2019/03/07 by Umar Asif, Asif, Umar, Subhrajit Roy +5
Medicine · Neuroscience · #Brain Tumor Detection and Classification #EEG and Brain-Computer Interfaces #Epilepsy research and treatment #FOS: Biological sciences #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Neurons and Cognition (q-bio.NC)
paper · pdf · doi:10.48550/arxiv.1903.03232
openalex publication_date 2019/03/07 · openalex created_date 2022/07/29 · openalex updated_date 2026/07/28
Automatic classification of epileptic seizure types in electroencephalograms\n(EEGs) data can enable more precise diagnosis and efficient management of the\ndisease. This task is challenging due to factors such as low signal-to-noise\nratios, signal artefacts, high variance in seizure semiology among epileptic\npatients, and limited availability of clinical data. To overcome these\nchallenges, in this paper, we present SeizureNet, a deep learning framework\nwhich learns multi-spectral feature embeddings using an ensemble architecture\nfor cross-patient seizure type classification. We used the recently released\nTUH EEG Seizure Corpus (V1.4.0 and V1.5.2) to evaluate the performance of\nSeizureNet. Experiments show that SeizureNet can reach a weighted F1 score of\nup to 0.94 for seizure-wise cross validation and 0.59 for patient-wise cross\nvalidation for scalp EEG based multi-class seizure type classification. We also\nshow that the high-level feature embeddings learnt by SeizureNet considerably\nimprove the accuracy of smaller networks through knowledge distillation for\napplications with low-memory constraints.\n