A Comprehensive Review of Epileptic Seizure Prediction using Medical Internet of Things

A neurological disorder in the nervous system is a structural, biochemical or electrical abnormality in the brain or spinal cord. Epilepsy is the most common disorder among various neurological disorders, and recurrent epileptic seizures characterize it. A seizure is a period of symptoms due to excessive neuronal activity in an abnormal manner. Seizures are either due to provoked conditions like blood sugar, brain infections, and fever or unprovoked conditions like stress or sleeplessness. The reliable prediction of epileptic seizures involves pre-processing of signals from EEG sensors by enhancing the signal to noise ratio. By recording the brain's electrical activity using electrodes attached to the brain, epilepsy has been diagnosed. Nowadays, IoT based epilepsy detection is the most common real-time diagnostic procedure for detecting abnormalities in the brain using EEG sensor data. The involvement of sensors and IoT in medical healthcare research implementations can improve the diagnosis and detection of brain abnormalities which results in the enhancement of smart healthcare research. The high-end device utilization affects high energy consumption, leading to prolonged diagnostic decisions. The signal quality analysis process lags due to less concentrated research in classification and detection. The delay and error rate has to be reduced by concentrating more on energy management. An energy-efficient epileptic seizure prediction framework using EEG sensor data has been proposed to resolve and reduce such issues. The energy utilization of the epileptic seizure prediction using the Medical Internet of Things (MIoT) will be optimized. The trade-offs between accuracy, reliability, and delay of the effective energy-optimized epileptic seizure prediction using the MIoT framework will be enhanced.