TU‐A‐213CD‐11: A Synchronized Moving Grid (SMOG) to Detect and Correct Imaging Lag in Cone Beam Computed Tomography (CBCT): A Simulation Study

JianYue Jin, L. Ren, J. Kim, S. Kim, B. Movsas, D. Jaffrey, I. Chetty

Research output: Contribution to journalArticle

Abstract

Purpose: Imaging lag and scatter are 2 major problems in CBCT. Previously we demonstrated that our proposed synchronized‐moving‐grid (SMOG) system mitigates the scatter problem. The SMOG‐system is placed between the radiation source and patient. Multiple projections with alternating image and ‘shadow’ bands are taken at different grid positions at each gantry angle, and are merged together to form a full projection. Scatter can be physically reduced by the grid, and further correction is facilitated by scatter measurements in the shadow bands. We hypothesize that the imaging lag may also be detected in the shadow bands and thus be corrected. Methods: Three imaging lag models, which provide a good fit for the first 10 frames of measured lag, but have differences for larger frame numbers, were used for the simulation. The lag effect was considered as the lag sum of the 100 previous frames. With the SMOG‐system, a pixel in a projection is alternately in the shadow or image band for different exposures. The intensities of the pixel for different exposures were obtained from 680 CBCT projections of a rando phantom without or with a static grid. The lag effect was calculated for arbitrary pixels without the SMOG‐system, and for paired pixels in the neighboring image and shadow bands with the SMOG‐ system for 2 and 3 exposures. The residual lag of the SMOG‐system was calculated as the lag difference of the paired pixels. Results: The maximum lag was 17%, 23% and 38%, respectively, for the 3 models without SMOG. The maximum residual lag was 4.1%, 4.1% and 3.9% with the SMOG‐ system for 2‐exposures, and was 0.7%, 0.8% and 0.8% for 3‐exposures for the 3 models. Conclusions: Using a model to correct for the lag effect is not reliable. However, the SMOG‐system is able to correct for the lag effect effectively.

Original languageEnglish (US)
Number of pages1
JournalMedical Physics
Volume39
Issue number6
DOIs
StatePublished - Jan 1 2012

Fingerprint

Cone-Beam Computed Tomography
Radiation

ASJC Scopus subject areas

  • Biophysics
  • Radiology Nuclear Medicine and imaging

Cite this

TU‐A‐213CD‐11 : A Synchronized Moving Grid (SMOG) to Detect and Correct Imaging Lag in Cone Beam Computed Tomography (CBCT): A Simulation Study. / Jin, JianYue; Ren, L.; Kim, J.; Kim, S.; Movsas, B.; Jaffrey, D.; Chetty, I.

In: Medical Physics, Vol. 39, No. 6, 01.01.2012.

Research output: Contribution to journalArticle

Jin, JianYue ; Ren, L. ; Kim, J. ; Kim, S. ; Movsas, B. ; Jaffrey, D. ; Chetty, I. / TU‐A‐213CD‐11 : A Synchronized Moving Grid (SMOG) to Detect and Correct Imaging Lag in Cone Beam Computed Tomography (CBCT): A Simulation Study. In: Medical Physics. 2012 ; Vol. 39, No. 6.
@article{2c7aa34e0d574382bcc8ada3c4795244,
title = "TU‐A‐213CD‐11: A Synchronized Moving Grid (SMOG) to Detect and Correct Imaging Lag in Cone Beam Computed Tomography (CBCT): A Simulation Study",
abstract = "Purpose: Imaging lag and scatter are 2 major problems in CBCT. Previously we demonstrated that our proposed synchronized‐moving‐grid (SMOG) system mitigates the scatter problem. The SMOG‐system is placed between the radiation source and patient. Multiple projections with alternating image and ‘shadow’ bands are taken at different grid positions at each gantry angle, and are merged together to form a full projection. Scatter can be physically reduced by the grid, and further correction is facilitated by scatter measurements in the shadow bands. We hypothesize that the imaging lag may also be detected in the shadow bands and thus be corrected. Methods: Three imaging lag models, which provide a good fit for the first 10 frames of measured lag, but have differences for larger frame numbers, were used for the simulation. The lag effect was considered as the lag sum of the 100 previous frames. With the SMOG‐system, a pixel in a projection is alternately in the shadow or image band for different exposures. The intensities of the pixel for different exposures were obtained from 680 CBCT projections of a rando phantom without or with a static grid. The lag effect was calculated for arbitrary pixels without the SMOG‐system, and for paired pixels in the neighboring image and shadow bands with the SMOG‐ system for 2 and 3 exposures. The residual lag of the SMOG‐system was calculated as the lag difference of the paired pixels. Results: The maximum lag was 17{\%}, 23{\%} and 38{\%}, respectively, for the 3 models without SMOG. The maximum residual lag was 4.1{\%}, 4.1{\%} and 3.9{\%} with the SMOG‐ system for 2‐exposures, and was 0.7{\%}, 0.8{\%} and 0.8{\%} for 3‐exposures for the 3 models. Conclusions: Using a model to correct for the lag effect is not reliable. However, the SMOG‐system is able to correct for the lag effect effectively.",
author = "JianYue Jin and L. Ren and J. Kim and S. Kim and B. Movsas and D. Jaffrey and I. Chetty",
year = "2012",
month = "1",
day = "1",
doi = "10.1118/1.4735890",
language = "English (US)",
volume = "39",
journal = "Medical Physics",
issn = "0094-2405",
publisher = "AAPM - American Association of Physicists in Medicine",
number = "6",

