Method and system for using artificial intelligence to assign patients to cohorts and dynamically controlling a treatment apparatus based on the assignment during an adaptive telemedical session
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-12246222-B2
Publication Date
2025-03-11
Expiration Date
Abstract
A method includes receiving data pertaining to a user that uses a treatment apparatus to perform a treatment plan. The data includes characteristics of the user, the treatment plan, and a result of the treatment plan. The method includes assigning the user to a cohort representing people having similarities to the characteristics of the user. The method includes receiving second data pertaining to a second user, the second data comprises characteristics of the second user. The method includes determining whether at least some of the characteristics of the second user match with at least some of the characteristics of the user, assigning the second user to the first cohort, and selecting, via a trained machine learning model, the treatment plan for the second user, and controlling, based on the treatment plan, the treatment apparatus while the second user uses the treatment apparatus.
Core Innovation
The document describes systems and methods for telemedical rehabilitation and exercise that use an artificial intelligence engine with one or more machine learning models to assign users to cohorts and select treatment plans. First data pertaining to a first user is received, where the first data comprises characteristics of the first user, the treatment plan, and a result of the treatment plan. Based on the first data, the first user is assigned to a first cohort of a plurality of cohorts, and the first cohort represents people having at least one similarity to the characteristics of the first user.
Second data pertaining to a second user is received, and the system determines whether at least some of the characteristics of the second user match at least some of the characteristics of the first user assigned to the first cohort. Responsive to determining that the characteristics match, the one or more machine learning models assign the second user to the first cohort and select the treatment plan for the second user.
After selecting the treatment plan, control instructions are transmitted to an electromechanical machine used by the second user. A first control instruction electronically adjusts a pedal radius setting of the electromechanical machine to be in compliance with at least a first range of motion specified in the treatment plan, and third data is received comprising the first range of motion achieved by the second user performing the treatment plan. Based on the first range of motion achieved, a second control instruction is transmitted that electronically adjusts the pedal radius setting to be in compliance with at least a second range of motion specified in the treatment plan.
During a telemedicine session, a medical professional is provided with a recommendation for a treatment plan and a selected treatment plan is received from a medical professional computing device. The system then controls the electromechanical machine based on the selected treatment plan while the second user uses the machine. The architecture includes processing devices and interfaces used to implement the cohort-based assignment, treatment-plan selection, and control instructions.
Claims Coverage
The document includes three independent claims: a method claim, a tangible non-transitory computer-readable medium claim, and a system claim. Across these independent claims, the inventive features combine cohort-based assignment via machine learning with treatment-plan selection and dynamically controlling an electromechanical machine by electronically adjusting a pedal radius setting to ranges of motion specified in the treatment plan.
Cohort-based assignment using machine learning
assign, via one or more machine learning models and based on the first data, the first user to a first cohort of a plurality of cohorts, wherein the one or more machine learning models are trained to assign the first user to the first cohort by comparing the first data of the first user to other data of people previously assigned to the plurality of cohorts, and wherein the first cohort represents the people having an at least one similarity to the characteristics of the first user
Second-user characteristic matching and cohort assignment
determine whether at least some of the characteristics of the second user match with at least some of the characteristics of the first user assigned to the first cohort; responsive to determining at least some of the characteristics of the second user match at least some of the characteristics of the first user, assign, via the one or more machine learning models, the second user to the first cohort
Machine-learning treatment-plan selection
select, via the one or more machine learning models, the treatment plan for the second user
Dynamic electromechanical-machine control via pedal radius in compliance with treatment-plan ranges of motion
transmitting, from one or more processing devices, a first control instruction to the electromechanical machine, wherein the first control instruction electronically adjusts a pedal radius setting of the electromechanical machine, such adjustment to be in compliance with at least a first range of motion specified in the treatment plan; receiving third data comprising the first range of motion achieved; and transmitting a second control instruction that electronically adjusts the pedal radius setting to be in compliance with at least a second range of motion specified in the treatment plan
Telemedicine plan recommendation and selection prior to machine control
during a telemedicine session, provide a medical professional with a treatment-plan recommendation for use with an electromechanical machine, receive a selected treatment plan, and then control the electromechanical machine based on that selected plan while a second user uses the machine
Taken together, the independent claims cover receiving electromechanical-machine treatment-plan data for a first user; using machine learning to assign users to cohorts by comparing characteristics; matching second-user characteristics to assign the second user and select a treatment plan; and electronically adjusting a pedal radius setting of the electromechanical machine so the adjustment is in compliance with ranges of motion specified in the selected treatment plan, with updates based on range of motion achieved.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?