Inspiration
For a long time, late diagnosis of various medical conditions has been a major issue in the medical industry. Many people died from diseases such as cancer, or cardiovascular issues. We wanted to help mitigate this issue, so we developed this tool to help doctors quickly diagnose problems with high accuracy, and combat the issue of late diagnosis.
What it does
This website takes in basic medical data about a patient, and runs it through an AI model that we trained in order to determine the risk of cardiovascular disease for the patient. An integrated chatbot helper is also provided that walks the user through any points of confusion, allowing for seamless and effective use.
How we built it
We trained an AI model using flask, and developed the website using HTML, CSS, and Java Script
Challenges we ran into
One of the major challenges we encountered was the sudden shut down of one of our computers. This significantly reduced the pace of our progress, but we were able to quickly adapt to the situation and adjust our roles, temporarily placing one of our teammates in an advisory role, giving us various ideas for an elevator pitch and website content.
Accomplishments that we're proud of
We were able to quickly adapt to an unforeseen issue, thus keeping our progress from halting. In addition, we were able to quickly train an AI model to accurately diagnose cardiovascular diseases using provided sample datasets
What we learned
We learned how to quickly adapt to issues that would otherwise completely halt our progress. In addition, we learned to handle the continuous onslaught of merge conflicts and quickly resolve them.
What's next for CardiAct
At the moment, our project only estimates the risk factor of a patient. We would like to expand it so it gathers more detailed data for more descriptive diagnoses. In addition, we would like to obtain a web domain so we can expand and make it more accessible to various medical professionals
Built With
- css
- flask
- html
- javascript
- kaggle
- openrouterapi
- python
Log in or sign up for Devpost to join the conversation.