Open AccessA Review on Wearable Epileptic Seizure Prediction System
Author(s) -
V. Seethalakshmi,
P Naveenkumar,
G Kavin Prabu,
Saumya Kumaar
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1916/1/012075
Subject(s) - epileptic seizure , heartbeat , computer science , wearable computer , artificial intelligence , internet of things , machine learning , gsm , feature extraction , feature (linguistics) , epilepsy , computer security , medicine , psychiatry , embedded system , telecommunications , linguistics , philosophy
Epileptic is a neurological condition that affects approximately 50 million people worldwide. Epileptic seizure prediction lowers the risk of a patient’s life being endangered by a seizure that occurs unexpectedly. The latest seizure prediction methods are computationally intensive due to the complicated hand-crafted features they extract, and they take a lot of memory to store their parameters, which makes them Inappropriate for IoT and connected systems with limited capabilities. In this paper, a deep learning-based IoT framework for accurate epileptic seizure prediction is presented. The proposed method combines the feature extraction and classification stages into a single integrated system in which raw data heartbeat and temperature signals are implemented without any pre-processing, reducing computing complexity even further. A machine learning based prediction model is proposed that extracts the relevant information from the temperature, heartbeat and haemoglobin value using of machine learning algorithm The health condition of patient or person can be found and give some analysis result like normal or abnormal condition. If abnormal condition is observed then the system predicts some medicine or dosage based on health condition and also send alert message using of GSM. In this work, a location tracking of patient is also included and alert is sent to authorized person when the patient fall down or patient get panic or abnormal health