the first activity unit. Information received via the user interface is stored and associated with the activity unit in the database.

[0010] In one or more implementations, a system and method are provided that includes a computing device configured to access a non-transitory processor-readable storage medium having instructions, which, when executed by the computing device, cause the computing device to perform various steps. For example, the device is configured to receive, via at least one communication network, parameter information representing conditions for exchanging at least personal health information for value from each of a plurality of user computing devices operated by each of a plurality of users. The device is further configured, in response to a request automatically received via the at least one communication network by a decentralized, distributed computer program, to identify a transmission relating to a commitment to provide a respective value in exchange for at least personal health information of a user associated with one of the plurality of computing devices according to at least a portion of the parameter information associated with each user. Furthermore, the device is configured to access, from the at least one computing device via the at least one communication network, at least a portion of a private database associated with the user's personal health information, the database being populated from time to time with automatically collected, verified records representing at least sensed intrinsic medical information and sensed extrinsic activity information automatically collected from the device configured with a sensor, a communication module, and a microcontroller. Additionally, the device is configured to provide access to a portion of the private database to at least one computing device associated with the party and to provide a respective value to at least one computing device associated with the user.

[0011] In one or more implementations, a system and method are provided for dynamically tracking user-based activity in response to information received from multiple data sources. Tracking of user activity information related to the number of steps taken by a user over a period of time occurs through a first application, the first application executing on a mobile computing device and including a graphical user interface presenting at least a health score associated with the user. Additionally, gameplay information is communicated to a second application executed by the mobile computing device, the communication including information regarding user interaction with a gaming environment provided by the mobile computing device. Furthermore, the tracked user activity information, along with at least a portion of the user interaction information, is integrated into an activity value, and the user's health score is updated using the activity value. The updated health score is then stored in association with a record for the user, and notification of the updated health score occurs on the mobile computing device when the first application is not running.

[0012] Various features, aspects and advantages of the present invention can be understood from the following detailed description and accompanying drawings. [Brief explanation of the drawings]

[0013] [Figure 1] FIG. 1 is a diagram illustrating an example of a hardware configuration according to an implementation example. [Figure 2] 1 illustrates the functional elements of an example information processor and / or workstation according to an implementation of the present application. [Figure 3] FIG. 1 is a block diagram illustrating components associated with a self-learning activity identification system. [Figure 4] 1 illustrates an example flow of information and steps associated with a self-learning activity identification system according to one or more implementations of the present application. [Figure 5] 10 is an example display screen illustrating a determination of a user's activity according to the user's heart rate, according to an example implementation of the present application. [Figure 6-7]1 illustrates an example data entry display screen associated with an example implementation of the present application. [Figure 8] 1 illustrates an example interactive data entry display screen associated with an activity and associated data entry. [Figure 9] 1 is an example graph identifying opportunities that may be provided in accordance with one or more implementations of the present application. [Figure 10] 1 shows an example chart for identifying populations suitable for service in accordance with the present application. [Figure 11] 1 illustrates a traditional life insurance policy and an improved life insurance policy in accordance with one or more implementations of the present application. [Figure 12] 1 illustrates an example chart identifying insurance premium discount rates offered to users in accordance with the present application. [Figure 13] 1 illustrates an example display screen provided on a mobile computing device that identifies each user's current health score. [Figure 14] 1 illustrates an example data entry display form that includes options for users to select related to maintaining control over their information and privacy. [Figure 15-23] Identify one or more implementations for engaging users with the information, calculating health score ranges, and publicly using the information. [Figure 24] FIG. 1 is a block diagram illustrating an example source of validated records suitable for entry in a distributed, secure ledger, such as a blockchain. [Figure 25] FIG. 1 is a block diagram illustrating an example blockchain including a verified individual electronic medical record and a verified organizational electronic medical record. [Figure 26] FIG. 1 is a block diagram illustrating an example blockchain including a smart contract offer. [Figure 27] It shows an example network of interactions between counterparties, including a distributed ledger containing smart contract offers from individual owners of health information. [Figure 28] 1 illustrates an example decentralized marketplace including a distributed ledger of life according to an example implementation of the present application. [Figure 29] 10 illustrates an alternative representation of the present application including health information generated in a distributed manner in response to health appliances. [Figure 30] The present application provides example features and advantages where the use of blockchain or similar technologies can help transform various industries, such as insurance and other health-related industries. [Figure 31] 1 shows a flow chart of one implementation of the present application according to the exchange. [Figure 32] 1 illustrates a flow diagram of an alternative implementation of the present application according to at least a decentralized smart contract model. [Figure 33] FIG. 1 illustrates dynamic activity data integration and data flow according to an example implementation. [Figure 34] 1 is a flow chart illustrating steps associated with an example implementation of the present application. [Figure 35] FIG. 2 is a block diagram illustrating example modules executed by a processor in an example implementation. [Figure 36] 1 shows a graph depicting features associated with walking scores in accordance with the present application. [Figure 37] 1 shows example ambulation score values ​​corresponding to the number of days until ambulation score declines or drops precipitously, and monetary expressions associated with the health reservoir. [Figure 38] 1 illustrates an example implementation including graphical virtual mapping functionality in accordance with the present application. DETAILED DESCRIPTION OF THE INVENTION

[0014] By way of overview and introduction, the present application provides improved systems and methods for integrating information received from multiple separate tracking devices with historical and behavioral information associated with user activities and health status. A state-of-the-art machine learning health and activity classification engine is provided that analyzes information received from various sources, including fitness tracking devices, to classify a user's activities and health status. Information representing an individual's activity units can be flexibly combined to represent specific activities in new and accurate ways. Furthermore, rather than simply determining the specific activities a user may be engaging in, the present application provides a predictive element that defines activity goals that are modified over time to encourage or force the user to adapt to a more (or less) intense exercise regimen.

[0015] In one or more implementations, the classification engine is provided with a sequential and / or continuous stream of passively tracked user information. Such information may include sensor data from various sources, such as pulse monitoring devices, one or more accelerometer-related applications (e.g., pedometers or other motion-sensing applications), temperature detection devices, pulse oximeters, and / or global positioning system ("GPS") devices. In addition to such sensor data, user profile information representing, for example, the user's age, gender, height, and weight, as well as various biological and / or physiological characteristics specific to the individual, is accessed and / or managed. One or more modules may be configured by executing code to aggregate such user profile information. The results of one or more aggregation processes may provide demographic information, which may serve as input to other processes, such as predictors related to people's health and / or activity. Additionally, other sources of information that contribute to determining and / or predicting various forms of user activity and health status may be accessed or otherwise managed by the classification engine. For example, calendar information regarding a user's personal and / or professional schedules can be accessed from one or more computing devices, such as a mobile computing device, and managed by the classification engine. Such calendar information can include workout schedules, personal schedules, work schedules, or virtually any other calendar items. The classification engine can be further configured to access and process various external data sets, such as time of day and specific day, week, and month information, including local weather, to classify and predict user activities and health conditions.

[0016] In this manner, information from various sources, including sensors, user profile databases, and calendar-related information, can be received, managed, and processed. One or more processors can execute code, such as to provide respective modules, to process the information to understand activity patterns and predict and classify specific activities in which a person may be engaged. As a person engages in respective activities, such as one or more of multiple regular workouts, a determination module can assess that the user has reached a plateau and should participate in more vigorous and / or strenuous activities. As a person physically develops, such as as a result of participating in varied and regular exercise, their tolerance improves, reducing the effectiveness of their initial exercise regimen. Thus, one aspect of the classification engine is to access information associated with a user's workouts to detect when the person is ready to progress further in their exercise regimen. The information can then be used to appropriately coach the user, such as through one or more interactive graphical user interfaces provided by a coaching module.

[0017] In one or more implementations of the present application, a user's heart rate can be detected and monitored periodically or continuously. A baseline heart rate and / or calculated MET value can be established in response to such monitoring, and any detected heart rate within a respective range (e.g., 65-75 bpm) can be used to form a determination that the user is at rest. At another time, a determination that the user is performing some activity can be made when the user's heart rate rises, such as by 105 beats per minute ("bmp"). Upon detecting a change in heart rate, one or more algorithms can be executed by the computing device in accordance with the teachings herein to automatically determine and classify each activity and define an activity unit having a start time, an end time, and an activity name. In other implementations, activity units can be defined based on other information sources provided to the classification engine, such as location information.

[0018] Hardware / Software Modules and Implementations Referring now to the drawings, wherein like reference numerals refer to like elements, FIG. 1 illustrates an example hardware configuration for acquiring, processing, and outputting content, such as over a communications network, such as the Internet. As shown in FIG. 1 , one or more of a plurality of user computing devices, such as information processor(s) 102 and user computing device(s) 104, are configured to receive electronic content from and / or transmit electronic content to one or more health devices 105. In the example illustrated in FIG. 1 , such health devices 105 include, but are not limited to, a home medical test kit 105, a blood pressure monitor 105, a wearable band 105, a glucose meter 105, and a weight scale 105. Thus, as shown in FIG. 1 , the present application provides a decentralized system in which data acquisition, data storage, and data processing can occur according to independently operating and configurable devices.

[0019] In one or more implementations, information processed in accordance with the present application can be used to generate a numerical score as a basis for assessing a user's relative health status (i.e., generating a user health score). Particular implementations relating to features associated with calculating and delivering a user health score are more fully shown and described in U.S. Non-Provisional Patent Application No. 14 / 257,855, entitled "AUTOMATED HEALTH DATA ACQUISITION, PROCESSING AND COMMUNICATION SYSTEM," which is assigned to the assignee of the present invention.

[0020] The present application supports computer-based systems and methods for collecting user health-related parameters and a user interface for presenting (e.g., displaying) the data in response to one or more applications executing on one or more user computing devices 104 and / or information processor 102. The computer-based applications can be implemented in a microcontroller including a processor, memory, and code that executes to configure the processor to perform at least some of the functions described herein. The memory can store data and instructions suitable for controlling the operation of one or more processors. Memory implementations can include, by way of example and not limitation, random access memory (RAM), a hard drive, or read-only memory (ROM). One component stored in the memory is a program. Specific examples of verifying, storing, and transacting according to a secure ledger associated with a user's health information (including health score information) are described in further detail below.

[0021] FIG. 2 illustrates one or more functional elements associated with the information processor 102, user computing device 104, and / or fitness appliance(s) 105, including a processing subsystem having one or more central processing units (CPUs) 202 used to execute software code to control the operation of the devices 102, 104, and / or 105, as well as processor-readable media including, for example, read-only memory (ROM) 204, random access memory (RAM) 206, and one or more network interfaces 208 for transmitting data to and receiving data from other computing devices over a communications network, storage 210 such as solid-state memory, hard disk drive, CD ROM, or DVD ROM for storing program code, databases, and application data, one or more input devices 212 such as a keyboard, mouse, trackball, virtual keyboard, touchscreen, microphone, and the like, and a display 214.

[0022] Memory 204 and / or 206 can be persistent or non-persistent storage devices that function to store an operating system for the processor, as well as one or more software modules. According to one or more embodiments, the memory includes one or more volatile and non-volatile memory, such as read-only memory (“ROM”), random-access memory (“RAM”), electrically erasable programmable read-only memory (“EEPROM”), phase-change memory (“PCM”), single in-line memory (“SIMM”), dual in-line memory (“DIMM”), or other memory types. Such memory can be fixed or removable, such as through the use of removable media cards or modules, as known to those skilled in the art. Additionally, one or more remote or local storage devices, such as databases, can include caches, including database caches and / or web caches. Depending on the program, data storage can include flat-file data stores, relational databases, object-oriented databases, hybrid relational-object databases, key-value data stores such as HADOOP or MONGODB, as well as other systems for data structure and retrieval known to those skilled in the art.

[0023] As used herein, the terms "processor" or "computer" generally refer to one or more electronic devices (e.g., semiconductor-based microcontrollers) configured with code in the form of software to execute a given set of instructions. For example, the mobile computing device 104 may include one or more processing or computing elements running a commercially available or custom operating system, e.g., a MICROSOFT WINDOWS®, APPLE OS X, UNIX®, or Linux®-based operating system implementation, such as those found on APPLE IPAD / IPHONES®, ANDROID® devices, or other electronic devices. In other implementations, the mobile computing device 104 may be any custom or non-standard hardware, firmware, or software configuration. The mobile computing device 104 may communicate with one or more remote networks using USB, digital input / output pins, eSATA, parallel ports, serial ports, FIREWIRE®, Wi-Fi, Bluetooth®, RF transmitter / transponder, or other communication interfaces. In certain configurations, the mobile computing device(s) are also configured, through hardware and software modules, to connect to one or more remote servers, computers, peripherals, or other hardware, either over a local or remote network or over the Internet 106, using standard or custom communication protocols and settings (e.g., TCP / IP, etc.).

