Automated Method for Analysis of Mammographic Breast Density - A Technique for Breast Cancer Risk Estimation
Abstract
The goal of this proposed project is to develop an automated technique to assist radiologists in estimating mammographic breast density. The computerized image analysis tool can provide a consistent and reproducible estimation of percent dense area on routine clinical mammograms, thereby contributing to the understanding of the relationship of mammographic density to breast cancer risk, detection, and prognosis, and the prevention and treatment of breast cancer. During this project year, we have improved our automated mammographic density segmentation program, referred to as Mammographic Density ESTimator (MDEST), for both digital mammograms (DMs) and digitized film mammograms (DFMs). The performance of the re-trained MDEST system was evaluated and compared with manually segmented mammographic density by experienced radiologists. The improved system was found to provide higher correlation and lower RMS error than the previous system in 10 of the 12 comparisons. The results indicate that the MDEST system is useful for mammographic density estimation for both DMs and DFMs. To further improve the estimation of percent dense area, we have designed new techniques and refined the existing methods for automatically tracking the breast boundary and the pectoral muscle edge on MLO view mammograms. These new methods improve the segmentation of breast boundary and pectoral muscle edge on noisy images over the previous methods. We will continue to improve the accuracy of the segmentation program and complete the development next year.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 2005
- Accession Number
- ADA438212
Entities
People
- Heang-Ping Chan
Organizations
- University of Michigan