Machine Learning Skeleton Torque Analysis for Rehabilitation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current physical therapy methods face challenges in providing timely and accurate feedback, as they often rely on intermittent and subjective evaluations, are costly, and require expensive wearable sensors that can impede patient motion, limiting their effectiveness in monitoring musculoskeletal rehabilitation progress, especially in home-based settings.
Innovation Solution
A method using machine learning models to analyze images captured by a camera, identifying joints and determining torque values without wearable sensors, enabling continuous, real-time monitoring of range of motion and muscle strength from multiple joints, and generating risk assessment reports based on predetermined thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If wearable sensors are used to monitor musculoskeletal rehabilitation, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent replaces mechanical wearable sensors with a computer vision-based system using cameras and machine learning models. The system captures images or video of the patient's movements and uses pose estimation algorithms to track joint positions and calculate kinematic parameters, eliminating the need for physical sensors attached to the body.
Solution Approach 2:
The system creates a virtual 3D model of the patient's skeleton by mapping 2D image data through machine learning algorithms. This digital twin or virtual replica allows for continuous monitoring of joint angles, range of motion, and movement patterns without physical contact, providing the same measurement functionality as wearable sensors but through optical copying.
2Reliability
If intermittent evaluations are used in physical therapy, then cost is reduced, but reliability of monitoring deteriorates
Solution Approach 1:
The system enables continuous monitoring by capturing images or video streams throughout the rehabilitation exercise and processing them in real-time or near real-time. This continuous data stream provides uninterrupted tracking of joint angles, range of motion, and movement quality, eliminating the gaps inherent in intermittent manual evaluations.
Solution Approach 2:
The system provides automated analysis and feedback without requiring continuous physical therapist intervention. The machine learning models automatically process the visual data, calculate kinematic parameters, compare them against target ranges, and generate feedback reports, allowing the system to monitor itself and provide objective measurements consistently.
3Measurement precision
If multiple parameters are monitored simultaneously, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system is designed to simultaneously monitor multiple parameters including joint angles, range of motion, movement velocity, acceleration, and temporal-spatial characteristics all through a single integrated camera-based platform. The machine learning pipeline processes visual data to extract all these parameters concurrently, providing multi-parameter monitoring without requiring separate specialized devices for each measurement.
Data Source
AI summary
A method can include receiving (1) images of at least one subject and (2) at least one total mass value for the at least one subject. The method can further include executing a first machine learning model to identify joints of the at least one subject. The method can further include executing a second machine learning model to determine limbs of the at least one subject based on the joints and the images. The method can further include generating three-dimensional (3D) representations of a skeleton based on the joints and the limbs. The method can further include determining a torque value for each limb, based on at least one of a mass value and a linear acceleration value, or a torque inertia and an angular acceleration value. The method can further include generating a risk assessment report based on at least one torque value being above a predetermined threshold.