Shared Attention-based Autoencoder with Hierarchical Fusion-based Graph Convolution Network for sEEG SOZ Identification

TL;DR

This work tackles the challenge of identifying seizure onset zones (SOZ) from sEEG by addressing limited intra-patient representations and ignoring interdependencies across feature elements. It proposes a cross-patient, shared attention-based autoencoder (sATAE) to learn general epileptic patterns, coupled with a hierarchical fusion-based graph convolution network (HFGCN) that fuses static and dynamic network features via hierarchical weighting. Extensive experiments on 17 temporal lobe epilepsy patients show that sATAE-HFGCN achieves state-of-the-art performance (e.g., ACC up to 80.46%), surpassing various baselines and demonstrating robust SOZ identification across behavioral states and brain regions. The approach offers a principled framework for leveraging cross-patient information and dynamic-brain connectivity in sEEG-based SOZ localization, with clear ablations supporting its core design choices.

Abstract

Diagnosing seizure onset zone (SOZ) is a challenge in neurosurgery, where stereoelectroencephalography (sEEG) serves as a critical technique. In sEEG SOZ identification, the existing studies focus solely on the intra-patient representation of epileptic information, overlooking the general features of epilepsy across patients and feature interdependencies between feature elements in each contact site. In order to address the aforementioned challenges, we propose the shared attention-based autoencoder (sATAE). sATAE is trained by sEEG data across all patients, with attention blocks introduced to enhance the representation of interdependencies between feature elements. Considering the spatial diversity of sEEG across patients, we introduce graph-based method for identification SOZ of each patient. However, the current graph-based methods for sEEG SOZ identification rely exclusively on static graphs to model epileptic networks. Inspired by the finding of neuroscience that epileptic network is intricately characterized by the interplay of sophisticated equilibrium between fluctuating and stable states, we design the hierarchical fusion-based graph convolution network (HFGCN) to identify the SOZ. HFGCN integrates the dynamic and static characteristics of epileptic networks through hierarchical weighting across different hierarchies, facilitating a more comprehensive learning of epileptic features and enriching node information for sEEG SOZ identification. Combining sATAE and HFGCN, we perform comprehensive experiments with sATAE-HFGCN on the self-build sEEG dataset, which includes sEEG data from 17 patients with temporal lobe epilepsy. The results show that our method, sATAE-HFGCN, achieves superior performance for identifying the SOZ of each patient, effectively addressing the aforementioned challenges, providing an efficient solution for sEEG-based SOZ identification.

Figures (5)

• Figure 1: (A) sEEG processing and extraction of latent features through shared attention-based autoencdoer. (B) Concrete details of shared attention-based autoencoder(sATAE).
• Figure 2: Construction of graph structure data and the overall framework of hierarchical fusion-based graph convolution network (HFGCN).
• Figure 3: Two Fusion Strategies in HFGCN for sEEG SOZ Identification.
• Figure 4: The method performance with different Chebyshev filter size $F$ for sEEG SOZ idenification.
• Figure 5: The method performance with different $K$ value in DGCN for sEEG SOZ idenification.