Medical Images for Nucleus Segmentation

21,000 nuclei from several different organ types annotated by medical experts.

Please view this page on a desktop computer to see more information including CSV preview and download options.

About this data

This dataset contains annotated Hematoxylin & Eosin (H&E) images, one of the most commonly used image types in histopathology. There images were cropped from 30 whole slide images (WSIs) of digitized tissue samples of seven organs from The Cancer Genomic Atlas (TCGA). Further, only one WSI per patient was used in order to maximize nuclear appearance variation.

These images came from 18 different hospitals, which introduced another source of appearance variation due to the differences in the staining practices across labs. The size of each cropped images is 1000 x 1000 pixels which is cropped from dense region of tissue. To further ensure richness of nuclear appearances, the dataset covered seven different organs, which are breast, liver, kidney, prostate, bladder, colon, and stomach, including both benign and diseased tissue samples.

These 30 cropped images contained more than 21000 nuclei annotated and validated by medical experts.This dataset can be used by the research community to develop and benchmark generalized nuclear segmentation techniques that work on diverse nuclear types.

Nuclear segmentation in digital microscopic tissue images can enable extraction of high quality features for nuclear morphometric and other analyses in computational pathology. Nuclear morphometric and appearance features such as density, nucleus-to-cytoplasm ratio, size and shape features, and pleomorphism can be helpful for assessing not only cancer grades but also for predicting treatment effectiveness. This dataset can be used by the research community to develop and benchmark generalized nuclear segmentation techniques that work on diverse nuclear types.

Citation:

Kumar, N., Verma, R., Sharma, S., Bhargava, S., Vahadane, A. and Sethi, A., 2017. A dataset and a technique for generalized nuclear segmentation for computational pathology. IEEE transactions on medical imaging, 36(7), pp.1550-1560.

Author’s project webpage:

nucleisegmentationbenchmark.weebly.com

 

Raw Data

The input data for this job consist of an image name and a corresponding URL. Images are cropped from 30 whole slide images (WSIs) of a digitized tissue sample of seven organs from The Cancer Genomic Atlas (TCGA) and used only one WSI per patient to maximize nuclear appearance variation.

This data comes from an advanced pixel labeling semantic segmentation template. To duplicate this workflow, please get in touch with Figure Eight.

image_nameimage_url
TCGA-G9-6348-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-G9-6348-01Z-00-DX1.tif
TCGA-E2-A1B5-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-E2-A1B5-01Z-00-DX1.tif
TCGA-CH-5767-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-CH-5767-01Z-00-DX1.tif
TCGA-AR-A1AS-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-AR-A1AS-01Z-00-DX1.tif
TCGA-AY-A8YK-01A-01-TS1.tiffhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-AY-A8YK-01A-01-TS1.tiff
TCGA-G9-6336-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-G9-6336-01Z-00-DX1.tif
TCGA-RD-A8N9-01A-01-TS1.tiffhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-RD-A8N9-01A-01-TS1.tiff
TCGA-DK-A2I6-01A-01-TS1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-DK-A2I6-01A-01-TS1.tif
TCGA-HE-7130-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-HE-7130-01Z-00-DX1.tif
TCGA-A7-A13F-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-A7-A13F-01Z-00-DX1.tif
TCGA-G9-6356-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-G9-6356-01Z-00-DX1.tif
TCGA-HE-7128-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-HE-7128-01Z-00-DX1.tif
TCGA-G2-A2EK-01A-02-TSB.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-G2-A2EK-01A-02-TSB.tif
TCGA-38-6178-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-38-6178-01Z-00-DX1.tif
TCGA-A7-A13E-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-A7-A13E-01Z-00-DX1.tif
TCGA-HE-7129-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-HE-7129-01Z-00-DX1.tif
TCGA-G9-6363-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-G9-6363-01Z-00-DX1.tif
TCGA-21-5786-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-21-5786-01Z-00-DX1.tif
TCGA-18-5592-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-18-5592-01Z-00-DX1.tif
TCGA-B0-5698-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-B0-5698-01Z-00-DX1.tif
TCGA-KB-A93J-01A-01-TS1.tiffhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-KB-A93J-01A-01-TS1.tiff
TCGA-NH-A8F7-01A-01-TS1.tiffhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-NH-A8F7-01A-01-TS1.tiff
TCGA-G9-6362-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-G9-6362-01Z-00-DX1.tif
TCGA-AR-A1AK-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-AR-A1AK-01Z-00-DX1.tif
TCGA-B0-5711-01Z-00-DX1.tifhttps://requestor-proxy.figure-eight.com/figure_eight_datasets/TCGA_NucleiSegmentation/TissueImages/TCGA-B0-5711-01Z-00-DX1.tif
Download Options
TissueImages.zip | 119.6 MB
Annotations.zip | 16.1 MB
ImagesName.csv | 8.4 KB