Breast Image Computing Lab
Artificial Intelligence (AI) Models Suited to 2D and 3D Mammographic Images
Mammographic images present multiple technical challenges that go beyond fine-tuning AI models. The goal of this project is to develop new AI architectures specifically suited to 2D and 3D mammography (breast tomosynthesis), which are able to (a) process mammographic images at high resolution, (b) integrate view-to-view correlations, and (c) retain robustness over different vendors and image acquisition settings.
Image-Driven Breast Cancer Risk Assessment and Racial Disparities
Improved breast cancer risk assessment models are needed to enable personalized screening strategies that achieve better
harm-to-benefit ratio based on earlier detection and better breast cancer outcomes than existing screening guidelines.
The aim of this project is to leverage computational breast imaging analytics in breast cancer risk assessment, with a primary focus on racial disparities towards bringing effective risk models to a larger breast cancer screening population than being targeted today.
Precision Medicine and Integrated Diagnostics for Breast Cancer Diagnosis and Prognosis
Clinical diagnostics span across multiple scales defined by radiology, pathology and multi-omics, with each of them producing data and observations that complement each other. The goal of this project is to develop integrated diagnostics methodologies to effectively combine multi-modal data, while looking at interactions and associations between different data sources, towards early diagnosis, prognosis and response to treatment for breast cancer.