One-Class SVM for outliers detection in epileptic EEG
Matvey Khoymov 1, Vadim Grubov 1,2, Vladimir Maksimenko 1,2, Nikita Utashev 3, Denis Andrikov 4, Semen Kurkin 1,2
1 Baltic Center for Artificial Intelligence and Neurotechnology,
Immanuel Kant Baltic Federal University, Kaliningrad, Russia
2 Neuroscience and Cognitive Technology Laboratory, Innopolis University, Kazan, Russia
3 National Medical and Surgical Center named after N. I. Pirogov,
Ministry of Healthcare of the Russian Federation, Moscow, Russia
4 Research and Production Company “Immersmed”, Moscow, Russia
Abstract
The method of application of the machine learning method for anomaly detection – One Class SVM, for epileptic seizures detection in human EEG is considered. Collected EEG data of patients with confirmed focal epilepsy. The data were processed using continuous wavelet transform (CWT). The optimal combination of parameters was selected based on the precision and recall of classification. The obtained results are analyzed and the model with the best result is selected.
File with abstract
Speaker
Matvey Khoymov
Baltic Center for Artificial Intelligence and Neurotechnology
Russia
Discussion
Ask question
