Creating and sharing knowledge for telecommunications

Analysis of Electrocardiographic Patterns for Epileptic Seizure Prediction

Sargo, F. ; Fred, A. L. N. ; Silva, H. ; Bentes, C.

Analysis of Electrocardiographic Patterns for Epileptic Seizure Prediction, Proc Conf. on Telecommunications - ConfTele, Lisbon, Portugal, Vol. , pp. - , June, 2019.

Digital Object Identifier:

Download Full text PDF ( 397 KBs)

The gold standard for the diagnosis, automatic detection and prediction of epileptic seizures is based on data gathered by long term Electroencephalography (EEG). This modality requires highly intrusive hardware, with limiting aspects such as the great number of electrodes placed on the subject’s scalp. For this reason, it is typically performed in clinical settings.
There is, therefore, a need for an ambulatory and comfortable monitoring application to control seizures, consisting of a seizure prediction method such as single Lead Electrocardiography (ECG) acquisition, which may be performed with more comfortable hardware, and which may be easily integrated into a wearable system.
The purpose of this work is to propose and evaluate some of the required methodologies for the development of a seizure prediction algorithm based on ECG data in signals acquired by a Lead-I setup. It involves the application of noise and baseline wander removal techniques, detection of fiducial points, and robust computation of morphological and rhythmical features. The work culminates in the attempt to automatically distinguish between inter-ictal and pre-ictal moments, with the use of supervised learning classifiers, such as Support Vector Machines (SVM), K-Nearest Neighbours (KNN), and Gaussian Naive Bayes (GaussNB). Preliminary results of this study are presented, comprising N=44 seizures acquired from 5 patients admitted at Hospital de Santa Maria (HSM) in Lisbon.