🫁 Lung Tumor Detection & Segmentation

This project is a web-based application built with Dash that enables users to upload lung CT scan images, detect tumors, and view a segmented report in PDF format.

🖼️ Live Demo

🚀 Features

• Upload CT scan images using drag-and-drop or file selection.
• Automatic tumor detection using image processing and ML model.
• Visualize detection output (e.g., bounding boxes).
• Segmented tumor report generated and rendered in a PDF viewer.
• Option to download the generated report.

🧠 Technologies Used

• 🐍 Python
• 📊 Dash
• 🖼️ PIL for image processing
• 📄 FPDF for PDF generation

📦 Installation

1. Clone the repository:

   git clone https://github.com/yourusername/Lung-Tumor-Detection-and-Segmentation.git
   cd Lung-Tumor-Detection-and-Segmentation

2. Create and activate a virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows use `venv\Scripts\activate`

3. Install Required Libraries

pip install -r requirements.txt

🧾 requirements.txt Example

dash
dash-bootstrap-components
torch
torchvision
Pillow
fpdf
numpy

✨ You can update the file with additional model dependencies if needed (e.g., PyTorch, TensorFlow).

🧪 Running the App

python app.py

Then navigate to http://127.0.0.1:8050/ in your browser.

📁 Project Structure

LUNG-TUMOR-DETECTION-AND-SEGMENTATION/
│
├── app.py                             # Dash app entry point
├── README.md                          # Project documentation
├── requirements.txt                   # All required packages
│
├── data_preparation/
│   └── data_loader.py                # Data loading and preprocessing
│
├── object_detection_training/
│   └── train_detector.py             # Training script for object detection model
│
├── segmentation_cropped_training/
│   ├── cropped_dataset.py            # Dataset and preprocessing for cropped segmentation
│   └── train_cropped_unet.py         # Training cropped segmentation model
│
├── segmentation_full_image_training/
│   └── train_segmentation_full.py    # Training full image segmentation model
│
├── test_and_report/
│   ├── model_loader.py               # Load trained models
│   ├── report_generator.py           # Generate analysis report as PDF
│   └── test_pipeline.py              # Full testing and inference pipeline
│
├── validation/
│   ├── utils.py                      # Utility functions
│   ├── validate_detection.py         # Validate object detection model
│   ├── validate_segmentation_crop.py # Validate cropped segmentation
│   └── validate_segmentation_full.py # Validate full image segmentation
├── assets/
│   ├── demo.gif

🖼️ Example Output

• Tumor fragments identified and highlighted.
• Bounding boxes overlaid on the original scan.
• PDF report with:
  • Fragment count
  • Tumor size
  • Coordinate locations
• Embedded viewer for instant preview.

---

📥 PDF Report

Each report includes:

• 🧩 Number of detected tumor regions
• 🔲 Bounding box coordinates (x, y, width, height)
• 📏 Approximate size (in pixels or mm)
• 📤 Downloadable via the app

---

🙌 Contributing

1. Fork the project.
2. Create your branch (git checkout -b feature-name).
3. Commit your changes (git commit -m 'add new feature').
4. Push to the branch (git push origin feature-name).
5. Open a Pull Request.

---

📜 License

This project is licensed under the MIT License.

---