MarMar: Revolutionizing Medication Management
Inspiration
The inspiration for MarMar came from a personal experience. A close Family member that was managing multiple conditions, often struggled with managing multiple medications. Each doctor's visit seemed to end with a new prescription or a change in medication regimen, leading to confusion and potential health risks due to adverse drug interactions. This sparked my resolve to develop a solution that would simplify medication management and enhance patient safety.
What We Learned
Throughout the development of MarMar, we learned a great deal about the complexities of polypharmacy. We gained insights into the importance of user-friendly design, especially for elderly users who are the most affected by polypharmacy but often the least tech-savvy. We also learned the value of actionable insights derived from data, which can empower users and inform their healthcare decisions.
How We Built MarMar
We started by mapping out the user journey to understand the critical touchpoints where MarMar could intervene to provide value. We utilized the following tech stack:
- Frontend: Streamlit
- Backend: Python
- AI and Machine Learning: Python for developing the AI that analyzes drug interactions, using libraries like TensorFlow for building and training models.
We focused on developing a prototype that could handle basic functions like medication input, scheduling, and interaction alerts, followed by user testing to gather feedback for iterations.
Challenges We Faced
One of the biggest challenges was ensuring the accuracy and reliability of the drug interaction data. The health implications of incorrect information were significant, so we invested a lot of time in validating our sources and refining our prompts. Another challenge was designing an interface that was intuitive enough for users with minimal digital literacy. We overcame this by using Streamlit for now and now conducting multiple rounds of user tests to continuously refining the UI/UX based on real feedback.
Conclusion
Building MarMar has been a journey of learning and adaptation. We are proud of what we have achieved and remain committed to improving the platform to meet the needs of those juggling multiple medications. With MarMar, we're one step closer to making medication management safer and more intuitive for everyone.
Built With
- gpt3.5
- python
- streamlit


Log in or sign up for Devpost to join the conversation.