Digital Mammography: 9th International Workshop, IWDM 2008 Tucson, AZ, USA, July 20-23, 2008 Proceedings (Lecture Notes in Computer Science / Image Processing, ... Vision, Pattern Recognition, and Graphics) 9783540705376
This book constitutes the refereed proceedings of the 9th International Workshop on Digital Mammography, IWDM 2008, held
135 60 115MB
English Pages 769 [793] Year 2008
Title Page
Preface
Organization
Table of Contents
Improvements to Single Energy Absorptiometry Method for Digital Mammography to Quantify Breast Tissue Density
Background
Method
Results and Discussion
Thickness Validation
Breast Density Validation and Error Estimation
Conclusion
References
Breast Density Segmentation: A Comparison of Clustering and Region Based Techniques
Introduction
Breast Density Segmentation
Thresholding Based on Excess Entropy $\it {(EE)}$
Fuzzy C-Means Approach $\it {(FCM)}$
Fisherfaces Based Method $\it {(FF)}$
Experimental Results
Application to Breast Density Classification
Conclusions
References
Mammographic Segmentation Based on Texture Modelling of Tab$\'{a}$r Mammographic Building Blocks
Background
Data
Methods
Texton Ranking
Outliers Detection
Visual Assessment
Results
Discussion
References
Analyzing Tree-Like Structures in Biomedical Images Based on Texture and Branching: An Application to Breast Imaging
Introduction
Background
Method
Segmentation of Regions of Interest (ROIs) and Delineation of the Corresponding Tree Structures
Computation of ROI Texture Features Using VQ
Computation of Tree Branching Descriptors Using String Encoding
Estimate the Similarity between Texture and Branching Topology Measures
Results
Discussion
References
Volumetric Assessment of Breast Tissue Composition from FFDM Images
Introduction
Improving Highnam & Brady’s Volumetric Breast Density
Overall Algorithm Quality Metric
Data
Results
Discussion
References
Effect of Image Quality on Film Reading
Introduction
Measurements of Image Quality in the Breast Screening Programme
Method
Data
Reading
Results
Discussion and Conclusions
References
A Comparison between Film-Screen Mammography and Full-Field Digital Mammography Utilizing Phase Contrast Technology in Breast Cancer Screening Programs
Background
Methods
Results
Discussion
Conclusion
References
Impact of Digital Mammography in Breast Cancer Screening: Initial Experience in a National Breast Screening Program
Introduction
Materials and Methods
Results
Recall Rate
Cancer Detection Rate
Discussion
References
Clinical Performance of Breast Tomosynthesis as a Function of Radiologist Experience Level
Background
Method
Results
Area under the ROC Curve
Recall Rate
Discussion
Conclusion
References
BIRADS Classification in Breast Tomosynthesis Compared to Mammography and Ultrasonography
Introduction
Tomosynthesis
Malm$\"{o}$ Breast Tomosynthesis
The BIRADS System
Material and Methods
Patient Material
Image Acquisition
Viewing Conditions and BIRADS Classification
Statistical Analysis of the BIRADS Classifications
Results
Discussion
References
Stereoscopic Digital Mammography: Improved Accuracy of Lesion Detection in Breast Cancer Screening
Introduction
Methods
Subjects
Study Design
Results
Sensitivity of Lesion Detection
Specificity of Lesion Detection
Discussion
References
Potential Role of FDG-PET Imaging in Defining Biology of Primary Breast Lesions
Introduction
Materials and Methods
FDG PET Imaging Protocol
Image Analysis
Results
Discussion
Research Support
Clinical Usefulness of Super High-Resolution Liquid Crystal Displays Using Independent Sub-pixel Driving Technology
Introduction
Methods and Materials
Results
Discussion
References
The Effect of Tomosynthesis X-Ray Pulse Width on Measured Beam Quality
Background
Method
Results
Discussions
Conclusion
References
Tomographic Dual Modality Breast Scanner
Background
Methods
Scanner Description
Phantom Studies
Human Studies
Results
Discussion
References
Dual-Energy X-Ray Absorptiometry Method Using a Full Field Digital Mammography System
Background
Theory
Materials and Methods
Devices
Image Acquisition
Calibration Curve Acquisition
Results
Choice of the Filter
QA Study
Breast Samples
Conclusion
References
Optimization of a Dual-Energy Contrast-Enhanced Technique for a Photon Counting Digital Breast Tomosynthesis System
Introduction
Material and Methods
Imaging System
Image Processing
Theoretical Model
Optimal Weighting Factor $\it {w_{t}}$
Dose Allocation Optimization
Results and Discussion
Optimal Weighting Factor $\it {w_{t}}$
Dose Allocation Optimization
Conclusion
References
Simulation and Phantom Studies of Contrast-Enhanced Dual Energy Mammography (CEDEM)
Introduction
Methods and Materials
Results
Discussion
References
Preliminary Evaluation of a Phase Contrast Imaging with Digital Mammography
Intoroduction
Principle of Phase Contrast
Materials and Methods
Experimental Conditions
Evaluation of Sharpness Due to PCM
Evaluation of Image Quality in Terms of Power Spectral Analysis
Clinical Mammograms
Results and Discussion
Evaluation of Sharpness Due to PCM
Evaluation of Image Quality in Terms of Power Spectral Analysis
Clinical Mammograms
Conclusion
References
Prompting in Mammography: Reproducibility
Introduction
Method
Results
Prompting of Lesions
False Positive Prompts in Cancer Cases
False Positive Prompts in Normal Mammograms
Persistence of Prompts through Multiple Digitisations
Marker Size
Discussion
References
Synthesising Abnormal Structures in Mammograms Using Pyramid Decomposition
Background
Synthesising Abnormalities in Digital Mammograms
Multi-scale Texture Analysis
Method
Steerable Pyramid Decomposition
Mammogram Data
Building a Model of Pyramid Variation
Synthesising Regions from the Models
Results
Discussion and Conclusions
References
Computer-Aided Microcalcification Detection on Digital Breast Tomosynthesis Data: A Preliminary Evaluation
Introduction
Method
Results
Discussion
References
Temporal Analysis of Mammograms Based on Graph Matching
Background
Methods
Data
Segmentation
Mass-Like Score
Graph Matching
Results
Discussion
References
Temporal Subtraction Versus Dual-Energy Contrast-Enhanced Digital Breast Tomosynthesis: A Pilot Study
Introduction
Material and Methods
Results
Discussion
References
Texture-Based Simultaneous Registration and Segmentation of Breast DCE-MRI
Introduction
Method
Texture Based Segmentation
Registration Framework
Results
Discussion
References
A CDMAM Image Phantom Software Improvement for Human Observer Assessment
Introduction
Methods
Results
Discussion
Conclusion
References
Computerized Classification of Mammary Gland Patterns in Whole Breast Ultrasound Images
Introduction
Materials
Image Acquisition
Classification of Mammary Gland Patterns
Methods
Overview
Preprocessing
VOI Extraction
Image Feature Extraction
Classification of Mammary Gland Patterns
Results and Discussion
Conclusion
References
Comparison between Physical Image Quality as Measured by a Newly Developed Phantom Dedicated for Digital Mammography QC and That by European Guidelines Methods
Introduction
Method
CNR
Geometric Distortion
Homogeneity
MTF
Missed Tissue at Chest Wall Side
Threshold Contrast Visibility
Result
Discussion
CNR
Geometric Distortion
Homogeneity
Modulation Transfer Function (MTF)
Missed Tissue at Chest Wall Side
Threshold Contrast Visibility
Conclusion
References
Assuring Authenticity of Digital Mammograms by Image Watermarking
Background
Method
Results
Discussion
Conclusion
References
Different Search Patterns and Similar Decision Outcomes: How Can Experts Agree in the Decisions They Make When Reading Digital Mammograms?
