Abstract
The data that medical sensors collect can be overwhelming, making it challenging to glean the most relevant insights. An algorithm for a body sensor network is needed for the purpose of spotting outliers in the collected data. Methods of machine learning and statistical sampling can be used in the research process. Real-time response optimization is a growing field, as more and more computationally intensive tasks are offloaded to the backend. Optimizing data transfers is a topic of study. Computing power is dispersed across many domains. Computation will become a network bottleneck as more and more devices gain Internet-of-Things capabilities. It is crucial to employ both task-level parallelism and distributed computing. To avoid running down the battery, the typical solution is to send the processing to a server in the background. The widespread deployment of Internet-of-Things (IoT) devices has raised serious privacy and security concerns among people everywhere. The rapid expansion of cyber threats has rendered our current privacy and security measures inadequate. Machine learning (ML) methods are gaining popularity because of the reliability of the results that they produce, which can be used to anticipate and detect vulnerabilities in Internet-of-Things-based systems. Network response times are improved by edge computing, which also increases decentralization and security. Edge nodes, which frequently communicate with the cloud, can now handle a sizable portion of mission-critical computation. Real-time, highly efficient solutions are possible with the help of this technology. To this end, we use a distributed-edge-computing-based Internet-of-Things (IoT) framework to investigate how cloud and edge computing can be combined with ML. IoT devices with sensor frameworks can collect massive amounts of data for subsequent analysis. The front-end component can benefit from some forethought in determining what information is most crucial. To accomplish this, an IoT server in the background can offer advice and direction. The idea is to use machine learning in the backend servers to find data signatures of interest. We intend to use the following ideas in the medical field as a case study. Using a distributed-edge-computing-based Internet-of-Things (IoT) framework, we are investigating how to combine the strengths of both cloud and edge computing with those of machine learning.