Enhancing Health Risk Prediction in Internet of Medical Things: Leveraging Association Rule Mining

Abstract

Due to rapid advancements in the field of Internet of Medical Things (IoMT), a continuous influx of health data is being generated at a large scale. The primary objective of IoMT solutions is to transmit critical health data from patients to remote locations in real-time. Apart from remote patient monitoring, the extensive collection of health data offers opportunities for uncovering noteworthy patterns and potential risks associated with future diseases. This study introduces a novel risk prediction approach, namely Association Rule Mining for Risk Prediction (ARMR), which integrates an IoMT framework with the emerging machine learning technique known as Association Rule Mining (ARM). The proposed scheme employs a dataset obtained from various hospitals. The findings demonstrate that ARMR effectively extracts rules to identify a patient's risk of heart disease by considering demographic, physiological, and lifestyle data. Moreover, intriguing, and unexpected patterns and associations in the disease data can be identified, aiding medical professionals in guiding diagnosis and treatment decisions more efficiently. 

Keywords

• predictive analytics
• diabetes
• heart diseases
• da