A Novel Machine Learning Technique for Diabetic Prediction in IoT-based Healthcare Monitoring System

• Published in: 2023 International Conference on Sustainable Computing and Smart Systems (ICSCSS) pp. 1000 - 1005
• Main Authors: Jesuraj, S Aravinth Vijay, Ganesh, R., Manjramkar, Vinda, Sridharan, M., Dubey, Vivek Ravishankar, Arun, M. R.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 14-06-2023
• Subjects: Diabetes; Diabetic Prediction; Healthcare Monitoring System; Internet of Things; Internet of Things (IoT); Machine learning; Medical services; Sensor systems; Sensors; Systematics; Tuna Swarm Optimization
• DOI: 10.1109/ICSCSS57650.2023.10169449

The successful development of a diagnosis system to identify diabetes in the Internet of Things (IoT) e-healthcare scenario has gained considerable attention to implement accurate diabetes diagnosis. IoT is playing an increasingly important role in healthcare environment by providing a structure for evaluating medical information to detect diseases via data mining techniques. The existing diagnostic methods have some challenges, such as lengthy calculation times and inaccurate predictions. To evade the limitations, this article suggested an IOT-based diagnosis system that uses Machine Learning (ML) techniques. Through the dataset from UCI Repository with medical sensors, a novel systematic technique is utilized to treat diabetic disease, and relevant medical data is produced to precisely predict those who would be seriously impacted by the condition. For predicting the illness and its severity, a brand-new classification technique called Tuna Swarm optimization-Aided Neural Classifier (TSO-NN) is suggested (T. The experimentation is conducted in MATLAB and the performance is evaluated using accuracy, precision, and F1-score. Besides, the efficiency is tested and confirmed by comparing over SOTA methods.