Introduction
Materials and Methods
Observers and Cases
Experimental Set-Up
Experimental Protocol
How the Observers’ Responses Were Scored
Visual Search and Decision Making Parameters Measured
Individual Analysis of Visual Search Strategy
Paired Analysis of Visual Search Strategy
Coverage of the Image
Results
Analysis of Visual Search and Decision Making Parameters
Individual Analysis of Visual Search Strategy
Paired Analysis of Visual Search Strategy
Coverage of the Image
Discussion
References
Optimization of Tomosynthesis Acquisition Parameters: Angular Range and Number of Projections
Introduction
Materials and Methods
Image Acquisition System and Reconstruction
Three-Dimensional Cascaded Linear System Theory
Phantom Data
Clinical Data
Results and Discussion
Conclusion
References
Subtle Abnormalities in Highly Dense Breasts Detected by Use of a Digital Phase Contrast Mammography System: A Report of Three Invasive Cancer Cases in the Early Stage
Background
Methods
Results
Discussion
Conclusion
References
Image Quality Assessment and Equipment Optimisation with Automated Phantom Evaluation in Full Field Digital Mammography (FFDM)
Background
Method
Results and Discussion
Conclusions
References
Breast Mass Detection under Increased Ambient Lighting
Introduction
Materials and Methods
Results
Discussion
References
Reducing Noise of Medical Grade Liquid Crystal Displays (LCD) and Its Relation to the Detection of Micro-calcifications
Introduction
Display Noise and Its Relation to Noises Common in Digital Radiography
Methods of Procedure
Noise Reduction Procedure and Noise Reduction Results
Pilot-Study of the Viewing Model
Results
Discussion
References
An Ontology to Support Adaptive Training for Breast Radiologists
Introduction
Ontology Research
Methods and Materials
The Annotated Cases
Methods
An Ontology to Support Adaptive Training for Breast Radiologists
Structure of the Ontology
Representation of the Cases and Learning Points
Discussion and Conclusion
References
A Web Database for Computer-Aided Detection and Diagnosis of Medical Images
Introduction
Data Collection and Analysis
Features
Conclusions and Future Work
References
An Interactive Computer Aided Decision Support System for Detection of Masses in Mammograms
Background
Methods
Results
Discusion
References
Detection of Masses in Digital Breast Tomosynthesis Mammography: Effects of the Number of Projection Views and Dose
Introduction
Materials and Methods
Results
Discussion
Conclusion
References
Effect of Similarity Metrics and ROI Sizes in Featureless Computer Aided Detection of Breast Masses in Tomosynthesis
Background
Methods
Results
Discussion
References
Knowledge Transfer across Breast Cancer Screening Modalities: A Pilot Study Using an Information-Theoretic CADe System for Mass Detection
Introduction
Materials and Methods
Overview of the KB-CADe System
Datasets
Experimental Design
Results
Discussion
References
Gradient Vector Flow Fields and Spiculated Mass Detection in Digital Mammography Images
Introduction
Gradient Vector Flow Fields
Algorithm
Algorithm Description
Implementation
Experimental Results and Discussions
Conclusions
References
The Evaluation of Effects on Breast Cancer Diagnoses When Using Mammographic Semantic Information
Introduction
Material and Methods
Semantic Information and Image Features
Experimental Design
Automatic Classification
Results
Conclusion and Discussions
References
Computerized Scheme for Focal Asymmetric Densities on Mammograms by Use of Geometric and Texture Analyses
Introduction
Clinical Materials
Methods and Techniques
CAD Scheme
Creation of Breast Map
Density Correction
Determination of Initial FAD Candidates
Identification of FADs
Results
Discussion
References
Identifying Corresponding Lesions from CC and MLO Views Via Correlative Feature Analysis
Introduction
Database
Methods
Lesion Segmentation
Feature Extraction
Feature Selection and Classification
Evaluation
Results
Discussion
References
Case-Specific Reliability Assessment for Improved False Positive Reduction with an Information-Theoretic CAD System
Background
Methods
Results
Dataset and Experimental Design
Reliability as a Performance Predictor
Reliability-Based Reduction of False Positives
Conclusions
References
Computerized Detection and Classification of Malignant and Benign Microcalcifications on Full Field Digital Mammograms
Background
Methods
Results
Discussion
References
The Effect of Training Sample Size on Performance of Mass Detection
Background
Method
Results
Discussion
References
Multiple-Instance Learning Improves CAD Detection of Masses in Digital Mammography
Background
Method
Notation
Logistic Regression
Traditional Statistical Learning of Classifiers
Multiple Instance Learning for Classifiers
Feature Selection for Optimizing MIL-Classifiers
Results
Run-Time Efficiency
Accuracy
Discussion
References
Optimizing the CAD Process for Detecting Mammographic Lesions by a New Generation Algorithm Using Linear Classifiers and a Gradient Based Approach
Background
Methods
The Detection of Mass Lesions
The Detection of Clusters of Micro-calcifications
The Aggregation of Micro-calcifications into Clusters
Classification vs. Sequential Filtration
Results
Discussion
References
Reliability Assessment of Ensemble Classifiers: Application in Mammography
Background
Methods
Information-Theoretic Computer-Aided Decision System
Constructing Multi-classifier System
Reliability Evaluation
Experimental Design and Results
Conclusions
References
Breast Mass Classification on Full-Field Digital Mammography and Screen-Film Mammography
Introduction
Methods
Data Sets
CAD Systems
Results
LDA Classification
DT Classification
Comparison of LDA and DT Classification
Conclusions
References
Detection of Microcalcifications Using a Nonuniform Noise Model
Introduction
Methods
Nonuniform Noise Model
Detector Inhomogeneity
Microcalcification Detection
Materials
Results
Discussion
References
The Breast Biomechanics Reference State for Multi-modal Image Analysis
Background
Methods
Results
Discussion
Applications of the Biomechanical Model
Conclusions
References
An Ideal Observer for a Model of X-Ray Imaging in Breast Parenchymal Tissue
Introduction
Theory
Model of a Binary Object
The Task
Evaluation of the MCMC Approach
Conclusions
References
Statistical Appearance Models of Mammographic Masses
Introduction
Method
Data
Shape Model
Texture Model
Combined Mass Appearance Model
Results
Discussion and Conclusions
References
Validation of a Digital Mammography Image Simulation Chain with Automated Scoring of CDMAM Images
Introduction
Methodology
Validation
Results
Discussion
References
Comparison of Two Methods to Develop Breast Models for Simulation of Breast Tomosynthesis and CT
Background
Methods
Results
Discussion
Future Work
References
Statistical Deformation Models of Breast Compressions from Biomechanical Simulations
Background
Method
Biomechanical Breast Model
Statistical Deformation Models
Results
Discussion
References
Classification of Benign and Malignant Masses in Ultrasound Breast Image Based on Geometric and Echo Features
Introduction
Method
Ultrasound Data
Feature Extraction
Results
Discussion and Conclusion
References
Validation of Cone-Beam CT Measurements of Tumour Burden Using Three-Dimensional Histopathology: Initial Results For a Lumpectomy
Background
Method
Results
Discussion
References
A Tool for Temporal Comparison of Mammograms: Image Toggling and Dense-Tissue-Preserving Registration
Introduction
Methods
Image Toggling
Dense-Tissue-Preserving Registration
Results
Conclusion
References
Development of Whole Breast Ultrasound Viewer and Automated Mass Detection System
Introduction
Materials and Methods
Imaging Technique
Viewing Technique
CAD Technique
Results
References
Automated Registration of Volumes of Interest for a Combined X-Ray Tomosynthesis and Ultrasound Breast Imaging System
Introduction
Methods
Results
Discussion
References
Estimating Individual Cancer Risks in the UK National Breast Screening Programme: A Feasibility Study
Introduction
Method
Results
Consent Rate
Weight Estimation
Questionnaire Data
Correlation of Breast Density with Risk Factors
Discussion
References
Clinical Performance of Digital Breast Tomosynthesis Versus Full-Field Digital Mammography: Preliminary Results
Background
Method
Results
Discussion
References
Analysis of Anatomical Linear Structure Information in Mammographic Risk Assessment
Background
Method
Data
Density Segmentation
Linear Structure Detection
Classification
Results
Discussion and Conclusions
References
Evaluating the Effect of Tomosynthesis Acquisition Parameters on Image Texture: A Study Based on an Anthropomorphic Breast Tissue Software Model
Introduction
Methods
Results
Discussion
References
Computer Aided Detection (CAD) for Digital Mammography: A Retrospective Reading Study for Consideration on Utilizing CAD Most Effectively
Introduction
Method and Materials
Mammography Imaging Unit and Viewer
CAD System
Database for Reading Test
Reading Test
Results
Discussion
Conclusion
References
Does a Mammography CAD Algorithm with Varying Filtering Levels of Detection Marks, Used to Reduce the False Mark Rate, Adversely Affect the Detection of Small Masses?
Background
Method
Results
Discussion
References
Performance of CADx on a Large Clinical Database of FFDM Images
Introduction
Materials and Methods
Database
Computerized Analysis Method
Performance Evaluation and Statistical Analysis
Results
Summary
References
Expedited Breast Care: A New Model in Breast Health
Background
Methods
Results
Discussion
References
Effect of Using Tungsten-Anode X-Ray Tubes on Dose and Image Quality in Full-Field Digital Mammography
Background
Method
Results
Discussion
References
Assessment of Low Energies and Slice Depth in the Quantification of Breast Tomosynthesis
Background
Methods
Results
Discussion
Conclusions and Future Work
References
Comparison of Multiple View Strategies to Reduce False Positives in Breast Imaging
Background
Methods
Candidate Detection
Multiple View Workflow
Asymmetry
Corresponding View
Breast Atlas
Results
Cad Performance
Multiple View Features
Corresponding View Feature vs. CAD Performance
Conclusion
References
Image Correction and Reconstruction for Breast Biopsy
Background
Methodology
Flat Field Correction
Dead Pixel Detection and Correction
Results
Conclusions
References
Progress Toward a Quantitative Scale for Describing Radiodensity in Mammographic Images
Introduction
Method
Performance
Discussion
References
Systematic Performance Analysis of SART as Applied to Digital Breast Tomosynthesis
Introduction
Methods
Results
Two Dimensions
Three Dimensions
Conclusion
References
Optimizing the Target-Filter Combination in Digital Mammography in the Sense of Image Quality and Average Glandular Dose
Introduction
Materials and Methods
Study Objectives
The Incidence Spectra for Mo/Mo, Mo/Rh, W/Rh, W/Ag, W/Mo, and W/Sn
Results
Results of the Evaluation
Discussions and Conclusion
References
Generic Infrastructure for Medical Informatics (GIMI): The Development of a Mammographic Training System
Introduction
Background
Overview: GIMI Training System
Mammographic Image Database
Mammographic Auditing
GIMI Middleware
Training Modules
Didactic Materials
Training Database
Results
Conclusions
References
Evaluation of 3D Breast Surface Reconstruction Accuracy Using Non-contact Scanner Images: A Phantom Study
Introduction
Methods
Materials
Reconstruction of the Breast Phantom Surface from Laser Scanner Images
Generation of the Phantom Ground Truth from CT Images
Evaluation of the Reconstructed Phantom Surfaces
Results
Grouping and Selection of Individual Camera-Phantom Positions
Effects of Observer Variation on the Reconstructed Phantom Surfaces
Effects of the Compression Plate on the Reconstructed Surface
Effect of Different Acquisition Angles on the Surface Accuracy
Discussions and Conclusions
References
Investigation of Different PV Distributions in Digital Breast Tomosynthesis (DBT) Mammography
Introduction
Methods
DBT System
Different PV Distributions
Breast Phantom and Figures of Merit
Results
Discussion
References
Characterization of Projection Ordering in Iterative Reconstruction Methods for Breast Tomosynthesis
Introduction
Method
Simulation of Breast Tomosynthesis System
Projection Ordering Schemes
Figures of Merit
Results
Discussion
References
Effect of Scan Angle and Reconstruction Algorithm on Model Observer Performance in Tomosynthesis
Background
Method
Results
Discussion
References
A Novel Approach for Filtered Backprojection in Tomosynthesis Based on Filter Kernels Determined by Iterative Reconstruction Techniques
Introduction
Reconstruction Methods
Results and Discussion
Conclusion
References
3D Digital Breast Tomosynthesis Using Total Variation Regularization
Background
Method
Results
Discussion
References
Image Artifact in Digital Breast Tomosynthesis and Its Dependence on System and Reconstruction Parameters
Introduction
Methods and Materials
Results and Discussion
Conclusion
References
Multi-projection Correlation Imaging as a New Diagnostic Tool for Improved Breast Cancer Detection
Introduction
Materials and Methods
Results
Discussion
Conclusions
References
