A team of Stony Brook University researchers, led by two scientists in the department of biomedical informatics in the Renaissance School of Medicine (RSOM) and College of Engineering and Applied Sciences (CEAS), are developing a new way to analyze breast cancer imaging that incorporates mathematical modeling and deep learning. The approach will be much more interpretable and robust compared to previous methods. Their goal is to improve disease diagnosis and chart a treatment plan specific to the biomarker imaging and modeling findings.
To better understand breast cancer, researchers center on understanding breast tissue architecture and its changes over time. Breast tissue is composed of mixed cell types such as epithelial and adipose cells. Breast tissue composition directly influences tumor pathogenesis. While high breast density may be a risk factor for breast cancer, breast tissue complexity and changing architecture often makes subtle changes to tissue hard to detect by clinicians on standard imaging.
To tackle these hurdles, co-lead researchers Chao Chen, PhD, associate professor, and Prateek Prasanna, PhD, assistant professor, in the department of biomedical informatics at Stony Brook University, will develop “TopoQuant,” a suite of informatics tools for breast tissue images. TopoQuant is built on advanced mathematical modeling and machine learning. The team analyzes the structural complexity of breast parenchyma. They expect to use TopoQuant in collaboration with Stony Brook Medicine clinicians to uncover the intricate changes to tissue architecture that occur during cancer pathogenesis, disease progression and radiation treatment.
The work is supported by a new four-year National Cancer Institute (NCI) $1.2 million grant that runs through August 2028. Both Chen and Prasanna are affiliated with the Stony Brook Cancer Center’s Imaging, Biomarker and Discovery and Engineering Sciences Research Division.
“This research will offer new insights into how structural changes in breast tissue can influence cancer screening and treatment outcomes,” Chen said. “Topology is the area of mathematics that studies structures. By incorporating topology with deep learning in a seamless fashion, we can develop novel algorithms to capture structural changes in ways that were previously difficult with traditional techniques such as textural radiomics, potentially leading to better predictive models and treatment strategies.”
There are other machine learning-driven tools currently used by cancer imaging researchers, but the Stony Brook investigators say that existing tools cannot interpret or explain findings. However, with TopoQuant, clinicians will receive quantitative evidence of changes in breast tissue architecture and how that relates to cancer risk and treatment response.
In published preliminary findings in 2021, the team demonstrated the efficacy of the approach using one of the informatics tools in predicting a patient’s response to neoadjuvant chemotherapy in breast cancer. Additionally, the qualitative and quantitative results from that study suggested differential topological behavior of breast tissue characterized by patients who responded favorably to therapy and those who did not.
“Our prediction models will be unique in that they do not rely on traditional post-hoc interpretation but ensure interpretability by design,” Prasanna said. “The research is intended to not only benefit breast cancer diagnosis and treatment but will also have broader applications in fields like neuroscience. Therefore, we are excited about the cross-disciplinary collaborations this project will foster and the new avenues it will open for medical imaging research.”
Other collaborators from the RSOM include Alexander Stessin, a clinician in the department of radiation oncology; Wei Zhao, a breast cancer screening specialist in the department of radiology and Haibin Ling in the department of computer science within the CEAS.
Submitted by Stony Brook University
