Method and system for remotely monitoring the physical and psychological state of an application user using altitude and/or motion data and one or more machine learning models
Inventors
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11967432-B2
Publication Date
2024-04-23
Expiration Date
Abstract
Altitude and/or motion data is collected from one or more devices associated with multiple application users, and the altitude and/or motion data is utilized to generate machine learning-based predictive model training data. One or more machine learning-based predictive models are trained using the machine learning-based predictive model training data. Current altitude and/or motion data is collected from one or more devices associated with a current user of an application and is provided to the one or more trained machine learning-based predictive models. Predictions regarding the physical and/or psychological state of the user are received from the one or more trained machine learning-based predictive models and analyzed to identify changes or anomalies in the user's physical and/or psychological state. Upon identification of changes or anomalies in the user's physical and/or psychological state, one or more actions are taken to assist the user.
Core Innovation
The invention relates to a digital therapeutic application in which a computing system identifies one or more physical medical conditions of one or more users and uses objectively measured altitude data and/or objectively measured motion data collected from one or more remote monitoring devices associated with the users. Based on the measured altitude or motion data during a defined period of time, the system identifies a specific physical activity being performed and uses the number of occurrences of the specific physical activity in that defined period of time as an indication of a status of the identified physical medical conditions.
The system generates physical activity count data representing data associated with the number of occurrences of the specific physical activity performed by each user during the defined period of time. For each user, the system obtains mental state data and/or physical state data during the defined period of time in which the user is performing the specific physical activity, and correlates the mental state data or physical state data with the physical activity count data.
The system processes the correlated mental state data and physical activity count data to generate machine learning-based mental state prediction model training data, and/or processes the correlated physical state data and physical activity count data to generate machine learning-based physical state prediction model training data. The machine learning-based prediction model training data is provided to one or more machine learning-based prediction models to generate one or more trained machine learning-based mental state prediction models and/or one or more trained machine learning-based physical state prediction models.
Using the current user prediction data, the computing system modifies therapeutic treatment content provided to a current user through a graphical user interface of the digital therapeutic application. The current user is a user other than the one or more users used to train the model, and the therapeutic treatment content provides treatment for the one or more previously diagnosed physical medical conditions of the current user.
Claims Coverage
The independent claims include clm-00001, clm-00008, clm-00015, and clm-00022. Across these independent claims, there are inventive features that detect specific physical activities from objectively measured altitude or motion data, generate physical activity count data, correlate that activity data with mental state data and/or physical state data to create machine learning-based prediction model training data, train machine learning-based prediction models to output prediction data for a current user, and modify therapeutic treatment content in a digital therapeutics application based on the prediction data.
Altitude-based physical activity identification from objective monitoring data
Collecting objectively measured altitude data from one or more remote monitoring devices associated with users, identifying a specific physical activity based on the objectively measured altitude data during a defined period of time, and using a number of occurrences of the specific physical activity in the defined period of time as an indication of a status of identified physical medical conditions.
Motion-based physical activity identification from objective monitoring data
Collecting objectively measured motion data from one or more remote monitoring devices associated with users, identifying a specific physical activity based on the objectively measured motion data during a defined period of time, and using a number of occurrences of the specific physical activity in the defined period of time as an indication of a status of identified physical medical conditions.
Physical activity count data generation
Generating physical activity count data representing data associated with the number of occurrences of the specific physical activity being performed by each user in the defined period of time.
Mental state correlation for machine learning training
Obtaining mental state data for each user during the defined period of time in which the user is performing the specific physical activity, correlating that user's mental state data with that user's physical activity count data, and collecting and processing the correlated mental state data and physical activity count data to generate machine learning-based mental state prediction model training data.
Physical state correlation for machine learning training
Obtaining physical state data for each user during the defined period of time in which the user is performing the specific physical activity, correlating that user's physical state data with that user's physical activity count data, and collecting and processing the correlated physical state data and physical activity count data to generate machine learning-based physical state prediction model training data.
Machine learning model training to generate trained prediction models
Providing the machine learning-based mental state prediction model training data or the machine learning-based physical state prediction model training data to one or more machine learning-based prediction models to generate one or more trained machine learning-based mental state prediction models or one or more trained machine learning-based physical state prediction models.
Predictive monitoring for a current user using trained models
Collecting objectively measured current altitude data and identifying a specific physical activity currently being performed by the current user during a defined period of time, generating current physical activity count data, providing the current physical activity count data to the one or more trained machine learning-based mental state prediction models or the one or more trained machine learning-based physical state prediction models, and receiving current user mental state prediction data or current user physical state prediction data from the trained models.
Therapeutic treatment content modification based on prediction data
Providing a current user with a digital therapeutic application and a graphical user interface to the digital therapeutic application, providing therapeutic treatment content that provides treatment for one or more previously diagnosed physical medical conditions of the current user, and based at least in part on the received current user prediction data, modifying the therapeutic treatment content provided to the current user through the graphical user interface such that the modified therapeutic treatment content provides treatment for the one or more previously diagnosed physical medical conditions.
Across clm-00001, clm-00008, clm-00015, and clm-00022, the claim coverage centers on generating physical activity count data from objective altitude or motion measurements, correlating activity-derived count data with mental state data and/or physical state data to create machine learning training data, training machine learning-based prediction models, and using prediction outputs to modify therapeutic treatment content in a digital therapeutics application for a current user with previously diagnosed physical medical conditions.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?