IIRPMS: An AI and IoT Enabled Remote Patient Monitoring and Data Analytics System for Real-Time Diagnosis and Treatment

The Internet of Things (IoT) is an amalgamation of interdisciplinary technologies that paved the way for novel and unprecedented possibilities and use cases. Although IoT technology will impact every human on this planet in the future, it will have an impact with no uncertain terms in various domains. In the healthcare domain, the IoT has its influence on making once impossible things into reality. Remote patient monitoring (RPM) has the potential to save the lives of people and improve quality of service (QoS) in healthcare units. It also saves time, effort and money in addition to preventing the death of people with its real-time monitoring of patients’ vital signs and providing treatment without delay. This is the motivation behind taking up this research. The existing endeavours in realizing an RPM system are step forward toward this goal. However, there is more to be done to achieve a reliable and secure RPM that serves the intended purpose. In this paper, we strive to take this research forward by implementing an efficient RPM system with a functional prototype. The proposed system is known as the IoT-driven Intelligent Remote Patient Monitoring System (IIRPMS). It sheds light on the mechanisms of the RPM and its utility to people in general. The proposed RPM is implemented using interoperable programming practices. It has artificial intelligence (AI)-enabled methods for the diagnosis of diseases. The proposed method for heart disease prediction is known as the Hybrid Bioinspired Model, which exploits the adaptive neuro-fuzzy inference system (ANFIS) algorithm whose parameters are optimized by a modified salp swarm optimization (MSSO) technique. The system has a provision for patient attendants and doctors to monitor patient vital signs and diagnosis results in real time. On observation of discrepancies in the patient’s vital signs, doctors can take steps to treat patients in near real time, which will minimize the mortality rate caused by cardiovascular and other such diseases that need real-time monitoring and treatment. The proposed HBM outperforms the state-of-the-art methods with 99.45% accuracy