Self supervised contrastive learning for digital histopathology
By Ozan Ciga, Anne L. Martel, Tony Xu. Machine Learning with Applications, 2020
412 citations recorded by the source corpus.
Unsupervised learning has been a long-standing goal of machine learning and is especially important for medical image analysis, where the learning can compensate for the scarcity of labeled datasets. A promising subclass of unsupervised learning is self-supervised learning, which aims to learn salient features using the raw input as the learning signal. In this paper, we use a contrastive self-supervised learning method Chen et al. (2020a) that achieved state-of-the-art results on natural-scene images, and apply this method to digital histopathology by collecting and training on 60 histopathology datasets without any labels. We find that combining multiple multi-organ datasets with different types of staining and resolution properties improves the quality of the learned features. Furthermore, we find drastically subsampling a dataset (e.g., using ? 1% of the available image patches) does not negatively impact the learned representations, unlike training on natural-scene images. Linear classifiers trained on top of the learned features show that networks pretrained on digital histopathology datasets perform better than ImageNet pretrained networks, boosting task performances up to 7.5% in accuracy and 8.9% in F1. These findings may also be useful when applying newer contrastive techniques to histopathology data. Pretrained PyTorch models are made publicly available at https://github.com/ozanciga/self-supervised-histopathology.
Subfields: Artificial Intelligence, Oncology, Radiology, Nuclear Medicine and Imaging
View Paper: https://www.semanticscholar.org/paper/086dfa02dca8dd8cac68bf59719a60ffcb0f144b
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Sergey Levine
- Quoc V. Le
- Yann LeCun
- Mohammad Norouzi
- Ruslan Salakhutdinov
- Yoshua Bengio
- Mohit Bansal
- Pieter Abbeel
In Oncology
In Radiology, Nuclear Medicine and Imaging
- Mohammad Norouzi
- Ting-Li Chen
- Yann LeCun
- Tom Goldstein
- Richard J. Chen
- Faisal Mahmood
- Simon Kornblith
- Basil Mustafa
Citations in the corpus
- Big Self-Supervised Models Advance Medical Image Classification (2021, background): “While some works have attempted to design domain-specific pretext tasks [3, 39, 52, 51], other works concentrate on tailoring contrastive learning to medical data [11, 19, 26, 50].”
- A Cookbook of Self-Supervised Learning (2023, background): “SSL can be especially useful in domains such as medicine where labels are costly or the specific task can not be known a priori [Krishnan et al., 2022, Ciga et al., 2022].”
- A foundation model for clinical-grade computational pathology and rare cancers detection (2024, positive): “Foundation models generate data representations, called embeddings, that can generalize well to diverse predictive tasks 27 .”
- Artificial intelligence in histopathology: enhancing cancer research and clinical oncology (2022, background): “In case of continuous numerical labels, this task is termed a regression problem 97 .”
- Scaling Vision Transformers to Gigapixel Images via Hierarchical Self-Supervised Learning (2022, background): “Lastly, though recent MIL approaches have adopted self-supervised learning as a strategy for patch-level feature extraction (called tokenization in ViT literature), parameters in the aggregation layers still require training [8, 16, 18, 20, 43, 45, 62].”
- Recent advances and clinical applications of deep learning in medical image analysis (2021, positive): “SimCLR pre-training also yielded a significant performance boost for histopathological image classification (Ciga et al., 2020).”