Developement of Matlab-based Graphical User Interface (GUI) for detection of high frequency oscillations (HFOs) in epileptic patients

Sahbi Chaibi, Romain Bouet, Julien Jung, Tarek Lajnef, Mounir Samet, Olivier Bertrand, Abdennaceur Kachouri, Karim Jerbi

Abstract: High-Frequency Oscillations (HFOs) in the 80-500 Hz band are important biomarkers of epileptogenic brain areas and could have a central role in the process of epileptogenesis and seizure genesis. Visual marking of HFOs is highly time consuming and tedious especially for long electroencephalographic (EEG) recordings. Automated HFO detection methods are potentially more efficient, repeatable and objective. Therefore, numerous automatic HFOs detection methodshave been developed. To evaluate and compare the performance of these algorithms in an intuitive and user-friendly framework accessible to researchers, neurologists and students, it is useful to implement the various methodsusing a dedicated Graphical User Interfaces (GUI). In this paper we describe a GUI-based tool that contains three HFOs detection methods. It allows the user to test and runthree different methods based respectively on FIR filter, Complex MORLET Wavelet and matching pursuit (MP). We also show how the GUI can be used to measure the performance of each method. Generally, high sensitivity entrains high false-positive detection rates. For that, the developed GUI contains a supplementary module that allows an expert(e.g. neurologist) to reject false detected events and only save the clinically relevant (true) events. In addition, the GUI presented here can be used to perform classification, as well as estimation of duration, frequency and position of different events. The presented software is easy to use and can easily be extended to include further methods. We thus expect it to become a valuable clinical tool for diagnosis of epilepsy and research purposes.