Generalized seizures are classified based on motor symptoms and non-motor symptoms that involve movement (Fisher et al., 2017, Schachter and Sirven, 2017, Scheffer et al., 2017). Generalized seizures affect both sides of the brain at the same time and again divided into absence, tonic, atonic, clonic, and tonic–clonic, myoclonic seizures (Schachter & Sirven, 2017). Focal seizures start on one side of the brain and depending on the patient level of awareness during a seizure, it is again classified as simple partial and complex partial seizures (Fisher et al., 2017, Scheffer et al., 2017). Accurate classification of epileptic seizure type plays a crucial role in the treatment and disease management of epilepsy patients (Roy, Asif, Tang, & Harrer, 2019). It can be concluded that the EEG based classification of seizure type using CNN model could be used in pre-surgical evaluation for treating patients with epilepsy.Įpileptic seizures are caused by a disturbance in the electrical activity of the brain, which is classified into focal, generalized, and unknown (Ko, 2017, Schachter and Sirven, 2017). Comparison results showed that CNN based approach outperformed conventional feature and clustering based approaches. The highest classification accuracy of 82.85% (using Googlenet) and 88.30% (using Inceptionv3) was achieved using transfer learning and extract image features approach respectively. The following ten pretrained networks were used to identify the optimal network for the proposed study: Alexnet, Vgg16, Vgg19, Squeezenet, Googlenet, Inceptionv3, Densenet201, Resnet18, Resnet50, and Resnet101. The following two different modalities were proposed using CNN: (1) Transfer learning using pretrained network, (2) Extract image features using pretrained network and classify using the support vector machine classifier. The 19 channels EEG time series was converted into a spectrogram stack before feeding as input to CNN. The objective of our study is to perform a multi-class classification of epileptic seizure type, which includes simple partial, complex partial, focal non-specific, generalized non-specific, absence, tonic, and tonic–clonic, and non-seizures. Therefore, this study attempts to classify seven variants of seizures with non-seizure EEG through the application of convolutional neural networks (CNN) and transfer learning by making use of the Temple University Hospital EEG corpus. Though automated early recognition of seizures from normal electroencephalogram (EEG) was existing, no attempts have been made towards the classification of variants of seizures. Recognition of epileptic seizure type is essential for the neurosurgeon to understand the cortical connectivity of the brain.
0 kommentar(er)
