CT Image Processing, Segmentation, and Classification

This repository provides tools for processing, segmenting, evaluating, and classifying CT images, with a focus on atlas-based segmentation and random forest classification. The code is modular and uses Python with SimpleITK.

Features

• Pre-processing: Load and prepare CT images and masks.
• Registration: Linear and non-linear (Demons) registration using SimpleITK.
• Atlas-based Segmentation: Majority voting from registered masks.
• Evaluation: Computes Dice, Jaccard, and Hausdorff metrics.
• Classification: Random forest classifier to identify slices containing the pubic symphysis.
• Visualization: Plotting of images and segmentation overlays.

Directory Structure

. ├── main.py # Main script for running the pipeline ├── functions.py # Core functions for registration, segmentation, evaluation, and classification ├── requirements.txt # Python dependencies
├── README.md └── .gitignore

Getting Started

Prerequisites

• Python 3.8+
• See requirements.txt for dependencies

Installation

git clone https://github.com/yourusername/your-repo.git
cd Python-CT
pip install -r requirements.txt

Usage

1. Place your CT images and masks in the appropriate folders (see paths in Main.py).
2. Run the main script:

   python Main.py

3. The script will:
   • Load images and masks
   • Perform registration and segmentation
   • Evaluate segmentation quality
   • Train and apply a classifier
   • Display plots of results

File Descriptions

• Main.py: Entry point; orchestrates the workflow.
• functions.py: Implements registration, segmentation, evaluation, and classification functions.
• requirements.txt: Lists required Python packages.

Example Output

• Dice, Jaccard, and Hausdorff metrics printed to console.
• Plots showing reference images, segmentations, and overlays.
• Slice indices with highest probability for pubic symphysis.

Notes

• Image and mask file paths are hardcoded in Main.py. Adjust as needed for your dataset.
• The code assumes images are in NIfTI format (.nii.gz).
• For more advanced usage or batch processing, consider extending the scripts or modularizing further.

License

MIT License

---

Maintained by David Dashti. For questions or suggestions, please open an issue or contact me.