Skip to content

SCMIA-Lab/CT-Segmentation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 

Repository files navigation

CT-Segmentation (DICOM → NIfTI → Bone Segmentation GUI)

CT-Segmentation is a lightweight and practical medical imaging segmentation GUI built using PyQt5.
It provides an easy end-to-end workflow to:

  • ✅ Load a DICOM folder
  • ✅ Convert it into a NIfTI (.nii.gz) volume
  • ✅ Run automated segmentation using modern AI tools

This project is useful for researchers, students, and developers working on CT imaging pipelines, helping reduce the manual effort of preparing data and running segmentation from the command line.

It supports:

  • Skellytour ( bone segmentation)
  • TotalSegmentator (bone and organ segmentation tasks)

Requirements

  • Python 3.10
  • PyQt5
  • SimpleITK
  • Skellytour (install before running this GUI)
  • TotalSegmentator (install before running this GUI)

Installation

1) Create Conda environment (recommended)

conda create -y -n ctseg python=3.10
conda activate ctseg

2) Install GUI dependencies

pip install PyQt5 SimpleITK

3) Install Skellytour

git clone https://github.com/cpwardell/Skellytour.git
cd Skellytour/
python -m pip install .

Verify Skellytour:

skellytour --help

4) Install TotalSegmentator

pip install totalsegmentator

Verify TotalSegmentator:

TotalSegmentator --help

Run the GUI

python app.py

How to Use

  1. Click Select DICOM Folder

  2. Click Select Output Folder

  3. Click Convert DICOM → NIfTI

    Creates: ct.nii.gz inside the output folder

  4. Select Segmentation Method

Skellytour

Model: low / medium / high

Device: gpu / cpu

TotalSegmentator

Select a task from the dropdown

Click Run Segmentation

Output

NIfTI Output

<output_folder>/ct.nii.gz

Skellytour Output

Saved inside:

<output_folder>/

TotalSegmentator Output

Saved inside:

<output_folder>/segmentations_<task_name>/

Notes

  • Make sure skellytour and TotalSegmentator commands work in your terminal before using the GUI.
  • GPU is recommended for faster segmentation, but CPU also works.
  • TotalSegmentator may require a license depending on your usage:
    • A non-commercial license is available.
    • For commercial usage, please check the official TotalSegmentator license terms.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages