A new arXiv preprint introduces an instance-aware graph modeling approach for detecting and classifying cells in pathology images. The method aims to capture complex cellular interactions within the tumor microenvironment rather than relying only on visual appearance. Accurate cell detection matters for diagnostic accuracy and treatment planning.
A new arXiv paper examines how standardized image-derived data can be shared in computational pathology using DICOM. The authors note that computational pathology research depends on large, diverse datasets, and that past efforts have focused heavily on collecting and centralizing imaging data. The work argues for common standards to make such data easier to exchange and reuse across institutions.
A new arXiv paper introduces ProsMAE, a masked autoencoder pretraining approach that draws on multiple data sources to classify ISUP grades from whole slide images. The authors address common obstacles in computational pathology, including gigapixel image sizes, staining and scanner variability, tissue artifacts, and scarce expert annotations. The work is a replacement submission on arXiv's machine learning category.