DEVELOPMENT OF AN AFFECTIVE E-HEALTHCARE SYSTEM FOR EPILEPTIC PATIENTS

Abstract

The stigmatisation attached to epilepsy brings about the isolation of epileptic patients in our society, especially in rural areas coupled with the challenges of not having regular access to quality healthcare. The concept of managing seizures in epileptic patients through the use of affective computing came as result of the stigma associated with being an epilepsy patient and also to predict epilepsy seizures as epileptic patients suffer from the unawareness of when the next seizure sets in, and what the consequences will be. The goal of this research is to adopt technology that can determine when seizure is about to occur in an epileptic patient. The research study is accomplished by designing a seizure detection model which comprises of Galvanic Skin Response (GSR) Sensor, Temperature Sensor and Pulse Rate Sensor to gather physiological data from epileptic patients, and then analyse using fuzzy logic and classify them into various types of seizures and also detect and predict the occurrence of seizure. This affective system is in form of a non-invasive wearable device such as sensory headwear and wrist-wear to allow the epileptic patients carry on with their own life, performing their everyday activities. Therefore, the system developed shows that the integration of affective e- Healthcare system helps to monitor epileptic patient’s health and also help detect and predict when the next seizure will occur so that patients can take medication to avoid a total loss of consciousness during seizure. Seizure classification performance was evaluated through crossvalidation by randomly segmenting the data into train and test sets for total of 7500 seconds times, giving the system an overall accuracy of 87.09%.