TH‐C‐AUD C‐03: Patient Setup Using 2D Projected Images and Setup Error Estimation

Q. Zhao, Huanmei Wu, Colleen DesRosiers, M. Cao, R. Mcmahon, V. Thai, Mark Langer

Research output: Contribution to journalArticle

Abstract

Purpose: Fiducial markers or bone structure are widely used in patient setup for image guided therapies and is also important in interventional radiology. A general approach (“2D‐2D”) for patient setup is based on 2D projected images on the anterior‐posterior (AP) and lateral (LAT) directions. However, it is still an open question to decide the number of fiducial markers required for patient setup, the optimal locations for marker plantation, the selection of fiducial markers versus bone structure, and the error estimations of different setup methods. Our work will address some of these issues and provide quantified information. Method and Materials: Clinical patient setup procedure (“2D‐2D”) is simulated using rigid registration. Least square metric is applied to minimize the alignment error of markers. Registration under different degree of freedom (DOF) are performed, including 3DOF (translation only), 4DOF (translation and AP couch rotation) and 5DOF (translation and rotation on AP and LAT). Registration errors are calculated based on absolute or percentage of missed volume between estimation tumor location and real tumor location. Results: Patient setup errors were investigated using different registration methods under various tumor motion conditions. The results showed that back and forth shifts using alternating projection planes can worsen the true registration of the target center with each “correction” step given target rotation. Second, for off center rotations of five degrees, no alignment can result in smaller errors than 2D‐2D registrations, despite a true translation of the target center. In addition, a 3DOF alignment process performs better against rotational shifts than both 4DOF and 5DOF that allow rotations within the projection planes. Conclusion: The quantitative analysis and setup error estimation using “2D‐2D” registration will provide better guidance for patient setup, which is important for effective radiation treatment of cancer patients.

Original languageEnglish (US)
Number of pages1
JournalMedical Physics
Volume35
Issue number6
DOIs
StatePublished - Jan 1 2008

Fingerprint

Fiducial Markers
Neoplasms
Bone and Bones
Interventional Radiology
Least-Squares Analysis
Radiation
Therapeutics

ASJC Scopus subject areas

  • Biophysics
  • Radiology Nuclear Medicine and imaging

Cite this

TH‐C‐AUD C‐03 : Patient Setup Using 2D Projected Images and Setup Error Estimation. / Zhao, Q.; Wu, Huanmei; DesRosiers, Colleen; Cao, M.; Mcmahon, R.; Thai, V.; Langer, Mark.

In: Medical Physics, Vol. 35, No. 6, 01.01.2008.

Research output: Contribution to journalArticle

@article{27dfea31cbef496c977e368c5e7e7eee,
title = "TH‐C‐AUD C‐03: Patient Setup Using 2D Projected Images and Setup Error Estimation",
abstract = "Purpose: Fiducial markers or bone structure are widely used in patient setup for image guided therapies and is also important in interventional radiology. A general approach (“2D‐2D”) for patient setup is based on 2D projected images on the anterior‐posterior (AP) and lateral (LAT) directions. However, it is still an open question to decide the number of fiducial markers required for patient setup, the optimal locations for marker plantation, the selection of fiducial markers versus bone structure, and the error estimations of different setup methods. Our work will address some of these issues and provide quantified information. Method and Materials: Clinical patient setup procedure (“2D‐2D”) is simulated using rigid registration. Least square metric is applied to minimize the alignment error of markers. Registration under different degree of freedom (DOF) are performed, including 3DOF (translation only), 4DOF (translation and AP couch rotation) and 5DOF (translation and rotation on AP and LAT). Registration errors are calculated based on absolute or percentage of missed volume between estimation tumor location and real tumor location. Results: Patient setup errors were investigated using different registration methods under various tumor motion conditions. The results showed that back and forth shifts using alternating projection planes can worsen the true registration of the target center with each “correction” step given target rotation. Second, for off center rotations of five degrees, no alignment can result in smaller errors than 2D‐2D registrations, despite a true translation of the target center. In addition, a 3DOF alignment process performs better against rotational shifts than both 4DOF and 5DOF that allow rotations within the projection planes. Conclusion: The quantitative analysis and setup error estimation using “2D‐2D” registration will provide better guidance for patient setup, which is important for effective radiation treatment of cancer patients.",
author = "Q. Zhao and Huanmei Wu and Colleen DesRosiers and M. Cao and R. Mcmahon and V. Thai and Mark Langer",
year = "2008",
month = "1",
day = "1",
doi = "10.1118/1.2962849",
language = "English (US)",
volume = "35",
journal = "Medical Physics",
issn = "0094-2405",
publisher = "AAPM - American Association of Physicists in Medicine",
number = "6",

}

TY - JOUR

T1 - TH‐C‐AUD C‐03

T2 - Patient Setup Using 2D Projected Images and Setup Error Estimation

AU - Zhao, Q.

