Use this software for your next deep learning project with histopathological images.

If you want to:

• Load WSIs of different formats (ndpi, tiff, jp2..)
• Pan, zoom, view at different resolutions
• Overlay either segmentation masks, nuclei positions, tumor regions, heatmaps, etc
• Train your own deep learning network
• Test, and then visualise your trained network
• Visualise the training process
• Go through sample scripts for dealing with WSIs

then this software is for you.

This repository also contains the implementation for Representation-Aggregation Networks for Segmentation of Multi-Gigapixel Histology Images, published at Workshop on Deep Learning in Irregular Domains (DLID) in BMVC2017.

Refer to this presentation for a quick walkthrough.

For any queries, contact me or create an issue.

Installation Instructions

Install all packages mentioned in requirements.txt. If you plan to use JP2 images, enable MATLAB Engine for Python.

Sample Visualisation

Visualisation Pane	Segmentation Mask
Tumor Region	Heatmap
Nuclei Position

RAN-LSTM Results