Intelligent Upper Limb Assessment System for Stroke Patients Based on the Fugl-Meyer Assessment (FMA)

View at publisher Summarise with VIRAT
Authors: Xiao-Yun Zhang, Chien-Hsiou Liu, Pei-Hsuan Huang, Chih-Yu Wang
Cited by: 0
References: 6

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This study presents an intelligent upper-limb function assessment system for stroke patients. Using MediaPipe pose estimation technology, the system automatically extracts 3D coordinates of the shoulder, elbow, and wrist joints and calculates real-time angle variations to identify compensatory movements. A visualized analysis interface displays synchronized angle curves, motion stability, and compensation conditions, while exporting CSV files containing joint angles, spatial coordinates, and compensation labels for clinical evaluation or AI model training. In addition, the system integrates a generative AI module based on a large language model (LLM) that automatically produces evaluation reports and personalized training recommendations based on angular and compensatory features, assisting clinicians in monitoring rehabilitation progress. Built with a low-cost and non-contact design, the system effectively quantifies upper-limb movement performance and compensatory behavior, demonstrating its feasibility and potential for clinical application.

Fields of Study & Subjects

Stroke Rehabilitation and Recovery Human Pose and Action Recognition Medical Practices and Rehabilitation

Publication details

Journal / source
International Conference on Consumer Electronics
Pages
1–4
Publication date
3 February 2026
Language
EN