Robust Epileptic Seizure Classification
A lot of feature vectors and sub-band signals are considered for Epileptic seizure classification. Unfortunately, not all the feature vectors and sub-band signals contribute to the final result. In view of this limitation, we propose a modified Differenti
- PDF / 416,961 Bytes
- 11 Pages / 439.37 x 666.142 pts Page_size
- 14 Downloads / 233 Views
Abstract. A lot of feature vectors and sub-band signals are considered for Epileptic seizure classification. Unfortunately, not all the feature vectors and sub-band signals contribute to the final result. In view of this limitation, we propose a modified Differential Evolution Feature Selection algorithm (MDEFS), which searches the best feature vector subset and the sub-band signals to distinguish three groups of subjects (healthy, ictal and interictal). From the experiment results, it is observed that the bagging method based on the optimal feature subset (the standard deviation attribute in the delta sub-band signal, the time-lag attribute in the delta sub-band signal, fractal dimension in the alpha sub-band signal, the correlation dimension attribute in the alpha sub-band signal and the standard deviation attribute in the beta sub-band signal) selected by MDEFS results in highest classification accuracy of 98.67 %. Keywords: Epileptic seizure classification Feature selection evolution Bagging Discrete Wavelet Transform
Differential
1 Introduction Epilepsy is one of the major diseases in the world, especially for children in the developed countries. About 1 % of people worldwide (65 million) have epilepsy, and nearly 80 % of the cases occur in developing countries [1, 2]. Epilepsy is a group of neurological disorders characterized by epileptic seizures. Epileptic seizures are the result of excessive and abnormal cortical nerve cell activity in the brain. Approximately, one out of every three individuals with epilepsy continues to experience frequent epileptic seizure. This seizure poses a serious risk of injury, and also limits the mobility and independence of an individual [3]. Early detection of this disease is crucial for timely treatment to minimize further deterioration. Electroencephalogram (EEG) is usually used for epileptic seizure detection, and epilepsy diagnosis is performed through identification of EEG abnormalities [4]. Since EEG readings are checked by expert neurologist, there are some concerns about accuracy caused by visual fatigue, time and cost for regular examinations. Computer-aided diagnosis © Springer International Publishing Switzerland 2016 D.-S. Huang and K.-H. Jo (Eds.): ICIC 2016, Part II, LNCS 9772, pp. 363–373, 2016. DOI: 10.1007/978-3-319-42294-7_32
364
F. Alzami et al.
approaches by EEG time series data analysis offer an efficient solution to epilepsy diagnosis. EEG datasets for epileptic seizure usually contain three groups of subjects, such as (1) healthy subject, (2) epileptic subject during seizure (ictal), and (3) epileptic subject during seizure free interval (interictal). Due to the non-stationary nature of EEG signals, wavelet analysis is a suitable analysis approach since wavelet transform gives precise frequency information at low frequencies and precise time information at high frequencies [5]. Many research using wavelet transform, especially DWT, to preprocess the EEG signals and decompose them into sub-bands signals. Features which include energy [6, 7], skewne
Data Loading...
