A Meta-GNN approach to personalized seizure detection and classification

TL;DR

This paper introduces a Meta-GNN framework for personalized seizure detection that adapts quickly to new patients using limited data, leveraging graph neural networks and meta-learning to improve accuracy on epilepsy datasets.

Contribution

The paper presents a novel Meta-GNN approach that combines GNNs and meta-learning for rapid personalization in seizure detection, outperforming existing baselines.

Findings

Achieved 82.7% accuracy on TUSZ dataset after 20 iterations.

Outperformed baseline methods in seizure classification tasks.

Demonstrated rapid adaptation to new patients with limited samples.

Abstract

In this paper, we propose a personalized seizure detection and classification framework that quickly adapts to a specific patient from limited seizure samples. We achieve this by combining two novel paradigms that have recently seen much success in a wide variety of real-world applications: graph neural networks (GNN), and meta-learning. We train a Meta-GNN based classifier that learns a global model from a set of training patients such that this global model can eventually be adapted to a new unseen patient using very limited samples. We apply our approach on the TUSZ-dataset, one of the largest and publicly available benchmark datasets for epilepsy. We show that our method outperforms the baselines by reaching 82.7% on accuracy and 82.08% on F1 score after only 20 iterations on new unseen patients.