Big Data based Deep Learning for Predicting the Conditions of Patients from Medical IoT Data in Healthcare

Data from the Internet of Things (IoT) is being generated and captured at a rapid pace, much of which is relevant to healthcare systems. These pieces of information are in real time and are not structured. The storage and processing of real-time medical data in Internet of Things applications, on the other hand, continues to be a significant challenge. In this paper, a big data pipeline for processing the data from healthcare IoT devices. The study is about to monitor the mental and physical strength of patients admitted in intensive care units via IoT devices. The big data processing is conducted using Apache Flume tool that collects and transfers large data from IoT data to Hadoop Distributed File System from cloud server. The stability of the patients is conducted using a deep learning framework that finds the features and classifies the stability. The simulation is conducted on the data collected from the patients in real-time environment. The simulation shows accurate prediction of patient stability from large collected data. Thus, it is believed that the proposed method acts a better tool in providing viable solution for prediction of patients’ stability from IoT data.