Low-resolution segmentation
To quickly segment the low-resolution data, we leverage GPU-accelerated Pyclesperanto_prototype (Supplementary Figure 13). Conveniently, a Napari plugin called napari-pyclesperanto-assistant is available to optimize the segmentation pipeline and the used parameters.
Setting up segmentation
Specifically, we initialize the segmentation by setting the path to the data, where to save it and which channels to segment:
parentfolder = '~/test_dataset/low_resSegmentation/fish1'
resultsfolder = parentfolder + "_segmented"
channellist = ['1_CH488_000000', '1_CH552_000000']
It then iterates over a folder with the following structure:
.
|-- parentfolder
| |-- t00000
| | |-- 1_CH488_000000.tif
| | |-- 1_CH552_000000.tif
...
Output
It generates the segmentation labels and saves them as .tif files (8-bit or 16-bit, depending on the number of labels). Moreover, it saves the label statistics in an Excel document per timepoint.
.
|-- parentfolder
| |-- t00000
| | |-- 1_CH488_000000.tif
| | |-- 1_CH552_000000.tif
|-- parentfolder_segmented
| |-- 1_CH488_000000
| | |-- t00000
| | | |-- 1_CH488_000000sg.tif
| |-- 1_CH552_000000
| | |-- t00000
| | | |-- 1_CH552_000000sg.tif
|-- parentfolder_segmented_xlsx
| |-- 1_CH488_000000
| | |-- t00000.xlsx
| |-- 1_CH552_000000
| | |-- t00000.xlsx
...
Test data for low-resolution visualization
Test data is available for the low-resolution visualization in the folder Exemplary_Segmentation/low_resolution.
- on Synapse:
- or Zenodo: