Knowledge-Data Fusion Based Source-Free Semi-Supervised Domain Adaptation for Seizure Subtype Classification

TL;DR

This paper introduces KDF-MutualSHOT, a novel source-free semi-supervised domain adaptation method that fuses raw EEG data and expert knowledge to improve seizure subtype classification accuracy across different datasets.

Contribution

It proposes a new knowledge-data fusion approach and a domain adaptation algorithm specifically designed for privacy-preserving EEG seizure classification tasks.

Findings

Outperforms existing methods on TUSZ and CHSZ datasets

Effective fusion of raw data and expert knowledge improves accuracy

MutualSHOT achieves better domain alignment and pseudo-labeling

Abstract

Electroencephalogram (EEG)-based seizure subtype classification enhances clinical diagnosis efficiency. Source-free semi-supervised domain adaptation (SF-SSDA), which transfers a pre-trained model to a new dataset with no source data and limited labeled target data, can be used for privacy-preserving seizure subtype classification. This paper considers two challenges in SF-SSDA for EEG-based seizure subtype classification: 1) How to effectively fuse both raw EEG data and expert knowledge in classifier design? 2) How to align the source and target domain distributions for SF-SSDA? We propose a Knowledge-Data Fusion based SF-SSDA approach, KDF-MutualSHOT, for EEG-based seizure subtype classification. In source model training, KDF uses Jensen-Shannon Divergence to facilitate mutual learning between a feature-driven Decision Tree-based model and a data-driven Transformer-based model. To adapt KDF to a new target dataset, an SF-SSDA algorithm, MutualSHOT, is developed, which features a consistency-based pseudo-label selection strategy. Experiments on the public TUSZ and CHSZ datasets demonstrated that KDF-MutualSHOT outperformed other supervised and source-free domain adaptation approaches in cross-subject seizure subtype classification.