Emotion Classification from Multi-Channel EEG Signals Using HiSTN: A Hierarchical Graph-based Spatial-Temporal Approach

TL;DR

This paper presents HiSTN, a lightweight hierarchical graph-based neural network for emotion classification from multi-channel EEG data, achieving high accuracy with fewer parameters and improved prediction balance.

Contribution

The study introduces a novel hierarchical graph-based spatial-temporal network (HiSTN) with a unique label smoothing method for efficient emotion classification from EEG signals.

Findings

Achieves over 96% F1 score in subject-dependent valence classification.

Surpasses 81% F1 score in subject-independent arousal classification.

Model with ~1,000 parameters outperforms existing methods in balanced prediction accuracy.

Abstract

This study introduces a parameter-efficient Hierarchical Spatial Temporal Network (HiSTN) specifically designed for the task of emotion classification using multi-channel electroencephalogram data. The network incorporates a graph hierarchy constructed from bottom-up at various abstraction levels, offering the dual advantages of enhanced task-relevant deep feature extraction and a lightweight design. The model's effectiveness is further amplified when used in conjunction with a proposed unique label smoothing method. Comprehensive benchmark experiments reveal that this combined approach yields high, balanced performance in terms of both quantitative and qualitative predictions. HiSTN, which has approximately 1,000 parameters, achieves mean F1 scores of 96.82% (valence) and 95.62% (arousal) in subject-dependent tests on the rarely-utilized 5-classification task problem from the DREAMER dataset. In the subject-independent settings, the same model yields mean F1 scores of 78.34% for valence and 81.59% for arousal. The adoption of the Sequential Top-2 Hit Rate (Seq2HR) metric highlights the significant enhancements in terms of the balance between model's quantitative and qualitative for predictions achieved through our approach when compared to training with regular one-hot labels. These improvements surpass 50% in subject-dependent tasks and 30% in subject-independent tasks. The study also includes relevant ablation studies and case explorations to further elucidate the workings of the proposed model and enhance its interpretability.