MedGuard AI

 

 

 

 

Title of the Project

MedGuard AI

Students Details

202211182 Salma Roshdy

202110170 Maysoon Adil

202110784 Raghad Tarkhan

202111554 Sabrin Ibrahim

Abstract
MedGuard is an AI-powered clinical decision support system for predicting drug–drug and drug–food interaction risks. Unlike traditional lookup tools that mainly detect
documented interactions, MedGuard uses molecular deep learning to estimate potential side effects from drug structure and patient medication context. The system combines
GNN-based molecular modeling, ChemBERTa-2 embeddings, Morgan fingerprints, and a HeteroGNN drug–food module inside one role-based platform for patients, doctors,
and pharmacists.