Automated Wolf Motor Function Test (WMFT) for Upper Extremities Rehabilitation

Abstract

The Wolf Motor Function Test (WMFT), which involves real time timing measurements, is used to define the functional ability of the paretic limb of post stroke patients. The accuracy and consistency of the timing results are crucial for scoring, while manual stopwatch measuring introduces mistakes and uncertainty. A technique which uses camera and sensors to automate the test and achieve accurate results is proposed here. This system measures time for seven tasks with a monitor camera and a Xilinx Virtex II Pro Field Programmable Gate Array (FPGA). The FPGA was mainly chosen for its better performance compared to general purpose System on Chip (SoC) for computational intensive tasks, such as image processing in this system. The accuracy of the system has been validated in a user study on selected WMFT tasks with ground truth time measurement obtained from pressure sensors. Suggestions are also given on how the framework can be adapted to the remaining tasks.