pjl54 / WSI_handling

Class for getting ROIs and masks from annotated pyramidal images. A work in progress.

WSI_handling

Code for handling digital pathology pyramidal whole slide images (WSIs). Currently works with annotation XMLs from Aperio ImageScope or annotation json's from QuPath and image formats supported by Openslide.

Supports getting a tile from a WSI at the desired micron-per-pixel (mpp), getting either the whole WSI or an annotated region, generating a mask image for either a tile or the WSIs, and showing the location of a tile on the WSI.

Annotation format

Example annotations are provided in ./example_annotations

XML annotations must follow the AperioImagescope format:

<?xml version="1.0" encoding="UTF-8"?>
<Annotations>
<Annotation LineColor="65280">
<Regions>
<Region>
<Vertices>
<Vertex X="56657.4765625" Y="78147.3984375"/>
<Vertex X="56657.4765625" Y="78147.3984375"/>
<Vertex X="56664.46875" Y="78147.3984375"/>
</Region>
</Regions>
</Annotation>
</Annotations>

With more <Annotation> or <Region> blocks for additional annotations.

json annotations must follow QuPath's json export format:

[
  {
    "type": "Feature",
    "id": "PathAnnotationObject",
    "geometry": {
      "type": "Polygon",
      "coordinates": [
        [
          [76793.51, 4613.02],
          [76651.56, 4684],
          [76580.59, 4684],
          [76580.59, 4754.97]                   
        ]
      ]
    },
    "properties": {
      "classification": {
        "name": "Tumor",
        "colorRGB": -3670016
      },
      "isLocked": true,
      "measurements": []
    }
]

Installation

pip install WSI_handling

Usage

import matplotlib.pyplot as plt
from WSI_handling import wsi

img_fname=r'/mnt/ccipd_data/TCGA_PRAD/2018Jan14/TCGA-EJ-5519-01Z-00-DX1.svs'
xml_fname=r'./example_annotations/TCGA-EJ-5519-01Z-00-DX1.xml'
w = wsi(img_fname,xml_fname)

plt.imshow(w.get_wsi(desired_mpp=8));

plt.imshow(w.mask_out_annotation(desired_mpp=8));

plt.imshow(w.get_annotated_region(desired_mpp=2,colors_to_use='other',annotation_idx='largest')[0]);

plt.imshow(w.get_annotated_region(desired_mpp=2,colors_to_use='red',annotation_idx='largest')[0]);

img, mask = w.get_annotated_region(desired_mpp=8,colors_to_use='green',annotation_idx='largest',mask_out_roi=False)
plt.imshow(img);
plt.show()
plt.figure
plt.imshow(mask);

plt.imshow(w.get_tile(desired_mpp=1,coords=(3400,54000),wh=(2000,2000)));

plt.imshow(w.mask_out_tile(desired_mpp=1,coords=(3400,54000),wh=(2000,2000)));

plt.imshow(w.show_tile_location(desired_mpp=1,coords=(3400,54000),wh=(2000,2000)));

img_fname=r'/mnt/ccipd_data/TCGA_PRAD/2018Jan14/TCGA-EJ-5519-01Z-00-DX1.svs'
xml_fname=r'./example_annotations/TCGA-EJ-5519-01Z-00-DX1.json'
w = wsi(img_fname,xml_fname)

plt.imshow(w.mask_out_annotation(desired_mpp=8,colors_to_use=['tumor','other']));

plt.imshow(w.get_annotated_region(desired_mpp=2,colors_to_use=['tumor'],annotation_idx='largest')[0]);