Authors:
Leonor Almeida
1
;
Sem Hoogteijling
2
;
3
;
Inês Silveira
1
;
Dania Furk
1
;
Irene Heijink
2
;
3
;
Maryse Van’T. Klooster
2
;
Hugo Gamboa
1
;
Luís Silva
1
and
Maeike Zijlmans
2
;
3
Affiliations:
1
Laboratório de Instrumentação, Engenharia Biomédica e Física da Radiação (LIBPhys-UNL), Departamento de Física, Faculdade de Ciências e Tecnologia, FCT, Universidade Nova de Lisboa, 2829-516 Caparica, Portugal
;
2
Department of Neurology and Neurosurgery, Brain Center, University Medical Center Utrecht, Part of ERN EpiCARE, P.O. box 85500, 3508 GA Utrecht, The Netherlands
;
3
Stichting Epilepsie Instellingen Nederland (SEIN), The Netherlands
Keyword(s):
Synthetic Data, Epilepsy, Epileptiform Activity, Epileptogenic Tissue, ioECoG, GAN.
Abstract:
Epilepsy surgery is a viable option for treating drug-resistant cases where anti-seizure medications fail, but accurately localizing epileptic tissue remains challenging. This process can be guided by the visual assessment of intraoperative electrocorticography (ioECoG). Data scarcity limits developing machine learning (ML) models for automatic epileptic tissue classification. To address this, we propose a generative model based on Generative Adversarial Networks (GANs) to synthesize realistic ioECoG signals. Our approach identified three distinct ioECoG patterns using Agglomerative Clustering, which guided training individual Deep Convolutional Wasserstein GANs with Gradient Penalty (DCwGAN-GP). Synthetic data (SD) was evaluated across multiple dimensions: fidelity using temporal (e.g., Wasserstein distance (WD)), frequency and time-frequency metrics; diversity through dimensionality reduction; and utility by comparing ML performance with and without SD. It replicated temporal and f
requency characteristics of real signals (fidelity), though lacked variability (diversity) due to potential data misclassifications. Specifically, the WD between real and synthetic signals outperformed literature benchmarks (i.e., 0.043 ± 0.025 vs. 0.078). Classifiers trained on a combination of real and SD achieved 88% accuracy, compared to 85% with real data alone. These results demonstrate the potential of SD to replicate real signals, address data scarcity, augment ioECoG datasets, and advance ML-based epilepsy surgery research.
(More)