Histo-Miner: Use case CPI dataset
Authors/Creators
-
1.
University of Cologne
- 2. Department for Dermatology, University Hospital Cologne
- 3. Department of Dermatology and Allergy, School of Medicine, Technical University of Munich
- 4. Department of Dermatology, Venereology and Allergology, Helios St. Elisabeth Hospital Oberhausen, University of Witten/Herdecke
- 5. Department of Dermatology and Allergology, University Hospital of the Paracelsus Medical University Salzburg
Description
Use case dataset from Histo-Miner: Deep learning based tissue features extraction pipeline from H&E whole slide images of cutaneous squamous cell carcinoma paper. The list of feature is openly avaialble on the github repository.
The datasets consists of 45 WSI images in .ndpi format from 45 Cutaneous Squamous Cell Carcinamo skin cancer (cSCC) patients. The images are scan of skin tissue stained with H&E. WSIs come from 6 medical centers. The metadata file, containing the binary class labels of the images, is available as a CSV. A classifier was trained for classification and the features extracted from the slides to train the classifier are listed in List_of_features_including_morphology_features.json and List_of_features_without_morphology_features.json and the ranking of features by mRMR method during cross-validation are available in Ranking_of_features.json.
How to visualize the jsons files
If you are unfamiliar with this format, one simple way to visualize a JSON file is to paste all the content of the JSON file (that you can open with any text editors) into jsongrid.com or jsontotable.org. There is no need to save the data on these websites (Save or Share buttons).
The dataset size is 75.83 GB.
Files
cSCC_CPI.zip
Files
(75.8 GB)
| Name | Size | |
|---|---|---|
|
md5:ca3680a6b6d9abcad15cd6bc7ee3e362
|
75.8 GB | Preview Download |
|
md5:316cc9a998fc750ee75913f81f0d6d60
|
1.5 kB | Preview Download |
|
md5:864ac070256846ce79e1179771339404
|
10.9 kB | Preview Download |
|
md5:0d643297e63f9db7a7e4e6af3c757f0b
|
4.7 kB | Preview Download |
|
md5:e083907d5202ae2b0384d0c1d34c7ff9
|
4.1 kB | Preview Download |