Development and testing of utility and usability of a machine learning-based myoelectric hand orthosis for people with amyotrophic lateral sclerosis
amyotrophic lateral sclerosis, dynamic orthoses, artificial intelligence, usability testing
Introduction: Recent technological advances driven by science have reached the healthcare field in a promising way, particularly in the realm of assistive technologies (ATs). Among these, hand orthoses have been the focus of technological improvements, given the importance of manual function for human functionality. In this context, a crucial step in the maturation of ATs designed for users with neurological conditions is the evaluation of their utility and usability. These analyses consider factors such as ease of use, comfort, practicality, time and energy efficiency, and should be taken into account from the early stages of device development. Objective: To implement a neuromuscular analysis system based on machine learning (ML) in the operation of different myoelectric devices for hand function rehabilitation and to test its usability in individuals with amyotrophic lateral sclerosis (ALS). Methods: This study is a technological development project aimed at designing and evaluating the utility and usability of a hand orthosis. It is conducted through a collaboration between the Laboratory for Technological Innovation in Health (LAIS/HUOL/UFRN), the Movement Analysis and Intervention Laboratory (LIAM)/Department of Physiotherapy/UFRN, and the Santos Dumont Institute (ISD). The study was approved by the ethics committee of the Federal University of Rio Grande do Norte (UFRN), CAAE: 25687819.3.0000.5537. The myoelectric hand orthosis and the ML-based neuromuscular analysis system are being developed by an interdisciplinary team composed of engineers and physiotherapists. Once these ATs reach a sufficient level of maturity, 10 healthy volunteers and 10 volunteers with ALS will be recruited for usability testing. Initially, clinical data will be collected using the following instruments: personal and health information (custom form), fatigue assessment (Fatigue Severity Scale), pain assessment (Numerical Visual Scale), functionality assessment (Amyotrophic Lateral Sclerosis Functional Rating Scale - Revised), and usability assessment (Quebec User Evaluation of Satisfaction with Assistive Technology - QUEST 2.0). Subsequently, the system’s flexibility will be tested by integrating it with an existing hand rehabilitation device available on the market. Descriptive and comparative analyses of study variables will be performed using the Statistical Package for the Social Sciences (SPSS), version 20.0. Expected results: The myoelectric hand orthosis activated by the neuromuscular analysis system is expected to be a functional device that provides a good user experience with concrete results regarding its utility, usability, and applicability for improving the manual function of volunteers with ALS. Additionally, the system is expected to be flexible enough to be used with other wearable devices and volunteers with ALS presenting different manual function profiles.