Internet of Medical Things (IoMT) Fog-Based Smart Health Monitoring

Real-time monitoring is one of the Emerging duties in IoMT, and numerous systems have been developed to ensure the patient health monitor with efficient sensors capable of sensing, processing, and wireless communication may be used to collect data for environmental and smart health monitoring. These sensors are connected via wireless sensor networks. They transfer data to the cloud via IoT protocols and technology for storage and processing. This allows for the prediction of possible equipment breakdowns using past data. At times, the volume of data transmitted to the cloud or the time it takes to transfer data to the cloud and back to the sensors/actuators might be excessive. Moving some of the processing closer to the sensors can help minimize the amount of network and cloud resources consumed in these instances. In order to give cloud access, fog installations need to determine the architecture for joining sensors and gateways. Sensors often create data streams that may be pre-processed, aggregated, or filtered before they reach the cloud. Gateways can also be utilized to undertake data analytics. As a result, fog organization is critical for balancing computational load and network resource utilization on public clouds in order to reduce latency and save money. Service irregularity detection is a type of predictive maintenance that may be carried out even if no data from prior equipment failures is available. When available, machine-learning algorithms based on binary classification are used to predict breakdowns in the near future, allowing for repairs or replacements to be scheduled. The prediction models are trained and assessed using historical data, which includes information on prior equipment failures. Because historical data might be massive, real-time cloud storage is a possibility, leading in cloud-based predictive maintenance. This paper proposes a fog-based smart-health monitoring system with a single feed-forward neural network learning method to resolve this issue. An efficient cognitive-based machine learning method was compared to another efficient related work in this paper.