Multisensor Data Fusion Algorithm for Contactless 3D Position Measurement for Post-Stroke Hand Rehabilitation

• Main Authors: Sophian, Ali, Md Yusof, Hazlina, Sediono, Wahju, Sudirman, Sud
• Format: Monograph
• Language: English
• Published: 2018
• Subjects: T Technology (General); TK7885 Computer engineering
• Online Access: http://irep.iium.edu.my/62936/1/Final%20Project%20Report%20RIGS%2015-151-0151%20r4b.pdf; http://irep.iium.edu.my/62936/

Repetitive hand motion exercises help the patients regain their hand motor control. One of the widely used therapies of this type is the patient squeezing a flexible exercise ball in his/her hand repetitively. The exercise balls come at different levels of resistance to accommodate the different levels of limitation of the patients’ hands. However, one of the challenges is to measure objectively the progress that has been made without making any contact such that no additional weights loading the affected arm or hand of the patient. The presence of the exercise ball in the hand adds a degree of difficulty to the problem when an optical solution is adopted. This research attempted to investigate the enabler technology for contactless quantitative measurement system for monitoring the progress in such hand therapy. Evaluation of potential commercial-grade stereo-vision systems have been performed and fingertip detection algorithms have been proposed and evaluated. A total of 4200 images, 2100 fingertip images and 2100 non-fingertip images, were used in the experiment. Our results show that the success rates for the fingertip detection are higher than 94% which demonstrates that the proposed method produces a promising result for fingertip detection for therapy-ball-holding hands.