Sensitivity of Contrast-Enhanced Digital Breast Tomosynthesis to Changes in Iodine Concentration during Acquisition
Introduction
Method
Phantom Imaging
Image Acquisition and Reconstruction
Image Analysis
Results
Discussion
References
Quantifying Breast Thickness for Density Measurement
Introduction
Method
Compression Paddle Tilt
Lateral Deformation of Compression Paddle
Modelling Breast Thickness at the Breast Periphery
Results
Discussion
References
Effect of Tissue Thickness Variation in Volumetric Breast Density Estimation
Introduction
Materials and Methods
Imaging Model and VBD Estimation
Thickness Variation
Effect of Thickness Variation on the VBD
Results and Discussions
Imaging Model
Thickness Variation
Effect of Thickness Variation
Conclusion
References
Breast Abnormality Detection Incorporating Breast Density Information Based on Independent Components Analysis
Introduction
Method
Experimental Results and Discussion
Conclusions
References
Comparison of Breast Percent Density Estimated from Digital Mammograms and Central Reconstructed Tomosynthesis Slice Images
Introduction
Material and Methods
Results
Discussion
References
Digital Breast Tomosynthesis Parenchymal Texture Analysis for Breast Cancer Risk Estimation: A Preliminary Study
Introduction
Methods
Results
Discussion
References
Texture Based Segmentation of Breast DCE-MRI
Introduction
Method
Texture
Hidden Markov Random Field
Multiple Image Segmentation
Results
Discussion
References
Image Quality Measurements in Breast Tomosynthesis
Introduction
Method
Homogeneity in the Reconstructed Images
Evaluation of Artifacts
Detail Detection with a Uniform Background
Results
Homogeneity in the Reconstructed Images
Evaluation of Artifacts
Detail Detection with a Uniform Background
Discussion
References
One Year of Experience with Remote Quality Assurance of Digital Mammography Systems in the Flemish Breast Cancer Screening Program
Background
Methods and Materials
Results
Discussion
References
Automatic Exposure Control in Digital Mammography: Contrast-to-Noise Ratio Versus Average Glandular Dose
Introduction
Materials and Method
Results
Discussion
References
Effect of Anode/Filter Combination on the Dose and Image Quality of a Digital Mammography System Based on an Amorphous Selenium Detector
Introduction
MaterialandMethods
Results
Discussion
Conclusion
References
Comparative Technical Study of Two Generations of CR Plates for Digital Mammography
Background
Material and Methods
Contrast-Detail Analysis
Modulation Transfer Function
Noise Power Spectra
Results
Contrast-Detail Analysis
MTF
Noise Power Spectrum
Discussion
References
Comparing the Performance of Digital Mammography Systems
Background
Method
Results
Discussion
References
Cross-Calibration and Longitudinal Quality Control of Hologic Selenia Full-Field Digital Mammography Systems for Volumetric Breast Density Measurements
Background
Method
SXA Technique
Quality Control (QC)
Cross-Calibration
Longitudinal Monitoring
Results
Discussion
Conclusion
References
Classification of Artifacts in Clinical Digital Mammography
Background
Methods and Materials
Results
Discussion
References
Contrast Sensitivity in Mammographic Softcopy Reading – Determination with Psychophysical Procedures
Background
Method
Target Items: Gabor Patterns and Digits
Hypotheses
Observers and Experimental Setup
Results
Discussion
Hypothesis H1 – Homogeneous vs. Mammographic Image
Hypothesis H2 – Different Mammograms
Hypothesis H3 – Gabor Pattern vs. Digit
Hypothesis H4 – Masking
Semitransparent Masking
Conclusion
References
Author Index
Recommend Papers
- Author / Uploaded
- Krupinski
- Elizabeth A. 编
File loading please wait...
Citation preview
Lecture Notes in Computer Science Commenced Publication in 1973 Founding and Former Series Editors: Gerhard Goos, Juris Hartmanis, and Jan van Leeuwen
Editorial Board David Hutchison Lancaster University, UK Takeo Kanade Carnegie Mellon University, Pittsburgh, PA, USA Josef Kittler University of Surrey, Guildford, UK Jon M. Kleinberg Cornell University, Ithaca, NY, USA Alfred Kobsa University of California, Irvine, CA, USA Friedemann Mattern ETH Zurich, Switzerland John C. Mitchell Stanford University, CA, USA Moni Naor Weizmann Institute of Science, Rehovot, Israel Oscar Nierstrasz University of Bern, Switzerland C. Pandu Rangan Indian Institute of Technology, Madras, India Bernhard Steffen University of Dortmund, Germany Madhu Sudan Massachusetts Institute of Technology, MA, USA Demetri Terzopoulos University of California, Los Angeles, CA, USA Doug Tygar University of California, Berkeley, CA, USA Gerhard Weikum Max-Planck Institute of Computer Science, Saarbruecken, Germany
5116
Elizabeth A. Krupinski (Ed.)
Digital Mammography 9th International Workshop, IWDM 2008 Tucson, AZ, USA, July 20-23, 2008 Proceedings
13
Volume Editor Elizabeth A. Krupinski Department of Radiology Research University of Arizona Tucson, AZ, USA E-mail: [email protected]
Library of Congress Control Number: 2008930118 CR Subject Classification (1998): I.4, I.5-6, H.3, J.3 LNCS Sublibrary: SL 6 – Image Processing, Computer Vision, Pattern Recognition, and Graphics ISSN ISBN-10 ISBN-13
0302-9743 3-540-70537-6 Springer Berlin Heidelberg New York 978-3-540-70537-6 Springer Berlin Heidelberg New York
This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, re-use of illustrations, recitation, broadcasting, reproduction on microfilms or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer. Violations are liable to prosecution under the German Copyright Law. Springer is a part of Springer Science+Business Media springer.com © Springer-Verlag Berlin Heidelberg 2008 Printed in Germany Typesetting: Camera-ready by author, data conversion by Scientific Publishing Services, Chennai, India Printed on acid-free paper SPIN: 12438426 06/3180 543210
Preface
This volume (5116) of Springer’s Lecture Notes in Computer Science contains the proceedings of the 9th International Workshop on Digital Mammography (IWDM) which was held July 20 – 23, 2008 in Tucson, AZ in the USA. The IWDM meetings traditionally bring together a diverse set of researchers (physicists, mathematicians, computer scientists, engineers), clinicians (radiologists, surgeons) and representatives of industry, who are jointly committed to developing technologies to support clinicians in the early detection and subsequent patient management of breast cancer. The IWDM conference series was initiated at a 1993 meeting of the SPIE Medical Imaging Symposium in San Jose, CA, with subsequent meetings hosted every two years at sites around the world. Previous meetings were held in York, England; Chicago, IL USA; Nijmegen, Netherlands; Toronto, Canada; Bremen, Germany; Durham, NC USA and Manchester, UK. The 9th IWDM meeting was attended by a very international group of participants, and during the two and one-half days of scientific sessions there were 70 oral presentations, 34 posters and 3 keynote addresses. The three keynote speakers discussed some of the “hot” topics in breast imaging today. Karen Lindfors spoke on “Dedicated Breast CT: Initial Clinical Experiences.” Elizabeth Rafferty asked the question is “Breast Tomosynthesis: Ready for Prime Time?” Finally, Martin Tornai discussed “3D Multi-Modality Molecular Breast Imaging.” All three talks reflected the very strong influence that imaging modalities capable of providing 3D rather than 2D information, as with traditional mammography, are having on breast imaging today. Although these keynote addresses and some of the scientific presentations hinted at the clinical promise of these new technologies, it is still unknown whether there is going to be a true impact on earlier detection and hence treatment of breast cancer. Perhaps the answer will be more clearly provided at the 10th IWDM meeting! Full-field digital mammography (FFDM) has been an important topic at earlier IWDM meetings and it was still the focus of a number of talks this year. Perhaps this is due to the fact that although it has been around for a few years now, it is still being deployed rather slowly in many parts of the world, possibly due to cost-related issues. Even in the USA the conversion from film to digital is far less than complete and this may be a reflection of cost as it relates to declining reimbursement rates for mammography and fewer insurance companies paying for regular exams in younger women and those with no family history of breast cancer. Clearly this is a problem not only for mammography but also for women in general. Hopefully scientific evidence, such as that presented at the IWDM meeting, will prevail and women will continue to benefit from the advances in technology being made by dedicated breast cancer researchers and clinicians. The 2008 IWDM program reflected many of the current trends, advances and efforts being made to further improve digital mammography for the early detection of breast cancer and improved management. As in previous years, a number of papers dealt with the challenges of developing tools to analyze breast density and texture in order to better predict a woman’s breast cancer risk. Included this year, however, were studies that
VI
Preface
now employed volumetric assessments using FFDM images and digital tomosynthesis images. In general, there were significantly more papers at this meeting on volumetric imaging, digital breast tomosynthesis and breast CT. Although most presentations dealt with the more technical aspects of these volumetric imaging modalities, there were a few more clinically based studies that hinted at the significant potential in diagnostic accuracy that may be gained. These new techniques will certainly raise even more questions as they continue to develop, including their impact on the clinical reading environment and reading efficiency. Will the potential diagnostic benefits of volumetric and multiple-slice imaging data be outweighed by the increased time it takes to interpret the images? We don’t know the answer today, but as these images become integrated into the clinical routine on a more frequent basis it may be necessary to find the answer and concentrate research efforts on optimizing presentation modes. As in previous years computer-aided image analysis techniques were discussed in great detail, but there were some new and interesting trends. Although computeraided detection was still a focus, many groups have moved away from FFDM images and traditional mammography to ultrasound, MRI, CT and tomosynthesis. The focus seems to be more on an integrated approach to computer-aided decision tools that combine information from different modalities in order to improve not only lesion detection but also lesion discrimination. Temporal comparisons and registration of multi-modality images were also discussed as ways to improve computer-aided decision tools, with some very promising results. In general, it seems that breast imaging is on the cusp of some very significant changes in the ways that images are acquired, analyzed and integrated with other types of patient information. More studies need to be done to fully evaluate how these new technologies and analysis tools actually impact both diagnostic accuracy and diagnostic efficiency. On a broader level, it may also be necessary to conduct more cost–benefit analyses in order to convince regulatory and reimbursing agencies to approve and pay for these amazing advances in imaging and patient care. The benefit to society often seems clear to those working so closely in the development and evaluation of new technologies, but convincing the rest of society that the benefits are real seems to take longer. As with any scientific meeting, many people put in many long hours prior to the meeting to make it look effortless, and such was the case with the 9th IWDM meeting. Following the precedent set by the organizers of the 8th IWDM, presenters were required to submit a 4-page abstract for consideration by the Scientific Committee. The abstracts were reviewed by at least two members of the Scientific Committee and feedback was provided to the submitters. The rejection rate was about 20% this year, reflecting the high quality of abstracts that were accepted after diligent review by the Scientific Committee. The 104 final 8-page papers included in these proceedings represent the work of some of the finest and most dedicated researchers in breast imaging today. Many thanks and deep appreciation go to the Scientific Committee for the time taken from their busy schedules to review the abstracts and provide feedback to the authors for the final papers. The 9th IWDM also had the generous support from its industrial partners who both exhibited at the meeting and provided sponsorship of various conference events. Their participation in both the exhibit hall and the scientific meeting added considerably to the quality and success of the meeting. Many thanks go to June Stavem, who worked
Preface
VII
many hours to recruit the industrial partners and help them throughout the meeting. Many thanks are also extended to Michel Rogulski, who provided significant technical advice throughout the planning phases of the meeting as well as on-site support.
