VPSI 2.0: IoT-Based Hybrid Protocol with Simultaneous Equations for Real-Time Seizure Classification and False-Negative Mitigation

Non-epileptic seizures are a clinical symptom of abnormally high synchronous cortical activity known as psychogenic non-epileptic seizures (PNES) as they exhibit no outward signs of neurological damage. The need for differentiating PNES from full-body General Seizures (GTCS) decreases therapy time a...

Full description

Saved in:
Bibliographic Details
Published in:IEEE internet of things journal 2024-10, p.1-1
Main Authors: Gopalakrishnan, Karthikeyan, Balakrishnan, Arun Kumar, Govardhanan, Kousalya, Rajesh, Kandala NV P S
Format: Article
Language:English
Subjects:
Biomedical monitoring
Brain modeling
Electroencephalography
Epilepsy
GTCS
Internet of Things
IoMT
IoT
Mathematical models
Monitoring
PNES
Protocols
Seizures
Sensors
Vibrations
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Non-epileptic seizures are a clinical symptom of abnormally high synchronous cortical activity known as psychogenic non-epileptic seizures (PNES) as they exhibit no outward signs of neurological damage. The need for differentiating PNES from full-body General Seizures (GTCS) decreases therapy time and ensures proper hospice. Internet-of-Medical-Things (IoMT) provide a closed-loop mechanism to accurately measure seizures. The erratic nature of seizures has drawn the attention where false detection could have catastrophic impact. The paper discusses the VPSI 2.0 (Vibration Profile Seizure Identifier) where vibration profile analysis of an ictal patient is measured in realtime to classify seizures in an IoMT framework. The novel seizure detection model has been proposed for differentiating between multiple seizure types. The Simultaneous Equation (S.E) protocol is developed for noninvasive stigma-free monitoring of seizures for continual monitoring. S.E based IoT seizure classifier is helpful to mitigate challenges present in detecting real-time occurrences of seizure by 95.683% in a controlled environment.
ISSN:2327-4662
2327-4662
DOI:10.1109/JIOT.2024.3486991