Generating and searching data structures that facilitate measurement-informed treatment recommendation
Inventors
Larkin, Roland • Gottipati, Srikanth • Moghadam, Reza • Tyler, Carolyn • HO, Gregory
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11798675-B2
Publication Date
2023-10-24
Expiration Date
Abstract
Methods, systems, apparatus, including computer programs encoded on computer storage media for generating psychiatric treatment recommendations. In one aspect, the method includes actions of receiving, by a server and from a first user device, one or more data structures that collectively include fields structuring data that represents (i) current patient symptoms and (ii) current patient diagnoses, generating, by the server, rendering data structure that includes fields structuring data that represents rendering data that, when rendered on a display device, presents a patient dashboard based on the data that is received from the mobile device, and providing, by the server, the rendering data structure to a second user device that is different than the first user device, wherein the second user device is configured to render the rendering data structured by the fields of the rendering data structure to output the patient dashboard on the display of the second user device.
Core Innovation
The invention relates to generating measurement-informed psychiatric treatment recommendations for a user by using a first user device and one or more processing devices. The first user device provides a plurality of visual stimuli for output on a display and generates one or more first data structures that include first keyed data identifying the user and a first field structuring data representing a measurement of the user, the measurement representing a response of the user to the presentation of the plurality of visual stimuli.
The one or more processing devices access one or more second data structures that include second keyed data identifying the user and a second field structuring data representing a diagnosis for the user by a medical professional. The processing devices identify the first keyed data in the first data structures and the second keyed data in the second data structures, correlate the first data structures with the second data structures based on the keyed data, and extract from the correlated data data representing the measurement of the user and the diagnosis for the user.
The invention uses the extracted measurement and diagnosis to generate an input data structure for a machine learning model that outputs one or more psychiatric treatment recommendations. The machine learning model is trained based on a training data item including training data and a label, where the label indicates a selection of the one or more psychiatric treatment recommendations used as a label, and where the training data includes measurements of users and corresponding diagnoses to learn associations between measurements, diagnoses, and psychiatric treatment recommendations.
Output data generated by the machine learning model is used to determine a treatment recommendation for the user, and rendering data structure fields are generated so that a user device displays a dashboard showing the treatment recommendation based on the output data.
Claims Coverage
The independent claims are clm-00001, clm-00011, and clm-00016. Across these independent claims, three main inventive features are defined: presenting visual stimuli and receiving measurement data keyed to the user, correlating extracted measurements with diagnosis keyed to the user for machine learning input, and using model output to generate rendering data for a dashboard on a different user device.
Visual stimuli and user measurement keyed data structures
Provide a plurality of visual stimuli for output on a display of a first user device associated with a user; access one or more first data structures, generated by the first user device, that include first keyed data identifying the user and a first field structuring data that represents a measurement of the user, the measurement representing a response of the user to the presentation of the plurality of visual stimuli.
Correlated measurement and diagnosis for machine learning input
Access one or more second data structures that include second keyed data identifying the user and a second field structuring data that represents a diagnosis for the user by a medical professional; identify the first keyed data in the one or more first data structures and the second key data in the one or more second data structures; correlate the one or more first data structures with the one or more second data structures based on the first keyed data and the second key data; extract, from the correlated data, data representing the measurement of the user and the diagnosis for the user; generate an input data structure based on the extracted data for input to a machine learning model that outputs one or more psychiatric treatment recommendations for the user.
Trained machine learning model and dashboard rendering on a different user device
Train the machine learning model based on a training data item comprising training data and a label, the label indicating a selection of the one or more psychiatric treatment recommendations, and the training data comprising measurements of users and corresponding diagnoses to learn associations between the measurements of users, the diagnoses, and the psychiatric treatment recommendations; obtain output data generated by the machine learning model; determine a treatment recommendation for the user based on the output data; generate a rendering data structure that, when rendered by a user device, causes the user device to display a dashboard that displays the treatment recommendation; provide the rendering data structure to a second user device that is different than the first user device.
Taken together, the independent claims cover a measurement-informed psychiatric treatment recommendation workflow that collects user responses to visual stimuli as keyed measurement data structures, correlates extracted measurements with keyed clinician diagnoses as machine learning input, and renders treatment recommendations on a dashboard via a rendering data structure provided to a different user device. The independent claims also define that the machine learning model is trained using labeled training data with measurements and corresponding diagnoses to learn associations between measurements, diagnoses, and psychiatric treatment recommendations.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?