Precision Monitoring of Health-Care Using Big Data and Java from Social Networking and Wearable Devices

Abstract

Wearable sensors and social networking sites help gather participant information for healthcare monitoring. Wearable sensors generate a lot of healthcare data for continuous patient monitoring. Social networking sites' user-generated healthcare data is huge and unstructured. Existing healthcare monitoring systems have trouble gathering and evaluating sensor and social network data, and traditional machine learning methods are insufficient to anticipate abnormalities in big healthcare data. A novel cloud-based healthcare monitoring architecture is presented to properly save and evaluate healthcare data and increase classification results. A Wireless Sensor Network and Big Data Analytics based Intelligent Health Monitoring System (IHMS) is suggested in this paper. The suggested large data analytics engine uses data mining, ontologies, and Bidirectional Long Short-Term Memory (Bi-LSTM). Data processing approaches preprocess information and minimize dimensionality. Proposed ontologies give semantic information about diabetes and blood pressure entities, aspects, and relationships (BP). Bi-LSTM categorizes information properly to forecast medication side effects and patient abnormalities. The suggested approach defines patients' health using diabetes, BP, mental health, and medicine reviews, and this model uses Java and Protégé Web Ontology Language. The findings demonstrate that the suggested system accurately manages healthcare information and predicts pharmacological side effects.