🧠 Kidney Stone Detection
A sleek AI-powered web application for detecting kidney stones in CT images. Upload your scan, and our deep-learning model (custom CNN) will tell you “Stone” or “Non-stone” with confidence scores—complete with interactive visual feedback.
✨ Features
- 🔍 Single-image upload & instant prediction
- 📊 Confidence visualization via interactive bar charts
- 🔄 Real-time feedback with skeleton loader
- 📥 High-resolution download of results
- 📤 Shareable results via Web Share API or clipboard
🛠️ Technologies
Python
Flask
PyTorch
Tailwind CSS
html2canvas
Chart.js
🖼️ Preview
📁 Project Structure
kidney_stone_detection/
├── app.py
├── model.py
├── models/
│ └── kidney_stone_cnn.pth
├── requirements.txt
├── templates/
│ ├── index.html
│ ├── real_example.html
│ └── docs.html
└── static/
└── img/
├── preview-upload.png
└── preview-result.png
🚀 Installation Guide
- Clone the repo
git clone https://github.com/yourusername/kidney_stone_detection.git - Install dependencies
pip install -r requirements.txt - Run the server
python app.py - Open in browser:
http://127.0.0.1:5000
🧪 API Endpoints
| Method | Endpoint | Description |
|---|---|---|
| GET | / | Landing page |
| GET | /docs | Project documentation |
| GET | /real_example | Upload & predict UI |
| POST | /predict | Accepts image & returns JSON result |
👨💻 Developers
- Bander Sidiq – AI & Backend
- Nawzad Rasul – Frontend & UI/UX
📄 Licensed under the MIT License.