Genomics-Guided Representation Learning for Pathologic Pan-Cancer Tumor Microenvironment Subtype Prediction

Fangliangzi Meng; Hongrun Zhang; Ruodan Yan; Guohui Chuai; Chao Li; Qi Liu

doi:10.1007/978-3-031-72384-1_20

Genomics-Guided Representation Learning for Pathologic Pan-Cancer Tumor Microenvironment Subtype Prediction

Fangliangzi Meng
, Hongrun Zhang
, Ruodan Yan
, Guohui Chuai (Lead / Corresponding author)
, Chao Li (Lead / Corresponding author)
, Qi Liu (Lead / Corresponding author)

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

The characterization of Tumor MicroEnvironment (TME) is challenging due to its complexity and heterogeneity. Relatively consistent TME characteristics embedded within highly specific tissue features, render them difficult to predict. The capability to accurately classify TME subtypes is of critical significance for clinical tumor diagnosis and precision medicine. Based on the observation that tumors with different origins share similar microenvironment patterns, we propose PathoTME, a genomics-guided Siamese representation learning framework employing Whole Slide Image (WSI) for pan-cancer TME subtypes prediction. Specifically, we utilize Siamese network to leverage genomic information as a regularization factor to assist WSI embeddings learning during the training phase. Additionally, we employ Domain Adversarial Neural Network (DANN) to mitigate the impact of tissue type variations. To eliminate domain bias, a dynamic WSI prompt is designed to further unleash the model’s capabilities. Our model achieves better performance than other state-of-the-art methods across 23 cancer types on TCGA dataset. Our code is available at https://github.com/bm2-lab/PathoTME.

Original language	English
Title of host publication	Medical Image Computing and Computer Assisted Intervention – MICCAI 2024
Subtitle of host publication	27th International Conference, Marrakesh, Morocco, October 6–10, 2024, Proceedings, Part III
Editors	Marius George Linguraru, Qi Dou, Aasa Feragen, Stamatia Giannarou, Ben Glocker, Karim Lekadir, Julia A. Schabel
Place of Publication	Cham
Publisher	Springer
Chapter	20
Pages	206-216
Number of pages	11
Volume	15003
Edition	1
ISBN (Electronic)	9783031723841
ISBN (Print)	9783031723834
DOIs	https://doi.org/10.1007/978-3-031-72384-1_20
Publication status	Published - 2024
Event	27th International Conference on Medical Image Computing and Computer-Assisted Intervention - Palmeraie Conference Centre, Marrakesh, Morocco Duration: 6 Oct 2024 → 10 Oct 2024 Conference number: 27th https://conferences.miccai.org/2024/en/ (Conference Website)

Publication series

Name	Lecture Notes in Computer Science (LNCS)
Publisher	Springer
Volume	15003
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	27th International Conference on Medical Image Computing and Computer-Assisted Intervention
Abbreviated title	MICCAI 2024
Country/Territory	Morocco
City	Marrakesh
Period	6/10/24 → 10/10/24
Internet address	https://conferences.miccai.org/2024/en/ (Conference Website)

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Access to Document

10.1007/978-3-031-72384-1_20

Cite this

Meng, F., Zhang, H., Yan, R., Chuai, G., Li, C., & Liu, Q. (2024). Genomics-Guided Representation Learning for Pathologic Pan-Cancer Tumor Microenvironment Subtype Prediction. In M. G. Linguraru, Q. Dou, A. Feragen, S. Giannarou, B. Glocker, K. Lekadir, & J. A. Schabel (Eds.), Medical Image Computing and Computer Assisted Intervention – MICCAI 2024: 27th International Conference, Marrakesh, Morocco, October 6–10, 2024, Proceedings, Part III (1 ed., Vol. 15003, pp. 206-216). (Lecture Notes in Computer Science (LNCS); Vol. 15003). Springer . https://doi.org/10.1007/978-3-031-72384-1_20

Meng, Fangliangzi ; Zhang, Hongrun ; Yan, Ruodan et al. / Genomics-Guided Representation Learning for Pathologic Pan-Cancer Tumor Microenvironment Subtype Prediction. Medical Image Computing and Computer Assisted Intervention – MICCAI 2024: 27th International Conference, Marrakesh, Morocco, October 6–10, 2024, Proceedings, Part III. editor / Marius George Linguraru ; Qi Dou ; Aasa Feragen ; Stamatia Giannarou ; Ben Glocker ; Karim Lekadir ; Julia A. Schabel. Vol. 15003 1. ed. Cham : Springer , 2024. pp. 206-216 (Lecture Notes in Computer Science (LNCS)).

