Automated Video-EEG Analysis in Epilepsy Studies: Advances and Challenges

TL;DR

The paper surveys automated video-EEG analysis for epilepsy, highlighting real-time EEG-based detection, video-only seizure sensing, and the nascent integration of modalities. It emphasizes dataset limitations, especially the scarcity of open video-EEG benchmarks, and discusses 다양한 DL and transformer approaches, including foundations for video, EEG, and multimodal fusion. A novel concept-based learning pipeline for treatment-effect estimation from vEEG data is proposed to advance interpretable, personalized epilepsy care. Overall, the work identifies critical challenges—symptom diversity, labeling inconsistencies, and data availability—and charts a path toward robust multimodal models and synthetic data to improve diagnostic accuracy and treatment guidance.

Abstract

Epilepsy is typically diagnosed through electroencephalography (EEG) and long-term video-EEG (vEEG) monitoring. The manual analysis of vEEG recordings is time-consuming, necessitating automated tools for seizure detection. Recent advancements in machine learning have shown promise in real-time seizure detection and prediction using EEG and video data. However, diversity of seizure symptoms, markup ambiguities, and limited availability of multimodal datasets hinder progress. This paper reviews the latest developments in automated video-EEG analysis and discusses the integration of multimodal data. We also propose a novel pipeline for treatment effect estimation from vEEG data using concept-based learning, offering a pathway for future research in this domain.

Figures (1)

• Figure 1: Proposed pipeline for treatment effect estimation