The effects of layer-wise relevance propagation-based feature selection for EEG classification: a comparative study on multiple datasets

Abstract

IntroductionThe brain-computer interface (BCI) allows individuals to control external devices using their neural signals. One popular BCI paradigm is motor imagery (MI), which involves imagining movements to induce neural signals that can be decoded to control devices according to the user's intention. Electroencephalography (EEG) is frequently used for acquiring neural signals from the brain in the fields of MI-BCI due to its non-invasiveness and high temporal resolution. However, EEG signals can be affected by noise and artifacts, and patterns of EEG signals vary across different subjects. Therefore, selecting the most informative features is one of the essential processes to enhance classification performance in MI-BCI.MethodsIn this study, we design a layer-wise relevance propagation (LRP)-based feature selection method which can be easily integrated into deep learning (DL)-based models. We assess its effectiveness for reliable class-discriminative EEG feature selection on two different publicly available EEG datasets with various DL-based backbone models in the subject-dependent scenario.Results and discussionThe results show that LRP-based feature selection enhances the performance for MI classification on both datasets for all DL-based backbone models. Based on our analysis, we believe that it can broad its capability to different research domains.

Advancing Skin Cancer Diagnosis through the Comparison of SHAP and Layer-wise Relevance Propagation (LRP)

“…The right side breaks down the model output into individual contributions from each feature, while the left side represents the total model output. LRP [30], or Layer-wise Relevance Propagation, is a technique used in the field of explainable artificial intelligence (XAI) to interpret the decisions made by complex machine learning models, particularly neural networks. LRP assigns relevance scores to each input feature of a model, indicating the contribution of that feature to the model's output.…”

TSANN-TG: Temporal–Spatial Attention Neural Networks with Task-Specific Graph for EEG Emotion Recognition

“…With the aid of X l 1 and X l 2 , we undertake the computation of the adjacency matrix, which serves to represent the connectivity between various EEG channels or nodes within a graph-based brain network. Here, we have opted for four commonly employed functional connectivity methods, namely correlation (Cor), coherence (Coh), phase locking value (PLV), and phase lag index (PLI), all of which have shown promising results in EEG-based emotion recognition [48][49][50][51].…”