Training data for the "Integrative Analysis of Histological Textures and Lymphocyte Infiltration in Renal Cell Carcinoma using Deep Learning"

There are two ZIP-files consisting of small histological image tiles that have been used to detect and quantify distinct tissue textures and lymphocyte proportions from H&E-stained clear cell renal cell carcinoma (KIRC) digital tissue sections of the Cancer Genome Atlas (TCGA) image archive and the Helsinki dataset. The tissue_classification file contains 300x300px tissue texture image tiles (n=52,713) representing renal cancer (“cancer”; n=13,057, 24.8%); normal renal (“normal”; n=8,652, 16.4%); stromal (“stroma”; n= 5,460, 10.4%) including smooth muscle, fibrous stroma and blood vessels; red blood cells (“blood”; n=996, 1.9%); empty background (“empty”; n=16,026, 30.4%); and other textures including necrotic, torn and adipose tissue (“other”; n=8,522, 16.2%). Image tiles have been randomly selected from the TCGA-KIRC WSI and the Helsinki datasets. The binary_lymphocytes file contains mostly 256x256px-sized but also smaller image tiles of Low (n=20,092, 80.1%) or High (n=5,003, 19.9%) lymphocyte density (n=25,095). Image tiles have been randomly selected from the TCGA-KIRC WSI dataset. All accuracy of all annotations have been double-checked. However, the classification between multiple tissue textures or lymphocyte density can be sometimes ambiguous. The deep learning model parameters trained with the ResNet-18 infrastructure for (1) lymphocyte and (2) texture classification are named as (1) resnet18_binary_lymphocytes.pth and (2) resnet18_tissue_classification.pth. Codes and instructions to use these are found in https://github.com/vahvero/RCC_textures_and_lymphocytes_publication_image_analysis. If you use either work, please cite the publication by Brummer O et al (1) AND the TCGA Research Network (2): (1) Brummer O et al (unpublished) (2) The results shown here are in whole or part based upon data generated by the TCGA Research Network: https://www.cancer.gov/tcga