Self-Rule to Adapt: Learning Generalized Features from Sparsely-Labeled Data Using Unsupervised Domain Adaptation for Colorectal Cancer Tissue Phenotyping

Christian Abbet, Linda Studer, Andreas Fischer, Heather Dawson, Inti Zlobec, Behzad Bozorgtabar, Jean-Philippe Thiran
Proceedings of the Fourth Conference on Medical Imaging with Deep Learning, PMLR 143:5-21, 2021.

Abstract

Supervised learning is conditioned by the availability of labeled data, which are especially expensive to acquire in the field of medical image analysis. Making use of open-source data for pre-training or using domain adaptation can be a way to overcome this issue. However, pre-trained networks often fail to generalize to new test domains that are not distributed identically due to variations in tissue stainings, types, and textures. Additionally, current domain adaptation methods mainly rely on fully-labeled source datasets. In this work, we propose Self-Rule to Adapt (SRA) which takes advantage of self-supervised learning to perform domain adaptation and removes the burden of fully-labeled source datasets. SRA can effectively transfer the discriminative knowledge obtained from a few labeled source domain to a new target domain without requiring additional tissue annotations. Our method harnesses both domains’ structures by capturing visual similarity with intra-domain and cross-domain self-supervision. We show that our proposed method outperforms baselines across diverse domain adaptation settings and further validate our approach to our in-house clinical cohort.

Cite this Paper

BibTeX
@InProceedings{pmlr-v143-abbet21a, title = {Self-Rule to Adapt: Learning Generalized Features from Sparsely-Labeled Data Using Unsupervised Domain Adaptation for Colorectal Cancer Tissue Phenotyping}, author = {Abbet, Christian and Studer, Linda and Fischer, Andreas and Dawson, Heather and Zlobec, Inti and Bozorgtabar, Behzad and Thiran, Jean-Philippe}, booktitle = {Proceedings of the Fourth Conference on Medical Imaging with Deep Learning}, pages = {5--21}, year = {2021}, editor = {Heinrich, Mattias and Dou, Qi and de Bruijne, Marleen and Lellmann, Jan and Schläfer, Alexander and Ernst, Floris}, volume = {143}, series = {Proceedings of Machine Learning Research}, month = {07--09 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v143/abbet21a/abbet21a.pdf}, url = {https://proceedings.mlr.press/v143/abbet21a.html}, abstract = {Supervised learning is conditioned by the availability of labeled data, which are especially expensive to acquire in the field of medical image analysis. Making use of open-source data for pre-training or using domain adaptation can be a way to overcome this issue. However, pre-trained networks often fail to generalize to new test domains that are not distributed identically due to variations in tissue stainings, types, and textures. Additionally, current domain adaptation methods mainly rely on fully-labeled source datasets. In this work, we propose Self-Rule to Adapt (SRA) which takes advantage of self-supervised learning to perform domain adaptation and removes the burden of fully-labeled source datasets. SRA can effectively transfer the discriminative knowledge obtained from a few labeled source domain to a new target domain without requiring additional tissue annotations. Our method harnesses both domains’ structures by capturing visual similarity with intra-domain and cross-domain self-supervision. We show that our proposed method outperforms baselines across diverse domain adaptation settings and further validate our approach to our in-house clinical cohort.} }
Endnote
%0 Conference Paper %T Self-Rule to Adapt: Learning Generalized Features from Sparsely-Labeled Data Using Unsupervised Domain Adaptation for Colorectal Cancer Tissue Phenotyping %A Christian Abbet %A Linda Studer %A Andreas Fischer %A Heather Dawson %A Inti Zlobec %A Behzad Bozorgtabar %A Jean-Philippe Thiran %B Proceedings of the Fourth Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2021 %E Mattias Heinrich %E Qi Dou %E Marleen de Bruijne %E Jan Lellmann %E Alexander Schläfer %E Floris Ernst %F pmlr-v143-abbet21a %I PMLR %P 5--21 %U https://proceedings.mlr.press/v143/abbet21a.html %V 143 %X Supervised learning is conditioned by the availability of labeled data, which are especially expensive to acquire in the field of medical image analysis. Making use of open-source data for pre-training or using domain adaptation can be a way to overcome this issue. However, pre-trained networks often fail to generalize to new test domains that are not distributed identically due to variations in tissue stainings, types, and textures. Additionally, current domain adaptation methods mainly rely on fully-labeled source datasets. In this work, we propose Self-Rule to Adapt (SRA) which takes advantage of self-supervised learning to perform domain adaptation and removes the burden of fully-labeled source datasets. SRA can effectively transfer the discriminative knowledge obtained from a few labeled source domain to a new target domain without requiring additional tissue annotations. Our method harnesses both domains’ structures by capturing visual similarity with intra-domain and cross-domain self-supervision. We show that our proposed method outperforms baselines across diverse domain adaptation settings and further validate our approach to our in-house clinical cohort.
APA
Abbet, C., Studer, L., Fischer, A., Dawson, H., Zlobec, I., Bozorgtabar, B. & Thiran, J.. (2021). Self-Rule to Adapt: Learning Generalized Features from Sparsely-Labeled Data Using Unsupervised Domain Adaptation for Colorectal Cancer Tissue Phenotyping. Proceedings of the Fourth Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 143:5-21 Available from https://proceedings.mlr.press/v143/abbet21a.html.

Related Material