Automated physical activity classification
Inventors
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Assignees
MemberPear Health LabsPear Health LabsPear Health Labs develops digital health and fitness software for organizations. Its platforms combine health protocols, assessments, biometric and wearable data, and user feedback to provide adaptive exercise, nutrition, recovery, and wellness guidance. The company also describes integrations with mobile and web applications, electronic health records, and wearable devices. Applications include general wellness, healthy aging, Medicare fitness programs, GLP-1 treatment support, musculoskeletal health, maternal and postpartum health, and military readiness.
Pear Health Labs develops digital health and fitness software for organizations. Its platforms combine health protocols, assessments, biometric and wearable data, and user feedback to provide adaptive exercise, nutrition, recovery, and wellness guidance. The company also describes integrations with mobile and web applications, electronic health records, and wearable devices. Applications include general wellness, healthy aging, Medicare fitness programs, GLP-1 treatment support, musculoskeletal health, maternal and postpartum health, and military readiness.
Publication Number
US-9665873-B2
Publication Date
2017-05-30
Expiration Date
Abstract
The invention relates to a system and method for analyzing an activity session. Data from a monitoring device worn by a user of the system for example is received over a period of time or distance and a classification system of the invention is utilized in real time or post session to determine the activities performed by the user during the session. The data collected relates to multiple parameters monitored during the activity session. The classification system defines an activity using a set of threshold criteria for a combination of parameters and therefore identifies a particular activity performed during the activity session when a combination of monitored parameters satisfies the threshold criteria for a particular activity. Determining the activities performed in this way allows for more useful interpretation of the data which in turn leads to more effective coaching advice and feedback for the user.
Core Innovation
The invention provides a method of analyzing an activity session by receiving activity data for an activity, where the activity data corresponds to multiple parameters. A set of threshold criteria is retrieved for a particular user in view of the activity, with the threshold criteria corresponding to the multiple parameters and at least one threshold criterion being based on an anaerobic threshold of the particular user.
The anaerobic threshold is determined by collecting activity data of the particular user prior to the activity session, forming a plurality of incidence bins containing numbers of incidences of the activity data within specific ranges, and identifying at least one incidence bin with the highest change, where the anaerobic threshold is based on a range of the identified incidence bin(s). The method compares the activity data with the set of threshold criteria to identify an effort of the particular user, where the effort is defined by a difference between the activity data and the threshold criteria based on the anaerobic threshold of the particular user.
A resistance experienced by the particular user is determined, where the resistance comprises a measure of terrain. Multiple durations of the activity are then classified into one or more predefined training type segments based on the resistance experienced and the identified effort, and the invention is implemented by one or more processors as a method, a system, and non-transitory computer readable media.
Claims Coverage
The independent claims cover a method, a system, and non-transitory computer-readable media. Across these claims, three inventive features are present: user-specific anaerobic threshold determination using incidence bins with highest change, effort identification from comparison to threshold criteria, and terrain-based resistance with training type segment classification.
Incidence-bin-based user anaerobic threshold determination
The anaerobic threshold is determined by collecting activity data of the particular user prior to the activity session, forming a plurality of incidence bins containing numbers of incidences of the activity data within specific ranges, and identifying at least one incidence bin of the plurality of incidence bins with the highest change, wherein the anaerobic threshold is based on a range of the identified at least one incidence bin.
Effort identification from difference to anaerobic-threshold-based criteria
The activity data is compared with the set of threshold criteria to identify an effort of the particular user, wherein the effort is defined by a difference between the activity data and the threshold criteria based on the anaerobic threshold of the particular user.
Terrain-based resistance and training type segment classification
A resistance experienced by the particular user is determined, wherein the resistance comprises a measure of terrain, and multiple durations of the activity are classified into one or more predefined training type segments based on the resistance experienced and the identified effort.
The claims collectively cover user-specific anaerobic threshold determination using incidence bins with highest change, effort identification as a difference from anaerobic-threshold-based criteria, terrain-based resistance determination, and classification into predefined training type segments.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?