End-to-End Deep Transfer Learning for Calibration-free Motor Imagery Brain Computer Interfaces

TL;DR

This paper explores end-to-end deep transfer learning methods for calibration-free, subject-independent motor imagery brain-computer interfaces, comparing three models on raw EEG data to improve accessibility and reduce calibration time.

Contribution

It introduces an end-to-end deep learning approach for MI-BCI classification that bypasses traditional preprocessing and feature engineering, evaluating three models on a large dataset.

Findings

EEGNet and DeepConvNet outperform MIN2Net in subject-independent classification.

Median accuracies of EEGNet and DeepConvNet are 62.5% and 59.2%, respectively.

Models do not reach the 70% accuracy threshold for reliable control.

Abstract

A major issue in Motor Imagery Brain-Computer Interfaces (MI-BCIs) is their poor classification accuracy and the large amount of data that is required for subject-specific calibration. This makes BCIs less accessible to general users in out-of-the-lab applications. This study employed deep transfer learning for development of calibration-free subject-independent MI-BCI classifiers. Unlike earlier works that applied signal preprocessing and feature engineering steps in transfer learning, this study adopted an end-to-end deep learning approach on raw EEG signals. Three deep learning models (MIN2Net, EEGNet and DeepConvNet) were trained and compared using an openly available dataset. The dataset contained EEG signals from 55 subjects who conducted a left- vs. right-hand motor imagery task. To evaluate the performance of each model, a leave-one-subject-out cross validation was used. The results of the models differed significantly. MIN2Net was not able to differentiate right- vs. left-hand motor imagery of new users, with a median accuracy of 51.7%. The other two models performed better, with median accuracies of 62.5% for EEGNet and 59.2% for DeepConvNet. These accuracies do not reach the required threshold of 70% needed for significant control, however, they are similar to the accuracies of these models when tested on other datasets without transfer learning.