Drug-on-Drug Interaction Network

Register for an account with your information.
Check the compatibility of your medication.
Add or remove medication with name and dosage.
Update your profile information.

Inspiration

Every year, 16.3 million people misuse prescription medications, often unaware of the risks posed by drug interactions. Healthcare specialists agree that the widespread availability and high usage of prescription drugs contribute to abuse, addiction, and even overdose. Many of these tragedies could be prevented with greater awareness and smarter tools for managing medication safety. That’s why we created the Drug-on-Drug Interaction (DDI) Network—a powerful, AI-driven solution designed to help users identify potential drug interactions before they lead to serious health consequences.

What it does

Drug-on-Drug Interaction Network (DDI Network) is a safety guide that allows users to input their personal health information and medication details, including dosage. It then analyzes potential drug interactions using a comprehensive drug database and helps users determine if the medicines they are taking are compatible with each other.

How it was Built

We developed the DDI Network using the following.

Llama to deploy and run the AI model efficiently.
Python for backend logic and AI-driven drug interaction analysis.
Biogrid and Therapeutics Data Commons (TDC) databases for accurate and up-to-date drug interaction data.
Streamlit for a clean and user-friendly interface.

Challenges we ran into

Something that we struggled with was ensuring the LLM was trained on up-to-date drug interaction data and understood these chemical processes. To do so, we needed to fine-tune our LLM, which came with its own set of complications; however, we were able to debug these issues and develop real-time AI integration. Additionally, designing an intuitive and accessible UI for users of all demographics was challenging. With multiple iterations, we created and added specific features each time that made our project more accessible for all demographics.

Accomplishments that we're proud of

We’re proud of successfully building a functional prototype within 24 hours, and specifically integrating real-time drug interaction analysis using AI. We also created a seamless user experience with a clean, accessible design and developed a potentially life-saving tool that can benefit users worldwide.

What we learned

Through researching and building this project, we learned the true complexity of drug interactions and the importance of reliable medical data. We also gained experience in optimizing AI models for real-time processing and user-friendly outputs. Most importantly, our project experience shed light on accessibility within healthcare applications.

What's next for Drug on Drug Interaction Network

There are multiple next steps for DDI Network. We would like to implement voice input and chatbot features for easier accessibility and to make medical information more easily understandable for the public. Additionally, we’d like to not only inform what drugs do or do not interact well, but also provide substitutions for drugs that perform similarly without interacting.