To search, Click below search items.


All Published Papers Search Service


A Novel Epileptic Seizure Detection Using Fast Potential-based Hierarchical Agglomerative Clustering Based on EMD


Sabrina Belhadj, Abedlouaheb Attia, Bachir Ahmed Adnane, Zoubir Ahmed-Foitih, Abdelmalik Taleb Ahmed


Vol. 16  No. 5  pp. 7-12


Epilepsy is a chronic brain disorder that widely affects people. Mainly,it represents recurrent seizures that are brief episodes of involuntary movement. Detecting seizure is an important component in the diagnosis of epilepsy and for the seizures control. In the clinical practice, this detection basically involves visual scanning of Electroencephalogram (EEG).Many techniques have been developed for unscrambling the fundamental devices of the current seizures in EEGs. This paper presents a new framework using fast potential-based hierarchical agglomerative (PHA) Clustering Method and Empirical Mode Decomposition (EMD). The introduced algorithm first computes the Intrinsic Mode Functions (IMFs) of EEG signal, then calculates the Kolmogorov distance between each IMF and performs the detection based on PHA method. The evaluation results are very promising indicating an overall accuracy of 98.84%. A comparison between the proposed method and other existing methods in literature has been performed to show the advantage of proposed framework for detecting epileptic segments.


Electroencephalogram (EEG), EMD, Epilepsy, PHA clustering, Seizure detection.