NeuroMoTIC: An Smart Tool to Support Pediatric Epilepsy Diagnosis

Abstract

According to the World Health Organization, 50 million people have epilepsy with 80% of them living in low- and middle-income countries. Three quarters of these do not receive the treatment they need due to delays in interpreting electroencephalograms (EEGs). This paper presents a Machine learning model to support the diagnosis of pediatric epilepsy in semi-automatic way. The model was built from more than 100 pediatric EEGs, with a diagnosis of epileptic seizure. The results achieved using the software were compared with annotations made by a pediatric neurologist, reaching up to 85% agreement. In addition, the neurologists stated that, during the evaluation of a 30-min EEG, the system allowed them to save up to half of the time that usually takes. The tool herein presented facilitates the study and evaluation of pediatric EEGs using a semi-automatic classification of EEG signals and it can be used in the diagnosis of pediatric epilepsy.