Non-rigid registration of mammograms obtained with variable breast compression: A phantom study

Frédéric J.P. Richard, Predrag R. Bakić, Andrew D.A. Maidment

Research output: Chapter in Book/Report/Conference proceedingBook chapterResearchpeer-review

Abstract

The amount of breast compression applied during a mammographic exam affects the appearance of mammograms by introducing variations in the shape, position, and contrast of breast anatomical structures, which can conceal existing breast abnormalities or generate false alarms. Due to the complex tissue organization and elastic properties of the breast and the projective nature of mammography, rigid registration approaches are not useful in correcting these variations. We describe a non-rigid approach focused on registration of mammogram regions of interest, taking into account the changes in image contrast. This registration algorithm has been applied to synthetic mammograms generated using a deformable 3D anthropomorphic phantom and a model of breast deformation during mammographic compression.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
EditorsJames C. Gee, J. B. Antoine Maintz, Michael W. Vannier
PublisherSpringer
Pages281-290
Number of pages10
ISBN (Print)3540203435, 9783540203438
DOIs
Publication statusPublished - 2003
Externally publishedYes

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2717
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Subject classification (UKÄ)

  • Medical and Health Sciences

Free keywords

  • Breast compression
  • Finite elements
  • Image registration
  • Mammogram synthesis
  • Mammography
  • Multigrid optimization
  • Multimodality registration
  • Partial differential equations
  • Tissue modeling

Fingerprint

Dive into the research topics of 'Non-rigid registration of mammograms obtained with variable breast compression: A phantom study'. Together they form a unique fingerprint.

Cite this