[0024] Moreover, the information processor 102 referred to herein may include a server, a cloud computing platform, a microcomputing element, a computer(s) on a chip, a home entertainment console, a media player, a set-top box, a prototyping device, or a "hobby" computing element. Such described computing elements are connected, directly or indirectly, to one or more memory storage devices (memories) to form a microcontroller structure.

[0025] Computer memory may also include secondary computer memory, such as a magnetic or optical disk drive or flash memory, that provides long-term storage of data in a manner similar to persistent memory devices. In one or more embodiments, the processor's memory provides storage for application programs and data files, as needed.

[0026] The described processors or computers are configured to execute code written in a standard, custom, proprietary, or modified programming language, such as a standard set, subset, superset, or extension set of JavaScript, PHP, Ruby, Scala, Erlang, C, C++, Objective-C, Swift, C#, Java, Assembly, Go, Python, Pearl, R, Visual Basic, Lisp, or Julia or any other object-oriented, functional, or other paradigm-based programming language.

[0027] Each computer program may include instructions that cause the processor 202 to perform steps that implement the methods described herein. The program may be implemented as a single module or as multiple modules operating in cooperation with each other. The program may include software that can be used in connection with one or more implementations of the present application. For example, a communications subsystem may be provided to communicate information from the microprocessor to a user interface, such as an external device (e.g., a handheld device or computer connected to the communications subsystem through a network). Information may be communicated by the communications subsystem in a variety of ways, including Bluetooth, Wi-Fi, Wi-Max, RF transmission, near-field communication, or other suitable communications protocols. Several different network topologies may be utilized, such as wired, optical, 3G, 4G, 5G, or other suitable networking protocols.

[0028] The communications subsystem can be part of a communications electronic device, including, by way of example, a smartphone or mobile phone, a personal digital assistant (PDA), a tablet computer, a netbook, a laptop computer, or other computing device. For example, the communications subsystem can connect directly through a device such as a smartphone, such as an iPhone®, Google Android® Phone, Samsung Tizen, BlackBerry®, or Microsoft Windows® Mobile-enabled phone, or a device such as a heart rate or blood pressure monitor, a weighing scale, exercise equipment, or the like. One or more of these devices can include or otherwise interface with a module or subsystem communications unit to allow information and control signals to flow between the subsystem and an external user interface device. The communications subsystem can cooperate with a conventional communications device or can be part of a device specialized for communicating information processed by a microcontroller.

[0029] Content provided in accordance with devices 102, 104, and / or 105 can include, for example, numerical, textual, graphical, pictorial, audio, and video material. Communication of such content can occur by and between one or more of the respective devices 102, 104, and 105. Accordingly, in one or more implementations, any of the fitness instruments 105 can employ hardware and software modules to collect and / or receive information, process the information, and transmit the information to device 102 / 104. The information processor 102 and / or user computing device 104 can employ software to enable a communication session, e.g., an HTTP session, to be established between the information processor 102 / user workstation 104 and one or more respective devices, such as to effectuate secure blockchain transactions.

[0030] In addition to monitoring the user's biological and physiological characteristics, the fitness device 105 can be configured to monitor exercise, such as steps identified by interval, size, and / or intensity. For example, steps taken in rapid succession may be more strenuous than those taken over a longer period of time. Moreover, walking uphill may be more strenuous than walking downhill. These and other factors may be included in exercise / activity data from each fitness device 105. Additionally, each fitness device 105 may be configured with global positioning system ("GPS") technology, which may provide location information regarding the height, slope, or other relevant geographic information of the step. In another example, one or more of the fitness devices 105 may be configured with galvanic skin sensors, for example, to measure the wearer's stress level. Furthermore, the fitness device 105 may also be configured for sleep monitoring of the wearer. In one or more implementations, the fitness band 101 is configured with memory for storing information associated with the wearer. In addition to obtaining and / or generating information substantially automatically, information regarding exercise and / or other activities may be input by the wearer, such as via a user interface via BLUETOOTH or other suitable communication with the user computing device 104.

[0031] In addition to receiving and transmitting information associated with the wearer, one or more of the health devices 105 can be configured to process information, such as in connection with information collected and / or processed by the detection components. “Raw” data, which may be in the form of signals, can be collected by each health device 105 and processed to represent biological and / or physiological information. Such information can include personal identification information, blood type, DNA information, weight information, blood pressure information, and other medical, physiological, and / or biological information. Each health device 105 can further be configured with one or more algorithms stored in memory that, when executed by the processor, calculate information such as the wearer’s health score. The wearer’s health score (and / or processed biological / physiological information) can then be transmitted to the information processor 102 / user computing device 104 for further processing, such as establishing a secure private object for storage and verification in a verified ledger record (e.g., blockchain) or other suitable technology.

[0032] Self-learning activity identification Referring now to the drawings, in which like reference numerals refer to like elements, FIG. 3 is a block diagram illustrating components associated with a self-learning activity identification system 300 according to one implementation of the present application. As shown in FIG. 3 , primary data 302 can include static and semi-static values ​​associated with a user's age, gender, height, and weight, for example. Dynamic primary data can include sensed or detected dynamic values, such as time of day, season, heart rate (pulse) rate, and acceleration. Derived data values ​​304 can include profiles associated with heart rate, rate, and acceleration, variability, predictability measures, and cluster IDs. Activity class list 306 represents, for example, activities in which a user can engage and can be used by the classification engine to determine and classify the detected activity according to biological and / or physiological values, such as pulse rate, body temperature, blood pressure, or the like. Examples of user activities can include, for example, running, walking, swimming, or participating in various sports. In one or more implementations of the present application, the classification engine can be trained, such as for future classification of activities. For example, the training section 308 identifies the start and stop of training actions, allowing a user selectable options to identify each activity that has occurred or may be occurring. Machine learning methods 310 can include cluster-based learning, such as for pattern recognition (e.g., K-means clustering, self-organizing mapping (SOM, Kohonen), Gaussian mixture modeling (GMM), support vector machines (SVM), case-based reasoning (CBR), ANN, and guided trees).

[0033] FIG. 4 illustrates an example flow of information and steps associated with a self-learning activity identification system according to one or more implementations of the present application. Sensor information 402, user profile information 404, and external information 406 can be transmitted to or otherwise accessed by a classification engine 408. At step 410, a change is detected, for example, in a user's heart rate, respiratory rate, or other parameter. As previously described, a user's resting heart rate can be established as a parameter, and when one or more processors acquire detected information that the user's heart rate exceeds the resting parameter, a determination can be made that some user activity, such as exercise, is occurring. The activity is then classified (412) as a specific activity according to one or more machine learning methods 310, along with a start time and, ultimately, an end time after the end of the activity is detected. It should be understood that when one activity ends, logically, another activity begins, provided it is a "resting" activity. Whether the end and start of an activity are correctly determined or established by the algorithm is a facet of one or more implementations of the present application.

[0034] Classifying an activity is not trivial and can be the result of complex processing of dynamic information received from multiple sensing devices, along with profile and external data accessible to one or more computing devices. For example, using GPS and calendar information, one or more machine learning techniques detect a pattern of a user going to the gym between 4:00 and 5:00 PM on Sundays. The user may also provide information, via a feedback loop (shown and described herein), that the user most enjoys Body Pump (weightlifting) activity and participates in Body Pump challenges with other users. Upon detecting a change in heart rate from baseline, signals available from currently known or future-developed tracking devices can be processed to classify the respective activities in which the user is engaged. In another example, user profile information may indicate that the user participates in skiing and sailing activities. Various forms of dynamic and external information can be used to correctly classify the detected activities. For example, a computing device considering GPS coordinates identifying the user as being on top of a mountain, calendar information indicating that it is winter, and weather information indicating that it is snowing where the user is located would incorrectly classify the user's activity as sailing. Such an activity would take into account all of the relevant composite information and, in relation to the user's elevated heart rate, correctly classify the activity as skiing. Moreover, monitoring heart rate information (and other physiological and metabolic-based information) allows for the period during which the activity is occurring.

[0035] Thus, the present application provides sophisticated processing of activity-based information to correctly classify an activity. This can be the result of processing specific and distinct time periods and various information as shown and described herein. For example, a discrete determination that a user was very active for five minutes, relatively inactive for the next two minutes, and active again for the next five minutes can result in three activity units being defined, or three minor units of a major activity unit, each with its own start and end time. Location-based information over the same 12-minute period could indicate that the user traveled five miles. An incorrect classification of the activity would be that the user walked quickly to the bus stop, rode the bus for four or five miles, got off the bus, and continued walking quickly. However, the present application correctly classifies the user's activity as riding a bicycle, with the periods of vigorous activity representing pedaling uphill and the periods of relative inactivity representing coasting downhill. Rather than analyzing only one or two data points, the automated classification engine according to the present application provides comprehensive and complex processing of many sources of information, including extrinsic factors such as time of day, day of the week, location, and weather, to correctly classify an activity. For example, the same 12 minutes of activity information identified as occurring late on a winter night during heavy snowfall would not be classified as riding a bicycle. This represents a complex arrangement and cooperation-based system of individual hardware and software devices that contribute to the long-term analysis of distinct parts of an activity for improved and correct classification of such activity.

[0036] In addition to providing automatic classification of user activities, the present application provides fault tolerance and provides users with the ability to reclassify activities. In one or more implementations, a user interface is provided that includes interactive, selectable elements that represent user activities over time (e.g., a graphical representation of a timeline). For example, a timeline is generated and displayed that identifies when the user woke up, the duration, distance, and number of steps walked before getting to their car, the distance and time taken during motorized transportation, the distance, duration, and number of steps taken while running, the period during which the user was subsequently inactive (e.g., sitting behind a desk), the amount of food consumed at lunch, the duration, distance, and number of steps taken during an afternoon run, the period during which the user was again inactive (e.g., sitting behind a desk), and the duration and type of activity the user engaged in after work (e.g., 15 minutes, bike ride, 5 miles). In one or more implementations, each of the respective timeline items includes flexible activity units that are selectable and allow the user to take corrective action, such as to reclassify the activity, merge several activity units, correct the active period, or correct any other data elements associated with them. Each time a reclassification or other correction is made, such information is preferably provided back to the classification engine for further learning. Optionally, changes to an activity unit's label, duration, and attachment to other activity units can be included in an audit trail for reporting and compliance checking purposes.

[0037] 5 illustrates an example display screen 500, such as may be provided on a mobile computing device (e.g., a smartphone), in which user activity has been detected substantially automatically, such as during a specific time period (e.g., 0.36 h), by heart rate monitoring. The user's activity may be determined, for example, by machine learning in accordance with the teachings herein. The user may further be prompted to communicate with an avatar, for example, to indicate how the user felt during each activity. Other information shown in FIG. 5 includes a graphical representation of the user's heart rate over time, as well as a determination of average heart rate, average power output, and tracking source.

[0038] Continuing with reference to process / information flow 400, at step 414, an output representing the classified activity is provided, for example, to a computing device associated with the user. For example, a graphical representation of a timeline is generated and provided (416), a health score for the user (as shown and described herein) is generated and provided (418), an update is provided to one or more social networks (420), or other suitable output is provided. A determination is then made whether the activity information, as provided in the timeline, should be reclassified (422). If so, the process branches to 408, and information associated with the reclassified items is provided back to the classification engine. Alternatively, if the determination at 422 is that no reclassification is required, the process branches back to step 414, and an output is provided.

[0039] In one or more implementations, personal activity units representing discrete activity periods are automatically classified by a computing device according to information processed by a health and activity classification engine. For example, a user spends 10 minutes at the gym performing a body pump activity, causing their heart rate to rise to 110 bpm. After a while, the user's heart rate returns to a resting state (e.g., 70 bpm). In this example, the determination that the heart rate has returned to a resting state signals the end of the discrete activity unit. The user then spends 5 minutes on a treadmill, causing their heart rate to rise again to 105 bpm, thereby signaling another discrete activity unit to the computing device. The user then spends 5 minutes on a stationary bicycle, which is interpreted as a third discrete activity unit. Each of these activity units can be joined into a primary activity unit, e.g., "gym workout," having an overall duration spanning from the start of the body pump activity to the end of the treadmill activity. This joinder is distinct from merging, an alternative method of reducing the number of activity units by combining two or more units.

