Emergency data detection using Hidden Markov Model during temporary disconnection of Wireless Body Area Networks

Abstract

Wireless body area networks (WBANs) is a recently developing technology which will be playing a vital role in resolving some challenges faced in the healthcare sector. Energy-efficient solutions help to foster the acceptance of this technology by the patients. To solve the issues related to conservation of energy during temporary disconnection of sensor node from the sink, a solution based on hidden Markov Model (HMM) has been developed. Here a novel approach of predicting hypertension from heart rate data using Hidden Markov Models has been implemented. The model is using the concept that since the heart rate is a major correlate of blood pressure, it can predict the development of hypertension in patients with elevated blood pressure values. The simultaneous happening of tachycardia and hypertension may lead to cardiovascular problems. Here using Hidden Markov Model decoding the change of state happening over tachycardia is detected and emergency data loss is prevented considering the temporary disconnection for a small interval of time.