[M] Decoding EMG using A Riemannian Geometry approach

Master Assignment

Decoding EMG using A Riemannian Geometry approach

Type: Master EE/CS/HMI/others 

Period: TBD

Student: (Unassigned)

If you are interested please contact :

Description:

In the Netherlands alone, more than two thousand patients undergo a major amputation at the lower limb per year [1]. Amputees rehabilitate and learn to perform daily life activities using their prosthesis, although the prosthesis is not able to act as a complete replacement of the lost body part. The European Horizon 2020 project MyLeg aims at "developing a smart and intuitive osseointegrated transfemoral prosthesis embodying advanced dynamic behaviours [2]. The project consists of several European partners: University of Groningen, University of Bologna, University of Twente, Roessingh Research & Development (RRD), Radboud University Medical Center (RUMC) and Össur and one Australian partner: Norwest Advanced Orthopeadics. RRD is responsible for developing high-level control algorithms to be used for prosthetic control.

Assignment:

Individuals with an amputation do not have direct control over their prosthetic knee and/or ankle [1]. The movements are controlled indirectly by using the stump. This lack of control makes dynamic balance difficult and makes a prosthesis limited in intuitiveness [1]. Proportional or direct control using EMG is investigated in some studies [3]–[5], which allows patients to control their knee or ankle joint. It would be interesting to see if a patient could control his prosthetic device in non-weight bearing situations using EMG, while being robust.

The main question for this assignment is whether a Riemannian Geometry approach is useful for transforming (regressing) EMG signals to joint angle/velocity/acceleration. The Riemannian approach is state of the art of classifying EEG signals1 and the goal of the assignment would be to come up with an Riemannian Geometry approach for transforming EMG signals that would be suited to use in non-weight-bearing situations.

For this assignment we have data from 10 able-bodied subjects, each measured on four separate days. Each subject performed non-weight-bearing activities, such as knee flexion and extension seated on a stool.

References:

  1. Vera Kooiman, Henk Van der Meet, Nico Verdonschot, Thor Fridriksson, Knut Lechler, Magnus Oddsson, Erla Gudrun Olafsdottir, Freygardur Thorsteinsson, Erik Prinsen, Eline Van Staveren, and Raffaella Carloni.Deliverable D2.1 Amputees' requirements. cordis.europa.eu/project/rcn/213177, 2018.
  2. Raffaella Carloni. MyLeg: Smart and intuitive osseointegrated transfemoral prostheses embodying advanced dynamic behaviors. MyLeg.eu, 2018.
  3. Carl D. Hoover, George D. Fulk, and Kevin B. Fite. Stair ascent with a powered transfemoral prosthesis under direct myoelectric control. IEEE/ASME Transactions on Mechatronics, 18(3):1191-1200, 2013.
  4. Oliver A. Kannape and Hugh M. Herr. Volitional control of ankle plantar exion in a powered transtibial prosthesis during stair-ambulation. 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014, pages 1662-1665, 2014.
  5. Stephanie Huang, Jeffrey P. Wensman, and Daniel P. Ferris. Locomotor Adaptation by Transtibial Amputees Walking with an Experimental Powered Prosthesis under Continuous Myoelectric Control. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 24(5):573-581, 2016.