Introduction

Welcome to the pathology AI platform (PAIP). PAIP is an educational service platform for AI researchers.

PAIP provides high-quality pathologic slide images of six tumor types as well as region-of-interest annotations developed by pathology experts.

This project is funded by the Ministry of Health and Welfare, Republic of Korea (2018 ~ 2023).

About the Data

Data is provided by Seoul National University Hospital (SNUH), Seoul National University Bundang Hospital (SNUBH), and SMG-SNU Boramae Medical Center (BRMH).

Acknowledgement

Please acknowledge PAIP 2019 in your publications as follows:

“Deidentified pathology images and annotations used in this research were provided by Seoul National University Hospital by a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (grant number: HI18C0316)”

License

PAIP data set is available under a Creative Commons Attribution-Non Commercial 4.0 International License (CC BY-NC 4.0) Under the following terms:

Datasets

The PAIP uploads two types of pathology whole slide images(WSIs, SVS files); liver hepatocellular carcinoma and colorectal adenocarcinoma cases.

  • liver hepatocellular carcinoma cases
  • colorectal adenocarcinoma cases

The whole slide images include the corresponding annotations (XML files). However, please note that the annotation levels of two types of cancer are not the same due to the characteristics as follows.

  • Liver cancer cases have 2 annotation levels, only for the tumor area and the other for the normal area.
  • Colorectal cancer cases have 1 annotation level representing only the tumor area.

File Format
  • SVS
  • XML
File Types & Names
  • Organ
    01 liver
    02 biliary
    03 prostate
    04 kidney
    05 colon
    06 pancreas
  • Diagnostic Information
    00 none
    01 Lymphatic Invasion
    02 Vascular Invasion
    03 Perineural Invasion
  • Histology
    01 adenocarcinoma
    02 ductal adenocarcinoma
    03 hepatocellular carcinoma
    04 renal cell carcinoma
  • Staining
    01 HE (hematoxylin eosin)
    02 IHC (immunohistochemistry)