4/14🔬 We captured 16-to 22-hour 3D movies of this process. However, segmenting and tracking individual cells in a dense 3D tissue is a challenging task. We built a deep learning pipeline that combines the forces of #StarDist, #UNET, #Cellpose, and #TrackMate. 💪#ImageAnalysis

✍️ New in #eLife: #CellSeg3D introduces #WNet3D, a self-supervised 3D #segmentation method for #microscopy data — no labels needed. Claims to outperform #Cellpose/#StarDist on 4 datasets. Includes #opensource plugin (#Napari) + full 3D annotated #cortex dataset. Will test it later.
@napari.org

Back to #coding after weeks of manuscript writing; my model’s been training for almost 2hrs, with 4 more to go on my NVIDIA GeForce RTX- ROG Strix!
#StarDist #Bioimage #Bioimage_analysis #Cell_Biology

Subcellular #SpatialTranscriptomics Cell Segmentation
#TransformerNeuralNetwork
based on Image AND Transcriptome

For Single-Cell/Subcellular ST
Stereo-seq; Seq-scope

vs other #DeepLearning methods #StarDist #CellPose #DeepCell

#NatureMethods 2023
www.nature.com/articles/s41...