Selection of Machine Learning Models for Predicting Movement Intentions Related to Gait in Users Simulating Post-Stroke Rehabilitation Patients Using Inertial Sensors

• Grant number: 24/17013-2
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: December 01, 2024
• End date: November 30, 2025
• Field of knowledge: Engineering - Biomedical Engineering - Bioengineering
• Principal Investigator: Gustavo Jose Giardini Lahr
• Grantee: Felipe Iannarelli Martins
• Host Institution: Instituto Israelita de Ensino e Pesquisa Albert Einstein (IIEPAE). Sociedade Beneficente Israelita Brasileira Albert Einstein (SBIBAE). São Paulo , SP, Brazil

Abstract

Stroke often causes motor difficulties, particularly in locomotion, and proper rehabilitation is essential to restore the patient's autonomy. Post-stroke rehabilitation patients can benefit from devices that assist in mobility and prevent falls, especially while walking. However, developing these technologies requires predicting the patient's locomotion intention so that the devices can execute appropriate movements. This project aims to develop a wearable technology with a low usage burden using IMUs (Inertial Measurement Units) to collect user gait data and predict their movement intention. Sensors will be placed on different parts of the lower limbs to monitor important walking parameters. This data will train machine learning models that accurately predict patients' movement intentions. The model should be able to classify with high performance the user's intent to stop, continue, or turn left or right. The best model will then be selected, and feature selection tests will be conducted to reduce the number of wearables the user needs. At this project stage, tests will be conducted with healthy subjects whose lower limbs will be immobilized to simulate the hemiparetic gait of post-stroke patients. It is expected that the quality of the prediction of the user's intention will be comparable to computer vision-based models, which are limited in portability. This project will contribute to developing intelligent assistive technologies to aid in walking.