BrainGlobe version 1 is here! Head over to the blog to find out more

Analysing segmentation from other napari plugins#

Introduction#

While manual analysis of regions can be effective for various applications, there are instances where automated methods are necessary. In such cases, other napari plugins can be used.

Plugins exist to segment many features common to biomedical images, and their results can be incorporated with BrainGlobe to analyse the distribution of features of interest within the context of an anatomical atlas.

To be compatible with brainglobe-segmentation, the third-party napari plugin must be able to take a 3D image layer as input and return one of:

  • A 3D labels layer with either a 2 or 3D region labeled (e.g., segmenting bulk axonal projections)

  • A 3D points layer with a series of points corresponding to a trajectory through 3D space

Hint

It is possible to segment structures outside the brainglobe-segmentation plugin and import them in later (e.g., by saving to a .tiff and reloading). However, it is simpler to load the data using brainglobe-segmentation and then segment it using a third-party plugin. This ensures that the coordinate spaces you are using are consistent.

Instructions#

Segmenting your feature of interest#

Hint

There are many plugins available, and we can’t recommend a single tool for all data. For simple segmentation of 3D structures we often use napari-simpleitk-image-processing. Another useful tool (also by Robert Haase) is napari-segment-blobs-and-things-with-membranes.

Analyse the segmented layer#

Hint

At this point you could load a layer segmented previously and saved to disk

  • If required, rename the layer to be analysed

  • Click either the Track tracing or Region segmentation buttons to load the respective panels.

  • Highlight the layer to be analysed in the napari layers list:

Highlighting a specific layer

  • Click either the Add track from selected layer or Add region from selected layer button as applicable. This will add the region/points from the layer to the brainglobe-segmentation analysis list (in the same way as if it had been analysed manually within brainglobe-segmentation).

  • Follow the instructions for either:

Note

All data will be saved into your brainreg output directory