Dear all,

I would like to point out to our new paper: *Deep learning enabled 
multi-organ segmentation of mouse embryos *by Rolfe et al. It is publicly 
available at  
https://journals.biologists.com/bio/article/12/2/bio059698/287076/Deep-learning-enabled-multi-organ-segmentation-of

Paper describes and evaluates the outcomes of a deep-learning segmentation 
network we trained using diceCT scans of E15 wild-type mice from IMPC, and 
the published anatomical atlas. 

The companion tool, *MEMOS, *is now available as a part of the 3D Slicer 
extension catalog. MEMOS is independent of the existing SlicerMorph 
extension, and you need to install separately. When a high-end GPU (16GB 
minimum GPU RAM required) is available, segmentation of 50 organs takes 
about 40-60 secs. If a GPU is not available, it defaults to CPU based 
inference, which is substantially longer at 10-40 minutes, depending on the 
number of cpu cores available. 

Installation instructions, along with a sample E15 fetal scan from IMPC is 
available at the official repository https://github.com/SlicerMorph/MEMOS.

Here is a brief video of how the tool is used. 
https://twitter.com/i/status/1628172435030937600

-- 
You received this message because you are subscribed to the Google Groups 
"Morphmet" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to morphmet2+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/morphmet2/a5cd2e59-af39-4732-af31-941151da801an%40googlegroups.com.

Reply via email to