Inspiration

We wanted to create an accessible, AI-powered tool to help patients, especially the elderly, understand their pneumonia diagnosis through clear explanations.

What it does

The AI Pneumonia Detector analyzes chest X-rays, calculates the probability of pneumonia, and provides an easy-to-understand explanation. It can also read the explanation aloud for better accessibility.

How we built it

We used a pre-trained DenseNet-121 model to analyze X-rays and OpenAI's GPT to generate explanations. Streamlit powers the web interface, and Pyttsx3 enables text-to-speech functionality.

Challenges we ran into

Fine-tuning the model for accuracy and ensuring clear, simple explanations were some of the main challenges, especially when combining AI and medical language.

Accomplishments that we're proud of

We successfully built an app that not only detects pneumonia but also makes the diagnosis easily understandable, with an added text-to-speech feature for accessibility.

What we learned

We learned how to integrate AI with user-facing applications, fine-tune GPT for simple language, and add accessibility features like text-to-speech.

What's next for AI Pneumonia Detector

Next, we plan to improve the model's accuracy, add personalized patient information, and expand the app for mobile platforms to reach more people.

Built With

Share this project:

Updates