College of Education for Pure Sciences, University of Thi-Qar discussed a master dissertation on a proposed smart model for detecting epilepsy in EEG recordings based on the time-frequency image (TFI) and the fractal dimension (FD) by the postgraduate student, Mrs. Zaman Ghani Nasir.

The dissertation reviewed designing of a proposed model for a reliable method for detecting epilepsy in EEG recordings based on the time-frequency image (TFI) and the fractional dimension (FD), as all the EEG segments are converted into an image using short-term instantaneous transform (STFT), as they are considered as parameters. STFT is necessary for the spectral program to have an acceptable resolution and then the TFI output is converted to a binary image, where the feature extraction depends on the fractal dimension.

The dissertation the calculation of the fractal dimension features by the box calculation algorithm. This work aims to increase the classification accuracy by using fewer features. Experiments were performed first by (KNN), then by (DT), and finally by LS-SVM classifier which achieved the highest accuracy. Most of the cases were obtained with 100% accuracy. This study uses EGG signal data from the University of Bonn to perform an experimental evaluation of the proposed technique.

The dissertation concluded that the proposed technique provides a high-accuracy classification and helps neurologists diagnose epilepsy and suggest appropriate treatment.