Improving Clinical Diagnosis Through Change Detection in Mammography
Abstract
Temporal change of mass lesions overtime is a key piece of information in computer-aided diagnosis of breast cancer and treatment monitoring. For a specific patient, change detection depends on the ability to align the images of the mammogram sequence to a common axis, and the ability to build up memory about the image scene overtime. The process of aligning images to a common axis is termed image registration. The image scene representation is called site model. In the second year of this project, we developed a novel registration technique to align temporal sequences of the same patient, to construct a scene memory or site model, with the ultimate goal of performing change detection. We developed (1) a new hybrid registration algorithm aimed at the registration of non-rigid objects with minimal a prior knowledge; (2) a new change quantification metric based on the joint relative entropy between two images; (3) a patient specific site model concept to image-guided lesion monitoring; (4) a methodology to combine multiple transforms together to determine a composite image transform; and (5) an improved statistical segmentation algorithm for sequences of images.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 2000
- Accession Number
- ADA386825
Entities
People
- Yue-joseph Wang
Organizations
- The Catholic University of America