AI-Driven Privacy-Preserving Healthcare System for Disease Prediction and Personalized Care

DOI

https://www.doi.org/10.59256/indjcst.20260501C013

Conference

International Conference on Cyber and AI Security

Organised By

Madanapalle Institute of Technology & Science (MITS)

Conference Date

9 & 10 March 2026

Pages

78-84

Article Type

Research Article

Check for updates

Machine learning has been a very beneficial aspect in the healthcare sector when it comes to the diagnostic services, health management and personal recommendations. However, the existing systems typically presuppose already designed input mechanism, such as checkbox symptoms, and vision of which prevents the user to engage with it and reduces the quality of foresight. The given paper introduces a novel system that incorporates DistilBioBERT to obtain linguistic insight into the symptoms, CatBoost to forecast and Integrated Gradients to clarify and a Risk-Aware Layer to recommend what to do. The system also has the Nutrient-Scored Diet Engine to obtain personalized diet prescription. The user can more securely and accurately input natural language by accepting and work on the state-of-the-art machine learning models, interpret the obtained system more accurately, more understandably and privately view sensitive healthcare information.