Abstract:

This project aims to develop a machine learning-based system for predicting diabetes diagnoses using real-time physiological data collected from wearable sensors. We utilize pre-existing medical datasets to train machine learning models, focusing on key health indicators relevant to diabetes, such as blood glucose levels, heart rate, and physical activity. The core innovation lies in integrating these predictive models with data gathered from an external wearable device equipped with various sensors. The wearable device continuously collects and transfers health data to our system, where it is processed and analyzed using our pre-trained models. The system provides real-time feedback, assisting in early detection and monitoring of diabetes risk. This approach emphasizes the seamless integration of sensor technology with predictive algorithms, aimed at enhancing preventive healthcare through non-invasive, continuous monitoring.