This technology is an automated software platform that segments and analyzes digital breast tomosynthesis (DBT) images to provide objective and reproducible breast density assessments for cancer screening, risk stratification, and monitoring.
Breast density is an important predictor of breast cancer risk and is used to guide patient monitoring and screening decisions. Current breast density assessment relies heavily on the Breast Imagining Reporting and Data System (BI-RADS), which is based on radiologist interpretation and can vary with experience, training, and subjective judgment. In addition, breast density measurements from digital breast tomosynthesis (DBT) can be influenced by variations in imaging hardware, reconstruction algorithms, and post-processing techniques. These variabilities can limit consistency of longitudinal monitoring and breast cancer risk assessment. There is a need for standardized, reproducible methods to accurately quantify breast density across imagining platforms.
This technology is an automated platform that analyzes digital breast tomosynthesis (DBT) images to estimate volumetric breast density. Using image segmentation and tissue classification methods, it distinguishes dense breast tissue from fatty tissue and generates an objective measurement of breast density. The software incorporates image standardization and refinement steps to reduce variability caused by differences in imaging systems and processing workflows, enabling more consistent density assessment over time and across sites.
Preliminary validation cohorts demonstrated strong within-patient correlation of volumetric breast density measurements over time, with slightly reduced consistency across vendors, and strong agreement with BI-RADS categories.
IR CU26309
Licensing Contact: Joan Martinez