MedicalJawn

Patient Profile Graphs
Index For Patients
Search Function

Inspiration

We initially wanted to create a project featuring AI which could use some kind of document as input into our project, and we decided to choose to make a project which manages medical data. once we had an idea, we thought of possible ways that AI could be used in the medical field and we first came up with data processing, data visualization, and patient treatment plans.

What it does

Medical Jawn stores information in its database for patients information along with entries for their blood pressure and cholesterol levels. Medical Jawn then plots this information out on a graph while the Google Gemini API analyzes the patients data and creates a prediction of their information for the next few years if trends continue. Further data processing with NumPy is used to provide a patient with tips on how to increase or lower their blood pressure depending on what the patient needs.

How we built it

We used an SQLite database to store patient information and medical history. This information was called using SQL queries to get the data used in API endpoints which we made using FastAPI. NumPy was used in a few of these endpoints to process the data to be used in providing tips for patients. We used the React framework for our frontend to create each graph as a component to display multiple graphs on the patient info page at once.

Challenges we ran into

Creating the SQL queries and the API endpoints proved to be a challenge because of how interconnected they were. If a single endpoint failed the project wouldn't function. Another challenge was calling the API endpoints on the frontend and parsing the information to be displayed on the graph. Figuring out how to visualize the data in an appealing way was difficult since we also needed to focus on it being functional. A final challenge that we faced was the limit of Gemini API uses during production because there were only so many times that we could test our code before being locked out of using one of the AI models.

Accomplishments that we're proud of

We are proud to have created a project with an idea that could have a real world benefit if it was continued to be developed. Since our entire team was made of first years at RIT this is one of our first projects which we put serious effort into and we are proud to have it completed.

What we learned

During WiCHacks we learned how to manage our time and divide our work load to be more productive while also learning more about a few of the systems used in software engineering such as APIs and SQL.