Non-circular scans and image reconstruction for breast CT

Junguo Bian, Nathan J. Packard, Kai Yang, Dan Xia, John M Boone, Xiaochuan Pan

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

1 Citation (Scopus)

Abstract

Current dedicated, cone-beam breast CT scanners generally use a circular scanning configuration largely because it is relatively easy to implement mechanically. It is also well-known, however, that a circular scanning configuration produces insufficient cone-beam data for reconstrucing accurate 3D breast images. Approximate algorithms, such as FDK has been widely applied to reconstruct images from circular cone-beam data. In the FDK reconstruction, it is possible to observe artifacts such as intensity decay for locations that are not within the plane containing the circular source trajectory. Such artifacts may potentially lead to false positive and/or false negative diagnosis of breast cancer. Non-circular imaging configurations may provide data sufficient for accurate image reconstruction. In this work, we implement, investigate innovative, non-circular scanning configurations such as helical and saddle configurations for data acquisition on a dedicated, cone-beam breast CT scanner, and develop novel algorithms to reconstruct accurate 3D images from these data. A dedicated, cone-beam breast CT scanner capable of performing non-circular scanning configurations was used in this research. We have investigated different scanning configurations, including helical and saddle configurations. A Defrise disk phantom and a dead mouse were scanned by use of these configurations. For each configuration, cone-beam data were acquired at 501 views over each turn. We have reconstructed images using our BPF algorithm from data acquired with the helical scanning configuration.

Original languageEnglish (US)
Title of host publicationProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume6913
DOIs
StatePublished - 2008
EventMedical Imaging 2008 - Physics of Medical Imaging - San Diego, CA, United States
Duration: Feb 18 2008Feb 21 2008

Other

OtherMedical Imaging 2008 - Physics of Medical Imaging
CountryUnited States
CitySan Diego, CA
Period2/18/082/21/08

Fingerprint

Image reconstruction
Cones
Scanning
Data acquisition
Trajectories
Imaging techniques

Keywords

  • Back-projection filtration
  • Cone beam projection
  • Dedicated breast CT
  • Exact image reconstruction
  • Non-circular scan

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Bian, J., Packard, N. J., Yang, K., Xia, D., Boone, J. M., & Pan, X. (2008). Non-circular scans and image reconstruction for breast CT. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE (Vol. 6913). [69134G] https://doi.org/10.1117/12.772093

Non-circular scans and image reconstruction for breast CT. / Bian, Junguo; Packard, Nathan J.; Yang, Kai; Xia, Dan; Boone, John M; Pan, Xiaochuan.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 6913 2008. 69134G.

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

Bian, J, Packard, NJ, Yang, K, Xia, D, Boone, JM & Pan, X 2008, Non-circular scans and image reconstruction for breast CT. in Progress in Biomedical Optics and Imaging - Proceedings of SPIE. vol. 6913, 69134G, Medical Imaging 2008 - Physics of Medical Imaging, San Diego, CA, United States, 2/18/08. https://doi.org/10.1117/12.772093
Bian J, Packard NJ, Yang K, Xia D, Boone JM, Pan X. Non-circular scans and image reconstruction for breast CT. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 6913. 2008. 69134G https://doi.org/10.1117/12.772093
Bian, Junguo ; Packard, Nathan J. ; Yang, Kai ; Xia, Dan ; Boone, John M ; Pan, Xiaochuan. / Non-circular scans and image reconstruction for breast CT. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 6913 2008.
@inproceedings{efae3ad1593745198fc60ae7a05427ce,
title = "Non-circular scans and image reconstruction for breast CT",
abstract = "Current dedicated, cone-beam breast CT scanners generally use a circular scanning configuration largely because it is relatively easy to implement mechanically. It is also well-known, however, that a circular scanning configuration produces insufficient cone-beam data for reconstrucing accurate 3D breast images. Approximate algorithms, such as FDK has been widely applied to reconstruct images from circular cone-beam data. In the FDK reconstruction, it is possible to observe artifacts such as intensity decay for locations that are not within the plane containing the circular source trajectory. Such artifacts may potentially lead to false positive and/or false negative diagnosis of breast cancer. Non-circular imaging configurations may provide data sufficient for accurate image reconstruction. In this work, we implement, investigate innovative, non-circular scanning configurations such as helical and saddle configurations for data acquisition on a dedicated, cone-beam breast CT scanner, and develop novel algorithms to reconstruct accurate 3D images from these data. A dedicated, cone-beam breast CT scanner capable of performing non-circular scanning configurations was used in this research. We have investigated different scanning configurations, including helical and saddle configurations. A Defrise disk phantom and a dead mouse were scanned by use of these configurations. For each configuration, cone-beam data were acquired at 501 views over each turn. We have reconstructed images using our BPF algorithm from data acquired with the helical scanning configuration.",
keywords = "Back-projection filtration, Cone beam projection, Dedicated breast CT, Exact image reconstruction, Non-circular scan",
author = "Junguo Bian and Packard, {Nathan J.} and Kai Yang and Dan Xia and Boone, {John M} and Xiaochuan Pan",
year = "2008",
doi = "10.1117/12.772093",
language = "English (US)",
isbn = "9780819470973",
volume = "6913",
booktitle = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",