[0040] An advantage of the present application is that activity units are provided in a flexible format that allows a user to reclassify activity units and combine multiple activity units into a unified group, such as multiple "minor" activity units into a single "major" activity unit, or merge two or more units into a single major or minor unit. Continuing with the previous example, a user reviews a timeline of activities that occurred while at the gym. Upon detecting three respective activity units (body pump, treadmill, and stationary bike), the user selects, using one or more graphical screen controls provided in the user interface, to classify the multiple activity units into a single activity unit, "gym workout." In this way, in accordance with the present application, minor activity units can be classified as part of a single major activity unit. Similarly, a user can be provided with the option to break down a single activity unit into multiple activity units. Once an activity is reclassified, including into a single (major) unit or multiple (minor) units, information is provided back to the classification engine to be used for future classification of the user's activities and for audit trail purposes.

[0041] While much of the description herein identifies changes in a user's heart rate to trigger the determination of a particular user activity, the application is not so limited. As described herein, the classification engine is configured to access or otherwise receive information from sensors, personal, and external sources. Any one or combination of information sources can be used as a trigger by the classification engine to detect and / or classify each user activity. For example, information from a GPS device and a calendar source can be processed to determine that the user arrived at the gym at a certain time on a certain day of the week. This combination of data points can be used to define the boundaries of a major activity unit: gym workout. While a user engages in various athletic activities, such as body pumping, stationary biking, and running on a treadmill, the classification engine can define minor activity units associated with each and generate and provide one or more summaries representing the particular activity, the duration of such activity, a graphical representation of the user's heart rate over the duration of such activity, or the like. In this manner, the classification engine can be used to broadly classify multiple activities into a single major activity unit while maintaining and providing information about each of the minor activities that comprise the major unit.

[0042] In one or more implementations, the present application provides computer-implemented systems and methods configured to acquire health-related and / or medical-related data and process the data. For example, the systems and methods herein provide feedback substantially in real time via online and / or mobile platforms. Using the systems and methods disclosed herein, information can be received from a user device, and the information can be processed to provide various forms of feedback, such as alerts and notifications. Upon accessing one or more modules associated with the present application, such as a user submitting initial registration information, a baseline assessment in the form of a health score can be calculated and provided to the user, at no cost or a nominal fee. This results in the user engaging with useful health-related information, increasing the likelihood that the user will continue to engage with and regularly provide health-related information that can be used, for example, in connection with a metric health model score and / or a quality of life model score.

[0043] In one or more implementations, a baseline assessment (e.g., health score) is based on four data points: age, height, sex, and weight (e.g., instant health score). In one or more implementations, an organization's health score distribution is based on exciting data, etc. Health scores can be calculated across an enterprise, such as through social networks, gamification, or in a stateless manner that excludes other personal identifiers.

[0044] The present application provides a distributed system in which data acquisition, data storage, and data processing can be used to generate a numerical score as a basis for assessing a user's relative health status. In one implementation, a computer-based application is provided for collecting a user's health-related parameters, and a user interface is provided for presenting (e.g., displaying) the data. The computer-based application can be implemented on a microcontroller including a processor, memory, and code executing thereon to configure the processor to perform at least some of the functions described herein. The memory can store data and instructions suitable for controlling the operation of one or more processors. Memory implementations can include, by way of example and not limitation, random access memory (RAM), a hard drive, or read-only memory (ROM). One component stored in the memory is a program. The program includes instructions that cause the processor to perform steps that implement the methods described herein. The program can be implemented as a single module or as multiple modules operating in cooperation with each other. The program can include software that can be used in connection with one or more implementations of the present application.

[0045] A communications subsystem may be provided for communicating information from the microprocessor to a user interface, such as an external device (e.g., a handheld device or computer connected to the communications subsystem through a network). Information may be communicated by the communications subsystem in a variety of ways, including Bluetooth, Wi-Fi, Wi-Max, RF transmission, near field communication, or other suitable communications protocols. Several different network topologies may be utilized, such as conventional wired, optical, 3G, 4G, or other suitable networking protocols.

[0046] The communications subsystem can be part of a communications electronic device, including, by way of example, a smartphone or mobile phone, a personal digital assistant (PDA), a tablet computer, a netbook, a laptop computer, or other computing device. For example, the communications subsystem can be directly connected through a device such as a smartphone, such as an iPhone®, Google Android® Phone, BlackBerry®, or Microsoft Windows® Mobile-enabled phone, or a device such as a heart rate or blood pressure monitor, a weighing scale, exercise equipment, or the like. One or more of these devices can include or otherwise interface with a module or subsystem communications unit to allow information and control signals to flow between the subsystem and an external user interface device. The communications subsystem can cooperate with a conventional communications device or can be part of a device specialized for communicating information processed by a microcontroller.

[0047] It is recognized that users may become confused as to which of each device to activate and when. Commonly referred to herein as "double counting," users may inadvertently count the same activity and / or health-related event more than once, which can result in errors and calculations that distort the user's overall health score. Moreover, some users may provide unclear or incomplete replies and even intentionally misrepresent the facts. The automated processes and devices shown and described herein reduce or eliminate the opportunity for users to accidentally or intentionally distort or misinterpret various activity information. Moreover, as described, audit trails can be created to assist users in classification and enforce compliance.

[0048] When communication electronics such as those of the type described herein are used as external user interface devices, the display, processor, and memory of such devices can be used to process health-related information, such as to calculate and provide a numerical assessment. Otherwise, the system can include a display and memory associated with the external device and used to support real-time or other data communication. More generally, the system includes a user interface, which can be implemented, in part, by software modules running within a microcontroller's processor or under the control of an external device. In part, the user interface can also include an output device, such as a display (e.g., a monitor). Displays can include, for example, organic light-emitting diode (OLED) displays, thin-film transistor liquid crystal displays, and plasma displays.

[0049] Additionally, one or more computing devices, including a server, smartphone, laptop, tablet, or other computing device, can transmit electronic content to and / or receive electronic content from a health band worn by a user. The content can include, for example, numerical, textual, graphical, pictorial, audio, and video material. Such communication can occur directly and / or indirectly between the server and the health band, such as via a mobile computing device, such as a smartphone, tablet computer, or other device. Alternatively, such communication can occur between the server and the health band without the use of a computing device. Thus, in one or more implementations, the health band can employ hardware and software modules to collect and / or receive information, process information, and transmit information between the health band and a server and / or between the health band and a mobile device.

[0050] In addition to communication-related modules, each fitness device 105 can be configured with one or more hardware and / or software modules that configure the fitness device 105 to track various forms of the wearer's biological / physiological characteristics, movement, and / or activity. For example, the fitness device 105 can be configured with one or more sensors and software that collect information regarding the wearer's biological / physiological characteristics, such as body temperature, heart rate, heart rate variability (“HRV”), blood pressure, sweating, or the like. The sensors can be used to directly collect health information about the user and transmit that information to the user computing device 104 and / or the information processor 102. A biosensor can be placed in contact with the user's body to measure vital signs or other health-related information from the user. For example, the biosensor can be a pulsometer, heart rate monitor, electrocardiogram device, pedometer, blood glucose monitor, or one of many other devices or systems worn by the user in contact with the user's body to detect the user's pulse. The biosensor can include a communication module (e.g., a communication subsystem) to communicate sensed data, either wired or wirelessly. The biosensor can communicate the sensed data to a user interface device, which then communicates the information to a microcontroller. Optionally, the biosensor can communicate the sensed data directly to the microprocessor. The use of a biosensor provides a degree of reliability to the reported data, as it eliminates user error associated with manually self-reported data.

[0051] In one or more implementations of the present application, one or more algorithms can be implemented by a server, a mobile device, and / or a health band based on a user's heart rate. One or more modules can automatically detect the user's average heart rate during the day and convert it into daily METs. For example, the user's heart rate can be monitored over time and then averaged. User activity can then be determined and / or detected during periods when the user's heart rate is elevated.

[0052] In one or more implementations, the present application provides a health platform, such as for calculating a health score and for implementing many of the features shown and described herein. A health platform system according to the present application can be accessed via an internet web browser software application (e.g., CHROME, FIREFOX, SAFARI, INTERNET EXPLORER), by using a desktop or laptop computer, and from a mobile device, such as a smartphone or tablet, via a mobile-optimized version of a website. Additionally, one or more graphical user interfaces provide access to users, such as via screen controls (e.g., buttons, icons, drop-down lists, radio buttons, check boxes, text boxes, or the like) for entering, viewing, modifying, and / or deleting information. For example, information, such as that related to indoor and outdoor activities, can be inserted manually via web forms or via a mobile platform, and users can also choose to upload images along with information associated with their activities. Alternatively (or additionally), data input can occur substantially automatically, such as via an import process of one or more files formatted in one of a variety of file types (e.g., TXT, DOC, PNG, JPEG, GIF, GPX, and TCX).

[0053] The health platform provided in accordance with the present application may comprise a smartphone software application, generally referred to herein as a "tracker application," for tracking fitness activities in an easy and automatic manner (in addition to providing manual input), and the recorded / tracked activities may be automatically uploaded to the health platform. The tracker application may be provided for devices running the IOS, Android®, and BlackBerry® operating systems and may be provided free of charge to users. For example, workout data may be uploaded by the tracker application.

[0054] In one or more implementations, the present application provides a tracker application for tracking a user's fitness activity and can be implemented on devices running iOS, ANDROID®, WINDOWS® PHONE, BLACKBERRY®, and other suitable mobile device operating systems. Outdoor and indoor activities can be tracked, and data uploaded to a server computer or other device can be provided in a secure format. The data can be seamlessly and automatically integrated to calculate a user's health score. For example, daily activity measured by a step counter / pedometer or other similar device can be integrated using the systems and methods shown and described herein.

[0055] Passive Tracking and Feedback Loop Module In one or more implementations, an interactive interface is provided to interact with the user, including to prompt the user to provide health-related information. Generally, notifications can include surveys or prompts for information entry and can be presented by avatars or virtually any other interactive mechanism.

[0056] As previously described, the present application can be configured to provide a representation of "minor" activity units categorized under each "major" activity unit. Additionally, as a user participates in an exercise regimen, one or more activity goals can be defined that are modified over time to encourage or compel the user to participate in more strenuous activities. In one or more implementations, an avatar or other interactive interface can prompt the user to adjust their major activity units to encourage them to participate in exercise activities that may require more effort or energy. Upon receiving an indication that the user is agreeable, the present application can dynamically adjust minor activity units, such as increasing the length of time for performing one or more of the activity units or an adjustable setting associated with a particular piece of equipment (e.g., treadmill resistance). In this manner, the present application can be configured to monitor and classify each activity and adjust such activity in one or more ways to enhance the exercise regimen and improve fitness.

[0057] In addition to the baseline calculated health score, the present application includes one or more modules that provide a form of health score screener that can provide health score values ​​beyond the individual level. Providing health-related information associated with populations in a concise, anonymous, and meaningful format can assist many industries and technologies. For example, insurance companies can use population-based health scores for risk modeling, helping employers reduce their employee insurance premiums and improve their employee health by identifying specific at-risk populations. Populations can be defined in myriad ways, such as by geography, socioeconomic strata, gender, and occupation, to name just a few. Health-related information, including population-based health scores, can identify return on investment ("ROI") and the associated costs of inactivity.

[0058] In a broad aspect, a method according to the present invention can be understood as providing for collecting health-related information, processing the information into a health score, and publishing the health score. A system for implementing the method can include a computer having a processor, memory, and code modules executing on the processor for collecting, processing, and publishing information. The baseline information can be relatively small and determined and provided in response to an initial response received from the user, which can engage the user in the future. Furthermore, population health score information can be provided, such as in response to demographics. Additionally, changes in heart rate information can be detected and used to determine (including automatically) the user's respective activity.

[0059] In one or more implementations, a daily, substantially real-time calculated health score is provided and integrated in conjunction with feedback loops and artificial intelligence (e.g., avatars) that enable users to learn about their health and lifestyle. The health score can be provided with content in conjunction with activity tracking, sleep monitoring, and / or stress exposure. In conjunction with activity monitoring, the user's heart rate and / or calculated METs can be based on information detected by a chest heart band, smartwatch, soft sensors, contact devices, nanosensors, and / or other suitable technology.

