Improving Stain Invariance of CNNs for Segmentation by Fusing Channel Attention and Domain-Adversarial Training

Kudaibergen Abutalip, Numan Saeed, Mustaqeem Khan, Abdulmotaleb El Saddik
Medical Imaging with Deep Learning, PMLR 227:1176-1198, 2024.

Abstract

Variability in staining protocols, such as different slide preparation techniques, chemicals, and scanner configurations, can result in a diverse set of whole slide images (WSIs). This distribution shift can negatively impact the performance of deep learning models on unseen samples, presenting a significant challenge for developing new computational pathology applications. In this study, we propose a method for improving the generalizability of convolutional neural networks (CNNs) to stain changes in a single-source setting for semantic segmentation. Recent studies indicate that style features mainly exist as covariances in earlier network layers. We design a channel attention mechanism based on these findings that detects stain-specific features and modify the previously proposed stain-invariant training scheme. We reweigh the outputs of earlier layers and pass them to the stain-adversarial training branch. We evaluate our method on multi-center, multi-stain datasets and demonstrate its effectiveness through interpretability analysis. Our approach achieves substantial improvements over baselines and competitive performance compared to other methods, as measured by various evaluation metrics. We also show that combining our method with stain augmentation leads to mutually beneficial results and outperforms other techniques. Overall, our study makes significant contributions to the field of computational pathology.

Cite this Paper

BibTeX
@InProceedings{pmlr-v227-abutalip24a, title = {Improving Stain Invariance of CNNs for Segmentation by Fusing Channel Attention and Domain-Adversarial Training}, author = {Abutalip, Kudaibergen and Saeed, Numan and Khan, Mustaqeem and Saddik, Abdulmotaleb El}, booktitle = {Medical Imaging with Deep Learning}, pages = {1176--1198}, year = {2024}, editor = {Oguz, Ipek and Noble, Jack and Li, Xiaoxiao and Styner, Martin and Baumgartner, Christian and Rusu, Mirabela and Heinmann, Tobias and Kontos, Despina and Landman, Bennett and Dawant, Benoit}, volume = {227}, series = {Proceedings of Machine Learning Research}, month = {10--12 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v227/abutalip24a/abutalip24a.pdf}, url = {https://proceedings.mlr.press/v227/abutalip24a.html}, abstract = {Variability in staining protocols, such as different slide preparation techniques, chemicals, and scanner configurations, can result in a diverse set of whole slide images (WSIs). This distribution shift can negatively impact the performance of deep learning models on unseen samples, presenting a significant challenge for developing new computational pathology applications. In this study, we propose a method for improving the generalizability of convolutional neural networks (CNNs) to stain changes in a single-source setting for semantic segmentation. Recent studies indicate that style features mainly exist as covariances in earlier network layers. We design a channel attention mechanism based on these findings that detects stain-specific features and modify the previously proposed stain-invariant training scheme. We reweigh the outputs of earlier layers and pass them to the stain-adversarial training branch. We evaluate our method on multi-center, multi-stain datasets and demonstrate its effectiveness through interpretability analysis. Our approach achieves substantial improvements over baselines and competitive performance compared to other methods, as measured by various evaluation metrics. We also show that combining our method with stain augmentation leads to mutually beneficial results and outperforms other techniques. Overall, our study makes significant contributions to the field of computational pathology.} }
Endnote
%0 Conference Paper %T Improving Stain Invariance of CNNs for Segmentation by Fusing Channel Attention and Domain-Adversarial Training %A Kudaibergen Abutalip %A Numan Saeed %A Mustaqeem Khan %A Abdulmotaleb El Saddik %B Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2024 %E Ipek Oguz %E Jack Noble %E Xiaoxiao Li %E Martin Styner %E Christian Baumgartner %E Mirabela Rusu %E Tobias Heinmann %E Despina Kontos %E Bennett Landman %E Benoit Dawant %F pmlr-v227-abutalip24a %I PMLR %P 1176--1198 %U https://proceedings.mlr.press/v227/abutalip24a.html %V 227 %X Variability in staining protocols, such as different slide preparation techniques, chemicals, and scanner configurations, can result in a diverse set of whole slide images (WSIs). This distribution shift can negatively impact the performance of deep learning models on unseen samples, presenting a significant challenge for developing new computational pathology applications. In this study, we propose a method for improving the generalizability of convolutional neural networks (CNNs) to stain changes in a single-source setting for semantic segmentation. Recent studies indicate that style features mainly exist as covariances in earlier network layers. We design a channel attention mechanism based on these findings that detects stain-specific features and modify the previously proposed stain-invariant training scheme. We reweigh the outputs of earlier layers and pass them to the stain-adversarial training branch. We evaluate our method on multi-center, multi-stain datasets and demonstrate its effectiveness through interpretability analysis. Our approach achieves substantial improvements over baselines and competitive performance compared to other methods, as measured by various evaluation metrics. We also show that combining our method with stain augmentation leads to mutually beneficial results and outperforms other techniques. Overall, our study makes significant contributions to the field of computational pathology.
APA
Abutalip, K., Saeed, N., Khan, M. & Saddik, A.E.. (2024). Improving Stain Invariance of CNNs for Segmentation by Fusing Channel Attention and Domain-Adversarial Training. Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 227:1176-1198 Available from https://proceedings.mlr.press/v227/abutalip24a.html.

Related Material