Title
Development of a peripheral thickness estimation method for volumetric breast density measurements in mammography using a 3d finite element breast model
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.
Year
DOI
Venue
2010
10.1007/978-3-642-13666-5_63
Digital Mammography / IWDM
Keywords
Field
DocType
peripheral thickness estimation method,volumetric breast density,breast image,digital mammograms,peripheral thickness,extreme periphery,finite element breast model,volumetric breast density measurement,simulated mammograms,central region,thickness error,breast thickness,average absolute thickness error,physical model,finite element
Digital mammography,Biomedical engineering,Mammography,Peripheral,Signal intensity,Computer vision,Compression (physics),Radial line,Finite element method,Artificial intelligence,Percentage point,Materials science
Conference
Volume
ISSN
ISBN
6136
0302-9743
3-642-13665-6
Citations 
PageRank 
References 
0
0.34
2
Authors
7
Name
Order
Citations
PageRank
Olivier Alonzo-Proulx122.21
James G. Mainprize2149.82
Nathan J. Packard342.02
John M. Boone44310.07
Adil Al-Mayah561.65
Kristy Brock661.32
Martin J. Yaffe710023.70