July 2008
Elizabeth Krupinski
Scientific Meeting Preliminary Program (Subject to Change)
Monday July 21, 2008 8:00 Welcome: Elizabeth Krupinski 8:15 Keynote Address: Karen Lindfors, MD "Dedicated Breast CT: Initial Clinical"
Session 1: Breast Density, Texture & Risk I Chair: Michael Brady z z z z z
8:45 Serghei Malkov "Improvements to Single Energy Absorptiometry Method for Digital Mammography to Quantify Breast Tissue Density" 9:00 Arnau Oliver Breast Density Segmentation: a Comparison Between Clustering and Region Based Techniques" 9:15 Wenda He "Mammographic Segmentation Based on Texture Modeling of Tabar Mammographic Building Blocks" 9:30 Michael Barnathan "Analyzing Tree-Like Structures in Biomedical Images Based on Texture and Branching: an Application to Breast Imaging" 9:45 Keith Hartman "Volumetric Assessment of Breast Tissue Composition from FFDM Images"
10:00 Coffee Break with Exhibitors & Poster Viewing
Session 2: Clinical Experiences Chair: Etta Pisano z z
z z z z z
10:30 Sue Astley "Effect of Image Quality on Recall Rates 10:45 Takako Morita "A Comparison Between Film-Screen Mammography and Full-Field Digital Mammography Utilizing Phase Contrast Technology in Breast Cancer Screening Programs" 11:00 Niamh Hambly "Impact of Digital Mammography in Breast Cancer Screening: Initial Experience in a National Breast Screening Program" 11:15 Andrew Smith "Clinical Performance of Breast Tomosynthesis as a Function of radiologist Experience Level" 11:30 Anders Tingberg "BIRADS Classification in Breast Tomosynthesis Compared to Mammography and Ultrasonography" 11:45 David Getty "Stereoscopic Digital Mammography: Improved Accuracy of Lesion Detection in Breast Cancer Screening" 12:00 Abass Alavi "Potential Role of FDG-PET Imaging in Defining Biology of Primary Breast Lesions"
X
Scientific Meeting
12:15 Lunch Break
Session 3: Breast Imaging Physics Chair: Martin Yaffe z z z z z
z z
1:15 Katsuhiro Ichikawa "Clinical Usefulness of Super High-Resolution Liquid Crystal Displays Using Independent Sub-Pixel Driving Technology" 1:30 Baorui Ren "The Effect of Tomosynthesis X-Ray Pulse Width on Measured Beam Quality" 1:45 Mark Williams "Tomographic Dual Modality Breast Scanner" 2:00 Aurelie Laidevant "Dual-Energy X-Ray Absorptiometry Method Using a Full Field Digital Mammography System" 2:15 Ann-Katherine Carton "Optimization of a Dual-Energy ContrastEnhanced Technique for a Photon Counting Digital Breast Tomosynthesis System" 2:30 Shih-Ying Huang "Simulation and Phantom Studies of Contrast-Enhanced Dual Energy Mammography (CEDEM)" 2:45 Satoru Matsuo "Evaluation of a Phase Contrast Imaging with Digital Mammography"
3:00 Coffee Break with Exhibitors & Poster Presentation Session #1 Chair: Hiroshi Fujita z z z z z z z z z
z z
z
Sue Astley "Prompting in Mammography: Reproducibility" Michael Berks "Synthesising Abnormal Structures in Mammograms Using Pyramid Decomposition" Sylvain Bernard "Computer-Aided Microcalcification Detection on Digital Breast Tomosynthesis data: a preliminary Evaluation" Murk Bottema "Temporal Analysis of Mammograms Based on Graph Matching: Ann-Katherine Carton "Temporal Subtraction versus Dual-Energy ContrastEnhanced Digital Breast Tomosynthesis: a Pilot Study" Yang Gong "Texture-Based Simultaneous Registration and Segmentation of Breast DCE-MRI" Eduardo Guibelalde "A CDMAM Image Phantom Software Improvement for Human Observer Assessment" Yuji Ikedo "Computerized Classification of Whole Breast Ultrasound Images Based on Mammary Gland Patterns" Tetsuro Kusunoki "Comparison Between Physical Image Quality as Measured by a Newly Developed Phantom Dedicated for Digital Mammography QC and that by European Guidelines Methods" Anthony Maeder "Assuring Authenticity of Digital Mammograms by Image Watermarking" Claudia Mello-Thoms "Different Search Patterns and Similar Decision Outcomes: How Can Experts Agree in the Decisions They Make When Reading Digital Mammograms?" Thomas Mertelmeier "Optimization of Tomosynthesis Acquisition Parameters: Angular Range and Number of Projections"
Scientific Meeting z
z
z z z
XI
Takako Morita "Subtle Abnormalities in Highly Dense Breasts Detected by use of a Digital Phase Contrast Mammography System: a Report of three Invasive Cancer Cases in the Early Stage" Nadia Oberhofer "Image Quality Assessment and Equipment Optimisation with Automated Phantom Evaluation in Full Field Digital Mammography (FFDM)" Benjamin Pollard "Effect of Increased Ambient Lighting on Mass Detection in Mammograms" Hans Roehrig Reducing Noise of Medical Grade Liquid Crystal Displays (LCD) and its Relation to the Detection of Micro-Calcifications Shanghua Sun "An Ontology to Support Adaptive Training for Breast Radiologists"
Session 4: Image Analysis and CAD I Chair: Maryellen Giger z z z z z
z z z z
4:00 Dave Tahmoush "A Web Database for Computer-Aided Detection and Diagnosis of Medical Images" 4:15 Nico Karssemeijer "An Interactive Computer Aided Decision Support System for Detection of Masses in Mammograms" 4:30 Heang-Ping Chan "Detection of Masses in Digital Tomosynthesis Mammography: Effects of the Number of Projection Views and Dose" 4:45 Swatee Singh "Effect of Similarity Metrics and ROI Sizes in Featureless Computer Aided Detection of Breast Masses in Tomosynthesis" 5:00 Georgia Tourassi "Knowledge Transfer Across Breast Cancer Screening Modalities: a pilot Study Using an Information-Theoretic CADe System for Mass Detection" 5:15 Fengmei Zou "Gradient Vector Flow Fields and Spiculated Mass Detection in Digital Mammography Images" 5:30 Da Qi "The Evaluation of Effects on Breast Cancer Diagnosis Using the Mammographic Semantic Information" 5:45 Shinsuke Katsuhara "Computerized SCheme for Focal Asymmetric Densities on Mammograms by use of Geometric and Texture Analysis" 6:00 Yading Yuan "Identifying Corresponding Lesions from CC and MLO Views via Correlative Feature Analysis"
7:00 Dinner Tuesday July 22, 2008
Session 5: Image Analysis and CAD II Chair: Nico Karssemeijer z z z
8:00 Piotr Habas "Case-Specific Reliability Assessment for Improved False Positive Reduction with an Information-Theoretic CAD System" 8:15 Lubomir Hadjiiski "Computerized Detection and Classification of Malignant and Benign Microcalcifications on Full Field Digital Mammograms" 8:30 Michiel Kallenberg "The Effect of Training Sample Size on Performance of Mass Detection"
XII
Scientific Meeting z z
z z z
8:45 Balaji Krishnapuram "Multiple-Instance Learning Improves CAD Detection of Masses in Digital Mammography" 9:00 Isaac Leichter "Optimizing the CAD Process for Detecting Mammographic Lesions by a New Generation Algorithm Based on linear Classifiers and a Gradient Based Method" 9:15 Maciej Mazurowski "Reliability Assessment of Ensemble Classifiers: Application in Mammography" 9:30 Berkman Sahiner "Breast Mass Classification on Full-Field Digital Mammography and Screen-Film Mammography" 9:45 Guido Van Schie "Detection of Microcalcifications Using a Nonuniform Noise Model"
10:00 Coffee Break with Exhibitors & Poster Viewing
Session 6: Modeling and Simulation Chair: Ulrich Bick z z z z z z
10:30 Vijay Rajagopal "The Biomechanical Breast Reference State for MultiModal Image Analysis" 10:45 Craig Abbey "An Ideal Observer for a Model of X-Ray Imaging in Breast Parenchymal Tissue" 11:00 Michael Berks "Statistical Appearance Models of Mammographic Masses" 11:15 Mary Yip "Validation of a Digital Mammography Image Simulation Chain with Automated Scoring of CDMAM Images" 11:30 Michael O'Connor "Comparison of Two Methods to Develop Breast Models for Simulation of Breast Tomosynthesis and CT" 11:45 Christine Tanner "Statistical Deformation Models of Breast Compressions from Biomechanical Simulations"
12:00 Lunch Break
Session 7: Image Analysis and CAD III Chair: Sue Astley z z z z z
1:15 Gobert Lee "Classification of Benign and Malignant Masses in Ultrasound Breast Image Based on Geometric and Echo Features" 1:30 Gina Clarke "Validation of Tumor Burden Measurements Using ThreeDimensional Histopathology" 1:45 Akira Hasegawa "A Tool for Temporal Comparison of Mammograms: Image Toggling and Dense-Tissue-Preserving Registration" 2:00 Takeshi Hara "Development of Whole Breast Ultrasound Viewer and Automated Mass Detection System" 2:15 Mitchell Goodsitt "Automated Registration of Volumes of Interest for a Combined X-Ray and Ultrasound Breast Imaging System"
Scientific Meeting
XIII
2:30 Keynote Address: Betty Rafferty, MD "Breast Tomosynthesis: Ready for Prime Time?" 3:00 Coffee Break with Exhibitors & Poster Presentation Session #2 Chair: Elizabeth Krupinski z z z z
z z
z z z z z z z z z z z