@inproceedings{fa52d65b0dfc428289f2aa15fb4a7243,

title = "Genomics-Guided Representation Learning for Pathologic Pan-Cancer Tumor Microenvironment Subtype Prediction",

abstract = "The characterization of Tumor MicroEnvironment (TME) is challenging due to its complexity and heterogeneity. Relatively consistent TME characteristics embedded within highly specific tissue features, render them difficult to predict. The capability to accurately classify TME subtypes is of critical significance for clinical tumor diagnosis and precision medicine. Based on the observation that tumors with different origins share similar microenvironment patterns, we propose PathoTME, a genomics-guided Siamese representation learning framework employing Whole Slide Image (WSI) for pan-cancer TME subtypes prediction. Specifically, we utilize Siamese network to leverage genomic information as a regularization factor to assist WSI embeddings learning during the training phase. Additionally, we employ Domain Adversarial Neural Network (DANN) to mitigate the impact of tissue type variations. To eliminate domain bias, a dynamic WSI prompt is designed to further unleash the model{\textquoteright}s capabilities. Our model achieves better performance than other state-of-the-art methods across 23 cancer types on TCGA dataset. Our code is available at https://github.com/bm2-lab/PathoTME.",

author = "Fangliangzi Meng and Hongrun Zhang and Ruodan Yan and Guohui Chuai and Chao Li and Qi Liu",

year = "2024",

doi = "10.1007/978-3-031-72384-1\_20",

language = "English",

isbn = "9783031723834",

volume = "15003",

series = "Lecture Notes in Computer Science (LNCS)",

publisher = "Springer ",

pages = "206--216",

editor = "Linguraru, \{ Marius George\} and Qi Dou and Aasa Feragen and Stamatia Giannarou and Ben Glocker and Karim Lekadir and Schabel, \{Julia A.\}",

booktitle = "Medical Image Computing and Computer Assisted Intervention {\textendash} MICCAI 2024",

edition = "1",

note = "27th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2024 ; Conference date: 06-10-2024 Through 10-10-2024",

url = "https://conferences.miccai.org/2024/en/",

}

Meng, F, Zhang, H, Yan, R, Chuai, G, Li, C & Liu, Q 2024, Genomics-Guided Representation Learning for Pathologic Pan-Cancer Tumor Microenvironment Subtype Prediction. in MG Linguraru, Q Dou, A Feragen, S Giannarou, B Glocker, K Lekadir & JA Schabel (eds), Medical Image Computing and Computer Assisted Intervention – MICCAI 2024: 27th International Conference, Marrakesh, Morocco, October 6–10, 2024, Proceedings, Part III. 1 edn, vol. 15003, Lecture Notes in Computer Science (LNCS), vol. 15003, Springer , Cham, pp. 206-216, 27th International Conference on Medical Image Computing and Computer-Assisted Intervention, Marrakesh, Morocco, 6/10/24. https://doi.org/10.1007/978-3-031-72384-1_20

Genomics-Guided Representation Learning for Pathologic Pan-Cancer Tumor Microenvironment Subtype Prediction. / Meng, Fangliangzi; Zhang, Hongrun; Yan, Ruodan et al.
Medical Image Computing and Computer Assisted Intervention – MICCAI 2024: 27th International Conference, Marrakesh, Morocco, October 6–10, 2024, Proceedings, Part III. ed. / Marius George Linguraru; Qi Dou; Aasa Feragen; Stamatia Giannarou; Ben Glocker; Karim Lekadir; Julia A. Schabel. Vol. 15003 1. ed. Cham: Springer , 2024. p. 206-216 (Lecture Notes in Computer Science (LNCS); Vol. 15003).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Genomics-Guided Representation Learning for Pathologic Pan-Cancer Tumor Microenvironment Subtype Prediction

