PhD scholarship in Machine Learning and Accessible Music Interfaces for Neurorehabilitation

Stroke patients who suffer from hemiparesis often reduce the use of the affected extremity, consequently leading to the further deterioration of the related motor function. Although traditional physiotherapy techniques have been shown to be effective in treating hemiparesis, such techniques present several limitations. In this PhD project, we propose to investigate novel music-based rehabilitation approaches in which patients control digital music instruments using motion capture sensors (e.g. webcam, mobile devices gyroscopes and accelerometers) and electromyogram sensors (e.g. Myo). This approach to stroke rehabilitation can be significantly more efficient than traditional physiotherapy as has been shown in a (randomized, double-blind, controlled) pilot study we conducted in the hospital Forum Consorci MAR. In the proposed project, we will investigate the benefits of different interventions for recovery in chronic stroke patients, explore interventions for motor function recovery of different parts of the body, target other conditions in addition to stroke, and potentiate the use of mobile devices’ sensors to enable at-home rehabilitation.

Topic: Machine learning, signal processing, accessible music interfaces

- Master degree in computer science, sound music computing or related area, excellent programming skills, music knowledge
- Admission to the PhD program of the Department of Information and Communication Technologies at UPF is a prerequisite to enjoy the contract.