Cognitive platform for deriving effort metric for optimizing cognitive treatment
Inventors
Alailima, Titiimaea • DeLoss, Denton J. • Heusser, Andrew C. • Villena, Claudia A.
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12026636-B2
Publication Date
2024-07-02
Expiration Date
Abstract
Adaptive modification and presentment of user interface elements in a computerized therapeutic treatment regimen. Embodiments of the present disclosure provide for non-linear computational analysis of cData and nData derived from user interactions with a mobile electronic device executing an instance of a computerized therapeutic treatment regimen. The cData and nData may be computed according to one or more artificial neural network or deep learning technique to derive patterns between computerized stimuli or interactions and sensor data. Patterns derived from analysis of the cData and nData may be used to define an effort metric associated with user input patterns in response to the computerized stimuli or interactions being indicative of a measure of user engagement or effort. A computational model or rules engine may be applied to adapt, modify, configure or present one or more graphical user interface elements in a subsequent instance of the computerized therapeutic treatment regimen.
Core Innovation
The invention relates to a cognitive therapeutic platform that configures an instance of a cognitive training application for a first end user and an instance of a companion application for a second end user. The cognitive training application presents computerized stimuli or interactions configured to elicit a specified response from the first end user, and the companion application includes a graphical user interface. The first end user and the second end user are linked in an application database by enabling at least one data transfer interface between the cognitive training application and the companion application.
User activity data is received at an application server, including input sensor data indicative of user-generated responses to the stimuli or interactions. The application server processes the user activity data according to a machine learning framework comprising at least one machine learning model that classifies stimulus-response patterns to generate a classified dataset comprising one or more data labels for one or more attributes of the user activity data. The classified dataset is stored in the application database.
The companion application is configured or modified at the graphical user interface according to datapoints from the classified dataset, and it presents graphical user interface elements to the second end user. The graphical user interface elements comprise at least one computerized adjustable element configured to provide quantitative metrics for the first end user according to the classified dataset, including a graphical indication that a degree of effort for the first end user is at or below a target threshold based on an output of the machine learning model.
Based on processing the user activity data, the application server generates recommendations for improving the degree of effort for the first end user in a subsequent instance of the cognitive training application and presents the recommendations to the second end user. User-generated input is received via the graphical user interface in accordance with the recommendations, and the application server processes the received input to configure or modify subsequent computerized stimuli or interactions for the subsequent instance of the cognitive training application.
Claims Coverage
The partial content explicitly includes three independent claims, a method, a system, and a non-transitory computer-readable medium, that share the same core inventive concept. Four inventive features are repeatedly emphasized: linkage through a data transfer interface, machine-learning classification of stimulus-response patterns into a labeled dataset, companion GUI metrics with an effort threshold indication, and recommendation-driven modification of subsequent cognitive training stimuli.
Cognitive training and companion linkage via data transfer interface
Configuring an instance of a cognitive training application for a first end user and an instance of a companion application for a second end user; linking the first end user and the second end user in an application database by enabling at least one data transfer interface between the cognitive training application and the companion application.
Machine-learning classification of stimulus-response patterns into labeled dataset
Receiving user activity data comprising input sensor data indicative of user-generated responses; processing the user activity data according to a machine learning framework comprising at least one machine learning model configured to classify stimulus-response patterns to generate a classified dataset comprising one or more data labels; storing the classified dataset in the application database.
Companion GUI quantitative adjustable elements with effort threshold indication
Presenting the companion application with a graphical user interface; configuring or modifying graphical user interface elements according to datapoints from the classified dataset; presenting at least one computerized adjustable element configured to provide quantitative metrics for the first end user; and providing a graphical indication that a degree of effort for the first end user is at or below a target threshold based on an output of the machine learning model.
Recommendation-driven modification of subsequent stimuli based on second end user input
Processing the user activity data to generate recommendations for improving the degree of effort for the first end user in a subsequent instance of the cognitive training application; presenting the recommendations to the second end user; receiving user-generated input in accordance with the recommendations; processing the received input; and configuring or modifying subsequent computerized stimuli or interactions in response to the processed input.
System implementation with application server and executable instructions
A computer-implemented system comprising a first end user computing device, a second end user computing device, and an application server communicably engaged with the first and second end user computing devices; the application server comprising at least one processor and a non-transitory computer readable medium encoded with processor-executable instructions to perform the recited operations.
Non-transitory computer-readable medium implementing the effort classification and recommendation loop
A non-transitory computer-readable medium with processor-executable instructions stored thereon that, when executed, command one or more processors to perform the recited operations for configuring the applications, classifying stimulus-response patterns into a classified dataset, configuring companion graphical user interface elements, generating recommendations, receiving second end user input, and modifying subsequent computerized stimuli or interactions.
Across the independent claims, the shared inventive concept is classifying stimulus-response patterns from user activity data into a labeled classified dataset using a machine learning framework, then using datapoints from that dataset to drive companion application graphical user interface elements with quantitative metrics and a target-threshold effort indication, while using recommendations plus second end user input to configure or modify subsequent cognitive training stimuli or interactions.
Stated Advantages
Provides recommendations for improving the degree of effort for the first end user in a subsequent instance of the cognitive training application.
Derives a degree of effort for the first end user by analyzing stimulus-response patterns and a temporal relationship of input sensor data.
Provides quantitative metrics for the first end user through at least one computerized adjustable element in the companion application.
Provides a graphical indication to the second end user that a degree of effort for the first end user is at or below a target threshold.
Documented Applications
No documented applications found
Interested in licensing this patent?