Skip to main content Skip to navigation


Page Contents

  1. Abstract
  2. Download weights
  3. Download WSI-level results
  4. Sample Data
  5. License


One Model is All You Need: Multi-Task Learning Enables Simultaneous Histology Image
Segmentation and Classification

The recent surge in performance for image analysis of digitised pathology slides can
largely be attributed to the advances in deep learning. Deep models can be used to
initially localise various structures in the tissue and hence facilitate the extraction of
interpretable features for biomarker discovery. However, these models are typically
trained for a single task and therefore scale poorly as we wish to adapt the model for an
increasing number of different tasks. Also, supervised deep learning models are very
data hungry and therefore rely on large amounts of training data to perform well. In this
paper, we present a multi-task learning approach for segmentation and classification of
nuclei, glands, lumina and different tissue regions that leverages data from multiple in-
dependent data sources. While ensuring that our tasks are aligned by the same tissue
type and resolution, we enable meaningful simultaneous prediction with a single net-
work. As a result of feature sharing, we also show that the learned representation can
be used to improve the performance of additional tasks via transfer learning, including
nuclear classification and signet ring cell detection. As part of this work, we train our
developed Cerberus model on a huge amount of data, consisting of over 600K objects
for segmentation and 440K patches for classification. We use our approach to process
599 colorectal whole-slide images from TCGA, where we localise 377 million, 900K
and 2.1 million nuclei, glands and lumina respectively. We make this resource available
to remove a major barrier in the development of explainable models for computational

Download Weights

  • Download the Cerberus weights
  • Download pretrained weights (per fold) optimised using:
  • Download pretrained ResNet34 (torchvision) weights for transfer learning

The included Cerberus weights perform simultaneous:

  • Gland instance segmentation
  • Gland semantic segmentation (classification)
  • Nuclear instance segmentation
  • Nuclear semantic segmentation (classification)
  • Lumen instance segmentation
  • Tissue type patch classification

Other included weights are for pretraining and not for inference.

Download WSI-level Results

Information on how to download WSI-level results on 599 CRC images from TCGA can be found in the Download TCGA CRC Results tab.

Sample Data

We provide a sample dataset to help with preparation of data for training Cerberus. For this, we have two zipped directories:

  • Original Data
  • Patches

Download the sample dataset here.


Model weights and data are licensed under Attribution-NonCommercial-ShareAlike 4.0 International. Please consider the implications of using the weights under this license.