AU - Meng, Fangliangzi

AU - Zhang, Hongrun

AU - Yan, Ruodan

AU - Chuai, Guohui

AU - Li, Chao

AU - Liu, Qi

N1 - Conference code: 27th

PY - 2024

Y1 - 2024

N2 - The characterization of Tumor MicroEnvironment (TME) is challenging due to its complexity and heterogeneity. Relatively consistent TME characteristics embedded within highly specific tissue features, render them difficult to predict. The capability to accurately classify TME subtypes is of critical significance for clinical tumor diagnosis and precision medicine. Based on the observation that tumors with different origins share similar microenvironment patterns, we propose PathoTME, a genomics-guided Siamese representation learning framework employing Whole Slide Image (WSI) for pan-cancer TME subtypes prediction. Specifically, we utilize Siamese network to leverage genomic information as a regularization factor to assist WSI embeddings learning during the training phase. Additionally, we employ Domain Adversarial Neural Network (DANN) to mitigate the impact of tissue type variations. To eliminate domain bias, a dynamic WSI prompt is designed to further unleash the model’s capabilities. Our model achieves better performance than other state-of-the-art methods across 23 cancer types on TCGA dataset. Our code is available at https://github.com/bm2-lab/PathoTME.

AB - The characterization of Tumor MicroEnvironment (TME) is challenging due to its complexity and heterogeneity. Relatively consistent TME characteristics embedded within highly specific tissue features, render them difficult to predict. The capability to accurately classify TME subtypes is of critical significance for clinical tumor diagnosis and precision medicine. Based on the observation that tumors with different origins share similar microenvironment patterns, we propose PathoTME, a genomics-guided Siamese representation learning framework employing Whole Slide Image (WSI) for pan-cancer TME subtypes prediction. Specifically, we utilize Siamese network to leverage genomic information as a regularization factor to assist WSI embeddings learning during the training phase. Additionally, we employ Domain Adversarial Neural Network (DANN) to mitigate the impact of tissue type variations. To eliminate domain bias, a dynamic WSI prompt is designed to further unleash the model’s capabilities. Our model achieves better performance than other state-of-the-art methods across 23 cancer types on TCGA dataset. Our code is available at https://github.com/bm2-lab/PathoTME.

U2 - 10.1007/978-3-031-72384-1_20

DO - 10.1007/978-3-031-72384-1_20

M3 - Conference contribution

SN - 9783031723834

VL - 15003

T3 - Lecture Notes in Computer Science (LNCS)

SP - 206

EP - 216

BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2024

A2 - Linguraru, Marius George

A2 - Dou, Qi

A2 - Feragen, Aasa

A2 - Giannarou, Stamatia

A2 - Glocker, Ben

A2 - Lekadir, Karim

A2 - Schabel, Julia A.

PB - Springer

CY - Cham

T2 - 27th International Conference on Medical Image Computing and Computer-Assisted Intervention

Y2 - 6 October 2024 through 10 October 2024

ER -

Meng F, Zhang H, Yan R, Chuai G, Li C, Liu Q. Genomics-Guided Representation Learning for Pathologic Pan-Cancer Tumor Microenvironment Subtype Prediction. In Linguraru MG, Dou Q, Feragen A, Giannarou S, Glocker B, Lekadir K, Schabel JA, editors, Medical Image Computing and Computer Assisted Intervention – MICCAI 2024: 27th International Conference, Marrakesh, Morocco, October 6–10, 2024, Proceedings, Part III. 1 ed. Vol. 15003. Cham: Springer . 2024. p. 206-216. (Lecture Notes in Computer Science (LNCS)). Epub 2024 Oct 3. doi: 10.1007/978-3-031-72384-1_20