/mdss_seg

MDSS segmentation utilities

Primary LanguageJupyter NotebookGNU General Public License v3.0GPL-3.0

Medical Image Segmentation using Unet and DoubleUnet

This is a comparative study between Unet and DoubleUnet on medical image segmentation. For the purposes of this study, the CVC dataset was used (contains colonoscopy images and their corresponding segmented masks).

How to

  • Clone this project

  • Download CVC dataset here and extract it in the project's root folder.

  • Convert images from .tif format to .jpeg format (use a program like irfanview).

  • Download pretrained DoubleUnet model for CVC dataset here

  • Create a virtual environment and run:

    pip install -r requirements.txt

Citations

Special Thanks To

  • Dr. Dimitris Iakovidis: Professor, Dept of Computer Science and Biomedical Informatics, University of Thessaly, Lamia, Greece.

  • Dimitra-Christina Koutsiou: Ph.D. Student, Dept of Computer Science and Biomedical Informatics, University of Thessaly, Lamia, Greece.