Seizure Activity Monitoring System

EasyChair Preprint 12488

Abstract

Epilepsy stands as one of the prevailing neurological disorders. This enduring ailment, marked by recurrent, unforeseeable, and unprovoked seizures, impacts a substantial global population. The transitory disruption in typical brain activity induced by this persistent condition can significantly impact the health of individuals affected by it. Detecting epileptic seizures before their onset proves invaluable. To streamline such diagnostic processes, contemporary research has put forth machine learning methodologies that amalgamate statistical principles with computer science.

Machine learning, a facet of artificial intelligence, empowers machines to autonomously acquire new knowledge. This technology, fueled by actionable data, enhances efficiency. Within the realm of healthcare, machine learning, along with computational techniques, is employed to forecast epileptic seizures based on electroencephalogram (EEG) recordings.

To study or predict a scenario, however, analyzing this data on its own is insufficient. This study’s objectives include providing full versions of machine learning prediction models for detecting epileptic seizures as well as identifying various types of predictive models and their applications in the field of healthcare.

Keyphrases: ECG, Electroencephalography(EEG), Epilepsy, Neurological disorders, Random Forest, Realtime Detection, Seizure, algorithm, feature selection, machine learning, seizure prediction

@booklet{EasyChair:12488,
  author    = {Jay Pathar and Meet Raychura and Vinit Tavde and Swapneel Trivedi and Ankit Chouhan},
  title     = {Seizure Activity Monitoring System},
  howpublished = {EasyChair Preprint 12488},
  year      = {EasyChair, 2024}}