Adaptive Neuro-Fuzzy Inference System As New Real-Time Approach For Parkinson Seizures Prediction

Abstract

Parkinson's disease, as a definition, is a neurological condition that affects the brain and causes tremors, stiffness, and difficulties walking, balancing, and coordinating. Symptoms of Parkinson's disease normally appear gradually and worsen with time. People with Parkinson's disease may have difficulties walking and speaking as the condition develops. Numerous recent studies have shown a direct association between Parkinson's disease and the incident of having epileptic seizures, which is defined to be a burst of the uncontrollable electrical activity of the brain cells, that is associated with an increased risk of sudden unexplained deaths. This project aims to obtain a real-time seizure prediction system for Parkinson's disease patients based on the electroencephalogram (EEG) signals, enabling the detection of a seizure before it happens. This will hopefully save them from risky situations or sudden death, as they will be alerted and have the time enabling them to be prepared and take the needed precautions and steps to prevent the seizure from happening. For this project, we've used the Neural Network and the ANFIS (“udaptive neuro-fuzzy inference system ‘’) to process and analyze the electroencephalogram (EEG) data signals of the Parkinson patients to detect seizures starting point.