Jennifer Diffey "Estimating Individual Cancer Risk in the UK Screening Programme: a Feasibility study" Gisella Gennaro "Clinical Performance of Digital Breast Tomosynthesis versus Full-Field Digital Mammography: Preliminary Results: Edward Hadley "Analysis of Using Anatomical Linear Structure Information in Mammographic Risk Assessment" Despina Kontos "Evaluating the Effect of Tomosynthesis Acquisition Parameters on Image Texture: a Study Based on an Anthropomorphic Breast Tissue Software Model" Yoshifumi Kuroki "Computer Aided Detection (CAD) for Digital Mammography: What Kind of reader Group Does CAD Have an Effect in?" Isaac Leichter "Does a Mammography CAD Algorithm with Varying Filtering Levels of Detection Marks, Used to Reduce the False Mark Rate, Adversely Affect the Detection of Small Masses?" Hui Li "Performance of CADx on a Large Clinical Database of FFDM Images" Ana Maria Lopez "Expedited Breast Care: a New Model in Breast Health" Jennifer Oduko "Effect of Tungsten-Anode X-Ray Tubes on Dose and Image Quality in Full-Field Digital Mammography" Christina Shafer "Assessment of Low Energies and Slice Depth in the Quantification of Breast Tomosynthesis" Joerg Teubl "Comparison of Multiple View Strategies to Reduce False Positives in Breast Imaging" Meritxell Tortajada "Image Correction and Reconstruction for Breast Biopsy" Chris Tromans "Progress Toward a Quantitative Scale for Describing Radiodensity in Mammographic Images" Dominique Van de Sompel "Systematic Performance Analysis of SART as Applied to Digital Breast Tomosynthesis" Mari Varjonen "Optimizing the target-Filter Combination in Digital Mammography in the Sense of Image Quality and Average Glandular Dose" Moi Hoon Yap "Generic Infrastructure for Medical Informatics (GIMI): the Development of a Mammographic Training System" Cuiping Zhang "Evaluation of 3D Breast Reconstruction Accuracy Using Non-Contact Scanner Images: a Phantom Study"
Digital Breast Tomosynthesis Investigation of Different PV Distributions in Digital Tomosynthesis Mammography (DTM) Y Zhang, HP Chan, MM Goodsitt, A Schmitz, JW Eberhard, BEH Claus
XIV
Scientific Meeting
Characterization of Projection Ordering in Iterative Reconstruction for Breast Tomosynthesis G Wu, J Mainprize, M Yaffe Effect of Scan Angle and reconstruction Algorithm on Model Observer Performance in Tomosynthesis I Reiser, BA Lau, RM Nishikawa A Novel Approach for Filtered Backprojection in Tomosynthesis Based on Filter Kernels Determined by Iterative Reconstruction Techniques J Ludwig, T Mertelmeier, H Kunze, W Härer 3D Digital Breast Tomosynthesis Using Total Variation Regularization I Kastanis, S Arridge, A Stewart, S Gunn, C Ullberg, T Francke Image Artifact in Digital Breast Tomosynthesis and its Dependence on System and Reconstruction Parameters YH Hu, W Zhao, T Mertelmeier, J Ludwig Multi-Projection Correlation Imaging as a New Diagnostic Tool for Improved Breast Cancer Detection AS Chawla, E Samei, JY Lo, T Mertelmeier Sensitivity of Contrast-Enhanced Digital Breast Tomosynthesis to Changes in Iodine Concentration During Acquisition ML Hill, JG Mainprize, MJ Yaffe
Breast Density, Texture & Risk II Quantifying Breast Thickness for Density Measurement J Diffey, A Hufton, C Beeston, J Smith, T Marchant, S Astley Effect of Tissue Thickness Variation in Volumetric Breast Density Estimation Alonzo-Proulx, Ah Tyson, J Mainprize, Ge Mawdsley, Mj Yaffe Breast Abnormality Detection Incorporating Breast Density Information Based on Independent Components Analysis S Petroudi, N Nicolaou, J Georgiou, M Brady Comparison of Breast Percent Density Estimated from Digital Mammograms and Central Reconstructed Tomosynthesis Slice Images PR Bakic, D Kontos, AB Troxel, A Maidment Digital Breast Tomosynthesis Parenchymal Texture Analysis for Breast Cancer Risk Estimation: a Preliminary study D Kontos, PR Bakic, AB Troxel, EF Conant, ADA Maidment Texture Based Segmentation of Breast DCE-MRI YC Gong, M Brady
Scientific Meeting
XV
Physics, Image Quality & Quality Assurance Image Quality Measurements in Breast Tomosynthesis R van Engen, R Bouwman, R van der Burght, B Lazzari, DR Dance, P Heid, M Aslund, KC Young One Year of Experience with remote Quality Assurance of Digital Mammography System in the Flemish Breast Cancer Screening Program J Jacobs, K Lemmens, J Nens, K Michielsen, G Marchal, H Bosmans Automatic Exposure Control in Digital Mammography: Contrast-to-Noise Ratio versus Average Glandular Dose G Gennaro, P Golinelli, E Bellan, P Colombo, L D'Ercole, A Di Nallo, L Gallo, C Giordano,G Meliadò, B Morri, E Nassivera, N Oberhofer, D Origgi, M Paolucci, N Paruccini, M Piergentili, E Rizzi, R Rossi Effect of Anode Target/Filter Combination on the Dose and Image Quality of a Digital Mammography System Based on an Amorphous Selenium Detector P Baldelli, N Phelan, G Egan Comparative Technical Study of Two Generations of CR Plates for Digital Mammography H Bosmans, K Lemmens, J Jacobs, B Verbrugge, K Michielsen, F Zanca, J Nens, C Van Ongeval, G Marchal Comparing the Performance of Digital Mammography Systems KC Young, JM Oduko, O Gundogdu, A Alsager Cross-Calibration of Hologic Selenia Full-Field Digital Mammography Systems for Volumetric Breast Density Measurements J Wang, S Malkov, B Fan, J Shepherd Classification of artifacts in Clinical Digital Mammography C Van Ongeval, J Jacobs, A Van Steen, F Zanca, H Bosmans, G Marchal Contrast Sensitivity in Mammographic Softcopy Reading - Determination with Psychophysical Procedures D Apelt, HO Peitgen
Table of Contents
Breast Density, Texture and Risk I Improvements to Single Energy Absorptiometry Method for Digital Mammography to Quantify Breast Tissue Density . . . . . . . . . . . . . . . . . . . . Serghei Malkov, Jeff Wang, and John Shepherd Breast Density Segmentation: A Comparison of Clustering and Region Based Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Torrent, A. Bardera, A. Oliver, J. Freixenet, I. Boada, M. Feixes, R. Mart´ı, X. Llad´ o, J. Pont, E. P´erez, S. Pedraza, and J. Mart´ı Mammographic Segmentation Based on Texture Modelling of Tab´ ar Mammographic Building Blocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wenda He, Izzati Muhimmah, Erika R.E. Denton, and Reyer Zwiggelaar Analyzing Tree-Like Structures in Biomedical Images Based on Texture and Branching: An Application to Breast Imaging . . . . . . . . . . . . . . . . . . . . Michael Barnathan, Jingjing Zhang, Despina Kontos, Predrag Bakic, Andrew Maidment, and Vasileios Megalooikonomou Volumetric Assessment of Breast Tissue Composition from FFDM Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Keith Hartman, Ralph Highnam, Ruth Warren, and Valerie Jackson
1
9
17
25
33
Clinical Experiences Effect of Image Quality on Film Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . Susan Astley, Neil Prasad, Eve Allcock, Jenny Diffey, Yit Yoong Lim, and Caroline Boggis
40
A Comparison between Film-Screen Mammography and Full-Field Digital Mammography Utilizing Phase Contrast Technology in Breast Cancer Screening Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Takako Morita, Maya Yamada, Akiko Kano, Sumiya Nagatsuka, Chika Honda, and Tokiko Endo
48
Impact of Digital Mammography in Breast Cancer Screening: Initial Experience in a National Breast Screening Program . . . . . . . . . . . . . . . . . . N. Hambly, N. Phelan, G. Hargaden, A. O’Doherty, and F. Flanagan
55
XVIII
Table of Contents
Clinical Performance of Breast Tomosynthesis as a Function of Radiologist Experience Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Andrew P. Smith, Elizabeth A. Rafferty, and Loren Niklason BIRADS Classification in Breast Tomosynthesis Compared to Mammography and Ultrasonography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Anders Tingberg, Ingvar Andersson, Debra M. Ikeda, Mark Ruschin, Tony Svahn, and Pontus Timberg
61
67
Stereoscopic Digital Mammography: Improved Accuracy of Lesion Detection in Breast Cancer Screening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . David J. Getty, Carl J. D’Orsi, and Ronald M. Pickett
74
Potential Role of FDG-PET Imaging in Defining Biology of Primary Breast Lesions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . S. Basu, A. Mavi, T. Cermik, R. Kumar, and A. Alavi
80
Breast Imaging Physics Clinical Usefulness of Super High-Resolution Liquid Crystal Displays Using Independent Sub-pixel Driving Technology . . . . . . . . . . . . . . . . . . . . Katsuhiro Ichikawa, Hiroko Kawashima, Naohiro Kimura, and Mikio Hasegawa The Effect of Tomosynthesis X-Ray Pulse Width on Measured Beam Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Baorui Ren, Andrew Smith, David Aizer, Don Kennedy, Jeffrey Yorker, and Zhenxue Jing Tomographic Dual Modality Breast Scanner . . . . . . . . . . . . . . . . . . . . . . . . . Mark B. Williams, Patricia G. Judy, Mitali J. More, Jennifer A. Harvey, Stan Majewski, James Proffitt, John McKisson, Alexander Stolin, Brian Kross, Alexander Stewart, Edward Bullard, Manish Kankaria, and Roman Janer Dual-Energy X-Ray Absorptiometry Method Using a Full Field Digital Mammography System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Aurelie Laidevant, Serghei Malkov, Alfred Au, and John Shepherd Optimization of a Dual-Energy Contrast-Enhanced Technique for a Photon Counting Digital Breast Tomosynthesis System . . . . . . . . . . . . . . . Ann-Katherine Carton, Christer Ullberg, Karin Lindman, Tom Francke, and Andrew Maidment Simulation and Phantom Studies of Contrast-Enhanced Dual Energy Mammography (CEDEM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shih-Ying Huang, John M. Boone, Dandan Zheng, Kai Yang, Nathan J. Packard, and George Burkett Jr.