[0060] Moreover, one or more goals and / or challenges can be tailored to an individual's lifestyle. The result can be a relevant, easy, and enjoyable factor that indicates the degree to which a user and / or organizational members are committed to maintaining good health in accordance with the teachings herein. Moreover, reporting modules can be provided for individuals and / or various groups of people, such as company divisions, profit centers, the entire company, and even geographically (e.g., across one or more countries). The result can be significant savings in healthcare-related costs.

[0061] In a non-limiting use, the passive tracking device can include one or more passive tracking devices. In a similarly non-limiting use, the processor can include one or more processors. In a still further non-limiting use, the database can include one or more databases.

[0062] As described herein, sensors can be used to collect information, and activities can be classified at least in part based thereon. For example, referring to display screen 600 shown in FIG. 6 , graphical screen controls can be provided for a user to enable one or more sensors. Selections can be made to activate and / or identify each active sensor. Heart rate 602 identifies information from a suitable device capable of detecting and / or storing periodic, daily heart rate information, such as by an APPLE WATCH. Step count 604 identifies step count information from any suitable device capable of detecting and / or storing periodic, daily step count information, such as by an APPLE WATCH and an IPHONE®. Travel activity 606 identifies travel information that can be received from a suitable device, such as an IPHONE®. In one or more implementations, determinations can be made to detect, for example, whether a user is stationary, walking, biking, traveling in a vehicle, or engaged in other activities. Location 608 identifies location information that can be received from a suitable device, such as by an IPHONE®. In one or more implementations, a low-precision, background mode is used to conserve battery power and rely less on GPS. Such a mode can include location cues from cell tower changes and visits to locations as determined by the respective device, such as an IPHONE®.

[0063] In one or more implementations of the present application, the data is broken down into 10-minute episodes. This results in 6 episodes per hour, with a maximum of 144 episodes per day. An episode can only be generated if heart rate data and / or step count data is present. Recent episode data (e.g., the last 1,000 episodes) can be supported. Various episode data can include:

[0064] Date. The start of each episode.

[0065] Heart rate. Average heart rate during each episode.

[0066] Steps. Step count for the episode. If multiple devices report step data, the app selects the highest contributing device for each episode.

[0067] Movement activities. Movement activities for each episode.

[0068] Position. The course position for each episode.

[0069] Distance. The change in course distance from each episode.

[0070] Inferred activities. Activities that are statistically inferred by a trained classification model.

[0071] Certainty. The certainty of the above guess, 0-1.

[0072] Confirmed Activity. The confirmed activity of the episode. This activity is manually set by the user to form the ground truth for training the classification model.

[0073] In one or more implementations, the user is prompted to confirm the number of data associated with the episode (e.g., 50) that is available to train an initial classification model. Episodes are preferably confirmed against the classification model for future prediction of such episodes. For example, if the user takes a few steps into an office building, the user may confirm the episode as "none" rather than "walking" because the user does not want to classify such limited exercise as a walking workout. As a result, the value "none" represents inactivity.

[0074] Additionally, the present application supports mapping terms that would not otherwise have an associated meaning. For example, a user may play badminton but not tennis. In such a case, the user may identify the episode of playing badminton by selecting the value representing tennis. In this way, the movements occurring while playing badminton are recognized and described.

[0075] In one or more implementations, the user is prompted to identify episodes in which the activity occurs for a continuous period of time, such as 10 minutes. For example, a user may go for a walk from 14:15 to 14:45 and identify episodes at 14:20 and 14:30 as walking, but not at 14:10 and 14:40. Such episodes are considered "dirty" and not suitable as ground truth for training the classification model. In such cases, the user is prompted to classify the exercise as "unconfirmed." In one or more implementations, the classification model can infer an activity in response to receiving several user classifications associated with a particular activity, some of which may be inconsistent, and then infer a classification based on the majority of the received user-based classifications.

[0076] If a user unintentionally acknowledges an episode, one or more options may be provided for the user to clear the acknowledgement and subsequently re-acknowledge the episode and / or acknowledge future episodes associated with the same activity.

[0077] After a predetermined number of episodes (e.g., 50 episodes) have been reviewed by the user and at least 25 of them have been classified as "none," the user may be prompted to further train the classification model. Options provided in an example training interface are shown in display screen 700 of FIG. 7. The trained model may be implemented with an artificial neural network ("ANN"). The model is based on and immediately applied to up to 10,000 recent episodes in the app, both reviewed and unreviewed. New episodes may be automatically classified based on the model.

[0078] Periodically, such as when the user first interacts with the system and / or reviews one or more episodes, the user may be prompted to retrain the model. In such cases, the user may be prompted to reexamine previously classified episodes and revise such classifications as necessary. For example, the user may be prompted to determine whether an inference is incorrect based on the episode's sensor data points. The user may further be prompted to determine whether additional reviewed episodes for one or more activities could improve accuracy.

[0079] An example interactive data entry display screen 802 is shown in Figure 8, which may include, for example, graphical screen controls within a user interface provided on device 104, allowing a user to enter activity information, schedule information, and respective values ​​associated with duration, distance, heart rate, energy, and images, which may be used, for example, for manual data entry in connection with the teachings herein.

[0080] Once a user has confirmed a predetermined number of episodes, such as at least 200 episodes, and at least 100 of them are classified as none, the user can engage in a process to cross-validate the episodes. For example, a 10-fold cross-validation can be performed using 90% of the confirmed episodes as a training set and the remaining 10% as a testing set. The process can be repeated 10 times, each time using a different 10% of the data as the testing set. After this cross-validation process is complete, the accuracy of the model is displayed and formatted as a percentage. Because there may be some randomness, especially for one or more activities with few confirmed episodes, the user may sometimes repeat the cross-validation step several times to properly measure accuracy.

[0081] If the user sees low accuracy (e.g., <90%), the user may attempt to configure the ANN's hyperparameters: one hyperparameter is the hidden layer width, which is the relative width of the hidden layer compared to the input layer, between 0 and 1. Wider hidden layers allow the algorithm to learn more complex patterns, but may lead to overfitting and therefore reduced accuracy.

[0082] Another example hyperparameter is hidden layer count, which is the number of hidden layers, between 0 and 4. More hidden layers allow the algorithm to learn more complex patterns, but can lead to overfitting and therefore reduced accuracy.

[0083] Another example hyperparameter is the mean squared error, which is a stopping threshold, between 1 and 0.00001. Once the mean squared error on the training set falls below that threshold, network training stops. A lower threshold allows the algorithm to train to higher accuracy, but can also lead to overfitting, and therefore reduced accuracy.

[0084] After changing one or more hyperparameters, the user can re-run cross-validation one or more times to see if there is an improvement or degradation in accuracy. If changing the mean squared error, the user may first try orders of magnitude, i.e., 1, 0.1, 0.01, ..., 0.00001. In one or more implementations, the default values ​​for the hyperparameters can be 0.3 hidden layer width, 1 hidden layer, and 0.001 mean squared error threshold.

[0085] "Pay as You Live" and Insurance Implementation In one or more implementations, health scores can be integrated into insurance products, such as health insurance, vehicle insurance, life and / or health insurance. Using the systems and methods disclosed herein, such as in connection with generating and receiving information substantially in real time via online and / or mobile platforms, insurance companies can identify risk exposure and provide pricing controls to users accordingly. For example, one or more sensors, such as biosensors, can be used to collect and transmit health information about a user purchasing health insurance. One or more sensors can be configured within a health band, as shown and described herein, to collect health and activity information about the user. Furthermore, a tracker application can track fitness activities and generate and transmit activity information to one or more computing devices. Introducing such technology into the insurance and / or reinsurance industry can foster entirely new user experiences and provide significant cost savings for both insurance companies and insureds. For example, information related to respiratory activity, step counts and other physical activity, heart rate, sleep, nutritional intake, stress, blood sugar, and food intake can be generated and transmitted to one or more computing devices and used to generate notifications for respective parties, such as the insured user. Such notifications can alert users to various conditions, leading to changes in behavior.

[0086] FIG. 9 is an example graph 900 identifying six opportunities that may be provided in accordance with one or more implementations of the present application. For example, opportunities may be provided in accordance with the present application by providing offers that align incentives for long-term behavioral change, such as in connection with chronic disease and pain management. One or more modules implemented by the present application support new insurance offers based on outcome-based payment models. This opportunity provides for the development of new insurance products that are contingent on outcomes and corresponding value to users. Additionally, mobile health ("mHealth" or "m-health") technology as shown and described herein may be employed to provide insights into insured behavior and provide actionable information that can improve outcomes and reduce costs.

[0087] Thus, the teachings herein provide a complex data core in response to integrated monitoring techniques, encouraging insureds to improve their behavior and reduce health risks. Furthermore, as more information attributable to each lifestyle is acquired and processed by the present application, a better understanding of the insured's health status can be achieved. The results can be, for example, improved income streams to supplement insurance income, reduced pressure on insurance underwriting policies, and significant cost savings for insured users.

[0088] FIG. 10 shows an example chart 1000 that broadly identifies populations that can be successfully served by the present application owner. For example, as shown in FIG. 10, a relatively large portion of the population, 40-60%, is successfully served by artificial intelligence-based interactions and has fully automated outreach. Because members are relatively healthy, the risk classification for this proportion of the population is relatively low. As a result, this proportion of the population incurs relatively low costs for health and / or life and / or health insurance. Also successfully served by the teachings herein is a significant portion of the population, roughly shown as 20-25%, that has some health risk and accounts for a somewhat larger proportion of insurance-related costs, such as 15-20% of total costs. Outreach to this proportion of the population is successfully automated, such as in response to email, mobile phone communications, chatbots, SMS texts, and online mobile applications. Also represented in Chart 1000 is a smaller percentage of the population, such as between 5-15%, who are at higher risk and likely to suffer from chronic health and pain issues, and who bear a majority of total insurance-related costs, such as 30-40% of the total. This percentage of users may be reachable by some form of technology, with a significant portion of the information representing this percentage of users provided by care management centers and / or medical professionals. Moreover, a relatively small percentage of the population, shown in Chart 1000 as between 2-3%, is not necessarily well served by the technical features shown and described herein. This small percentage of the population may bear 40-50% of total insurance-related costs, suffer from chronic conditions, and require active care and case management.

[0089] Thus, the present application provides opportunities and implementations for new insurance-related products configured to incorporate M-health information, including health scoring, and moving toward real-time lifestyle-based underwriting and new products. Insured individuals can benefit from the teachings herein because their behavior and activities are monitored and, with substantially no human interaction, information is generated and transmitted to one or more computing devices associated with the insurance company, improving speed and efficiency. Insurance companies similarly benefit because the health and condition of the insured party (and the insured object) remains healthy, thereby extending the period over which premiums are paid.

[0090] FIG. 11 is an exemplary diagram 1100 illustrating traditional life and / or health insurance and the improved life and / or health insurance offering of the present application, plotting insurance costs (Y-axis) and premium payments over time (X-axis). In traditional insurance models, results are relatively linear and static. However, the improved life and / or health insurance of the present application represents the dynamic nature of real-time lifestyle information. The dynamic relationships depicted in the improved life and / or health insurance graph represent events and illnesses in response to information regularly and automatically received from one or more online and mobile platforms. Additionally, health scores are calculated substantially in real time, and insurance rates are determined accordingly, such as for new insurance products and / or to adjust rates on existing ones.

[0091] As a result, in one or more implementations, people's behavior and health status can be automatically monitored, and information can be submitted dynamically and regularly, such as to calculate a real-time health score. This enables insurance companies to develop lifestyle-based insurance products, such as those related to health insurance and life and / or health insurance, that are much more accurate, provide better value to users, and increase profitability for the company. For example, the present application implements blockchain and electronic information transfer technologies in a substantially automated insurance offering. The present application enables companies to assess risks that may otherwise go undetected, allowing improved underwriting and pricing resources to be developed to ensure comprehensive underwriting and pricing strategies that ensure sustainable, long-term profitability and growth. Using the technology shown and described herein, users can access real-time information that helps them reduce risk and positively impact outcome-based pricing and / or profits.

[0092] In one or more implementations, the information received and processed in accordance with the teachings herein can be applied to determine discounts, rebates, or other financial benefits associated with insurance premiums. Figure 12 shows an example chart 1200 that identifies discount percentages provided to users in response to information actively provided by the user, such as in response to a questionnaire or one or more interactive mechanisms (e.g., data entry forms). Upon reaching a certain level, such as by completing 13 or more personal questionnaires over time, the user's status increases (e.g., to silver, gold, platinum), and the benefits provided by the present application, such as discount percentages on insurance premiums, may increase accordingly.

