This repository contains MRArt code for the paper:
S. Boudissa, G. Kanli, D. Perlo, T. Jaquet and O. Keunen. "ADDRESSING ARTEFACTS IN ANATOMICAL MR IMAGES: A K-SPACE-BASED APPROACH", 2024 IEEE International Symposium on Biomedical Imaging (ISBI), Athens, Greece, 2024, pp. 1-5, doi: 10.1109/ISBI56570.2024.10635199. https://ieeexplore.ieee.org/document/10635199
This paper was accepted to the 21st IEEE International Symposium on Biomedical Imaging . You can find the poster presented to the ISBI24 conference under Poster_ISBI24_542.pdf.
We propose a library named MRArt, for MRI Artefacts, that simulates realistic primary artefacts by manipulating the k-space signal, using standard image processing techniques. MRArt focuses on three degradations commonly encountered in anatomical images:
- Gaussian noise
- Blurriness effect
- Motion artefact
Each of the above degradation are generated with varying levels from low to severe. The image below illustrates the image domain and the corresponding k-space domain of each degradation types.
In order to use the library you need to install the following packages:
- pip install matplotlib
- pip install numpy
- pip install pydicom
- pip install scipy
- pip install skimage
You can find instructions on loading the data and utilizing the MRArt library in the Jupyter notebook titled Examples.ipynb.
If you have any questions, feel free to reach out to our support team at imaging@lih.lu.
We plan to introduce a 3D version of the library in the future, featuring additional degradation types.
Selma BOUDISSA, Georgia KANLI, Daniele PERLO, Thomas JAQUET and Olivier KEUNEN.
If you find MRArt useful in your research, please use the following for citation.
S. Boudissa, G. Kanli, D. Perlo, T. Jaquet and O. Keunen, "Addressing Artefacts in Anatomical MR Images: A k-space-Based Approach," 2024 IEEE International Symposium on Biomedical Imaging (ISBI), Athens, Greece, 2024, pp. 1-5, doi: 10.1109/ISBI56570.2024.10635199.