84
91
99
108
116
124
Table of Contents
Preliminary Evaluation of a Phase Contrast Imaging with Digital Mammography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Satoru Matsuo, Hiroshi Fujita, Junji Morishita, Tetsuro Katafuchi, Chika Honda, and Junko Sugiyama
XIX
130
Poster Session I Prompting in Mammography: Reproducibility . . . . . . . . . . . . . . . . . . . . . . . Stephanie Tiew, Susan Astley, Bernice Dillon, Julie Morris, and Caroline Boggis
137
Synthesising Abnormal Structures in Mammograms Using Pyramid Decomposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Michael Berks, Chris Rose, Caroline Boggis, and Susan Astley
143
Computer-Aided Microcalcification Detection on Digital Breast Tomosynthesis Data: A Preliminary Evaluation . . . . . . . . . . . . . . . . . . . . . . Sylvain Bernard, Serge Muller, and Jon Onativia
151
Temporal Analysis of Mammograms Based on Graph Matching . . . . . . . . Fei Ma, Mariusz Bajger, and Murk J. Bottema Temporal Subtraction Versus Dual-Energy Contrast-Enhanced Digital Breast Tomosynthesis: A Pilot Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ann-Katherine Carton, Jean Anne Currivan, Emily Conant, and Andrew Maidment
158
166
Texture-Based Simultaneous Registration and Segmentation of Breast DCE-MRI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yang Can Gong and Michael Brady
174
A CDMAM Image Phantom Software Improvement for Human Observer Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gabriel Prieto, Margarita Chevalier, and Eduardo Guibelalde
181
Computerized Classification of Mammary Gland Patterns in Whole Breast Ultrasound Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yuji Ikedo, Takako Morita, Daisuke Fukuoka, Takeshi Hara, Hiroshi Fujita, Etsuo Takada, and Tokiko Endo Comparison between Physical Image Quality as Measured by a Newly Developed Phantom Dedicated for Digital Mammography QC and That by European Guidelines Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tetsuro Kusunoki, Masahiro Tagi, Kenji Yosikawa, Hiroshi Arai, Nachiko Uchiyama, Chieko Nagashima, and Hiroyuki Kobayashi
188
196
XX
Table of Contents
Assuring Authenticity of Digital Mammograms by Image Watermarking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Anthony Maeder, Jason Dowling, Anthony Nguyen, Emma Brunton, and Phuong Nguyen Different Search Patterns and Similar Decision Outcomes: How Can Experts Agree in the Decisions They Make When Reading Digital Mammograms? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Claudia Mello-Thoms, Marie Ganott, Jules Sumkin, Christiane Hakim, Cynthia Britton, Luisa Wallace, and Lara Hardesty
204
212
Optimization of Tomosynthesis Acquisition Parameters: Angular Range and Number of Projections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Thomas Mertelmeier, Jasmina Ludwig, Bo Zhao, and Wei Zhao
220
Subtle Abnormalities in Highly Dense Breasts Detected by Use of a Digital Phase Contrast Mammography System: A Report of Three Invasive Cancer Cases in the Early Stage . . . . . . . . . . . . . . . . . . . . . . . . . . . . Takako Morita, Maya Yamada, Akiko Kano, Sumiya Nagatsuka, Chika Honda, and Tokiko Endo
228
Image Quality Assessment and Equipment Optimisation with Automated Phantom Evaluation in Full Field Digital Mammography (FFDM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nadia Oberhofer, Nicoletta Paruccini, and Ehrenfried Moroder Breast Mass Detection under Increased Ambient Lighting . . . . . . . . . . . . . Benjamin J. Pollard, Amarpreet S. Chawla, Noriyuki Hashimoto, and Ehsan Samei Reducing Noise of Medical Grade Liquid Crystal Displays (LCD) and Its Relation to the Detection of Micro-calcifications . . . . . . . . . . . . . . . . . . Hans Roehrig, Jiahua Fan, William J. Dallas, Elizabeth A. Krupinski, and Jeffrey Johnson An Ontology to Support Adaptive Training for Breast Radiologists . . . . . Shanghua Sun, Paul Taylor, Louise Wilkinson, and Lisanne Khoo
235 243
249
257
Image Analysis and CAD I A Web Database for Computer-Aided Detection and Diagnosis of Medical Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Dave Tahmoush and Hanan Samet An Interactive Computer Aided Decision Support System for Detection of Masses in Mammograms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nico Karssemeijer, Andrea Hupse, Maurice Samulski, Michiel Kallenberg, Carla Boetes, and Gerard den Heeten
265
273
Table of Contents
Detection of Masses in Digital Breast Tomosynthesis Mammography: Effects of the Number of Projection Views and Dose . . . . . . . . . . . . . . . . . . Heang-Ping Chan, Jun Wei, Yiheng Zhang, Berkman Sahiner, Lubomir Hadjiiski, and Mark A. Helvie Effect of Similarity Metrics and ROI Sizes in Featureless Computer Aided Detection of Breast Masses in Tomosynthesis . . . . . . . . . . . . . . . . . . Swatee Singh, Georgia D. Tourassi, and Joseph Y. Lo Knowledge Transfer across Breast Cancer Screening Modalities: A Pilot Study Using an Information-Theoretic CADe System for Mass Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Georgia D. Tourassi, Amy C. Sharma, Swatee Singh, Robert S. Saunders, Joseph Y. Lo, Ehsan Samei, and Brian P. Harrawood Gradient Vector Flow Fields and Spiculated Mass Detection in Digital Mammography Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fengmei Zou, Yufeng Zheng, Zhengdong Zhou, and Kwabena Agyepong The Evaluation of Effects on Breast Cancer Diagnoses When Using Mammographic Semantic Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Da Qi, Erika R.E. Denton, Joanna M.E. Leason, Diaa Othman, and Reyer Zwiggelaar Computerized Scheme for Focal Asymmetric Densities on Mammograms by Use of Geometric and Texture Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . Shinsuke Katsuhara, Hitoshi Futamura, Satoshi Kasai, Takako Morita, and Tokiko Endo Identifying Corresponding Lesions from CC and MLO Views Via Correlative Feature Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yading Yuan, Maryellen Giger, Hui Li, Li Lan, and Charlene Sennett
XXI
279
286
292
299
307
315
323
Image Analysis and CAD II Case-Specific Reliability Assessment for Improved False Positive Reduction with an Information-Theoretic CAD System . . . . . . . . . . . . . . . Piotr A. Habas, Jacek M. Zurada, and Georgia D. Tourassi Computerized Detection and Classification of Malignant and Benign Microcalcifications on Full Field Digital Mammograms . . . . . . . . . . . . . . . . Lubomir Hadjiiski, Peter Filev, Heang-Ping Chan, Jun Ge, Berkman Sahiner, Mark A. Helvie, and Marilyn A. Roubidoux The Effect of Training Sample Size on Performance of Mass Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Michiel Kallenberg and Nico Karssemeijer
329
336
343
XXII
Table of Contents
Multiple-Instance Learning Improves CAD Detection of Masses in Digital Mammography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Balaji Krishnapuram, Jonathan Stoeckel, Vikas Raykar, Bharat Rao, Philippe Bamberger, Eli Ratner, Nicolas Merlet, Inna Stainvas, Menahem Abramov, and Alexandra Manevitch
350
Optimizing the CAD Process for Detecting Mammographic Lesions by a New Generation Algorithm Using Linear Classifiers and a Gradient Based Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Philippe Bamberger, Isaac Leichter, Nicolas Merlet, Eli Ratner, Glenn Fung, and Richard Lederman
358
Reliability Assessment of Ensemble Classifiers: Application in Mammography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Maciej A. Mazurowski, Jacek M. Zurada, and Georgia D. Tourassi
366
Breast Mass Classification on Full-Field Digital Mammography and Screen-Film Mammography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jiazheng Shi, Berkman Sahiner, Heang-Ping Chan, Lubomir M. Hadjiiski, Jun Ge, and Jun Wei Detection of Microcalcifications Using a Nonuniform Noise Model . . . . . . Guido van Schie and Nico Karssemeijer
371
378
Modeling and Simulation The Breast Biomechanics Reference State for Multi-modal Image Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Vijay Rajagopal, Martyn P. Nash, Ralph P. Highnam, and Poul M.F. Nielsen An Ideal Observer for a Model of X-Ray Imaging in Breast Parenchymal Tissue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Craig K. Abbey and John M. Boone Statistical Appearance Models of Mammographic Masses . . . . . . . . . . . . . Michael Berks, Steven Caulkin, Rumana Rahim, Caroline Boggis, and Susan Astley Validation of a Digital Mammography Image Simulation Chain with Automated Scoring of CDMAM Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mary Yip, Abdulaziz Alsager, Emma Lewis, Kevin Wells, and Kenneth C. Young Comparison of Two Methods to Develop Breast Models for Simulation of Breast Tomosynthesis and CT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . J. Michael O’Connor, Mini Das, Clay Didier, Mufeed Mah’D, and Stephen J. Glick