[0093] In addition to the information received in response to the questionnaire, other information can be attributed to define each user's status and influence benefits, such as insurance premium discount rates. FIG. 13 shows an example display screen 1300 provided on a mobile computing device that identifies each user's current health score, as well as a bar graph that identifies the user's daily health score progress over the past week. Additionally, as shown in display screen 1300, graphical screen controls are shown that provide feedback associated with lifestyle, body, and emotions. Furthermore, display screen 1300 includes a status line that indicates long-term financial benefits and goals attributed to the health score and its impact on the user's insurance premiums.

[0094] The inventors have recognized that some users of the present application may be willing to accept some financial benefit, such as a reward or insurance discount, in exchange for greater control over the user's personal health information and data privacy. FIG. 14 shows an example data entry display form 1400 that includes options for the user to select in connection with maintaining control over their information and privacy. As shown in display screen 1400, options are available for the user to maintain data privacy, with one option disclosing the user's health score, as used in calculating the user's health score, a limited number of times (e.g., twice), but not disclosing any health data. Another selectable option allows the user to indicate a willingness to share the health score for a given period, such as a full year, but not to share any of the health data used in calculating the health score. Yet another selectable option shown in example display screen 1400 allows the user to identify their family members to share their family health score, which may be a calculated aggregate value representing the family's health status as a whole, without identifying and disclosing any individual members of the family's health score. Also shown on display screen 1400 is a chart identifying the percentage of premiums, including 90% of twice-yearly index-based and family-based policies, that are identified as directly affected by the amount of information users are willing to share or otherwise, or in limited contexts, publicly disclose. As a result, the present application provides substantially persistent, real-time information that improves automated underwriting / risk modeling and rewards customers accordingly.

[0095] In one or more implementations, the present application processes health scores and / or related information associated with multiple people as a whole. For example, information processor 102 and / or user computing device 104 can process user information associated with multiple users as a whole and define bands or tiers of users accordingly. Examples of such bands can be defined as shown in display screen 100 (FIG. 10) and display screen 1100 (FIG. 11).

[0096] For example, a band of individuals deemed to be very healthy and extremely low risk, at least according to their respective health scores, may meet the "Platinum" user band and be offered a maximum discount on their insurance premiums, such as 30%. Each other user band or tier, as defined or represented relative to the user's health score, may be offered other corresponding benefits, such as a discount on their insurance premiums. In various implementations, other types of information, such as representing user responsiveness, lifestyle, activity, or the like, can be incorporated into the aggregate user group to define a user band and determine each individual within it. Processing each user's information as a whole, such as related to their health score, improves upon known traditional computing and technical processes associated with underwriting / risk modeling to provide customers with benefits directly attributable thereto.

[0097] Various implementations of the present application, including pay-for-life insurance models and processing information for multiple people as a whole, as shown and described herein, can provide improved profitability for insurance companies while passing on cost savings associated with improved technical features to customers. For example, 30 bands or tiers of users can be defined according to various medical criteria, including the user's health score and / or health score range, as shown and described herein. Users can be informed that multiple tiers or tiers exist and that various degrees of benefit (e.g., premium discounts, coverage extensions, or other benefits) are associated with each tier. This motivates users to improve their health score, for example, by eating right, exercising regularly, and following medical advice, thereby lowering their overall risk level to insurance companies.

[0098] Once a user is classified into a band or stage, mobility to another band, such as one offering greater or lesser benefits, is available. However, while users may engage in health activities in an effort to improve their respective health scores in hopes of moving to a more beneficial band, it is recognized that such improved behavior often fades over time. It would be inefficient to repeatedly process information associated with thousands or millions of users to reclassify users back and forth into various bands as behavioral habits that affect health scores change. Accordingly, the present application supports a mechanism for implementing a module to detect when a user's behavior improves (e.g., engaging in activities that improve their health score) and then delay changing the user's classified user band for a period of time or according to other predetermined criteria. In one or more implementations, a mechanism is supported for detecting when a user's health score improves and initiating a process to track the user over time for continued and consistent improvement. Additionally, the present application can be configured to recognize patterns of behavior over time that may inaccurately represent a user's behavior and health trends. For example, a user may consistently exercise, eat right, and follow medical advice at the beginning of each month, but by the end of each month, that same user is not exercising, eating unhealthy foods, and not taking prescribed medications. Because the user will repeat this behavior, it would be inefficient to move this user in and out of higher stages or bands each month.

[0099] The present application improves computing and processing technology by including and using a form of dampening mechanism to prevent changes in a user's respective band or stage. For example, after a user has actively trended for a period of time, e.g., one month, three months, six months, or more, the user can transition from their current user band to a different one. An example of activity detection is provided in display screen 1300 (FIG. 13), which shows a spike in the user's health score at the beginning of the week, followed by a decrease in the user's health score over time. Using the dampening mechanism of the present application, inconsistencies in the user's health score become apparent so the user is not classified into different stages or bands.

[0100] In one or more implementations, the present application supports a messaging interface that can be configured for each stage and / or band of a user, as well as for users engaged in various activities or lack thereof. For example, a database accessible by the information processor 102 and / or the user computing device 104 can store messages or elements associated with each user band. Criteria associated with user behavior, user health scores, user bands and / or stages can be used to identify the respective messages and / or elements in the database that are retrieved, assembled, and / or sent to the user. For example, a message can indicate that a user is relatively active in various ways and is improving their health score. The messages can be provided via a feedback loop and can be formatted in many ways, such as text, chatbot, voice, or other suitable mechanisms, as shown and described herein.

[0101] Thus, as shown and described herein, the technical features and advantages shown and described herein are used to support outcome-based insurance. For example, the use of health scores may be particularly essential for identifying risk and calculating initial and / or adjusted premiums for users over time. This application provides users and insurance companies with a competitive advantage depending on our underwriting research and development services and participation in feedback that helps users make better decisions to reduce risk. In one or more implementations, computer transmissions are limited over time based on reclassification of each of a plurality of users, depending at least on the masked numerical scores received over the data communications network.

[0102] Health score range implementation In one or more implementations, the present application supports the generation and display of ranges associated with health scores. Health score ranges can be provided in a software application that is integrated with one or more publisher websites, such as, for example, a social network. One goal is to raise awareness and / or promote a user's health score in a public forum and encourage further social and user interaction, such as at no cost and / or for a paying subscriber-based model. Figures 15-23 identify one or more implementations for calculating and publicly using health score ranges to allow users to provide various information.

[0103] FIG. 15 illustrates an example flow of information and steps associated with generating and using a health score range according to one or more implementations of the present application. In step 1502, a “landing” page is provided, which may be via an online internet-based site, a mobile app, or other suitable interactive platform. An example landing page display screen 1600 is shown in FIG. 16A , including one or more graphical screen controls for a user to initiate the health score range generation process. The landing page may introduce an option for logging in to the user's account via one or more social networks (e.g., Facebook, Twitter, Google, or LinkedIn). Alternatively, the user may skip the login process and proceed to enter information for generating a health score range. An example login data entry display screen 1650 is shown in FIG. 16B . The landing page may be optimized with “Open Graph” and “Twitter Card” tags, for example, to ensure proper rendering when shared on social networks. Examples include ogp.me, dev.twitter.com / cards / overview, or the like. Furthermore, the landing page can be opened via a simple uniform resource locator ("URL") or via a more complex URL formatted to include the respective user or other ID. Each ID can be referenced in one or more databases, and personalization information can be accessed and entered into Open Graph and / or Twitter Card tags. This allows personal information to be rendered when a user shares their ID on their respective social network. When other users then select (e.g., click on) the shared link, they can be "sent" to the landing page, where they can be prompted to enter information for their respective health score ranges.One or more implementations require only a minimal amount of information about the user, such as their respective IDs, the identification of one or more "friends" who are actively using the health score range application, their usernames, and / or profile pictures.

[0104] At step 1504 (FIG. 15), a determination is made whether the user has already submitted information suitable for calculating a health score range (taken a "test") If not, the process proceeds to step 1506, where one or more questions are presented to the user.

[0105] In one or more implementations, a random variable is assigned to the test, e.g., value "1" has values ​​"A" through "C," and the variable can be used to determine the title and description used in Open Graph and Twitter Card tags. Table 1 shows the variables in an example implementation that uses a test ID in accordance with the present application. [Table 1]

[0106] Once the appropriate login procedure is complete, a series of prompts can be generated on the mobile computing device, and the process can proceed to step 1506 (FIG. 15) where the user can be prompted to begin the process of calculating their health score range for a social network or other publisher site, as shown in display screens 1700 and 1800 (FIGS. 17 and 18, respectively). Multiple data entry display screens containing questions for the user to respond to can then be provided. Example display screens 1900 (FIG. 19A), 1950 (FIG. 19B), 1970 (FIG. 19C), and 1990 (FIG. 19D) show example data entry interfaces where the user is requested to respond with information, such as information related to age, gender, weight, and height. Other user information requested can include habit-based information, such as smoking and drinking habits, types of food consumed, and subjective emotional information, such as whether the user feels anxious or hopeful.

[0107] Table 2 shows an example implementation including questions, conditions, and data input types according to which a "test" is presented to a user in relation to generating a health score range. [Table 2]

[0108] Upon receiving answers to one or more of the respective questions via an interactive data interface, such as those shown in Figures 19A-19D, one or more modules operate to generate and display a health score range (step 1508, Figure 15) substantially without human interaction. For example, as shown in example display screen 2000 (Figure 20), a range of 600 to 650 is generated and displayed for each user. Additionally, a vertical chart is provided, as shown in display screen 2000, that identifies the quality of the health score range, and in display screen 2000, the user's health score is between average and very good.

[0109] Continuing with the example shown in Figure 15, a population comparison of health score ranges can be provided to the user, such as in the form of a bar graph (display screen 2100, Figure 21). Providing the user with a population comparison of health score ranges can be useful feedback, such as to identify where each user's health score range falls within the population. In the example shown in display screen 2100, the user's health score range is within the top 65% of the overall population.

[0110] In addition to identifying a group comparison for the user, a social comparison can be provided (step 1512, FIG. 15). For example, as shown in display screen 2200 (FIG. 22A), a user's health score range can be displayed in a table along with other users (e.g., "friends"), which can encourage competition and motivate users to improve their health score range and, consequently, their overall health. The social aspect of step 1512 can be expanded in various ways, including as shown in example display screen 2250 (FIG. 22B), where a text box can be configured to allow posts (e.g., tweets) to other users within a social network. In one or more implementations, tweets can be automatically generated regarding the user's health score range for ease and convenience.

[0111] Continuing with reference to the flow shown in Figure 15, another opportunity provided in accordance with the present application is upsale (step 1514), which can encourage and enhance interactivity, subscriptions, and other social-based online activities. For example, Figure 23 shows example display screen 2300 that includes multiple prompts for further engagement, such as to download and install additional software (e.g., apps), to subscribe to newsletters or other appropriate content, etc.

[0112] Neuro-Linguistic Programming Implementation In one or more implementations, one or more modules executing neuro-linguistic programming (NLP) code, for example, on a mobile computing device, provide voice-enabled feedback. In one or more implementations of the present application, feedback such as various types of useful health-related information, including warnings and notifications, encouragement, and / or a user's health score, can be provided using voice-based technology. Such NLP modules strengthen the connection between a user's neurology and their language and experiences (e.g., cultural processes), thereby encouraging behavioral changes in the user to achieve specific goals, including activity-based and health-based goals.

[0113] Moreover, NLP modules executing on one or more computing devices that provide voice-based interactivity are particularly useful for reinforcing patterns of behavior, such as in response to auditory-based expressions, verbal communication, and positive reinforcement. Such sensory-based reinforcement results in behavior modification, even at a subconscious or unconscious level.

[0114] In one or more implementations, a module executing NLP programming code "learns" from the user's behavior and provides the user with a subjective experience that contributes to modifying the user's behavior, if necessary. The user's mental state can be determined in response to behavior monitored using the various components shown and described herein, and one or more verbal interventions or suggestions can be made to help guide the user to behave in a specific way. For example, the user may be reminded to take medication, check their blood pressure, get up and exercise, or even resist the urge to consume something considered unhealthy, based on patterns learned by the module. The information detected regarding the user's specific state, combined with the user's desired goal, can result in verbal encouragement, warnings of undesired behavior, or other feedback that similarly influences the achievement of the respective outcome.

