The performance of machine learning algorithms used for the segmentation of 3D biomedical images lags behind that of the algorithms employed in the classification of 2D photos. This may be explained by the comparative lack of high-volume, high-quality training datasets, which require state-of-the art imaging facilities, domain experts for annotation and large computational and personal resources to create. The HR-Kidney dataset presented in this work bridges this gap by providing 1.7 TB of artefact-corrected synchrotron radiation-based X-ray phase-contrast microtomography images of whole mouse kidneys and validated segmentations of 33 729 glomeruli, which represents a 1-2 orders of magnitude increase over currently available biomedical datasets. The dataset further contains the underlying raw data, classical segmentations of renal vasculature and uriniferous tubules, as well as true 3D manual annotations. By removing limits currently imposed by small training datasets, the provided data open up the possibility for disruptions in machine learning for biomedical image analysis.
Download