DeepAI AI Chat
Log In Sign Up

Self-Supervision Closes the Gap Between Weak and Strong Supervision in Histology

by   Olivier Dehaene, et al.

One of the biggest challenges for applying machine learning to histopathology is weak supervision: whole-slide images have billions of pixels yet often only one global label. The state of the art therefore relies on strongly-supervised model training using additional local annotations from domain experts. However, in the absence of detailed annotations, most weakly-supervised approaches depend on a frozen feature extractor pre-trained on ImageNet. We identify this as a key weakness and propose to train an in-domain feature extractor on histology images using MoCo v2, a recent self-supervised learning algorithm. Experimental results on Camelyon16 and TCGA show that the proposed extractor greatly outperforms its ImageNet counterpart. In particular, our results improve the weakly-supervised state of the art on Camelyon16 from 91.4 98.7 99.3 trained via self-supervised learning can act as drop-in replacements to significantly improve existing machine learning techniques in histology. Lastly, we show that the learned embedding space exhibits biologically meaningful separation of tissue structures.


page 6

page 14

page 15


Revisiting Weakly Supervised Pre-Training of Visual Perception Models

Model pre-training is a cornerstone of modern visual recognition systems...

A Histopathology Study Comparing Contrastive Semi-Supervised and Fully Supervised Learning

Data labeling is often the most challenging task when developing computa...

Human-machine Interactive Tissue Prototype Learning for Label-efficient Histopathology Image Segmentation

Recently, deep neural networks have greatly advanced histopathology imag...

Self-supervised Multi-scale Consistency for Weakly Supervised Segmentation Learning

Collecting large-scale medical datasets with fine-grained annotations is...

Curriculum Learning Meets Weakly Supervised Modality Correlation Learning

In the field of multimodal sentiment analysis (MSA), a few studies have ...

ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders

Driven by improved architectures and better representation learning fram...

Histopathological Image Classification based on Self-Supervised Vision Transformer and Weak Labels

Whole Slide Image (WSI) analysis is a powerful method to facilitate the ...