[0115] In one or more implementations, the present application integrates with existing voice-based systems, such as through an open application programming interface ("API"), software development kit ("SDK"), or other suitable form of connectivity. Using one or more open platforms, the present application can integrate voice-based health scoring, lifestyle navigation, coaching, and feedback. Additionally, the NLP module can implement voice-based interactivity in any of the various implementations shown and described herein, where information is provided to the user. This provides an effective approach to communication, self-improvement, and healthy behaviors, such as in each implementation shown and described herein.

[0116] For example, the systems and methods herein provide audio-based feedback in substantially real time via NLP. Using the systems and methods disclosed herein, information can be received from a user device and the information can be processed to provide various forms of feedback, such as alerts and notifications. Related information can then be received and processed for further feedback (e.g., in the form of a feedback loop).

[0117] In one or more implementations, one or more rule engines can be provided that periodically and / or continuously process information and generate notifications for users, as shown and described herein. The implementation can depend on the respective subsystem (e.g., data collection subsystem, data communication subsystem, data processing subsystem) and one or more corresponding notification features. Moreover, one or more notification generation rule engines can be part of the individual subsystems that generate notifications. The notification features can include core information elements useful for the feedback process. In general, notifications can include surveys and / or prompts for information and can be presented through an interactive interface. The result can include an infrastructure configured for scheduling, processing, and delivery of notifications via various communication channels and formats.

[0118] Encrypted health information exchange implementation In one or more implementations, the present application can include a decentralized, distributed system in which health information acquisition, health information storage, health information processing, and secure communication are used in a dynamic barter platform. The present application can include computer-implemented systems and methods configured to acquire health-related and / or medical-related information from one or more sources and securely store the information after it has been verified. Moreover, according to the present application, the user represented by the information controls whether and how the information is sold, exchanged, shared, or otherwise accessed.

[0119] In one or more implementations of the present application, personal health information is securely stored according to shared and distributed ledger technology to ensure reliable, transparent, and immutable records. For example, at least a portion of a user's health information is recorded and verified according to an algorithm that provides consensus among multiple computing devices, such as a peer-to-peer (P2P) network. Distributed ledger security can be provided by a protocol that can include private key and public key cryptography. According to each implementation of the present application, one or more public ledgers can be provided in an open, permissionless blockchain and / or in a secure, permissioned shared or private blockchain that allows users to define specific access levels for each party. The use of such an implementation, such as in one or more respective blockchains, protects each party's respective health information without incurring the risk that such information is invalid or insecure. Thus, according to one or more implementations of the present application, a public / private "hybrid" blockchain model is provided to provide access to information according to each health information and according to various conditions defined by each user.

[0120] It is recognized by the present inventors that creating secure, private objects representing health-related information, as collected, processed, and / or transmitted by one or more of the health appliances 105, for verification and storage in a blockchain or other suitable technology may be considered a source of a verified “personal” electronic medical record. Control of such personal electronic medical records may be held by the individual users of the information processors 102 and / or user computing devices 104. Methods for verifying that the information is accurate may include verifying certain identifying values, such as machine access control addresses (“MAC” addresses), Internet Protocol (“IP”) addresses, secure signatures (including keys, certificates, and / or other trusted mechanisms) that confirm unaltered, accurate information received via the respective devices and / or during the respective data sessions. In accordance with the present application, such personal electronic medical records may be stored in a blockchain or other suitable format in accordance with the teachings herein and available in connection with transactions between counterparties.

[0121] As shown and described herein, a person (i.e., a human) is provided with information that has been verified and stored in a secure, private object on a blockchain or similar technology, such as in a securely stored ledger. The information can be collected from various health instruments 105 (e.g., digital medical devices), and once verified and securely stored, the person can define a set of rules that allow public and / or private access to at least certain portions of the blockchain according to one or more parameters defined by the person.

[0122] For example, a user can selectively license or otherwise consent to the secure dissemination of verified health-related information, such as pursuant to one or more smart contracts, whose terms and parameters are defined by the owner of the information (i.e., that person) and approved by one or more parties seeking access to that information (e.g., health information requesters), as known in the art. Such requesting parties may desire access to specific health information, for example, for clinical trials, physician evaluations, health assessments, health scoring, insurance policies, and insurance risk modeling. In the context of DNA sequencing, for example, a pharmaceutical company may wish to research a particular disease, DNA, or the like, and request access to a particular genome or disease from an individual it knows to be the “owner” of such particular data point. In one or more implementations, access to information in a secure ledger (e.g., a blockchain via opt-in or other suitable acceptance technology) can be offered in exchange for something of value. While monetary gain is considered a suitable consideration for exchange for access to health-related information stored within a blockchain, the present application contemplates exchange for other forms of value and alternative payment models, such as insurance premiums, healthcare visits, treatments and medications, and other healthcare-related value, which are generally referred to herein as "economic something for something."

[0123] In addition to individual electronic medical records, such as those achieved by one or more of the respective devices 105 shown and described herein, organizational-based electronic medical records can be verified and securely stored in a distributed ledger (e.g., a blockchain) in accordance with the features herein. Professionally generated electronic medical records can be verified according to respective certificates, digital signatures, or other cryptographic means (e.g., hashes) that indicate the records are authentic, accurate, and up-to-date. Once added to the blockchain or other suitable technology, each medical record is tamper-proof and cannot be altered. This is supported, at least in part, within a blockchain network, where multiple participating nodes share access to the blockchain to verify transactions and data sources over time.

[0124] Figure 24 is a block diagram illustrating sources of example verified records 2400 suitable for entry in a distributed, secure ledger, such as a blockchain. In the example shown in Figure 24, a "personal" verified electronic medical record 2402 is shown, which includes information received directly or indirectly from various devices 105. Additionally, a verified electronic medical record 2404 is shown, which includes verified information from, for example, a lab, from radiology (e.g., MRI / X-ray), and from a clinic.

[0125] FIG. 25 is a block diagram illustrating an example blockchain 2500 including a verified individual electronic medical record 2402 and a verified organizational electronic medical record 2404 that are securely stored within the blockchain and therefore invulnerable to tampering or alteration. The items within the blockchain are then available in connection with one or more transactions, pursuant to terms agreed upon by the user and the other party (e.g., a pharmaceutical company seeking research information). In the example shown in FIG. 25, portions of the verified individual electronic medical record 2402 and the verified organizational electronic medical record 2404 are defined, in part, as public and in part as private (denoted as "public / private"). This further illustrates the functionality and flexibility of implementations of the present application, whereby a user can define that a particular electronic medical record (or portion thereof) securely stored within the blockchain is limited to private access (e.g., permission-based) and / or, more generally, which portions the user chooses to make publicly accessible. This illustrates the hybrid nature of blockchain technology in various implementations of the present application and further illustrates the flexibility of implementations of the technology, such as in relation to ownership and access rights to a person's health information and in exchange for the value of that information.

[0126] Additionally, in one or more implementations, an individual may provide access to their health information anonymously, for example, for statistical testing or specific research that the individual wishes to support without any specific compensation or benefit.

[0127] FIG. 26 is a block diagram illustrating an example blockchain including a smart contract offer 2602 according to one or more criteria and / or parameters specified by a user in a license offer for health information. Such parameters can be used, for example, to define the user's terms and can be stored on a validating node within the blockchain. Once all of the terms are accepted by the requesting party, such as a pharmaceutical company, research institution, insurance company, or other interested party, a transaction can occur in accordance with the terms. In the example shown in FIG. 26, a smart contract offer to grant permission to use a user's health information is validated and securely stored within the blockchain. In each blockchain record shown in FIG. 26, multiple parameters are defined in the offer that represent the terms specified by the licensor. For example, parameter 2604 represents the classification of health information the licensor is willing to offer for exchange. Health information can be public or private, as specified in each record within the secure ledger, which can directly affect the value the licensor expects to receive in exchange for that information. For example, certain medical information that may be considered highly sensitive and private may be considered more valuable by the licensor and therefore directly affect the amount of the value parameter 2612 in the offer. Other parameters include the frequency of transmission of the respective health information, such as daily, weekly, monthly, or some other period. Additionally, the example license provision 2602 shown in FIG. 26 includes a parameter 2608 that describes the nature of the licensee's business. For example, the licensor may not want to grant permission to use health information to businesses known to engage in unsavory business practices. Furthermore, a parameter 2610 is shown in the example license provision 2602 that requires transparency of the use of the licensed information. For example, the licensor may specify a condition that it wants to know whenever health information is being used in a particular context.Various details can be defined regarding value 2612, such as that in exchange for the use of the user's health information, lifestyle information (including, for example, nutritional intake), discounts or full payments on insurance premiums, health care visits, medical tests, medications, nutrition or other health-related parameters of value will be received, including over a long period of time.

[0128] Thus, as shown in the licensing example 2602 in each blockchain 2500, the conditions and parameters that are requirements for the smart contract to be entered into can be defined by the owner of the health information.

[0129] FIG. 27 illustrates an example network of interactions between counterparties, including a distributed ledger containing smart contract offers 2500A, 2500B, 2500C, and 2500D from respective individual owners of health information (e.g., license offers for specific health information), and various counterparties, generally referred to herein as requesters, requesting access to such health information. In the example shown in FIG. 27, requesters, such as pharmaceutical companies, doctors, research institutions, insurance companies, and hospitals, submit respective blockchains 2704A, 2704B, 2704C, 2704D, and 2704E containing parameters defined by the respective counterparties. Of course, other interested potential licensees are also contemplated herein. Similarly, the example configuration of FIG. 27 includes an exchange 2702 in which one of the smart contract offers defined in blockchain 2500C from health information owners can be matched with an offer defined in blockchain 2704D from a requester of such information. In one or more implementations, to update a particular ledger with one or more transactions, each participating node in the blockchain processes the transaction(s) according to the conditions and logic specified in a smart contract. Transaction processing in a decentralized architecture is more efficient, meets the scalability needs of each participant, and ensures confidence between counterparties that the information specified in the blockchain is accurate, valid, and secure.

[0130] As described herein, the present application can be implemented in a decentralized architecture in which a verified, secure, distributed ledger ("ledger of life") representing information sent and received from various parties and devices supports a health sharing economy. Figure 28 illustrates an example decentralized marketplace 2802 including a distributed ledger of life according to an example implementation of the present application. Smart contracts support a value-based, market-priced economy based on health information and healthcare in which tangible outcomes are provided. Such outcomes can include, for example, health scores representing large numbers of individuals, as well as specific research and health insurance modeling, in exchange for various values ​​that can include healthcare, treatments, medications, or other value sources. Implementing opt-in in an open, sharing, decentralized platform, such as that defined in a blockchain, allows records of transactions to be verified, ensuring that exchanges can be guaranteed over time.

[0131] While many of the examples and discussions described herein consider an exchange of value for verified health information securely stored and provided on one or more distributed ledgers, the present application supports examples of accessing information using blockchain or other similar technologies without requiring negotiation for the exchange or further exchange of value. For example, an individual with a securely stored electronic medical record at their leisure provides access to their physician to expedite treatment options. Such access does not necessarily stem from a desire for monetary value or other types of benefit, but rather, more generally, to improve access to information and medical care. Furthermore, this example demonstrates the flexibility of the present application to support one-to-one procedures without requiring a virtual "exchange" with multiple participants.

[0132] 29 shows an alternative representation of the present application including health information generated in a distributed manner in response to health devices 105 (represented as the "Internet of Things"). Additionally, a blockchain 2500 is illustrated that includes personal identification information, blood type information, DNA information, weight information, blood pressure information, and other information associated with the user's health.

[0133] Examples of types of information that can be included in the blockchain 2500 include: The user's date of birth User's gender User age User height User weight User's blood pressure User's blood type User DNA User's heart rate Your heart rate variability (“HRV”) Acid on the user's skin User's breathing rate User's skin temperature User's blood sugar The user's blood cholesterol User's ECG The number of steps the user has taken User posture / fat percentage / BMI Detecting user falls User sleep time / quality / wake-up User Activity User stress monitor User energy consumption The user's daily nutritional intake

[0134] The present inventors have recognized financial constraints on the insurance industry, such as with regard to life and health insurance, particularly as sectors experience an aging population. For example, nutritional deficiencies are a contributing factor to lifestyle and healthcare and treatment needs, and the industry desires improved, predictable health outcomes and more accurate risk models. For example, a person is diagnosed with type 2 diabetes. By providing access to information collected and processed by health appliances 105, such as in a verified, secure, distributed ledger, insurance companies can offer quality insurance product(s) at highly competitive prices. Because insurance companies have access to accurate information showing, for example, that a person is eating well and otherwise engaging in a healthy lifestyle, the likelihood of claims for treatment is significantly reduced, and insurance companies can pass on savings to insureds.