385
393 401
409
417
Table of Contents
Statistical Deformation Models of Breast Compressions from Biomechanical Simulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Tanner, J.H. Hipwell, and D.J. Hawkes
XXIII
426
Image Analysis and CAD III Classification of Benign and Malignant Masses in Ultrasound Breast Image Based on Geometric and Echo Features . . . . . . . . . . . . . . . . . . . . . . . Gobert N. Lee, Daisuke Fukuoka, Yuji Ikedo, Takeshi Hara, Hiroshi Fujita, Etsuo Takada, Tokiko Endo, and Takako Morita Validation of Cone-Beam CT Measurements of Tumour Burden Using Three-Dimensional Histopathology: Initial Results for a Lumpectomy . . . G.M. Clarke, J.G. Mainprize, C. Peressotti, L. Sun, D. Wang, J.T. Zubovits, C.M.B. Holloway, G.J. Czarnota, and M.J. Yaffe A Tool for Temporal Comparison of Mammograms: Image Toggling and Dense-Tissue-Preserving Registration . . . . . . . . . . . . . . . . . . . . . . . . . . . Akira Hasegawa, Huzefa Neemuchwala, Hiroko Tsunoda-Shimizu, Satoru Honda, Kazuo Shimura, Minoru Sato, Tomomi Koyama, Mari Kikuchi, and Sonoe Hiramatsu Development of Whole Breast Ultrasound Viewer and Automated Mass Detection System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Takeshi Hara, Daisuke Fukuoka, Yuji Ikedo, Etsuo Takada, Hiroshi Fujita, Tokiko Endo, and Takako Morita Automated Registration of Volumes of Interest for a Combined X-Ray Tomosynthesis and Ultrasound Breast Imaging System . . . . . . . . . . . . . . . Mitchell M. Goodsitt, Heang-Ping Chan, Lubomir Hadjiiski, Gerald L. LeCarpentier, and Paul L. Carson
433
440
447
455
463
Poster Session II Estimating Individual Cancer Risks in the UK National Breast Screening Programme: A Feasibility Study . . . . . . . . . . . . . . . . . . . . . . . . . . Jennifer Diffey, Alan Hufton, Susan Astley, Claire Mercer, and Anthony Maxwell Clinical Performance of Digital Breast Tomosynthesis Versus Full-Field Digital Mammography: Preliminary Results . . . . . . . . . . . . . . . . . . . . . . . . . Gisella Gennaro, Enrica Baldan, Elisabetta Bezzon, Manuela La Grassa, Luigi Pescarini, and Cosimo di Maggio Analysis of Anatomical Linear Structure Information in Mammographic Risk Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Edward M. Hadley, Erika R.E. Denton, Josep Pont, Elsa P´erez, and Reyer Zwiggelaar
469
477
483
XXIV
Table of Contents
Evaluating the Effect of Tomosynthesis Acquisition Parameters on Image Texture: A Study Based on an Anthropomorphic Breast Tissue Software Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Despina Kontos, Cuiping Zhang, Nicole Ruiter, Predrag R. Bakic, and Andrew D.A. Maidment
491
Computer Aided Detection (CAD) for Digital Mammography: A Retrospective Reading Study for Consideration on Utilizing CAD Most Effectively . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yoshifumi Kuroki, Shigeru Nawano, Seiko Suzuki, Hideya Takeo, and Shigeru Saotome
499
Does a Mammography CAD Algorithm with Varying Filtering Levels of Detection Marks, Used to Reduce the False Mark Rate, Adversely Affect the Detection of Small Masses? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Isaac Leichter, Richard Lederman, Eli Ratner, Nicolas Merlet, Glenn Fung, Balaji Krishnapuram, and Philippe Bamberger
504
Performance of CADx on a Large Clinical Database of FFDM Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hui Li, Maryellen L. Giger, Yading Yuan, Li Lan, and Charlene A. Sennett Expedited Breast Care: A New Model in Breast Health . . . . . . . . . . . . . . . A.M. Lopez, G. Barker, A. Bhattacharryya, K. Scott, M. Descour, L. Richter, J. Davenport, S. Lazarus, L. Kreykes, A. Valencia, and R. Weinstein Effect of Using Tungsten-Anode X-Ray Tubes on Dose and Image Quality in Full-Field Digital Mammography . . . . . . . . . . . . . . . . . . . . . . . . . Jennifer M. Oduko, Kenneth C. Young, Ozcan Gundogdu, and Abdulaziz Alsager Assessment of Low Energies and Slice Depth in the Quantification of Breast Tomosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Christina M. Shafer, Ehsan Samei, Thomas Mertelmeier, Robert S. Saunders, Moustafa Zerhouni, and Joseph Y. Lo Comparison of Multiple View Strategies to Reduce False Positives in Breast Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Joerg Teubl and Horst Bischof Image Correction and Reconstruction for Breast Biopsy . . . . . . . . . . . . . . . Meritxell Tortajada, Robert Mart´ı, Jordi Freixenet, Josep Fern´ andez, and Melcior Sent´ıs Progress Toward a Quantitative Scale for Describing Radiodensity in Mammographic Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Christopher Tromans, Sir Michael Brady, Dominique Van de Sompel, Michele Lorenzon, Massimo Bazzocchi, and Chiara Zuiani
510
515
522
530
537 545
553
Table of Contents
XXV
Systematic Performance Analysis of SART as Applied to Digital Breast Tomosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Dominique Van de Sompel and Michael Brady
561
Optimizing the Target-Filter Combination in Digital Mammography in the Sense of Image Quality and Average Glandular Dose . . . . . . . . . . . . . . Mari Varjonen and Pekka Str¨ ommer
570
Generic Infrastructure for Medical Informatics (GIMI): The Development of a Mammographic Training System . . . . . . . . . . . . . . . . . . . Moi Hoon Yap, Alastair G. Gale, and Hazel J. Scott
577
Evaluation of 3D Breast Surface Reconstruction Accuracy Using Non-contact Scanner Images: A Phantom Study . . . . . . . . . . . . . . . . . . . . . Cuiping Zhang, Predrag R. Bakic, Shugao Xia, Fengshan Liu, and Andrew D.A. Maidment
585
Digital Breast Tomosynthesis Investigation of Different PV Distributions in Digital Breast Tomosynthesis (DBT) Mammography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yiheng Zhang, Heang-Ping Chan, Mitchell M. Goodsitt, Andrea Schmitz, Jeffrey W. Eberhard, and Bernhard E.H. Claus
593
Characterization of Projection Ordering in Iterative Reconstruction Methods for Breast Tomosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gang Wu, James Mainprize, and Martin Yaffe
601
Effect of Scan Angle and Reconstruction Algorithm on Model Observer Performance in Tomosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I. Reiser, B.A. Lau, and R.M. Nishikawa
606
A Novel Approach for Filtered Backprojection in Tomosynthesis Based on Filter Kernels Determined by Iterative Reconstruction Techniques . . . Jasmina Ludwig, Thomas Mertelmeier, Holger Kunze, and Wolfgang H¨ arer 3D Digital Breast Tomosynthesis Using Total Variation Regularization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Iason Kastanis, Simon Arridge, Alex Stewart, Spencer Gunn, Christer Ullberg, and Tom Francke Image Artifact in Digital Breast Tomosynthesis and Its Dependence on System and Reconstruction Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yue-Houng Hu, Wei Zhao, Thomas Mertelmeier, and Jasmina Ludwig
612
621
628
XXVI
Table of Contents
Multi-projection Correlation Imaging as a New Diagnostic Tool for Improved Breast Cancer Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Amarpreet S. Chawla, Ehsan Samei, Joseph Y. Lo, and Thomas Mertelmeier Sensitivity of Contrast-Enhanced Digital Breast Tomosynthesis to Changes in Iodine Concentration During Acquisition . . . . . . . . . . . . . . . . . Melissa L. Hill, James G. Mainprize, and Martin J. Yaffe
635
643
Breast Density, Texture and Risk II Quantifying Breast Thickness for Density Measurement . . . . . . . . . . . . . . . Jennifer Diffey, Alan Hufton, Christine Beeston, Julia Smith, Tom Marchant, and Susan Astley Effect of Tissue Thickness Variation in Volumetric Breast Density Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Olivier Alonzo-Proulx, Albert H. Tyson, Gordon E. Mawdsley, and Martin J. Yaffe Breast Abnormality Detection Incorporating Breast Density Information Based on Independent Components Analysis . . . . . . . . . . . . . Styliani Petroudi, Nicoletta Nicolaou, Julius Georgiou, and Michael Brady Comparison of Breast Percent Density Estimated from Digital Mammograms and Central Reconstructed Tomosynthesis Slice Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Predrag R. Bakic, Despina Kontos, Andrea B. Troxel, and Andrew Maidment Digital Breast Tomosynthesis Parenchymal Texture Analysis for Breast Cancer Risk Estimation: A Preliminary Study . . . . . . . . . . . . . . . . . . . . . . . Despina Kontos, Predrag R. Bakic, Andrea B. Troxel, Emily F. Conant, and Andrew D.A. Maidment Texture Based Segmentation of Breast DCE-MRI . . . . . . . . . . . . . . . . . . . . Yang Can Gong and Michael Brady
651
659
667
674
681
689
