CT Scan Analysis

Overview

This project is designed for the analysis of CT scan images, specifically focusing on airway segmentation and blockage quantification. The analysis includes:

• Loading and visualizing DICOM images
• Segmenting airways based on Hounsfield Units (HU)
• Measuring airway volume and cross-sectional area
• Detecting potential airway obstructions
• 3D visualization of airway structures

Directory Structure

ct_scan/
├── raw_data/              # Stores raw DICOM files (ignored by .gitignore)
│   ├── dicom/             # Subdirectory for DICOM images
│   ├── metadata/          # Stores extracted metadata from DICOM files
│   └── uploads/           # Temporary directory for uploaded files
├── analysis/              # Jupyter notebooks and scripts for processing
│   ├── 01_preprocessing.ipynb  # Load and preprocess DICOM images
│   ├── 02_segmentation.ipynb   # Segment airways
│   ├── 03_visualization.ipynb  # 3D rendering and volume calculations
│   ├── 04_quantification.ipynb # Airway blockage analysis
│   └── utils/                  # Helper scripts for processing
├── results/               # Processed data and analysis results
│   ├── plots/             # Figures and visualizations
│   ├── volumes/           # Computed airway volume data
│   └── annotations/       # CSV files with extracted metadata or results
├── environment.yml        # Conda environment definition
├── LICENSE                # Project license (BSD-3-Clause)
├── README.md              # Project overview and usage instructions
└── .gitignore             # Files and directories to ignore in version control

Installation

1. Create the Conda Environment

Run the following to set up the environment:

conda env create -f environment.yml
conda activate ct_analysis

2. Install Jupyter Kernel (if not installed)

python -m ipykernel install --user --name=ct_analysis

Usage

1. Run Jupyter Lab

jupyter lab

2. Open Notebooks

• 01_preprocessing.ipynb → Load and explore CT scan images
• 02_segmentation.ipynb → Segment airways
• 03_visualization.ipynb → Generate 3D airway models
• 04_quantification.ipynb → Analyze airway obstruction

Contribution

If you'd like to contribute, fork the repository, create a feature branch, and submit a pull request.

License

This project is licensed under the BSD-3-Clause License.

Contact

Sebastian Eguiguren
Email: sebastian.eguiguren@tu-dresden.de
GitHub: jseguiguren