基于模式识别的脑卒中患者上肢运动功能自动化评估系统
By combining Hidden Markov Models and Support Vector Machines, a refined and automated assessment of upper limb motor function in stroke patients was achieved, solving the problem of coarse assessment granularity in existing technologies and improving the accuracy and consistency of the assessment.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- TIANJIN UNIV OF TRADITIONAL CHINESE MEDICINE
- Filing Date
- 2026-06-16
- Publication Date
- 2026-07-17
AI Technical Summary
Existing methods for assessing upper limb motor function in stroke patients cannot achieve fine segmentation and neglect the time alignment between surface electromyography signals and joint angle trajectories, resulting in coarse assessment granularity and difficulty in capturing the heterogeneity of motor control ability and functional impairments in specific motor segments.
By acquiring surface electromyography signal sequences and joint angle trajectory sequences, a hidden Markov model is used for motion phase segmentation. Multimodal feature vectors are extracted using a clustering algorithm, and a support vector machine is used for grade recognition to generate a comprehensive evaluation result.
It enables refined and automated assessment of upper limb motor function in stroke patients, improving the objectivity and consistency of the assessment and accurately identifying the level and stage of abnormal motor function.
Smart Images
Figure CN122398285A_ABST