Contextual Unsupervised Deep Clustering in Digital Pathology

Mariia Sidulova, Seyed Kahaki, Ian Hagemann, Alexej Gossmann
Proceedings of the fifth Conference on Health, Inference, and Learning, PMLR 248:558-565, 2024.

Abstract

Clustering can be used in medical imaging research to identify different domains within a specific dataset, aiding in a better understanding of subgroups or strata that may not have been annotated. Moreover, in digital pathology, clustering can be used to effectively sample image patches from whole slide images (WSI). In this work, we conduct a comparative analysis of three deep clustering algorithms – a simple two-step approach applying K-means onto a learned feature space, an end-to-end deep clustering method (DEC), and a Graph Convolutional Network (GCN) based method – in application to a digital pathology dataset of endometrial biopsy WSIs. For consistency, all methods use the same Autoencoder (AE) architecture backbone that extracts features from image patches. The GCN-based model, specifically, stands out as a deep clustering algorithm that considers spatial contextual information in predicting clusters. Our study highlights the computation of graphs for WSIs and emphasizes the impact of these graphs on the formation of clusters. The main finding of our research indicates that GCN-based deep clustering demonstrates heightened spatial awareness compared to the other methods, resulting in higher cluster agreement with previous clinical annotations of WSIs.

Cite this Paper

BibTeX
@InProceedings{pmlr-v248-sidulova24a, title = {Contextual Unsupervised Deep Clustering in Digital Pathology}, author = {Sidulova, Mariia and Kahaki, Seyed and Hagemann, Ian and Gossmann, Alexej}, booktitle = {Proceedings of the fifth Conference on Health, Inference, and Learning}, pages = {558--565}, year = {2024}, editor = {Pollard, Tom and Choi, Edward and Singhal, Pankhuri and Hughes, Michael and Sizikova, Elena and Mortazavi, Bobak and Chen, Irene and Wang, Fei and Sarker, Tasmie and McDermott, Matthew and Ghassemi, Marzyeh}, volume = {248}, series = {Proceedings of Machine Learning Research}, month = {27--28 Jun}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v248/main/assets/sidulova24a/sidulova24a.pdf}, url = {https://proceedings.mlr.press/v248/sidulova24a.html}, abstract = {Clustering can be used in medical imaging research to identify different domains within a specific dataset, aiding in a better understanding of subgroups or strata that may not have been annotated. Moreover, in digital pathology, clustering can be used to effectively sample image patches from whole slide images (WSI). In this work, we conduct a comparative analysis of three deep clustering algorithms – a simple two-step approach applying K-means onto a learned feature space, an end-to-end deep clustering method (DEC), and a Graph Convolutional Network (GCN) based method – in application to a digital pathology dataset of endometrial biopsy WSIs. For consistency, all methods use the same Autoencoder (AE) architecture backbone that extracts features from image patches. The GCN-based model, specifically, stands out as a deep clustering algorithm that considers spatial contextual information in predicting clusters. Our study highlights the computation of graphs for WSIs and emphasizes the impact of these graphs on the formation of clusters. The main finding of our research indicates that GCN-based deep clustering demonstrates heightened spatial awareness compared to the other methods, resulting in higher cluster agreement with previous clinical annotations of WSIs.} }
Endnote
%0 Conference Paper %T Contextual Unsupervised Deep Clustering in Digital Pathology %A Mariia Sidulova %A Seyed Kahaki %A Ian Hagemann %A Alexej Gossmann %B Proceedings of the fifth Conference on Health, Inference, and Learning %C Proceedings of Machine Learning Research %D 2024 %E Tom Pollard %E Edward Choi %E Pankhuri Singhal %E Michael Hughes %E Elena Sizikova %E Bobak Mortazavi %E Irene Chen %E Fei Wang %E Tasmie Sarker %E Matthew McDermott %E Marzyeh Ghassemi %F pmlr-v248-sidulova24a %I PMLR %P 558--565 %U https://proceedings.mlr.press/v248/sidulova24a.html %V 248 %X Clustering can be used in medical imaging research to identify different domains within a specific dataset, aiding in a better understanding of subgroups or strata that may not have been annotated. Moreover, in digital pathology, clustering can be used to effectively sample image patches from whole slide images (WSI). In this work, we conduct a comparative analysis of three deep clustering algorithms – a simple two-step approach applying K-means onto a learned feature space, an end-to-end deep clustering method (DEC), and a Graph Convolutional Network (GCN) based method – in application to a digital pathology dataset of endometrial biopsy WSIs. For consistency, all methods use the same Autoencoder (AE) architecture backbone that extracts features from image patches. The GCN-based model, specifically, stands out as a deep clustering algorithm that considers spatial contextual information in predicting clusters. Our study highlights the computation of graphs for WSIs and emphasizes the impact of these graphs on the formation of clusters. The main finding of our research indicates that GCN-based deep clustering demonstrates heightened spatial awareness compared to the other methods, resulting in higher cluster agreement with previous clinical annotations of WSIs.
APA
Sidulova, M., Kahaki, S., Hagemann, I. & Gossmann, A.. (2024). Contextual Unsupervised Deep Clustering in Digital Pathology. Proceedings of the fifth Conference on Health, Inference, and Learning, in Proceedings of Machine Learning Research 248:558-565 Available from https://proceedings.mlr.press/v248/sidulova24a.html.

Related Material