.png){kind=icon}
.png){kind=icon}
Return
Decoding 3D cell morphology with interpretable point cloud models
.png){kind=icon}
PRE
AI
DOI：10.1038/s41592-025-02752-w.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The morphology of fragmented cellular structures can be captured by converting microscopy images to 3D point clouds and analyzing them with rotation-invariant deep-learning models. These compact models represent a powerful alternative to conventional pixel-based analysis pipelines because they achieve high reconstruction and classification accuracies while remaining fast and biologically interpretable.