All Published Papers Search Service

Title

Identification and Monitoring the Details of COMA Patient using Resilient Propagation Basis Function Neural Network Classifier Algorithm for Monitoring Brain Signals through Wireless Sensor Network

Author

M. Mohamed Zamam Nazar and Dr. M. Mohamed Surputheen

Vol. 22  No. 6  pp. 485-492

Abstract

Coma is an unconscious state wherein the patient is unable to respond. Therefore, the IoT sensor can be used to monitor the patient's coma brain solution where a wave of effective signal data is considered. The processing of brain waves based on the coma patient monitoring may be a very difficult analysis of the previous system's signal. The method proposed in the investigation, through the IoT sensors, is based on the positive signal changes found in the brain waves in the coma patients and is analyzed based on machine learning. In this article, a brainwave monitoring system developed in three steps using a Wireless Sensor Network (WSN), is continuously monitored for a patient's physical parameters in coma. The proposed Resilient Propagation Basis Function Neural Network (RPBFNN) classification algorithm identifies the coma patient’s state and alerts the patient. The system has followed three stages: Preprocessing, feature selection and classification. The first stage is Ensemble Filter-based preprocessing to remove the noise from the sensor data. Gaussian random frequency domain wavelet distribution (GRFDW) based feature selection is the second stage to collect the features (sub-band EEG signals, Standard deviation, variance Skewness). Then the final stage is the implementation of Resilient Propagation Basis Function Neural Network to identify the coma patient state level and alert the patient or system.

Keywords

Resilient Propagation Basis Function Neural Network (RPBFNN), Gaussian random frequency domain wavelet distribution (GRFDW), coma patient state level, Ensemble Filter, sub-band EEG signals.

URL

http://paper.ijcsns.org/07_book/202206/20220662.pdf