AdvDINO: Domain-Adversarial Self-Supervised Representation Learning for Spatial Proteomics

Stella Su, Marc Harary, Scott J. Rodig, William Lotter
Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, PMLR 315:3704-3722, 2026.

Abstract

Self-supervised learning (SSL) has emerged as a powerful approach for learning visual representations without manual annotations. However, the robustness of standard SSL methods to domain shift—systematic differences across data sources—remains uncertain, posing an especially critical challenge in biomedical imaging where batch effects can obscure true biological signals. We present AdvDINO, a domain-adversarial SSL framework that integrates a gradient reversal layer into the DINOv2 architecture to promote domain-invariant feature learning. Applied to a real-world cohort of six-channel multiplex immunofluorescence (mIF) whole slide images from lung cancer patients, AdvDINO mitigates slide-specific biases to learn more robust and biologically meaningful representations than non-adversarial baselines. Across more than 5.46 million mIF image tiles, the model uncovers phenotype clusters with differing proteomic profiles and prognostic significance, and enables strong survival prediction performance via attention-based multiple instance learning. The improved robustness also extends to a breast cancer cohort. While demonstrated on mIF data, AdvDINO is broadly applicable to other medical imaging domains, where domain shift is a common challenge.

Cite this Paper

BibTeX
@InProceedings{pmlr-v315-su26a, title = {AdvDINO: Domain-Adversarial Self-Supervised Representation Learning for Spatial Proteomics}, author = {Su, Stella and Harary, Marc and Rodig, Scott J. and Lotter, William}, booktitle = {Proceedings of The 9th International Conference on Medical Imaging with Deep Learning}, pages = {3704--3722}, year = {2026}, editor = {Huo, Yuankai and Gao, Mingchen and Kuo, Chang-Fu and Jin, Yueming and Deng, Ruining}, volume = {315}, series = {Proceedings of Machine Learning Research}, month = {08--10 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v315/main/assets/su26a/su26a.pdf}, url = {https://proceedings.mlr.press/v315/su26a.html}, abstract = {Self-supervised learning (SSL) has emerged as a powerful approach for learning visual representations without manual annotations. However, the robustness of standard SSL methods to domain shift—systematic differences across data sources—remains uncertain, posing an especially critical challenge in biomedical imaging where batch effects can obscure true biological signals. We present AdvDINO, a domain-adversarial SSL framework that integrates a gradient reversal layer into the DINOv2 architecture to promote domain-invariant feature learning. Applied to a real-world cohort of six-channel multiplex immunofluorescence (mIF) whole slide images from lung cancer patients, AdvDINO mitigates slide-specific biases to learn more robust and biologically meaningful representations than non-adversarial baselines. Across more than 5.46 million mIF image tiles, the model uncovers phenotype clusters with differing proteomic profiles and prognostic significance, and enables strong survival prediction performance via attention-based multiple instance learning. The improved robustness also extends to a breast cancer cohort. While demonstrated on mIF data, AdvDINO is broadly applicable to other medical imaging domains, where domain shift is a common challenge.} }
Endnote
%0 Conference Paper %T AdvDINO: Domain-Adversarial Self-Supervised Representation Learning for Spatial Proteomics %A Stella Su %A Marc Harary %A Scott J. Rodig %A William Lotter %B Proceedings of The 9th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2026 %E Yuankai Huo %E Mingchen Gao %E Chang-Fu Kuo %E Yueming Jin %E Ruining Deng %F pmlr-v315-su26a %I PMLR %P 3704--3722 %U https://proceedings.mlr.press/v315/su26a.html %V 315 %X Self-supervised learning (SSL) has emerged as a powerful approach for learning visual representations without manual annotations. However, the robustness of standard SSL methods to domain shift—systematic differences across data sources—remains uncertain, posing an especially critical challenge in biomedical imaging where batch effects can obscure true biological signals. We present AdvDINO, a domain-adversarial SSL framework that integrates a gradient reversal layer into the DINOv2 architecture to promote domain-invariant feature learning. Applied to a real-world cohort of six-channel multiplex immunofluorescence (mIF) whole slide images from lung cancer patients, AdvDINO mitigates slide-specific biases to learn more robust and biologically meaningful representations than non-adversarial baselines. Across more than 5.46 million mIF image tiles, the model uncovers phenotype clusters with differing proteomic profiles and prognostic significance, and enables strong survival prediction performance via attention-based multiple instance learning. The improved robustness also extends to a breast cancer cohort. While demonstrated on mIF data, AdvDINO is broadly applicable to other medical imaging domains, where domain shift is a common challenge.
APA
Su, S., Harary, M., Rodig, S.J. & Lotter, W.. (2026). AdvDINO: Domain-Adversarial Self-Supervised Representation Learning for Spatial Proteomics. Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 315:3704-3722 Available from https://proceedings.mlr.press/v315/su26a.html.

Related Material