}

TY - GEN

T1 - Non-circular scans and image reconstruction for breast CT

AU - Bian, Junguo

AU - Packard, Nathan J.

AU - Yang, Kai

AU - Xia, Dan

AU - Boone, John M

AU - Pan, Xiaochuan

PY - 2008

Y1 - 2008

N2 - Current dedicated, cone-beam breast CT scanners generally use a circular scanning configuration largely because it is relatively easy to implement mechanically. It is also well-known, however, that a circular scanning configuration produces insufficient cone-beam data for reconstrucing accurate 3D breast images. Approximate algorithms, such as FDK has been widely applied to reconstruct images from circular cone-beam data. In the FDK reconstruction, it is possible to observe artifacts such as intensity decay for locations that are not within the plane containing the circular source trajectory. Such artifacts may potentially lead to false positive and/or false negative diagnosis of breast cancer. Non-circular imaging configurations may provide data sufficient for accurate image reconstruction. In this work, we implement, investigate innovative, non-circular scanning configurations such as helical and saddle configurations for data acquisition on a dedicated, cone-beam breast CT scanner, and develop novel algorithms to reconstruct accurate 3D images from these data. A dedicated, cone-beam breast CT scanner capable of performing non-circular scanning configurations was used in this research. We have investigated different scanning configurations, including helical and saddle configurations. A Defrise disk phantom and a dead mouse were scanned by use of these configurations. For each configuration, cone-beam data were acquired at 501 views over each turn. We have reconstructed images using our BPF algorithm from data acquired with the helical scanning configuration.

AB - Current dedicated, cone-beam breast CT scanners generally use a circular scanning configuration largely because it is relatively easy to implement mechanically. It is also well-known, however, that a circular scanning configuration produces insufficient cone-beam data for reconstrucing accurate 3D breast images. Approximate algorithms, such as FDK has been widely applied to reconstruct images from circular cone-beam data. In the FDK reconstruction, it is possible to observe artifacts such as intensity decay for locations that are not within the plane containing the circular source trajectory. Such artifacts may potentially lead to false positive and/or false negative diagnosis of breast cancer. Non-circular imaging configurations may provide data sufficient for accurate image reconstruction. In this work, we implement, investigate innovative, non-circular scanning configurations such as helical and saddle configurations for data acquisition on a dedicated, cone-beam breast CT scanner, and develop novel algorithms to reconstruct accurate 3D images from these data. A dedicated, cone-beam breast CT scanner capable of performing non-circular scanning configurations was used in this research. We have investigated different scanning configurations, including helical and saddle configurations. A Defrise disk phantom and a dead mouse were scanned by use of these configurations. For each configuration, cone-beam data were acquired at 501 views over each turn. We have reconstructed images using our BPF algorithm from data acquired with the helical scanning configuration.

KW - Back-projection filtration

KW - Cone beam projection

KW - Dedicated breast CT

KW - Exact image reconstruction

KW - Non-circular scan

UR - http://www.scopus.com/inward/record.url?scp=43149088528&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=43149088528&partnerID=8YFLogxK

U2 - 10.1117/12.772093

DO - 10.1117/12.772093

M3 - Conference contribution

AN - SCOPUS:43149088528

SN - 9780819470973

VL - 6913

BT - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

ER -