Renal Cell Carcinoma Region Detection and Subtyping Dataset
Complete Region Annotation
Areas that lie inside Red Lines are cancerous regions and vice versa
Minimal Point Based Annotation
The Red Point represents the marked position in the cancerous region
The Green Point stands for the non-cancerous region
Minimal Point-Based Annotation Rules
Equally mark points on both cancerous and non-cancerous regions. We set this number to 5 for RCC datasets.
Evenly distribute the points within the whole image.
Do not mark points on the blank, edge, badly stained, damaged (man-made), and other abnormal areas.
Min-Point annotation can reduce the annotation time to roughly one-twentieth when compared to the complete annotation. Noisy annotation is a annotation method between Min-Point and Complete Region.
A large RCC classification dataset with two types of annotation, the test set is composed by two parts, i.e., patch-level test set for cancer region detection, WSI-level test set for subtyping
Cancer Region Detection Results (Heat Maps)
Subtyping Results - Blue(ccRCC), Red(chRCC), Green(pRCC)
This dataset is composed of two parts and each sample is stored in an individual folder (named as the TCGA id of each svs file):
l_label: The complete region annotation is saved as a png file, the region marked by green is the cancer region, and marked by black is the background or normal tissue region.
p_label: The Min-Point annotation is saved as two folders, i.e., cancer and normal, each contains five png files indicate the point marked regions.
Note that, Only annotation data is available here, the original WSIs (.svs file) need to be downloaded from the TCGA portal. The labeled and unlabeled image patches need to be cropped from the original WSIs based on the annotation data.