Description du projet
Un nouveau système de contrôle des neuroprothèses
Les dispositifs neuroprosthétiques utilisent des électrodes pour créer une interface avec le système nerveux et tenter de restaurer la perte de fonction ou de mouvement. Les scientifiques du projet Neuroprosthesis-UI, financé par l’UE, proposent de développer une interface utilisateur permettant aux personnes présentant un handicap des membres supérieurs de contrôler la neuroprothèse grâce à des compétences motrices résiduelles. Cette interface comprendra plusieurs capteurs pour capturer les contractions musculaires, et grâce à l’apprentissage automatique, elle décodera les entrées de l’utilisateur en intentions de mouvement. Ce système hybride aidera les patients présentant des handicaps des membres supérieurs tels qu’une lésion de la moelle épinière, un accident vasculaire cérébral et une sclérose en plaques à réaliser des activités quotidiennes de manière indépendante.
Objectif
In this project, I will develop a user interface that will allow persons with upper limb disabilities to control neuroprosthesis using their residual motor skills. This interface will consist of inertial sensors (IMU) and electromyography (EMG) that are capable of capturing movements and muscle contraction that even persons with high tetraplegia still can control. The interface will also be able to learn different inputs, customizing the system for each user. This requires techniques of machine learning, making it flexible and indicated for users with different upper limb disabilities, such as spinal cord injury, stroke and multiple sclerosis. The machine learning techniques will classify the user inputs into desired commands, working as an intention decoder. The interface will be used to control a hybrid upper limb neuroprosthesis based on surface functional electrical stimulation (FES) and a semi passive mechanical orthosis. The system will allow users to perform activities of daily life independently. To my knowledge, such a hybrid system with FES, and controlled by an interface based on IMUs, EMG and machine learning techniques is novel. I will be working with Christine Coste, an expert in neuroprosthesis for disabled persons, and her interdisciplinary team, which consist of engineers and health professionals with vast experience in neurorehabilitation. This fellowship will enable the transfer of knowledge between her team and me through experiments with real patients and mutual training. I can contribute to the team with my expertise in machine learning and control, whereas they have vast access to patients, medical doctors, mechanical designers, electrical stimulators and sensors. This project is going to be an important step in my career as expand my network in Europe, develop my skills as a biomedical engineer and improve my research experience towards becoming a world-leading expert in neurorehabilitation engineering.
Champ scientifique
- medical and health sciencesbasic medicineneurologymultiple sclerosis
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringsensors
- medical and health sciencesbasic medicineneurologystroke
- natural sciencescomputer and information sciencesartificial intelligencemachine learning
Programme(s)
Régime de financement
MSCA-IF - Marie Skłodowska-Curie Individual Fellowships (IF)Coordinateur
78153 Le Chesnay Cedex
France