A Wearable Wireless Sensor System Using Machine Learning Classification to Detect Arrhythmia

Health care is becoming a public concern and has given intensifying attention in recent years considering the aspects such as an increase in population, urbanization and globalization. (a). Good quality and effective health care system is although low in cost but its ability to detect abnormalities and anomalies is not compromised. The objective of this research work is to introduce a novel cost-effective technique that allows the measured ECG waveform to get classified with the help of the LabVIEW. Using the combination of the sensor system, first, the input ECG sensor signal is collected and then processed in LabVIEW to get classified. (b). A LabVIEW based simulation is presented in this article which classifies the heart ECG signal to be as healthy, non-healthy and not defined. Moreover, the relevant hardware details are also discussed. The classification system is trained using the machine learning (ML) technique (K-mean clustering). (c). The findings from the work include classification of heart health status, timely detection of anomalies and (various) arrhythmia conditions at their preliminary stages. Further discoveries contain performance evaluation resulting in response time lesser than half a minute and accuracy estimation from the experiment on three patients. (d). The system can be useful for detecting the COVID-19 breathing issues at their early stage and an automatic appointment can be set with the available scheduled heart professional based on the severity of the detected arrhythmia condition. The system allows early access to the hospital support system and can help to reduce the crowds in the medical centers.