PanNuke: An Open Pan-Cancer Histology Dataset for
Nuclei Instance Segmentation and Classification



Illustration of PanNuke label generation and verification. In stage A a Fully Convolutional Neural Network (FCNN) trained on public and internal nuclei classification and detection datasets was used for nuclei dotting to assist clinical experts in labeling. In stage B we used NuClick to create segmentation masks for every nuclei. These masks were verified and corrected by clinical experts.

Semi automatically generated nuclei instance segmentation and classification dataset with exhaustive nuclei labels across 19 different tissue types. The dataset consists of 481 visual fields, of which 312 are randomly sampled from more than 20K whole slide images at different magnifications, from multiple data sources. In total the dataset contains 205,343 labeled nuclei, each with an instance segmentation mask. Models trained on pannuke can aid in whole slide image tissue type segmentation, and generalise to new tissues. PanNuke demonstrates one of the first succesfully semi-automatically generated datasets.

[Cite]
[Download Data]
For enquiries, please contact Jevgenij Gamper (j.gamper@warwick.ac.uk)


News



Papers

Jevgenij Gamper, Navid A. Koohbanani, Simon Graham, Mostafa Jahanifar, Syed Ali Khurram, Ayesha Azam, Katherine Hewitt, Nasir Rajpoot

PanNuke Dataset Extensions, Insights and Baselines

Arxiv, 2020. Work in progress.
Jevgenij Gamper, Navid A. Koohbanani, Ksenija Benes, Ali Khuram, Nasir Rajpoot

PanNuke: An Open Pan-Cancer Histology Dataset for
Nuclei Instance Segmentation and Classification

ECDP, 2019.



Ground Truth Demo

<
Samples from exhaustively annotated PanNuke dataset, that contains image patches from 19 tissue types for nuclei instance segmentation and classification (Red: Neoplastic; Green: Inflammatory; Dark Blue: Connective; Yellow: Dead; Orange: Epithelial)


Nuclei Type Statistics


A comparative plot of class distributions per tissue. Numbers in parenthesis represent the total number of nuclei within that category or tissue type.


PanNuke Trained Model Applied to WSI


Visualisation of applying nuclei segmentation and classification network trained on PanNuke to unseen whole-slide images. Top row: Cervix tissue, with a visible differentiation between tumor and other tissue types. Bottom row: Prostate tissue.




This webpage template was borrowed from Richard Zhang.