Video Demo
https://drive.google.com/file/d/1s-f8HBks5tT72nTEvYlYm4FdU1ryDCNA/view?usp=sharing
We were inspired by the urgent need to find faster treatments for rare diseases, where traditional drug discovery is too slow and costly. Our goal was to use AI to intelligently repurpose existing FDA-approved drugs by analyzing real scientific research. We collected detailed disease information from PubMed, gathering about 20 scientific papers per disease, and built a structured drug database with mechanisms of action and repurposing potential. Using a combination of Flask, React.js, Pandas, and Material-UI, we developed a full-stack platform that applies Natural Language Processing (NLP) to understand diseases, runs a multi-factor scoring algorithm to match drugs, and delivers explainable, evidence-based recommendations. Throughout the project, we learned how to handle complex biomedical data, balance AI scoring systems, optimize backend performance, and design a transparent user experience. Some challenges we faced included parsing dense medical language, ensuring accurate disease classification, and maintaining fast, real-time response times — but ultimately, we built a solution that transforms months of research into minutes, offering faster hope for millions of rare disease patients.
Log in or sign up for Devpost to join the conversation.