Edge Analytics with TinyML Technique for MIoT Applications for Personalized Healthcare in Smart Nation

Analytics on the edge of an IoT device is utilized in a variety of settings when immediate decisions are required to be made. In order to carry out cloud-based analytics on data collected from the Internet of Things (IoT), the information must first be transmitted from IoT devices and stored in a centralized location for further processing which leads to high turnaround time in order to make necessary decision making. However, in applications that deal with matters of life and death, such as healthcare, this huge turnaround time will not be acceptable. As a result, it is absolutely necessary for analytics to be carried out at the edge in these MIoT applications in order to avoid hazards to human lives. As the resources of IoT devices are very limited, it is impossible to store the actual Machine Learning algorithms to do the processing. Hence, TinyML based applications are becoming increasingly popular in the edge analytics field as a way to alleviate power and resource restrictions. The purpose of this research paper is to shed light on the use of TinyML in MIoT applications for the purpose of conducting effective analyses of patient data in order to deliver personalized healthcare solutions.