Recent searches

No recent searches

Search options

Only available when logged in.
eupolicy.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
This Mastodon server is a friendly and respectful discussion space for people working in areas related to EU policy. When you request to create an account, please tell us something about you.

Administered by:

Server stats:

206
active users

eupolicy.social: AboutProfiles directoryPrivacy policy

Mastodon: AboutGet the appKeyboard shortcutsView source codev4.4.3

#cellseg3d

0 posts0 participants0 posts today
Fabrizio Musacchio<p>To wrap this up: Both tools are easy to test. I highly recommend trying them on your own data to see what works best for your use case.</p><p>I’ll include <a href="https://sigmoid.social/tags/CellSeg3D" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>CellSeg3D</span></a> in our next <a href="https://sigmoid.social/tags/Napari" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Napari</span></a> <a href="https://sigmoid.social/tags/bioimage" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>bioimage</span></a> analysis course (<a href="https://www.fabriziomusacchio.com/teaching/teaching_bioimage_analysis/" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://www.</span><span class="ellipsis">fabriziomusacchio.com/teaching</span><span class="invisible">/teaching_bioimage_analysis/</span></a>). Curious what impressions and feedback the students will share. 🧪🔍</p><p>What I really like about <span class="h-card" translate="no"><a href="https://fosstodon.org/@napari" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>napari</span></a></span> is how well it integrates modern <a href="https://sigmoid.social/tags/Python" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Python</span></a> tools. Great to have such a flexible, evolving <a href="https://sigmoid.social/tags/opensource" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>opensource</span></a> platform for (bio) <a href="https://sigmoid.social/tags/imageanalysis" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>imageanalysis</span></a>! 👌</p>
Fabrizio Musacchio<p>Tried the same with a more realistic 3D stack from the <a href="https://sigmoid.social/tags/ImageJ" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ImageJ</span></a> sample library.<a href="https://sigmoid.social/tags/Cellpose" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Cellpose</span></a> runs fast and segments very well out of the box.<a href="https://sigmoid.social/tags/CellSeg3D" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>CellSeg3D</span></a> takes considerably longer and seems to segment decently, but I couldn’t get a proper instance <a href="https://sigmoid.social/tags/segmentation" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>segmentation</span></a> in the post-processing step (which is recommended as part of its workflow). However, <a href="https://sigmoid.social/tags/CellSeg3D" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>CellSeg3D</span></a> looks very promising — just needs some more time and parameter exploration, I guess. </p><p>I’d recommend giving it a try 👌</p>
Fabrizio Musacchio<p>Tested <a href="https://sigmoid.social/tags/CellSeg3D" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>CellSeg3D</span></a> and <a href="https://sigmoid.social/tags/Cellpose" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Cellpose</span></a> on their example c5image dataset. Both segmentations look reasonable out-of-the-box, without any deep parameter tuning. With some extra effort, one could likely push either further I guess. Overall, both tools perform quite well on this small sample data set.</p>
Fabrizio Musacchio<p>✍️ New in <a href="https://sigmoid.social/tags/eLife" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>eLife</span></a>: <a href="https://sigmoid.social/tags/CellSeg3D" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>CellSeg3D</span></a> introduces <a href="https://sigmoid.social/tags/WNet3D" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>WNet3D</span></a>, a self-supervised 3D <a href="https://sigmoid.social/tags/segmentation" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>segmentation</span></a> method for <a href="https://sigmoid.social/tags/microscopy" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>microscopy</span></a> data — no labels needed. Claims to outperform <a href="https://sigmoid.social/tags/Cellpose" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Cellpose</span></a>/#StarDist on 4 datasets. Includes <a href="https://sigmoid.social/tags/opensource" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>opensource</span></a> plugin (<a href="https://sigmoid.social/tags/Napari" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Napari</span></a>) + full 3D annotated <a href="https://sigmoid.social/tags/cortex" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>cortex</span></a> dataset. Will test it later. </p><p>🌍 <a href="https://elifesciences.org/articles/99848" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">elifesciences.org/articles/998</span><span class="invisible">48</span></a></p><p><a href="https://sigmoid.social/tags/DeepLearning" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>DeepLearning</span></a> <a href="https://sigmoid.social/tags/Neuroscience" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>Neuroscience</span></a></p>
TrendingLive feeds
About