Physics, Image Quality and Quality Assurance Image Quality Measurements in Breast Tomosynthesis . . . . . . . . . . . . . . . . Ruben van Engen, Ramona Bouwman, Roeland van der Burght, Barbara Lazzari, David R. Dance, Patrice Heid, Magnus Aslund, and Kenneth C. Young
696
Table of Contents
One Year of Experience with Remote Quality Assurance of Digital Mammography Systems in the Flemish Breast Cancer Screening Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jurgen Jacobs, Kim Lemmens, Joris Nens, Koen Michielsen, Guy Marchal, and Hilde Bosmans Automatic Exposure Control in Digital Mammography: Contrast-to-Noise Ratio Versus Average Glandular Dose . . . . . . . . . . . . . . Gisella Gennaro, Paola Golinelli, Elena Bellan, Paola Colombo, Loredana D’Ercole, Anna Di Nallo, Lara Gallo, Carlo Giordano, Gabriele Meliad` o, Barbara Morri, Elvina Nassivera, Nadia Oberhofer, Daniela Origgi, Massimiliano Paolucci, Nicoletta Paruccini, Michela Piergentili, Elisa Rizzi, and Raffaella Rossi Effect of Anode/Filter Combination on the Dose and Image Quality of a Digital Mammography System Based on an Amorphous Selenium Detector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Paola Baldelli, Niall Phelan, and Gillian Egan Comparative Technical Study of Two Generations of CR Plates for Digital Mammography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hilde Bosmans, Kim Lemmens, Jurgen Jacobs, Beatrijs Verbrugge, Koen Michielsen, Federica Zanca, Joris Nens, Chantal Van Ongeval, and Guy Marchal Comparing the Performance of Digital Mammography Systems . . . . . . . . Kenneth C. Young, Jennifer M. Oduko, Ozcan Gundogdu, and Abdulaziz Alsager Cross-Calibration and Longitudinal Quality Control of Hologic Selenia Full-Field Digital Mammography Systems for Volumetric Breast Density Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jeff Wang, Serghei Malkov, Bo Fan, and John Shepherd Classification of Artifacts in Clinical Digital Mammography . . . . . . . . . . . Chantal Van Ongeval, Jurgen Jacobs, Andr´e Van Steen, Federica Zanca, Hilde Bosmans, and Guy Marchal
XXVII
703
711
716
724
732
740 748
Contrast Sensitivity in Mammographic Softcopy Reading – Determination with Psychophysical Procedures . . . . . . . . . . . . . D¨ orte Apelt and Heinz-Otto Peitgen
756
Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
765
Improvements to Single Energy Absorptiometry Method for Digital Mammography to Quantify Breast Tissue Density Serghei Malkov, Jeff Wang, and John Shepherd University of California, San Francisco, Department of Radiology 1635 Divisadero Street, Suite 525, San Francisco, CA 94115 [email protected]
Abstract. We report on the improvements of the design and characteristics of a single energy absorptiometry (SXA) method to quantify the volume of dense breast tissue from mammograms. The method consists of including a breast tissue equivalent phantom in the unused portion of the mammogram as a reference for breast composition. The phantom contains nine lead positioning markers which solve for the orientation of the paddle by a 3D reconstruction least square technique. The volumetric breast density was calculated by comparing image pixel values to the fat/fibroglandular references at the same thickness and using an elaborated calibration map. The breast density precision and accuracy were determined by multiple scans of a test phantom at different compression thicknesses and paddle-tilt angles. We also estimated the error contribution due to thickness and tilt angle variations. Keywords: full-field digital mammography, volumetric breast density, single x-ray absorptiometry, breast cancer.
1 Background Breast density, which has been studied extensively in large prospective and casecontrol studies, has been shown to be one of the strongest indicators of breast cancer risk. Volumetric breast density is considered to be a more accurate and precise measurement of true breast composition and a better cancer risk marker than commonly used areal mammographic breast density. Two volumetric density approaches for film mammography have been developed and previously described [1, 2]. Digital mammography has made the processing of mammograms more efficient, stable, and potentially higher accurate and precise. Socalled “phantomless” methods that use self-referencing standards in the breast to calibrate the image and incorporate a full physics model have also been developed [3, 4]. Extensive calibration information, rather than a heavy dependence on models has also been used in [5]. However, the calibration of a mammogram is known to be very sensitive to errors in breast thickness at each image pixel. There is also much interest in measuring true volumetric fibroglandular density (%FGV) for greater precision and accuracy in individuals. E.A. Krupinski (Ed.): IWDM 2008, LNCS 5116, pp. 1–8, 2008. © Springer-Verlag Berlin Heidelberg 2008
2
S. Malkov, J. Wang, and J. Shepherd
The SXA method [6, 7] is meant to individually calibrate each image by including a reference phantom. The objective has been to develop, validate and clinically test this new technology for quantifying volumetric compositional breast density. As part of this work, we have quantified the errors in %FGV associated with differences in compression paddle tilt and compression thickness. We utilized the in-mammogram phantom approach to calibrate pixel grey-scale values to the unique volume of fat and fibroglandular tissue for the pixel’s breast thickness.
2 Method The SXA method models the compressed breast as a tissue mass of two materials, adipose and fibroglandular tissue. The SXA technique relies on a novel reference phantom which allows derivation of adipose and fibroglandular tissue reference values and the breast thickness map. Percent glandular volumetric density %FGV(x,y) at any pixel including breast tissue (x, y) can be calculated by comparing adipose and fibroglandular tissue reference attenuations and the global volumetric fibroglandular density %FGV is estimated by volume weighted breast density summing up of all breast area pixels. The current design of the in-mammogram phantom is a multi-step phantom made of breast tissue equivalent material and containing nine lead positioning markers. The phantom adhered to the top of the mammography compression paddle and phantom image are displayed on Fig. 1.
Fig. 1. SXA phantom atop mammographic compression paddle with validation phantom (left) and the phantom image (right)
The phantom is stepped so that it provides an entire reference curve of attenuation pixel values versus breast thickness and composition from 1 to 7 cm. Thus the height and orientation of the phantom’s base is the height and orientation (tilt) of the compression paddle. By describing its position and orientation, and the extended shape (planar or curved) of the paddle, we can describe the thickness of the breast at all imaged locations. A 3-D reconstruction technique was used to find the breast compression thickness and the paddle tilt. One has to know the origin coordinates (x, y, z) and the orientation angles (α, β, θ) of the phantom coordinate system in respect to a world system related
Improvements to Single Energy Absorptiometry Method for Digital Mammography
3
to the detector. These six parameters are found by an unconstrained nonlinear least square method. The phantom’s lead sphere locations are identified using opening and subtracting, dilating and eroding, thresholding and labeling morphological operations. The step attenuation values are quantified automatically by applying the translation and rotation matrix and Boolean expressions to a 3-D region of interest (ROI) template of the step tops and base. The sequence of image analysis procedures are demonstrated in Fig. 2.
a
b
c
d
e
Fig. 2. a) initial image, b) step isolation, c) step subtraction, d) artifact removal, e) RIO and lead marker location
Two novel phantoms were constructed to calibrate the SXA method. The density step phantom, DSP7, was manufactured by CIRS (CIRS, Norfolk, VA) and has constant thickness density steps at seven fibroglandular densities. The TILT phantom is best described as two DSP7s that have been machined to form a partial wedge with angles from -10 to +10 degrees. To test the SXA method, the DSP7 and TILT phantoms were placed at the breast position with the reference phantom at its usual position in the corner of the paddle. The measurements of the DSP7 with thicknesses from 0.2 to 8 cm and for different x-ray techniques (24 – 35 kVp, and 20 – 250 mAs) were made using a Selenia mammography machine under screening conditions. Table 1 shows the combinations of screening conditions used in clinical hospital practice at the California Pacific Medical Center and is based on 9816 clinical images. Table 1. Screening condition statistics kVp 24 25 26 27 28 29 30 31 32 33 34
Number 298 469 721 866 1452 1323 1269 1150 1906 308 54
Thickness, mm