Cross-Paradigm Pretraining of Convolutional Networks Improves Intracranial EEG Decoding

Abstract

When it comes to the classification of brain signals in real-life applications, the training and the prediction data are often described by different distributions. Furthermore, diverse data sets, e.g., recorded from various subjects or tasks, can even exhibit distinct feature spaces. The fact that data that have to be classified are often only available in small amounts reinforces the need for techniques to generalize learned information, as performances of brain-computer interfaces (BCIs) are enhanced by increasing quantity of available data. In this paper, we apply transfer learning to a framework based on deep convolutional neural networks (deep ConvNets) to prove the transferability of learned patterns in error-related brain signals across different tasks. The experiments described in this paper demonstrate the usefulness of transfer learning, especially improving performances when only little data can be used to distinguish between erroneous and correct realization of a task. This effect could be delimited from a transfer of merely general brain signal characteristics, underlining the transfer of error-specific information. Furthermore, we could extract similar patterns in time-frequency analyses in identical channels, leading to selective high signal correlations between the two different paradigms. Classification on the intracranial data yields in median accuracies up to (81.50±9.49) %. Decoding on only 10% of the data without pre-training reaches performances of (54.76±3.56) %, compared to (64.95 ± 0.79) % with pre-training.

Improving Multichannel Raw Electroencephalography-based Diagnosis of Major Depressive Disorder via Transfer Learning with Single Channel Sleep Stage Data*

“…Nevertheless, relatively few pretraining approaches for raw EEG have been developed. Of those that have been developed, they often involve event-related potentials [11] or task data [12] that are very different from resting state data or the use of self-supervised learning [13]. One study of particular interest used a pretrained architecture originally developed in the imaging domain to classify raw EEG data [14].…”

A causal perspective on brainwave modeling for brain–computer interfaces

“…Unlike data augmentation though, using pre-training does not mean to create or obtain more examples from the joint distribution P(X, Y). This technique is based on the assumption that if a ML model is trained on a huge, diverse and close to the targeted brainwave task dataset, it will capture extra information from all of these various distributions, can benefit both causal and anti-causal models and achieve stronger BCI generalization as it is demonstrated in [63] for enhanced intracranial EEG decoding or in [64] for improved MI EEG classification (voluntarily engaged and endogenous task).…”

Interpretable functional specialization emerges in deep convolutional networks trained on brain signals

“…We selected a specific CNN design (Braindecode Deep4 network) [10], as the Deep4 network was already successfully applied to a wide range of EEG and iEEG classification problems [8,10,11,43,44] and also allowed first insights into the internal representations of learned EEG features [33]. In summary, the Deep4 network consists of an input layer, four hidden layers and an output layer (see supplementary figure S2).…”