QMed

Inspiration

Our team was inspired and motivated by a mix of reasons to build Questiomedica (or QMed for short). A couple of us had a prior interest or even first-hand experience in the healthcare industry and were, therefore, very keen on building something related to that. Others were instead enticed by the potential to incorporate data science and/or hands-on building. Nevertheless, we were all united by our shared passion in building impactful projects. And, knowing how hectic the medical industry can, knowing how an early diagnosis can save a life, we knew that this project would certainly be impactful.

What it does

Questiomedica is a medical questionaire for patients to fill out prior to their doctor's appointments. It starts by asking patients several general questions about who they are and how they're feeling to cover the main bases of medical information. Then, based on said patient's system, it asks custom follow-up questions to gather useful, specific data and narrow down potential ailments. Once completed, a PDF file is produced for the doctor with all the medical information formmatted in user-friendly charts and tables as well as possible diagnoses for them to consider. It ensures that each patient's concerns and symptoms are be properly heard. It also allows the doctor to prepare for the patient's visits and frees up time during the appointment that can be allocated to detailed discussion about treatment options, etc. By providing a thorough inspection and analysis of the patient's well-being, this questionaire assists both the patient and the doctor.

How we built it

After choosing a project, we quickly designed a simple project idea and plan (a medical questionare that uses AI to create next questions), installed all the neccesary softwares, and immediately got started on coding the AI program and backend. And, along the way, we developed, refined, and experimented with our ideas for the frontend and final data visualization, which led to us building several dummy frontend pages and visuals. Lastly, we combined the frontend, AI and backend, and data visualization code.

Challenges we ran into

Of course, this process wasn't without its roadblocks. There were several bugs, especially merging the frontend, backend, and data visualization. And, yet, with the help of our trusty friends Copilot, Claude, and Cursor, we stuck through the debugging and finished victorusly on the other side!

Accomplishments that we're proud of a

We are very, very proud of Questiomedica. She is our child. We love her UI, her AI-generated follow-up questions, her data visuals — we're proud of every part of her and all the struggles that went into making her.

What we learned

For all of us, this was either a first or second hackathon, so we learned a bunch! We learned about how to properly use GitHub, leveraging search engines and LLMs to learn the most we could about new languages and coding issues, and much more!

What's next for QMed

If we had more time to work on Questiomedica, we'd probably incorporate a database of medical conditions and symptoms to train the AI algorithim to create more effective follow-up questions and more accurate diagnoses. Furthermore, we'd create a database to securely store patient medical data. Lastly, we'd create a separate webpage where doctors can login and view patient medical data, graphs, and diagnoses.

Built With

• cursor
• node.js
• openai-api
• react
• typescript