LLM as Clinical Graph Structure Refiner: Enhancing Representation Learning in EEG Seizure Diagnosis

2026-04-30

AI Summary

This study addresses the challenge that noise in EEG signals often introduces redundant or irrelevant edges into graph structures, thereby degrading seizure detection performance. To mitigate this issue, the authors propose a two-stage graph refinement framework: an initial graph is first constructed and scored using a Transformer combined with a multilayer perceptron; subsequently, for the first time in this domain, a large language model (LLM) is leveraged to semantically refine the edge set by integrating textual descriptions with statistical features. Evaluated on the TUSZ dataset, the proposed method significantly improves seizure detection accuracy while yielding sparser, more interpretable, and semantically meaningful graph structures, thus pioneering the application of LLMs in clinical EEG-based graph construction.

Abstract

Electroencephalogram (EEG) signals are vital for automated seizure detection, but their inherent noise makes robust representation learning challenging. Existing graph construction methods, whether correlation-based or learning-based, often generate redundant or irrelevant edges due to the noisy nature of EEG data. This significantly impairs the quality of graph representation and limits downstream task performance. Motivated by the remarkable reasoning and contextual understanding capabilities of large language models (LLMs), we explore the idea of using LLMs as graph edge refiners. Specifically, we propose a two-stage framework: we first verify that LLM-based edge refinement can effectively identify and remove redundant connections, leading to significant improvements in seizure detection accuracy and more meaningful graph structures. Building on this insight, we further develop a robust solution where the initial graph is constructed using a Transformer-based edge predictor and multilayer perceptron, assigning probability scores to potential edges and applying a threshold to determine their existence. The LLM then acts as an edge set refiner, making informed decisions based on both textual and statistical features of node pairs to validate the remaining connections. Extensive experiments on TUSZ dataset demonstrate that our LLM-refined graph learning framework not only enhances task performance but also yields cleaner and more interpretable graph representations.

Problem

• EEG
• seizure detection
• graph construction
• noise
• redundant edges

Innovation

• Large Language Model
• Graph Refinement
• EEG Seizure Diagnosis
• Representation Learning
• Edge Pruning