Meeting Banner
Abstract #2111

Accuracy of Multi-Expert algorithm for segmenting the breast

Artem Mikheev1, Laura Heacock1, Jean Logan1, and Henry Rusinek1

1Radiology Dept., NYU School of Medicine, New York, NY, United States

Breast density, defined as fraction of fibroglandular tissue (FGT), and post-contrast FGT enhancement (background parenchymal enhancement) are considered cancer risk factors. These MRI measures are recommended for radiologic reports and are promising cancer biomarkers. There is a general agreement that isolating the breasts from the chest wall (CW) is the most difficult to automate step in the FGT segmentation pipeline. Various methods for this task have been reported, but all show significant limitations. We have previously developed a semi-automated FGT segmentation tool that required approximately 7 min per case. We are reporting a new algorithm based on six overlapping Experts that significantly improves segmentation speed and accuracy.

This abstract and the presentation materials are available to members only; a login is required.

Join Here