Abstract
We develop and evaluate an ideal observer for model of the 3D spatial distribution of x-ray attenuation coefficients in the breast. This model relies on thresholding of an underlying Gaussian random field to generate binary objects representing the distribution of adipose and glandular tissue in the breast parenchyma. Our motivation is to evaluate an emerging breast CT device for breast cancer screening. We show how the thresholded Gaussian model fits into the Markov-Chain Monte-Carlo (MCMC) approach for evaluating ideal-observer performance devised by Kupinski et al. [JOSA-A, 2003], and we show some preliminary results indicating that the procedure can be made to generate qualitatively realistic simulations. We demonstrate improved performance of the MCMC ideal observer over a Hotelling linear filter in a small-scale simulation.
| Original language | English (US) |
|---|---|
| Title of host publication | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
| Pages | 393-400 |
| Number of pages | 8 |
| Volume | 5116 LNCS |
| DOIs | |
| State | Published - 2008 |
| Event | 9th International Workshop on Digital Mammography, IWDM 2008 - Tucson, AZ, United States Duration: Jul 20 2008 → Jul 23 2008 |
Publication series
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| Volume | 5116 LNCS |
| ISSN (Print) | 03029743 |
| ISSN (Electronic) | 16113349 |
Other
| Other | 9th International Workshop on Digital Mammography, IWDM 2008 |
|---|---|
| Country | United States |
| City | Tucson, AZ |
| Period | 7/20/08 → 7/23/08 |
Fingerprint
Keywords
- Binary texture
- Breast CT
- Ideal observer
- Screening mammography
ASJC Scopus subject areas
- Computer Science(all)
- Biochemistry, Genetics and Molecular Biology(all)
- Theoretical Computer Science
Cite this
An ideal observer for a model of x-ray imaging in breast parenchymal tissue. / Abbey, Craig K.; Boone, John M.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 5116 LNCS 2008. p. 393-400 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5116 LNCS).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - An ideal observer for a model of x-ray imaging in breast parenchymal tissue
AU - Abbey, Craig K.
AU - Boone, John M
PY - 2008
Y1 - 2008
N2 - We develop and evaluate an ideal observer for model of the 3D spatial distribution of x-ray attenuation coefficients in the breast. This model relies on thresholding of an underlying Gaussian random field to generate binary objects representing the distribution of adipose and glandular tissue in the breast parenchyma. Our motivation is to evaluate an emerging breast CT device for breast cancer screening. We show how the thresholded Gaussian model fits into the Markov-Chain Monte-Carlo (MCMC) approach for evaluating ideal-observer performance devised by Kupinski et al. [JOSA-A, 2003], and we show some preliminary results indicating that the procedure can be made to generate qualitatively realistic simulations. We demonstrate improved performance of the MCMC ideal observer over a Hotelling linear filter in a small-scale simulation.
AB - We develop and evaluate an ideal observer for model of the 3D spatial distribution of x-ray attenuation coefficients in the breast. This model relies on thresholding of an underlying Gaussian random field to generate binary objects representing the distribution of adipose and glandular tissue in the breast parenchyma. Our motivation is to evaluate an emerging breast CT device for breast cancer screening. We show how the thresholded Gaussian model fits into the Markov-Chain Monte-Carlo (MCMC) approach for evaluating ideal-observer performance devised by Kupinski et al. [JOSA-A, 2003], and we show some preliminary results indicating that the procedure can be made to generate qualitatively realistic simulations. We demonstrate improved performance of the MCMC ideal observer over a Hotelling linear filter in a small-scale simulation.
KW - Binary texture
KW - Breast CT
KW - Ideal observer
KW - Screening mammography
UR - http://www.scopus.com/inward/record.url?scp=50949108726&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=50949108726&partnerID=8YFLogxK
U2 - 10.1007/978-3-540-70538-3_55
DO - 10.1007/978-3-540-70538-3_55
M3 - Conference contribution
AN - SCOPUS:50949108726
SN - 3540705376
SN - 9783540705376
VL - 5116 LNCS
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 393
EP - 400
BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
ER -