NA-MIC Project Weeks
Back to Projects List
SlicerImageAugmenter : evolution and new features
Key Investigators
- Ciro Benito Raggio (Karlsruhe Institute of Technology, Germany)
- Paolo Zaffino (Magna Græcia University of Catanzaro, Italy)
- Maria Francesca Spadea (Karlsruhe Institute of Technology, Germany)
Project Description
ImageAugmenter is a 3D Slicer extension that provides a simple interface to apply over 20 simultaneous MONAI transforms (spatial, intensity, etc.) to medical image datasets without programming.
It makes medical image augmentation more accessible, allowing a wider range of users to improve the performance of DL models in medical image analysis by increasing the number of samples available for training.
Since the extension is officially available in the Extension Manager since version 5.7.0 (current Preview Release), the idea is to fix known bugs, improve various aspects and add new features before the extension will be available in the stable release of 3D Slicer.
Objective
- Faster loading of the interface on first start
- Improve error handling and view management of 3D Slicer
- Improving the application of some transformations
- Possibility to select the sample to be used for the preview function
- Make the community aware of the extension and release it in the stable release of 3D Slicer
Approach and Plan
- Check all GUI components and/or dynamically load some components
- Better handling of generic exceptions by preventing the interface from crashing and requiring a manual restart
- Analyze MONAI documentation regarding available transformations for better interpretation of parameters
- Added a drop-down menu in preview mode to select the sample to which the selected transformations will be applied and previewed
- Networking with other people during the PW
Progress and Next Steps
New Features
- Enabled selection of specific samples for transformation preview
- Introduced regex support for defining patterns to locate images and masks
Revisions and Bug Fixes
- Refined available transformations to better align with MONAI documentation
- Improved exception handling for enhanced robustness
- Revised the UI for better consistency in components and layout
- Optimised and enhanced extension loading on first startup
- Fixed minor bugs
- Updated the extension logo