Ilastik – practical
Trainer: Dominik Kutra
Overview: This practical is an introduction on how to analyse microscopy images with machine learning-based tools, no coding or prior machine learning expertise required. ilastik is a simple, user-friendly tool for interactive image classification, segmentation and analysis. It is built as a modular software framework, which currently has workflows for automated (supervised) pixel- and object-level classification, automated and semi-automated object tracking, automated and semi-automated segmentation of boundary based objects, and object counting without detection. ilastik can handle data with up to 5 dimensions (3D + time + channel). Its computational back-end runs operations on-demand wherever possible, allowing for interactive prediction on data larger than RAM. Once the classifiers are trained, ilastik workflows can be applied to new data from the command line without further user interaction.
Learning outcomes:
After this section you should be able to:
- Perform semantic segmentation in ilastik using shallow and deep machine learning classifiers
- Perform boundary-based segmentation in ilastik
Materials:
- Practical guide
- Platform: https://www.ilastik.org/
- Data
- Hysteresis thresholding in order to separate touching objects: documentation and a video turorial
- Forum post explaining the atlas input in ilastik
- Documentation for the workflow: https://www.ilastik.org/documentation/tda/tda