A Fog-Based Intelligent Secured IoMT Framework for Early Diabetes Prediction

Early prediction of diabetes is often needed for a clinically effective outcome due to the existence of a relatively long asymptomatic period. Because of its long-term asymptomatic period, about 50% of the people suffering from diabetes are unidentified. It is only possible to make an early diagnosis of diabetes by thoroughly examining both common and less common signs, which may be identified timely at different stages from initiation of the disease to diagnosis. To detect early diabetes and to avoid serious effects, it is possible to track in real time due to the advancement of information technologies. The chapter suggests the Internet of Medical Things (IoMT) with a fog-assisted healthcare system for early diabetes prediction (or emergency) and notification for remote patients. The system continuously monitors the information for data analysis to track physiological signals and contextual information through notified alert messages to service providers and end users in real time. Boosting techniques for the risk prediction model of the disease have been well considered by many researchers. In this work, gradient boosting (GB) algorithm is proposed to predict the early symptoms of diabetes with higher classification accuracy. The experimental results demonstrate the enhanced performance of the proposed system compared to machine learning (ML) in terms of precision, recall, and F1-score.