Abstract
The two main approaches in the study of breast cancer are histopathology (analyzing visual characteristics of tumors) and genomics. While both histopathology and genomics are fundamental to cancer research, the connections between these fields have been relatively superficial. We bridge this gap by investigating the Carolina Breast Cancer Study through the development of an integrative, exploratory analysis framework. Our analysis gives insights—some known, some novel—that are engaging to both pathologists and geneticists. Our analysis framework is based on angle-based joint and individual variation explained (AJIVE) for statistical data integration and exploits convolutional neural networks (CNNs) as a powerful, automatic method for image feature extraction. CNNs raise interpretability issues that we address by developing novel methods to explore visual modes of variation captured by statistical algorithms (e.g., PCA or AJIVE) applied to CNN features.
Funding Statement
Research reported in this publication was supported
by a Specialized Program of Research Excellence (SPORE) in breast cancer (P50
CA058223), an award from the Susan G. Komen Foundation (OGUNC1202), the North
Carolina University Cancer Research Fund and a Cancer Center Support Grant (P30
CA016086). Iain Carmichael and J. S. Marron were partially supported by NSF
Grant IIS-1633074, BIG DATA 2016–2019. Iain Carmichael is currently supported by
NSF MSPRF DMS-1902440. Katherine Hoadley was supported by Komen Career Catalyst
Grant (CCR16376756).
Acknowledgments
We thank the Carolina Breast Cancer Study participants and staff. We also want to acknowledge Robert C. Millikan, founder of the Carolina Breast Cancer Study Phase 3.
Funding Statement
Research reported in this publication was supported
by a Specialized Program of Research Excellence (SPORE) in breast cancer (P50
CA058223), an award from the Susan G. Komen Foundation (OGUNC1202), the North
Carolina University Cancer Research Fund and a Cancer Center Support Grant (P30
CA016086). Iain Carmichael and J. S. Marron were partially supported by NSF
Grant IIS-1633074, BIG DATA 2016–2019. Iain Carmichael is currently supported by
NSF MSPRF DMS-1902440. Katherine Hoadley was supported by Komen Career Catalyst
Grant (CCR16376756).
Acknowledgments
We thank the Carolina Breast Cancer Study participants and staff. We also want to acknowledge Robert C. Millikan, founder of the Carolina Breast Cancer Study Phase 3.
Citation
Iain Carmichael. Benjamin C. Calhoun. Katherine A. Hoadley. Melissa A. Troester. Joseph Geradts. Heather D. Couture. Linnea Olsson. Charles M. Perou. Marc Niethammer. Jan Hannig. J. S. Marron. "Joint and individual analysis of breast cancer histologic images and genomic covariates." Ann. Appl. Stat. 15 (4) 1697 - 1722, December 2021. https://doi.org/10.1214/20-AOAS1433
Information