Papillary Renal Cell Carcinoma subtyping Dataset

Re-diagnosis WSIs

These WSIs were re-diagnosed by two experienced pathologists.

Select Region of Interests from WSIs

The pathologists have selected some (5-10) ROIs for each WSI

  • The pRCC subtyping dataset comprised 171 diagnostic WSI from 171 patients (scanned at 40x), of which 62 of type 1, 109 of type 2.

  • These WSIs are downloaded from KIRP project of TCGA.

  • The pathologists have selected a total of 1162 (613 vs. 549) ROIs in 2000x2000 size, approximately 10 ROIs for every type 1 case, 5 for type 2.

  • We randomly divide the dataset into three subsets in patient-level, training (60%), validation (20%), and testing (20%) set.

  • The dataset provided here is for research purposes only. Commercial uses are not allowed.

  • If you intend to publish a research work that uses any of these datasets, you must cite our publication.

Type 1 Case

Type 1 pRCC is composed of papillae covered with a single layer of small cells and scant clear or pale cytoplasm and uniform nuclei with inconspicuous nucleoli lying near the basement membrane

Type 2 Case

Type 2 PRCC is composed of tumor cells with voluminous cytoplasm and pseudostratified high-grade nuclei with prominent nucleoli


Instance-based Vision Transformer for Subtyping of Papillary Renal Cell Carcinoma in Histopathological Image

Zeyu Gao, Bangyang Hong, Xianli Zhang, Yang Li, Chang Jia, Jialun Wu, Chunbao Wang, Deyu Meng, Chen Li

Data Format

Each sample of this dataset is composed of two parts:

  1. The original ROIs (image patches) with the diagnostic labels (type 1 or type 2).

    • Save as png files under the corresponding folder.

  2. The Corresponding nuclei segmentation and classification results of each ROI.

    • Save as mat files under the corresponding folder.

    • There are two keys in each mat file: 'Instance' and 'Type'.

    • The key 'Instance' saves a pixel matrix with 0 to N. 0 represents the background and 1-N represent the instance number of each nucleus.

    • The key 'Type' saves a pixel matrix with 0 to 4. 0 represents the background, 1-3 represent the grade of each tumor nucleus and 4 represents endothelial nuclei.