Work Package: WP7
Deliverable: 17.1: “EEG decoders to detect motor anticipatory potentials in motor therapies”
Deliverable due date: month 18
This document reports on the progress of work on Deliverable 17.1. All the planned tasks related to this Deliverable have been accomplished. Obtained results with healthy subjects were submitted to “Journal of Neural Engineering” peer-reviewed journal in December 2014 and published in April 2015 (http://iopscience.iop.org/article/10.1088/1741-2560/12/3/036007). The paper is also available in a non-formatted version under the following link: https://www.researchgate.net/publication/275525027_Continuous_detection_of_the_self-initiated_walking_pre-movement_state_from_EEG_correlates_without_session-to-session_recalibration.
Results for chronic stroke patients were submitted in April 2015 to “Journal of Neural Engineering and Rehabilitation” peer-reviewed journal and published in December 2015 under the Open Access policy, which makes it accessible with no restrictions under the following link: http://jneuroengrehab.biomedcentral.com/articles/10.1186/s12984-015-0087-4.
These publications are a scientific part of the Deliverable 17.2 (and the JNE also contains first results for Deliverable 17.2 ) and this “Project Deliverable Report” document focuses on its relation to the NETT proposal.
In the NETT 289146 Grant Annex, we stated in the context of Deliverable 17.1 and 17.2: “P17 is related to robot-mediated motor rehabilitation therapies for patients after stroke, where one of the key points is to proceed with the rehabilitation task only when the patient is engaged and thus using their (damaged) motor cortex. In this context the objective is the development of signal processing and machine learning techniques able to detect in real-time neural processes preceding movement and motor imagery, i.e. intention of movement.”
In the aforementioned publications we described two studies carried out to: 1) study the pre-movement neural correlates of gait intention; and 2) develop decoders that can detect the intention to move from this correlates. Detectors are based on regularized linear discriminant analysis to cope with the large number of features compared to the limited calibration examples. The results show that the combination of ERD and MRCP features can robustly detect intention of motion before it is actually measurable in the muscle activity. The decoders were evaluated for the first time on chronic stroke patients achieving a similar performance around 70% as for healthy subjects. A preliminary online integration with a real robot also showed the applicability in real rehabilitation scenarios.
In such a way we accomplish Deliverable 17.1 and “detect in real-time neural processes preceding movement and motor imagery, i.e. intention of movement “. The remaining part of P17, that deals with removing the need of calibration during neuro-rehabilitation interventions will be reported in Deliverable 10.2.
Contributors: Andreea Sburlea, Luis Montesano, Javier Minguez