HookNet

HookNet algorithm for the segmentation of histopathology breast tissue including ductal carcinoma in situ, invasive ductal carcinoma, invasive lobular carcinoma, non-malignant epithelium, fat and other breast tissue.

  1. Online examples
  2. More information
HookNet
Overview of the network architecture

Mart van Rijthoven

Online examples

The examples below show the output of HookNet, overlayed on the processed WSIs. You can zoom in by scrolling, moving around can be done with click&drag.

Algorithm

We have made the pretrained networks available via algorithms on the Grand Challenge Platform. You can try out the following algorithms:

HookNet-Lung

Segmentation algorithm for histopathology lung tissue.

HookNet-Breast

Segmentation algorithm for histopathology breast tissue.

Code

You can find the tensorflow implementation on our GitHub repository

Contact

Further questions regarding HookNet can be addressed to: Mart van Rijthoven.

People

Mart van Rijthoven

Mart van Rijthoven

PhD Candidate

Maschenka Balkenhol

Maschenka Balkenhol

Pathology Resident and Postdoctoral Researcher

Jeroen van der Laak

Jeroen van der Laak

Associate Professor

Francesco Ciompi

Francesco Ciompi

Assistant Professor

Publications

  • M. van Rijthoven, M. Balkenhol, K. Silina, J. van der Laak and F. Ciompi, "HookNet: Multi-resolution convolutional neural networks for semantic segmentation in histopathology whole-slide images", Medical Image Analysis, 2021;68:101890.