AU - Wu, Huanmei

AU - DesRosiers, Colleen

AU - Cao, M.

AU - Mcmahon, R.

AU - Thai, V.

AU - Langer, Mark

PY - 2008/1/1

Y1 - 2008/1/1

N2 - Purpose: Fiducial markers or bone structure are widely used in patient setup for image guided therapies and is also important in interventional radiology. A general approach (“2D‐2D”) for patient setup is based on 2D projected images on the anterior‐posterior (AP) and lateral (LAT) directions. However, it is still an open question to decide the number of fiducial markers required for patient setup, the optimal locations for marker plantation, the selection of fiducial markers versus bone structure, and the error estimations of different setup methods. Our work will address some of these issues and provide quantified information. Method and Materials: Clinical patient setup procedure (“2D‐2D”) is simulated using rigid registration. Least square metric is applied to minimize the alignment error of markers. Registration under different degree of freedom (DOF) are performed, including 3DOF (translation only), 4DOF (translation and AP couch rotation) and 5DOF (translation and rotation on AP and LAT). Registration errors are calculated based on absolute or percentage of missed volume between estimation tumor location and real tumor location. Results: Patient setup errors were investigated using different registration methods under various tumor motion conditions. The results showed that back and forth shifts using alternating projection planes can worsen the true registration of the target center with each “correction” step given target rotation. Second, for off center rotations of five degrees, no alignment can result in smaller errors than 2D‐2D registrations, despite a true translation of the target center. In addition, a 3DOF alignment process performs better against rotational shifts than both 4DOF and 5DOF that allow rotations within the projection planes. Conclusion: The quantitative analysis and setup error estimation using “2D‐2D” registration will provide better guidance for patient setup, which is important for effective radiation treatment of cancer patients.

AB - Purpose: Fiducial markers or bone structure are widely used in patient setup for image guided therapies and is also important in interventional radiology. A general approach (“2D‐2D”) for patient setup is based on 2D projected images on the anterior‐posterior (AP) and lateral (LAT) directions. However, it is still an open question to decide the number of fiducial markers required for patient setup, the optimal locations for marker plantation, the selection of fiducial markers versus bone structure, and the error estimations of different setup methods. Our work will address some of these issues and provide quantified information. Method and Materials: Clinical patient setup procedure (“2D‐2D”) is simulated using rigid registration. Least square metric is applied to minimize the alignment error of markers. Registration under different degree of freedom (DOF) are performed, including 3DOF (translation only), 4DOF (translation and AP couch rotation) and 5DOF (translation and rotation on AP and LAT). Registration errors are calculated based on absolute or percentage of missed volume between estimation tumor location and real tumor location. Results: Patient setup errors were investigated using different registration methods under various tumor motion conditions. The results showed that back and forth shifts using alternating projection planes can worsen the true registration of the target center with each “correction” step given target rotation. Second, for off center rotations of five degrees, no alignment can result in smaller errors than 2D‐2D registrations, despite a true translation of the target center. In addition, a 3DOF alignment process performs better against rotational shifts than both 4DOF and 5DOF that allow rotations within the projection planes. Conclusion: The quantitative analysis and setup error estimation using “2D‐2D” registration will provide better guidance for patient setup, which is important for effective radiation treatment of cancer patients.

UR - http://www.scopus.com/inward/record.url?scp=85024820992&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85024820992&partnerID=8YFLogxK

U2 - 10.1118/1.2962849

DO - 10.1118/1.2962849

M3 - Article

AN - SCOPUS:85024820992

VL - 35

JO - Medical Physics

JF - Medical Physics

SN - 0094-2405

IS - 6

ER -