GRAPE - Guy's bReast cAncer lymPh nodEs

This repo contains all documentation for the GRAPE dataset hosted on AWS

• Dataset size: 3TB
• Contact name: Gregory Verghese
• Institution name: Kings's College London
• Institution URL: http://cancerbioinformatics.co.uk/
• Download at AWS: https://registry.opendata.aws/guys-breast-cancer-lymph-nodes/
• Contact email: https://github.com/cancerbioinformatics/GRAPE/issues

Description

This is a retrospective dataset of 1523 H&E-stained whole slide image (WSI) lymph nodes from breast cancer patients. The cohort consisted of 177 patients (122 LN-positive - metastasis was reported in at least 1 LN - and 55 LN-negative patients) with invasive breast carcinoma treated between 1984 and 2002 at Guy’s Hospital London, UK. Slides were scanned and digitised at 40x magnification (0.23 µm/pixel), NanoZoomer H.T2.0 2.0-HT (Hamamatsu Photonics UK, Ltd, Welwyn Garden City, UK).

Data structure

Patients indexed with KCLCB_pa01 format. Each folder contains WSIs for a single patient.

.
└── GRAPE
    ├── KCLCB_pa01
    │   ├── KCLCB_pa01_img1.ndpi
    │   ├── KCLCB_pa01_img2.ndpi
    │   ├── KCLCB_pa01_img3.ndpi
    │   └── KCLCB_pa01_img4.ndpi
    ├── KCLCB_pa02
    │   ├── KCLCB_pa02_img1.ndpi
    │   ├── KCLCB_pa02_img2.ndpi
    │   ├── KCLCB_pa02_img3.ndpi
    ...

Example usage

Below code demonstrates how to load WSIs programmatically using Python Openslide package

import openslide
import numpy as np
import matplotlib.pyplot as plt

#Read WSI
wsi = openslide.OpenSlide("/Path/to/wsi.ndpi")

#Get slide properties
dims=wsi.dimensions
x_resolution=wsi.properties[openslide.PROPERTY_NAME_MPP_X]
y_resolution=wsi.properties[openslide.PROPERTY_NAME_MPP_X]
base_mag=wsi.properties[openslide.PROPERTY_NAME_OBJECTIVE_POWER]

#Display thumbnail
wsi_thumbnail = wsi.get_thumbnail((1000,1000))
wsi_thumbnail=np.array(wsi_thumbnail)
plt.imshow(wsi_thumbnail)
plt.axis('off')

Credits

If you find the data useful, please cite the below paper:

@inproceedings{,
    title={},
    booktitle={},
    author={},
    year={}
}

License

Shield:

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.