Exploiting Task Constraints for Self-Calibrated Brain-Machine Interface Control Using Error-Related Potentials.

Iñaki Iturrate,Jonathan Grizou,Jason Omedes,Manuel Lopes,Luis Montesano,Pierre-Yves Oudeyer,Mikhail A Lebedev

doi:10.1371/journal.pone.0131491

Abstract

This paper presents a new approach for self-calibration BCI for reaching tasks using error-related potentials. The proposed method exploits task constraints to simultaneously calibrate the decoder and control the device, by using a robust likelihood function and an ad-hoc planner to cope with the large uncertainty resulting from the unknown task and decoder. The method has been evaluated in closed-loop online experiments with 8 users using a previously proposed BCI protocol for reaching tasks over a grid. The results show that it is possible to have a usable BCI control from the beginning of the experiment without any prior calibration. Furthermore, comparisons with simulations and previous results obtained using standard calibration hint that both the quality of recorded signals and the performance of the system were comparable to those obtained with a standard calibration approach.

Highlights

The field of brain-computer interfaces (BCI) has witnessed largely successful applications in different contexts such as the control of assistive devices or communication restoration [1]
This paper presented an online study where eight subjects controlled a cursor using EEG errorrelated potentials with a self calibration of the decoder
All the subjects performed an online experiment, where the task and the classification parameters were learned in parallel, in an unsupervised way and transparent to the user

Summary

Introduction

The field of brain-computer interfaces (BCI) has witnessed largely successful applications in different contexts such as the control of assistive devices or communication restoration [1]. The brain signals used in BCI have a high variability caused by factors such as non-stationarities [2], task-dependent variations [3], and large subject specificity [4] To solve these problems, most applications rely on an initial calibration phase that needs to be frequently repeated due to the changing nature of the brain signals. Most applications rely on an initial calibration phase that needs to be frequently repeated due to the changing nature of the brain signals This calibration process is a tedious and time consuming part in most BCI applications that hinders their deployment. It impedes an out-of-the-box use of the BCI both in assistive or leisure scenarios. Calibration is an overload that affects the usability of the system in the long term and, the introduction and acceptance of such systems

Methods

Results

Conclusion