[0135] Continuing with the example shown in Figure 29, exchange 2702 includes an exchange where health information is licensed for use by one or more of pharmaceutical companies, health research organizations, and insurance companies (shown as "health metrics"). Additionally, outcome information is identified in section 2902, which includes health score information that can be used for, for example, outcomes-based research, preventative care, predictive analytics, cost modeling, and specific risk modeling. Figure 29 shows how each blockchain ("ledger of life") is connected within health metrics and health scoring.

[0136] In one or more implementations, the present application further supports the free donation of medical information, such as for non-profit research. Such donation may occur after a person's death, given a promise or other agreement set forth in the ledger. Example candidates for such agreements may include patients with terminal illnesses, specific diseases, or other health conditions.

[0137] Figure 30 illustrates examples of features and benefits of the present application in which the use of blockchain can help transform various industries, such as insurance and other health-related industries. For example, blockchain or similar technologies can serve to authenticate identity and value, transfer value (i.e., transaction-based), store value (e.g., health information), store underwriting value, exchange value, implement funding and investments, ensure value and manage risk, and account for value across industries.

[0138] The method and system of the present invention according to an exemplary embodiment is now described in conjunction with the exemplary computing system environment generally shown and described above. The methods described herein are illustrated with reference to flow diagrams to facilitate description of the major processes of the exemplary embodiment of the present invention, although it is understood that certain blocks can occur in any order, such as when events drive program flow, such as in an object-oriented program. As a result, the flow diagrams should be understood as example flows, and blocks can occur in orders different from those illustrated.

[0139] FIG. 31 illustrates a flow diagram of one example implementation of the present application in accordance with exchange 2702 in the implementation illustrated in FIG. 27. In step 3102, parameter information for a smart contract offer is received. The parameter information may include, for example, a classification of the type of health information being offered for exchange, a frequency of sending the information to the licensee, a preferred classification of the licensee's business, and an expression of value in exchange for the health information that is permitted to be used. Then, terms and agreements for providing value according to the terms are identified (step 3104). A portion of a verified secure ledger (e.g., a blockchain) containing the health information is accessed (step 3106). Then, in step 3108, access to the verified secure ledger according to the conditions specified by the parameters is provided. In exchange, access to the value according to the conditions is provided in step 3110. In one or more alternative implementations, such access to the health information can be donated or otherwise provided without a request for compensation, as described herein. The process then ends.

[0140] Figure 32 shows a flow diagram of an alternative implementation of the present application according to at least a decentralized smart contract model. In step 3202, a decentralized distributed ledger is provided that includes terms for exchanging health information for value. A decentralized distributed ledger is further provided that includes terms for exchanging value for the health information (step 3204). At least a portion of the distributed ledger that includes the health information is accessed (step 3206), and access to the portion of the distributed ledger is provided (step 3208). In exchange, value is provided for access to the portion of the distributed ledger (step 3210). The process then ends.

[0141] Thus, as shown and described herein, multiple interactive modules are provided that enable external parties seeking certain data points to receive such points in accordance with blockchain technology. A centralized ledger network supports the flow of information and operational control from one or more specific points. In addition, a decentralized ledger network allows nodes to make independent processing and computational decisions, distributing the computational workload across multiple nodes in the network. The blockchain network structure of the present application ensures proof of accuracy, accountability, and transparency of information that was previously unavailable. Additionally, data integrity, data confidentiality, and newly constructed mechanisms for contracting on behalf of users' personal (and public) health information are provided herein.

[0142] The present application will now be further described with reference to two examples: An individual has access to DNA data and stores the information in a blockchain. The individual decides to store information representing license terms for accessing the DNA data in the blockchain, the terms being specified in a smart contract. A pharmaceutical company desires access to the DNA data as offered by the individual and agrees to the license terms specified in the smart contract offer. In exchange, the individual receives free medication for 10 years in response to the data being provided for an innovative new drug. Essentially, the individual uses the systems and methods disclosed herein to participate in research and development projects, provide access to medical information on blockchain technology, and receive years of health care, being compensated in response to the blockchain verifying the transaction.

[0143] In a second example, an individual has multiple chronic conditions and is considered a high insurance risk. Insurance premiums for that individual are prohibitively expensive. Using the systems and methods disclosed herein, the individual agrees to participate in a professionally controlled program offered by a life and health insurance company, self-insurer, retailer, pharmacy, or other company. In exchange for access to that individual's health information, the individual receives significantly discounted premiums for insurance. Access to that individual's health information can include, for example, verified and secure information received via multiple biosensors, information from physicians, and big data analysis in response to laboratory investigations and tests.

[0144] Although the present invention has been described in relation to specific embodiments thereof, many other variations and modifications and other uses will become apparent to those skilled in the art. For example, in the context of P2P communities, cross-jurisdictional demographic access is available as counterparties can assess risk and pricing in their respective health insurance communities and opportunities become available in light of free market exchange.

[0145] Furthermore, in one or more implementations, the present application may include a decentralized, distributed system in which the acquisition and processing of completely different forms of information, such as information associated with health and fitness tracking, where the information is associated with an online gaming environment, is automatically provided in an integrated hardware and software platform. The present application may provide computer-implemented systems and methods configured to acquire health-related and / or medical-related information from multiple sources and calculate a health score based thereon. The information may be processed in accordance with the teachings herein to generate a numerical score as a basis for assessing a user's relative health status (i.e., generating a user health score). Particular implementations of features associated with the calculation and distribution of a user health score are more fully shown and described in U.S. Non-Provisional Patent Application No. 14 / 257,855, entitled "AUTOMATED HEALTH DATA ACQUISITION, PROCESSING AND COMMUNICATION SYSTEM," assigned to the assignee of the present invention. In one or more implementations, an initial health score may be calculated, for example, based on population or other demographic information. In some examples, only a few baseline points of information, such as an individual's height, age, weight, and gender, can be processed by the mobile computing device to generate an initial health score. A health score calculated from such baseline information is useful for quickly engaging a user and getting started with the systems and methods disclosed herein.

[0146] The present application addresses technological shortcomings to bridge the gap between disparate information sources and utilize information from such sources in an integrated and holistic manner. The inventors recognize that individuals, at least initially, often enthusiastically use technology to improve and monitor their health progress and keep up with good health habits. However, over time, individuals may find maintaining good habits and keeping up with fitness and health-based technology devices arduous or too demanding. Furthermore, it is recognized that useful health-related information can be collected or is already collected in one or more online and / or virtual gaming environments. For example, gaming environments implemented or supported by mobile computing devices generate or otherwise access information associated with user activity, such as distance traveled or other fitness-related information. Such information is often health and fitness-related but may not be accessible or otherwise available to fitness-based solutions.

[0147] The present application addresses these concerns and provides a solution for tracking, storing, and managing health information, such as in connection with gaming and other entertainment environments. An example hardware configuration according to such an implementation is shown in FIG. 33. In one or more implementations, gameplay information is used to augment other activity data, thereby enabling users to keep their health information up-to-date and accurate. The present application makes this possible by providing new communication between gaming environments and fitness tracking solutions.

[0148] In a particular implementation, a first application running on a mobile computing device tracks user activity information related to the number of steps taken by a user over a period of time, such as a day. The first application includes a graphical user interface running on the mobile computing device. The mobile computing device communicates with a second application running by the mobile computing device, the communication including information regarding user interactions with a gaming environment provided by the mobile computing device. One or more instruction modules are executed by the mobile computing device to integrate the tracked user activity information, along with at least a portion of the user interaction information, into an activity value using the first application. The activity value is automatically presented within the graphical user interface of the first application when the first application is running.

[0149] In an alternative implementation, dynamic tracking of user-based activity is provided in response to information received from multiple data sources. For example, a first application tracks user activity information related to the number of steps a user takes over a period of time. The first application running on the mobile computing device includes a graphical user interface that presents at least a health score associated with the user. The health score can be provided in response to criteria information received from the user. Alternatively, the health score can be calculated using multiple health activity parameters, such as those described in U.S. Non-Provisional Patent Application No. 14 / 257,855. Communication of gameplay information with a second application executed by the mobile computing device is provided by the mobile computing device, and the tracked user activity information, along with at least a portion of the user interaction information, is integrated into an activity value. Using the activity value, the user's health score is updated and stored in association with a record for the user. Furthermore, the updated health score is presented within the graphical user interface, such as by triggering a notification, independent of launching the first application.

[0150] In one or more implementations, the health score and / or updated health score includes a walking score, which broadly represents the inverse of the reduction in all-cause mortality risk associated with walking. Additionally, the walking score may include a buffer, generally referred to herein as a "health reservoir." The health reservoir is a numerical value that represents, for example, a user's ability to rest before their health score (e.g., walking score, as described in further detail herein) begins to decline. Health score decline is more specifically described in U.S. Non-Provisional Patent Application No. 14 / 257,855, such as with respect to the relative weight of extrinsic lifestyle components or physical activity information. According to this application, declines may occur over time with respect to the walking score, and a time or other measure may be provided by the health reservoir to indicate the buffer before such decline occurs. Furthermore, the walking score may include a buffer that allows the user to rest for a period of time without losing score. The size of this buffer may depend on the accumulated energy gained by the user, for example, from continued walking.

[0151] Additionally or alternatively, the reservoir associated with a user's walking score ("waScore") can be directly converted into redeemable value, such as offsetting the cost of items, insurance premiums, or other items of value. For example, discounts on insurance premiums can be:

number

[0152] In one or more implementations, inputs, such as tracked footprints, that perform a series of steps associated with the calculation of a walking score are applied to one or more algorithms. For example, the amount of daily energy consumed by the user throughout the day is calculated. A modulator risk based on the risk for all-cause mortality associated with obesity can further be calculated, and an energy amplifier from an obesity risk model can further be calculated. The results can be used to motivate overweight and / or obese users to walk more, such as by providing value in a user interface and / or for encouragement in various ways. Furthermore, a value representing net years gained due to the positive contribution from energy expended walking and, potentially, the negative contribution from excess weight can be calculated in accordance with the present application.

[0153] For example, in addition to the various scores resulting from the aforementioned calculations, the present application may access or otherwise include a rich gaming environment whose goal is to entice users to walk more and provide benefits such as the generation of points that can be redeemed for various types of rewards. Examples of such games are provided herein. In one or more implementations, the gaming environment may be generated and / or customized to implement each walking score for a specific purpose, such as a discount engine for a health or life insurance company. An example implementation of lifestyle-related payments related to insurance is more fully shown and described in U.S. Provisional Patent Application No. 62 / 341,421, entitled "AUTOMATED HEALTH DATA ACQUISITION, PROCESSING AND COMMUNICATION SYSTEM AND METHOD," which is assigned to the assignee of the present invention.

[0154] In one or more implementations, the present application provides access to new and challenging game environments that can include features such as adventures, along with various attractions to keep users engaged. For example, in-game challenges are provided that can include various game levels and corresponding rewards and bonuses upon completion of each level. Users can be represented as specific characters and / or have special character names, and social interaction is supported for users to anonymously participate in the levels. For example, users can select alias names that are represented in a shared implementation, such as on a map graphically represented within a social networking site or other socially accessible platform. Furthermore, in one or more implementations, upon completion of one or more game levels, users can select a team to play with, such as in adventures and team battles. Furthermore, the virtual representations provided on the mobile computing device can be implemented in the real world, such that quests for gems or other objects of virtual value can be pursued and collected.

[0155] In one or more implementations, activity related to a game environment is measured according to points (e.g., "experience points" ("XP")). Initially, experience points may reflect or be measured according to user activity. For example, one XP may be equal to one step taken by the user. However, over time, as the user participates in respective games, such as to complete an adventure or participate in team matches, the amount of XP earned may gradually increase. Furthermore, as the user completes levels within respective games, additional XP may be awarded, which may increase the amount of virtual value accumulated by the individual user over time. Experience points, in one or more implementations, may be redeemable for cash, such as for a lower premium or other value.

[0156] When a gaming environment is provided or information associated with the gaming environment is otherwise accessed, user-based activities such as walking are monitored and information associated therewith is used, such as to update the user's health score, increase a health reservoir, or the like.

