Data for McGovern et al, 2024: Finding and Following: A deep learning-based pipeline for tracking platelets during thrombus formation in vivo and ex vivo

Data for McGovern et al, 2024: Finding and Following: A deep learning-based pipeline for tracking platelets during thrombus formation in vivo and ex vivo

In these directories you will find example data to run the software described in the paper:

• segmentation
• • training_data: example frames (training_data/training_images) and corresponding ground truth segmentations (training_data/training_gt) that can be used to train the U-net described in the paper.
  • {exvivo,invivo}_example: example images with multiple matching corresponding manual segmentations that can be used to validate the U-net's performance.
• tracking image datasets that can be segmented with the U-net trained from the segmentation data, then tracked and analysed.

The data format is OME-NGFF v0.4, an emerging open format for bioimaging data and metadata. It can therefore be opened with open software in various ecosystems^[1]. To open the files in napari, install the napari-ome-zarr plugin and then (for example):

napari --plugin napari-ome-zarr tracking/mouse_invivo/200527_IVMTR73_Inj4_saline_exp3.ome.zarr

Note, however, that due to a current implementation issue with napari-ome-zarr, the opened segmentation files will not be manually editable with napari. For the moment, use the data loading widget from iterseg if you want to paint into the segmentation data.

1. https://ngff.openmicroscopy.org/tools/ ↩︎