Development of a peripheral thickness estimation method for volumetric breast density measurements in mammography using a 3D finite element breast model

Olivier Alonzo-Proulx, James G. Mainprize, Nathan J. Packard, John M. Boone, Adil Al-Mayah, Kristy Brock, Martin J. Yaffe

Research output: Chapter in Book/Report/Conference proceedingConference contribution

11 Scopus citations

Abstract

A method was developed to determine the area in a mammogram where the breast is not in contact with the compression paddle (the periphery), and to predict the breast thickness in that peripheral region. The periphery is determined by evaluating the variation of the signal intensity along radial lines, and the peripheral thickness is modeled assuming the breast has a semi-circular shape. The method was tested on 26 simulated mammograms for which the volumetric information was available. The mammograms were obtained using CT data that were deformed to simulate mammographic compression and then projected using a physical model. The method predicted the thickness in the periphery to within 3.3 mm of the actual value and the volumetric breast density within 4.3 percentage points. The method was also tested on 209 digital mammograms, and on average it was estimated that thickness errors occurred on 9% of the breast image, and the average absolute thickness error on those points was estimated to be approximately 2.0 mm in the periphery and central region of the breast but as large as 10.5 mm in the extreme periphery where the thickness is small.

Original languageEnglish (US)
Title of host publicationDigital Mammography - 10th International Workshop, IWDM 2010, Proceedings
Pages467-473
Number of pages7
DOIs
StatePublished - 2010
Externally publishedYes
Event10th International Workshop on Digital Mammography, IWDM 2010 - Girona, Catalonia, Spain
Duration: Jun 16 2010Jun 18 2010

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume6136 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other10th International Workshop on Digital Mammography, IWDM 2010
Country/TerritorySpain
CityGirona, Catalonia
Period6/16/106/18/10

Keywords

  • digital mammography
  • periphery detection
  • volumetric breast density

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Development of a peripheral thickness estimation method for volumetric breast density measurements in mammography using a 3D finite element breast model'. Together they form a unique fingerprint.

Cite this