}

TY - JOUR

T1 - TU‐A‐213CD‐11

T2 - A Synchronized Moving Grid (SMOG) to Detect and Correct Imaging Lag in Cone Beam Computed Tomography (CBCT): A Simulation Study

AU - Jin, JianYue

AU - Ren, L.

AU - Kim, J.

AU - Kim, S.

AU - Movsas, B.

AU - Jaffrey, D.

AU - Chetty, I.

PY - 2012/1/1

Y1 - 2012/1/1

N2 - Purpose: Imaging lag and scatter are 2 major problems in CBCT. Previously we demonstrated that our proposed synchronized‐moving‐grid (SMOG) system mitigates the scatter problem. The SMOG‐system is placed between the radiation source and patient. Multiple projections with alternating image and ‘shadow’ bands are taken at different grid positions at each gantry angle, and are merged together to form a full projection. Scatter can be physically reduced by the grid, and further correction is facilitated by scatter measurements in the shadow bands. We hypothesize that the imaging lag may also be detected in the shadow bands and thus be corrected. Methods: Three imaging lag models, which provide a good fit for the first 10 frames of measured lag, but have differences for larger frame numbers, were used for the simulation. The lag effect was considered as the lag sum of the 100 previous frames. With the SMOG‐system, a pixel in a projection is alternately in the shadow or image band for different exposures. The intensities of the pixel for different exposures were obtained from 680 CBCT projections of a rando phantom without or with a static grid. The lag effect was calculated for arbitrary pixels without the SMOG‐system, and for paired pixels in the neighboring image and shadow bands with the SMOG‐ system for 2 and 3 exposures. The residual lag of the SMOG‐system was calculated as the lag difference of the paired pixels. Results: The maximum lag was 17%, 23% and 38%, respectively, for the 3 models without SMOG. The maximum residual lag was 4.1%, 4.1% and 3.9% with the SMOG‐ system for 2‐exposures, and was 0.7%, 0.8% and 0.8% for 3‐exposures for the 3 models. Conclusions: Using a model to correct for the lag effect is not reliable. However, the SMOG‐system is able to correct for the lag effect effectively.

AB - Purpose: Imaging lag and scatter are 2 major problems in CBCT. Previously we demonstrated that our proposed synchronized‐moving‐grid (SMOG) system mitigates the scatter problem. The SMOG‐system is placed between the radiation source and patient. Multiple projections with alternating image and ‘shadow’ bands are taken at different grid positions at each gantry angle, and are merged together to form a full projection. Scatter can be physically reduced by the grid, and further correction is facilitated by scatter measurements in the shadow bands. We hypothesize that the imaging lag may also be detected in the shadow bands and thus be corrected. Methods: Three imaging lag models, which provide a good fit for the first 10 frames of measured lag, but have differences for larger frame numbers, were used for the simulation. The lag effect was considered as the lag sum of the 100 previous frames. With the SMOG‐system, a pixel in a projection is alternately in the shadow or image band for different exposures. The intensities of the pixel for different exposures were obtained from 680 CBCT projections of a rando phantom without or with a static grid. The lag effect was calculated for arbitrary pixels without the SMOG‐system, and for paired pixels in the neighboring image and shadow bands with the SMOG‐ system for 2 and 3 exposures. The residual lag of the SMOG‐system was calculated as the lag difference of the paired pixels. Results: The maximum lag was 17%, 23% and 38%, respectively, for the 3 models without SMOG. The maximum residual lag was 4.1%, 4.1% and 3.9% with the SMOG‐ system for 2‐exposures, and was 0.7%, 0.8% and 0.8% for 3‐exposures for the 3 models. Conclusions: Using a model to correct for the lag effect is not reliable. However, the SMOG‐system is able to correct for the lag effect effectively.

UR - http://www.scopus.com/inward/record.url?scp=85024789593&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85024789593&partnerID=8YFLogxK

U2 - 10.1118/1.4735890

DO - 10.1118/1.4735890

M3 - Article

AN - SCOPUS:85024789593

VL - 39

JO - Medical Physics

JF - Medical Physics

SN - 0094-2405

IS - 6

ER -