IOT Based Biomedical Wireless Sensor Networks and Machine Learning Algorithms for Detection of Diseased Conditions

With rapidly increasing population and industrialization, the facilities available to rural areas are diminishing every day. Immediate, feasible and accurate healthcare monitoring is a must, but it is often difficult to find doctors in such areas. One solution is an IoT based wireless sensor network for patient monitoring. It is a smart system, which uses real time data collected and stored in a secure database, for continuous monitoring, analyzing the physiological signals using signal processing techniques. The data collected can be correlated to predict the disorder underlying the abnormality present using LSTM recurrent neural network. The physiological signals monitored are electrocardiogram (ECG), body temperature, and blood pressure. Among the acquired physiological signals, the raw ECG data is filtered using hamming window FIR filter. A QRS detection wavelet transform algorithm is also used to make the data more reliable. This processed data along with the other two parameters, the body temperature and the blood pressure are first stored in the database, and then transmitted wirelessly to a doctor along with the location details of the patient. The paper describes a prototype patient monitoring system, through a common interface, an android application. This application will bridge the communication gap, and increase the accuracy of our smart monitoring system.