Disease Prediction Using Artificial Intelligence: A Case Study on Epileptic Seizure Prediction

Abstract

Artificial Intelligence uses statistical theory to generate mathematical models from samples. After a model is generated, its depiction and algorithmic solution for understanding require being competent as well. Biomedical data related to different diseases are recorded from a body, which can be at the organ level, cell level or molecular level. Biomedical data is mainly utilized to predict, diagnose or identify particular physiological or pathological conditions. The goal of biomedical data analysis is exact modelling of data by employing feature extraction, feature selection and dimension reduction for the prediction and detection of upcoming pathological problems by utilizing artificial intelligence algorithms. This chapter explains the steps of biomedical data analysis and how artificial intelligence techniques are utilized in disease prediction. An automated epileptic seizure prediction and detection approach based on deep learning is also presented. Since Deep Learning can automatically extract and learn features, the electroencephalography (EEG) time series are fed into the deep learning model. Deep Learning has been utilized in the prediction and detection of epileptic seizures. Since EEG recordings are high dimensional data, a Convolutional Neural Network (CNN) is suitable for this use. The results show that CNN achieved a testing accuracy of 99.09% accuracy for the prediction of epileptic seizures from EEG signals.