NA-MIC Project Weeks

Back to Projects List

MR Image Normalization

Key Investigators

• Michela Destito (University Magna Graecia of Catanzaro, Italy)
• Paolo Zaffino (University Magna Graecia of Catanzaro, Italy)
• Maria Francesca Spadea (Institute of Biomedical Engineering, KIT - Karlsruher Institut für Technologie, Germany)
• Petros Koutsouvelis (Maastricht University, Netherlands)

Presenter location: In-person

Project Description

A key step in medical image processing, particularly in MRI images, is normalization of gray level intensities. This normalization is important to ensure that images have a consistent intensity scale, facilitating any future analysis. The purpose is to create a targeted extension for normalization of MR images in Slicer.

Objective

• The aim of the project will be to provide an extension to normalize the intensities of MRI images
• It will be possible to choose different normalization methods.

Approach and Plan

1. Create an extension in which three normalization methods can be chosen: Zscore, WhiteStripe and Nyul.
2. To be able to compare the different gray levels of images normalized by multiple methods.

Progress and Next Steps

1. In this week I created the Extension for Normalization MRI Images with three normalization methods.
2. In the created extension you can choose which method to use.
3. Considering the first two proposed methods (Z-score and WhiteStripe) only the MRI image needs to be loaded and is normalized.
4. Considering the third method (Nyul) one must load in addition to the image to be Normalized, the MRI dataset and only then is the image normalized.
5. Future developments will be to implement new normalization methods proposed in the literature.

Illustrations

Background and References

1. https://github.com/Micheladestito/ImageNormalizationSlicer
2. https://github.com/jcreinhold/intensity-normalization