Machine Learning Model for Motor Function Variation Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for estimating motor function index value variation, particularly the TUG value, do not account for changes from the start of rehabilitation and after a predetermined period, failing to provide accurate assessments for individuals needing self-support nursing care.
Innovation Solution
A method and device that utilize machine learning to estimate motor function index value variation by acquiring physical strength measurement data, extracting relevant feature amounts, and inputting them into a learned estimation model, incorporating additional data such as TUG test results, one-leg stand test results, and basic information to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional methods are used to estimate motor function index value, then the estimation can be performed, but the accuracy of estimating motor function index value variation from start of rehabilitation and after predetermined period is insufficient
Solution Approach 1:
The patent applies preliminary action by collecting and storing physical strength measurement data at the start of rehabilitation before the actual rehabilitation process begins. This baseline data is then used in conjunction with machine learning models to predict and accurately estimate motor function index value variation after a predetermined period, enabling accurate assessment of rehabilitation progress without requiring multiple actual measurements during the rehabilitation process.
Solution Approach 2:
The patent replaces conventional mechanical measurement methods with a machine learning-based estimation system. Instead of relying on traditional physical testing and manual assessment methods, the system uses machine learning models trained on physical strength measurement data to automatically estimate motor function index value variation, significantly improving estimation accuracy and enabling reliable prediction of rehabilitation outcomes.
2Measurement precision
If physical strength measurement data is collected and machine learning is applied to estimate motor function index value variation, then estimation accuracy is improved, but data processing complexity and computational requirements increase
Solution Approach 1:
The patent reduces computational complexity during runtime by performing preliminary actions during the data collection phase. Physical strength measurement data is collected and pre-processed at the start of rehabilitation, and machine learning models are trained in advance. This allows the system to make accurate estimations later with minimal real-time computation, balancing accuracy requirements with system complexity.
Data Source
AI summary
An information processing device acquires physical strength measurement data of a target; extracts a plurality of feature amounts based on the physical strength measurement data having been acquired; calculates a motor function index value variation of the target by inputting the plurality of feature amounts having been extracted to a learned motor function index value variation estimation model obtained by machine learning using, as training data, the plurality of feature amounts and actual measurement data of the motor function index value variation from start of rehabilitation and after passage of a predetermined period of time, which pertain to each of a plurality of subjects; and outputs the motor function index value variation having been calculated.