pT1 Hotspot Tumor Budding T-cell Graph (pT1-HBTG) Dataset

<p>Dataset described in MIDL 2023 publication <a href="https://openreview.net/forum?id=ruaXPgZCk6i">&quot;Tumor Budding T-cell Graphs: Assessing the Need for Resection in pT1 Colorectal Cancer Patients&quot;</a>. For more information, please refer to the article.&nbsp;<strong>Please cite this article when using the data set.</strong></p> <p>The code for the Graph Neural Network experiments can be found on GitHub: <a href="https://github.com/digitalpathologybern/pT1-HBTG-MIDL2023">https://github.com/digitalpathologybern/pT1-HBTG-MIDL2023</a>.</p> <p><strong>Content:</strong></p> <ul> <li><strong>Zip file containing the different graph configurations in <a href="https://www.wikiwand.com/en/GXL">GXL format</a>.</strong> <ul> <li>Node labels: <ul> <li>Coordinates x and y in&nbsp;<span>\(\mu m\)</span> relative to the top-left of the hotspot (0/0).</li> <li>Type: lymphocyte/tumor bud (based on automated detection)</li> <li>ImageNet ViT-26 (DINO) features</li> </ul> </li> <li>Edge labels: <ul> <li>Distance between the nodes in <span>\(\mu m\)</span></li> </ul> </li> <li><br> &nbsp;</li> </ul> </li> <li><strong>JSON file with the class labels and cross-validation splits.</strong> The first level contains the split (5-folds, index 0-4), the second level the class labels (0/1), and the list of corresponding file-IDs.&nbsp; <pre><code>{ "0": { "0": [ "107_1", "107_2", ... ], "1": [ "160", "178_1", "178_2", ... ] }, ... "4": { "0": [ "10", "103", "104", ... ], "1": [ "117", "140_1", "140_2", ... ] } }</code></pre> <p>&nbsp;</p> </li> <li><strong>Zip file containing the PNGs of the ITBCC hotspots</strong> on which the graphs are based, extracted at full resolution (level 0, area of 0.785&thinsp;mm²). The WSIs were digitized using a Pannoramic 250 scanner at <span>\(0.243\mu m/pixel\)</span>.</li> <li>&nbsp;</li> <li><strong>Zip file containing the patches</strong> from which the ImageNet-based features were extracted (size 200x200 pixels, centered on the coordinates of the element)</li> </ul> <p><strong>File-ID nomenclature</strong><br> The first number indicates the patient/case (e.g. 10.gxl). If we have more than one WSI per patient, they are indicated by a second number (e.g. 13_1.gxl and 13_2.gxl). The file-ID is consistent between all data.</p> <p><strong>Bibtex for citation:</strong></p> <pre><code>@inproceedings{studer2023tumor, title={Tumor Budding T-cell Graphs: Assessing the Need for Resection in pT1 Colorectal Cancer Patients}, author={Studer, Linda and Bokhorst, John-Melle and Nagtegaal, Iris and Zlobec, Inti and Dawson, Heather and Fischer, Andreas}, booktitle={Medical Imaging with Deep Learning}, year={2023} }</code></pre>