Molecular Subtype Prediction for Breast Cancer Using H&E Specialized Backbone

Samaneh Abbasi-Sureshjani, Anıl Yüce, Simon Schönenberger, Maris Skujevskis, Uwe Schalles, Fabien Gaire, Konstanty Korski
Proceedings of the MICCAI Workshop on Computational Pathology, PMLR 156:1-9, 2021.

Abstract

Identifying the molecular markers to categorize breast cancer is one of the key steps in determining the prognosis and treatment strategy. The standard clinical practice is to do this analysis based on multiple immunohistochemistry (IHC) stainings for each biomarker, which is expensive and inconsistent when lacking resources. In this work, we investigated the predictiveness of morphological characteristics of Hematoxylin and Eosin (H&E) stained tissues for molecular subtype analysis, as an initial step for direct treatment response prediction based on H&E whole slide images (WSI). Transfer learning using backbones pre-trained on natural images is a common practice to deal with the challenge of lack of large and precisely annotated datasets. However, using pre-training on natural images is not optimal for clinical images. To deal with this challenge and leverage large pools of unlabeled data, we propose to use a specialized backbone pre-trained on H&E WSI in a self-supervised setting, i.e. without any labels. Our experiments show that this backbone is capable of learning discriminating morphological characteristics from H&E images which are well predictive of the molecular subtypes in weakly supervised settings. Also, the network performs better in terms of generalization to unseen data from new scanner types, despite the relatively small size of the dataset used for pre-training the backbone.

Cite this Paper

BibTeX
@InProceedings{pmlr-v156-abbasi-sureshjani21a, title = {Molecular Subtype Prediction for Breast Cancer Using H&E Specialized Backbone}, author = {Abbasi-Sureshjani, Samaneh and Y{\"u}ce, An{\i}l and Sch{\"o}nenberger, Simon and Skujevskis, Maris and Schalles, Uwe and Gaire, Fabien and Korski, Konstanty}, booktitle = {Proceedings of the MICCAI Workshop on Computational Pathology}, pages = {1--9}, year = {2021}, editor = {Atzori, Manfredo and Burlutskiy, Nikolay and Ciompi, Francesco and Li, Zhang and Minhas, Fayyaz and Müller, Henning and Peng, Tingying and Rajpoot, Nasir and Torben-Nielsen, Ben and van der Laak, Jeroen and Veta, Mitko and Yuan, Yinyin and Zlobec, Inti}, volume = {156}, series = {Proceedings of Machine Learning Research}, month = {27 Sep}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v156/abbasi-sureshjani21a/abbasi-sureshjani21a.pdf}, url = {https://proceedings.mlr.press/v156/abbasi-sureshjani21a.html}, abstract = { Identifying the molecular markers to categorize breast cancer is one of the key steps in determining the prognosis and treatment strategy. The standard clinical practice is to do this analysis based on multiple immunohistochemistry (IHC) stainings for each biomarker, which is expensive and inconsistent when lacking resources. In this work, we investigated the predictiveness of morphological characteristics of Hematoxylin and Eosin (H&E) stained tissues for molecular subtype analysis, as an initial step for direct treatment response prediction based on H&E whole slide images (WSI). Transfer learning using backbones pre-trained on natural images is a common practice to deal with the challenge of lack of large and precisely annotated datasets. However, using pre-training on natural images is not optimal for clinical images. To deal with this challenge and leverage large pools of unlabeled data, we propose to use a specialized backbone pre-trained on H&E WSI in a self-supervised setting, i.e. without any labels. Our experiments show that this backbone is capable of learning discriminating morphological characteristics from H&E images which are well predictive of the molecular subtypes in weakly supervised settings. Also, the network performs better in terms of generalization to unseen data from new scanner types, despite the relatively small size of the dataset used for pre-training the backbone. } }
Endnote
%0 Conference Paper %T Molecular Subtype Prediction for Breast Cancer Using H&E Specialized Backbone %A Samaneh Abbasi-Sureshjani %A Anıl Yüce %A Simon Schönenberger %A Maris Skujevskis %A Uwe Schalles %A Fabien Gaire %A Konstanty Korski %B Proceedings of the MICCAI Workshop on Computational Pathology %C Proceedings of Machine Learning Research %D 2021 %E Manfredo Atzori %E Nikolay Burlutskiy %E Francesco Ciompi %E Zhang Li %E Fayyaz Minhas %E Henning Müller %E Tingying Peng %E Nasir Rajpoot %E Ben Torben-Nielsen %E Jeroen van der Laak %E Mitko Veta %E Yinyin Yuan %E Inti Zlobec %F pmlr-v156-abbasi-sureshjani21a %I PMLR %P 1--9 %U https://proceedings.mlr.press/v156/abbasi-sureshjani21a.html %V 156 %X Identifying the molecular markers to categorize breast cancer is one of the key steps in determining the prognosis and treatment strategy. The standard clinical practice is to do this analysis based on multiple immunohistochemistry (IHC) stainings for each biomarker, which is expensive and inconsistent when lacking resources. In this work, we investigated the predictiveness of morphological characteristics of Hematoxylin and Eosin (H&E) stained tissues for molecular subtype analysis, as an initial step for direct treatment response prediction based on H&E whole slide images (WSI). Transfer learning using backbones pre-trained on natural images is a common practice to deal with the challenge of lack of large and precisely annotated datasets. However, using pre-training on natural images is not optimal for clinical images. To deal with this challenge and leverage large pools of unlabeled data, we propose to use a specialized backbone pre-trained on H&E WSI in a self-supervised setting, i.e. without any labels. Our experiments show that this backbone is capable of learning discriminating morphological characteristics from H&E images which are well predictive of the molecular subtypes in weakly supervised settings. Also, the network performs better in terms of generalization to unseen data from new scanner types, despite the relatively small size of the dataset used for pre-training the backbone.
APA
Abbasi-Sureshjani, S., Yüce, A., Schönenberger, S., Skujevskis, M., Schalles, U., Gaire, F. & Korski, K.. (2021). Molecular Subtype Prediction for Breast Cancer Using H&E Specialized Backbone. Proceedings of the MICCAI Workshop on Computational Pathology, in Proceedings of Machine Learning Research 156:1-9 Available from https://proceedings.mlr.press/v156/abbasi-sureshjani21a.html.

Related Material