An effectual IOT coupled EEG analysing model for continuous patient monitoring

• Published in: Measurement. Sensors Vol. 24; p. 100597
• Main Authors: Khiani, Simran, Mohamed Iqbal, M., Dhakne, Amol, Sai Thrinath, B.V., Gayathri, PG, Thiagarajan, R.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01-12-2022; Elsevier
• Subjects: Bi-LSTM; Cloud; EEG; IoT; Multidimensional data smart healthcare; Sensors; Bi-LSTM; Multidimensional data smart healthcare; Sensors; EEG; Cloud; IoT
• ISSN: 2665-9174; 2665-9174
• DOI: 10.1016/j.measen.2022.100597

A persistent neurological condition known as epilepsy is characterized by aberrant brain electrical activity. Epilepsy is a disorder characterized by sporadic symptoms and aberrant electrical brain activity that is spasmodic. Such aberrations are supported by non-stationary, non-linear, and multidimensional clinical data. Such data processing and analysis present a variety of computing research difficulties. One of the most often used signals obtained from the human brain in both science and therapeutic settings is the electroencephalogram (EEG). It is an effective resource for offering insightful knowledge of the mechanics of the brain. EEG signal analyses that are precise and thorough are crucial in the diagnosis of brain diseases like epilepsy. The EEG signal is well-liked in the field of biomedical study because of its notable temporal solution, non-invasive recording setup, and convenience with minimal costs. Currently, skilled neurologists make a diagnosis by visually examining EEG information to determine the most likely course of action. It takes time and is error-prone to visually inspect high-dimensional and non-stationary EEG recordings. An automated seizure detector is a crucial tool for better understanding how seizures are generated as well as a relief for medical experts who become exhausted when viewing a continuous, massive, long-term recording. As it records from at least 16 channels, the multichannel automatic seizure detection system is large. It is difficult to use such a method on distantly located subjects. The suggested system gathers data from IoT devices, and patient history-related electronic clinical data that are saved in the cloud were subjected to predictive analytics. The Bi-LSTM-based intelligent healthcare system for tracking and precisely forecasting brain risk. The trade-off between the current system and the current methods is anticipated in the results section.