Provisioning a Predictor Model for Alzheimers Prediction using Learning Approaches

One of the most common neurological disorders in people, epilepsy increases the risk of stroke, brain tumours, and severe head injuries. Early seizure detection can assist treat epilepsy patients and lessen the degree of injuries. An accurate identification of pre-ictal brain state before seizure episode is the goal of a seizure prediction system. Patient-independent seizure prediction models have been found as helpful solutions to the seizure prediction problem since they are created to give dependable performance over a dataset’s many patients. We concur that this typically occurs in Intensive Care Unit (ICU) monitoring circumstances when acquiring sufficient samples to train a prediction model is challenging. However, more effort must be put into designing models that can account for this. We suggested a Higher-Order Boosting Predictor (HOBP) to solve this issue. With improvements in MCC values of 96%, the proposed models outperform the existing approach and can correctly detect seizures. The prediction window of the proposed model performs well for seizure prediction with 98% accuracy, respectively.