Histological images for tumor detection in gastrointestinal cancer

<p>This is a set of 11977 image patches of hematoxylin &amp; eosin stained histological samples of human colorectal cancer. It is a subset of the data set &quot;100,000 histological images of human colorectal cancer and healthy tissue&quot; which is accessible at http://dx.doi.org/10.5281/zenodo.1214456. Further information on the samples is available there.</p> <p>This set contains three classes:</p> <p>ADIMUC - adipose tissue and mucus, i.e. loose non-tumor tissue</p> <p>STRMUS - stroma and muscle, i.e. dense non-tumor tissue</p> <p>TUMSTU - colorectal cancer epithelial tissue and stomach cancer epithelial tissue, i.e. tumor tissue</p> <p>All images are 512x512 px at 0.5 &micro;m/px</p> <p>We are using this data set to train a deep neural network to detect tumor cells in histological whole slide images of colorectal and stomach cancer.</p> <p>If you use these images in your research, please consider citing our previous [1] and upcoming publication.</p> <p>For information on ethics board approval, see [1].</p> <p>-------------</p> <p>[1] http://dx.doi.org/10.1371/journal.pmed.1002730</p>