[0157] For example, referring now to the drawings, like reference numbers refer to like elements as shown in FIG.

[0158] 34 and 35 , a mobile computing device 104 configured with one or more modules of a health application accesses user activity data, as in step 3402. In a particular implementation, an activity tracking module 3502 configures at least one processor of the mobile computing device 104 to access activity tracking data, where the activity tracking data may be stored user data, such as previous activity data stored in a database remote or local to the mobile computing device 104. Alternatively, the at least one processor of the mobile computing device 104 is configured with one or more sub-modules of the activity tracking module 3502 to access from one or more walking or pedometer user data, such as distance traveled over a given period of time.

[0159] The mobile computing device 104 is further configured to access data corresponding to user interactions within the gaming application, as shown in step 3404. Here, at least one processor of the mobile computing device 104 is configured by a communications module 3504 to access data regarding the user's current interaction with the gaming application. The communications module 3504 may include one or more sub-modules configured to access, process, transmit, or format data exchanged between the wellness application and the gaming application. The gaming application configures at least one processor of the mobile computing device 104 to access the user interaction data during excitation of the gaming module 3516. Here, the gaming module 3516 configures at least one processor of the mobile computing device 104 to initiate a series of activities incorporating the user interaction. In a further implementation, obtaining the interaction data includes accessing one or more GPS geolocation measurements from GPS satellites using the positioning module 3520. Such collected data is transmitted back to the wellness application using the communications module 3518.

[0160] It has been recognized by the present inventors that information collected by a pedometer or other tracking device may be a duplicate of information already described in connection with the gaming environment. Accordingly, to eliminate double-counting of user activity, the present application eliminates information collected either from the tracking device or from the gaming environment if duplicate information is detected. Examples of eliminating double-counting of information are also shown and described in commonly assigned U.S. Provisional Patent Application No. 62 / 341,421.

[0161] Upon obtaining the game state data using the communications module 3504, at least one processor of the mobile computing device 104 is configured to integrate the activity tracking data and the received interaction data. As shown in step 3406, the at least one processor of the mobile computing device 104 is configured by an integration generation module 3506 to generate an integrated value from the activity and interaction data. The communications module may include one or more sub-modules configured to access, process, transmit, or format data in connection with the described steps.

[0162] A status module 3508 configures at least one processor of the mobile computing device 104 to evaluate the current status of the wellness application. For example, as shown in step 3408, the runtime of the wellness application is evaluated.

[0163] If the health application is currently running, the combined score generated in step 3408 is presented to the user via one or more visual, auditory, tactile, or integrated notification protocols.

[0164] The aggregate value is presented to the user, as shown in step 3410. In particular implementations, at least one processor of the mobile computing device 104 is configured with an alert module 3510, where the alert module causes one or more graphical user interface elements of the mobile computing device 104 to generate a textual or iconic representation of the aggregate value.

[0165] The alert module 3510 may include one or more sub-modules configured to access, process, transmit, or format data exchanged between the health application and one or more system-level or application-level GUI applications.

[0166] If the health application is not currently running, the combined score generated in step 3408 is stored in one or more remote or local storage locations, such as a database. As shown in step 3412, at least one processor of the mobile computing device 104 is configured with a storage module 3512 to store the combined value and other user data in a user profile or account.

[0167] In one or more further implementations, the user's health score is updated using the tracked data using the activity values, as in step 3414. The mobile computing device 104, configured with an update module 3514, accesses one or more user profile data sets including health score data and updates the health score based at least on the activity values ​​tracked in step 3402.

[0168]

[00130] Figure 36 illustrates graphs associated with calculated walking scores in the context of the present application. For example, graph 3602 provides plot areas identifying walking scores along the y-axis and corresponding metabolic equivalent values ​​along the x-axis. Graph 3604 illustrates a comparison of raw information, as well as information stabilized with one or more filters, such as exponential moving averages ("EMA") and reservoir values, associated with walking scores. Figure 37 illustrates an example walking score value (498) corresponding to the number of days until a walking score declines or drops precipitously, and a corresponding monetary value ($3.55) that can be applied, such as to offset insurance premiums.

[0169] FIG. 38 shows an example implementation including a graphical virtual map 3802 that virtually represents locations in the real world, with a corresponding map portion 3804 including locations where points of interest (e.g., sought items) are shown and respective icons for locations where each player may be located.

[0170] Thus, as shown and described herein, an integration of two pillars is provided: a health pillar is implemented that can include a Walk Score value and a health reservoir to provide value, such as insurance discounts. Additionally, a game pillar is implemented or accessed that can be driven by experience points and game levels, where user interest is maintained through features such as adventure journeys, team battles, and collecting items at mapped points of interest. Users can maintain and monitor good health habits, such as to prevent a decline in their Walk Score, while simultaneously realizing value, such as cost savings.

[0171] The foregoing subject matter is provided by way of example only and should not be construed as limiting. Various modifications and changes can be made to the subject matter described herein without following the example embodiments and applications illustrated and described and without departing from the spirit and scope of the present invention, as defined in the following claims.

[0172] The various embodiments described above can optionally be combined in various ways depending on the desired implementation. It will be understood that other embodiments based on different combinations of features are possible. It will also be understood that more than one parameter can be used for a particular parameter type. None of the described features are mutually exclusive, and any combination can be deployed to achieve the described functionality.

[0173] Thus, as shown and described herein, multiple interactive modules are provided to encourage individual users to submit information activities and to passively track activity and health information. The information can be used to gain important insights into an individual's health, such as in connection with calculating a health score. In one or more implementations, pay-as-you-go insurance options can be provided in response to health score information and interactivity, such as via a social internet website, that encourages users to promote health information and, perhaps more importantly, improve their health. Such functionality has not been available heretofore.

[0174] Although the present invention has been described in relation to particular embodiments thereof, many other variations and modifications and other uses will become apparent to those skilled in the art, and it is therefore preferred that the invention not be limited by the specific disclosure herein.

Claims

1. 1. A system for classifying user activity, comprising: a computing device configured with at least one processor, a non-transitory processor-readable medium, and instructions stored on the non-transitory processor-readable medium; a first sensing subsystem for passively tracking a physiological and / or metabolic state of the person, the first sensing subsystem comprising a biosensor in contact with or within the body of the person and periodically transmitting sensed information associated with the physiological and / or metabolic state of the person; a second sensing subsystem for passively tracking a location of the person, the second sensing subsystem comprising a Global Positioning System ("GPS") receiver for receiving GPS information usable to determine a location of the person; and Equipped with The computing device includes: receiving and / or accessing information representative of the sensed information and determined location; classifying and predicting user activity and / or user health status according to one or more machine learning techniques; Detecting a change from a respective baseline state associated with at least some of the sensed information and i) at least some information stored in a user profile associated with the person, and ii) external information. configured to execute at least some of the instructions to cause the In response to detecting the change, the computing device: defining a first activity unit having a first start time corresponding to detecting a user engaging in the activity; monitoring the sensed information, the external information, or both; establishing a first end time for the first unit of activity using the monitored information; automatically classifying the first activity unit as true; outputting the classification of the first activity unit to a display; storing the classification of the first unit of activity in a database remote or local to the computing device; providing a user interface including selectable options associated with the first unit of activity; In response to at least a received selection of at least one of the selectable options, Combining or merging the first activity unit and a second activity unit having a second start time and a second end time, whereby the corrected classification has a start time equal to the first start time and an end time equal to the second end time; or Dividing the first activity unit into at least two activity units, each of the at least two activity units having a different respective start time and a different respective end time. Corrected by outputting the corrected classification to a display of a computing device; providing the corrected classification to a classification engine; Use the corrected classification for machine learning The system is configured as follows:

2. 10. The system of claim 1, wherein the computing device is configured to execute instructions that cause the computing device to generate and provide an audit trail that identifies at least differences between the classification and the corrected classification.

3. The system of claim 2 , wherein the audit trail is configured for compliance checking.

4. 2. The system of claim 1, wherein the first activity unit and the second activity unit are minor activity units, and combining the first activity unit and the second activity unit comprises including both the first activity unit and the second activity unit in a major activity unit.

5. 2. The system of claim 1, wherein the first activity unit and the second activity unit are minor activity units, and merging the first activity unit and the second activity unit comprises reducing a number of activity units to one minor activity unit comprising the first activity unit and the second activity unit.

6. The system of claim 1 , wherein the information stored in the user profile includes demographic information associated with at least one user.

7. The system of claim 1 , wherein the external information includes at least one of schedule information, calendar information, and weather information.

8. 10. The system of claim 1, wherein the classification engine is configured to execute instructions that cause the computing device to define a criterion representing a state of a person not engaged in an activity in response to a first portion of the sensed information from the first sensing subsystem, and wherein detecting the user engaged in the activity is based at least in part on detecting a change from the criterion.

9. 10. The system of claim 1, wherein the machine learning comprises at least one of cluster-based learning, including at least one of K-means clustering, self-organizing mapping, Gaussian mixture modeling, support vector machines, case-based reasoning, and guided trees.

10. The system of claim 1 , further comprising an artificial neural network for performing at least some of the machine learning.

11. 1. A method for classifying user activity, the method comprising: passively tracking a physiological and / or metabolic state of the person with a first sensing subsystem, the first sensing subsystem comprising a biosensor in contact with or within the body of the person and periodically transmitting sensed information associated with the physiological and / or metabolic state of the person; passively tracking a location of the person with a second sensing subsystem, the second sensing subsystem comprising a Global Positioning System ("GPS") receiver that receives GPS information usable to determine a location of the person; receiving, by a computing device remote from the first sensing subsystem and configured with a processor, a communications module, and instructions stored on a non-transitory processor-readable medium, the sensed information from the first sensing subsystem and the second sensing subsystem; classifying and predicting, by the computing device, user activity and / or user health status according to one or more machine learning techniques; detecting, by the computing device, a change from a respective baseline state associated with at least some of the sensed information and i) at least some information stored in a user profile associated with the person, and ii) external information; In response to detecting the change, defining, by the computing device, a first unit of activity having a first start time corresponding to detection of a user engaging in a unit of activity; monitoring, by the computing device, the sensed information, the external information, or both; establishing, by the computing device, a first end time for the first unit of activity using the monitored information; automatically determining, by the computing device, a classification of the first unit of activity; outputting, by the computing device, the classification of the first activity unit on a display; storing, by the computing device, the classification of the first unit of activity in a database remote or local to the computing device; providing, by the computing device, a user interface including selectable options associated with the first unit of activity; and, by the computing device, modifying the classification in response to at least a received selection of at least one of the selectable options. Combining or merging the first activity unit and a second activity unit having a second start time and a second end time, whereby the corrected classification has a start time equal to the first start time and an end time equal to the second end time; or Dividing the first activity unit into at least two activity units, each of the at least two activity units having a different respective start time and a different respective end time; and outputting, by the computing device, the corrected classification to a display of a computing device; providing, by the computing device, the corrected classification to a classification engine; using, by the computing device, the corrected classification for machine learning; and A method comprising:

12. The method of claim 11 , further comprising generating and providing, by the computing device, an audit trail that identifies at least differences between the classification and the corrected classification.

13. The method of claim 12 , wherein the audit trail is configured for compliance checking.

14. 12. The method of claim 11, wherein the first activity unit and the second activity unit are minor activity units, and combining the first activity unit and the second activity unit comprises including both the first activity unit and the second activity unit in a major activity unit.

15. 12. The method of claim 11, wherein the first activity unit and the second activity unit are minor activity units, and merging the first activity unit and the second activity unit comprises reducing a number of activity units to one minor activity unit comprising the first activity unit and the second activity unit.

16. The method of claim 11 , wherein the information stored in the user profile includes demographic information associated with at least one user.

17. The method of claim 11 , wherein the external information includes at least one of schedule information, calendar information, and weather information.

18. and defining, by the computing device, a baseline representative of a state of a person not engaged in an activity in response to a first portion of the sensed information from the first sensing subsystem, wherein detecting the user engaged in the activity is based at least in part on detecting a change from the baseline. The method of claim 11 further comprising:

19. 12. The method of claim 11, wherein the machine learning comprises at least one of cluster-based learning, including at least one of K-means clustering, self-organizing mapping, Gaussian mixture modeling, support vector machines, case-based reasoning, and guided trees.

20. The method of claim 11 , wherein at least some of the machine learning is performed within an artificial neural network.

Images