You are here: Home Research FAMOX: Adaptive Neuronal Signal …

FAMOX: Adaptive Neuronal Signal Processing Methods for the Facilitation of Attempted Motor Execution

Brain-computer interfaces for the motor rehabilitation after stroke-induced hemiplegia.

Overview

Project run time: 2013-2017

Funding

CogReha is kindly supported by

Description:

The execution of repetitive motor tasks is characterized typically by a significant performance variance. For two scenarios, this trial-by-trial performance is very pronounced:

  1. The decoding of motor commands (imagery or attempted execution) for brain-machine interfaces (BMI) / brain-computer interfaces (BCI).
  2. Motor tasks used in rehabilitation training e.g. after stroke.

In both scenarios, the execution quality of repeated tasks varies on several time scales, such that performance variation is observed on both, the course of a session and even from trial to trial. These performance fluctuations currently challenge a reliable and robust use of BCIs/BMIs in clinical or home use settings. They can not solely be explained by differing initial conditions of the experimental setup. In Famox, we investigate the hypothesis, that performance variations may partially be caused by the brain state of the user. Thus Famox investigates the novel methods to extract these brain states by multivariate data analysis methods with the goals:

  1. Contribute an understanding of the underlying brain processes by decoding features (neuronal markers) of a user's brain activity, which are informative in single trial about the current and the upcoming motor task performance.
  2. Evaluates novel closed-loop interaction principles, which may allow to manipulate the expected success of motor tasks.

To reach its project goals, adaptive data analysis methods are developed within Famox, which make heavy use of machine learning methods to decode the ongoing brain states in single trial.

Methods are designed to be applicable in closed-loop scenarios. Famox thrives to deliver a proof-of-principle for performance manipulation in the context of BCI/BMI systems and in novel motor rehabilitation paradigms for stroke-induced hemiplegia.

People

Contact person:

Collaborators:

Publications

Meinel A, Castaño-Candamil S, Reis J, Tangermann M, Pre-Trial EEG-based Single-Trial Motor Performance Prediction to Enhance Neuroergonomics for a Hand Force Task. Frontiers in Human Neuroscience, 10(170), 2016 [url]

Tangermann M, Reis J, Meinel A, Commonalities of Motor Performance Metrics are Revealed by Predictive Oscillatory EEG Components. Proceedings of the 3rd International Congress on Neurotechnology, Electronics and Informatics (NEUROTECHNIX 2015), p32-38, Lisbon, 2015

Castaño-Candamil S, Meinel A, Reis J, Tangermann M, P186. Correlates to influence user performance in a hand motor rehabilitation task. Clinical Neurophysiology, 126(8):e166 - e167, 2015 [url]

Meinel A, Castaño-Candamil JS, Dähne S, Reis J, Tangermann M, EEG Band Power Predicts Single-Trial Reaction Time in a Hand Motor Task. Proc. Int. IEEE Conf. on Neural Eng. (NER), IEEE, p182-185, 2015