Accuracy of an automatic patient-positioning system based on the correlation of two edge images in radiotherapy

Myonggeun Yoon, Minho Cheong, Jinsung Kim, Dong Ho Shin, Sung Yong Park, Se Byeong Lee

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

We have clinically evaluated the accuracy of an automatic patient-positioning system based on the image correlation of two edge images in radiotherapy. Ninetysix head & neck images from eight patients undergoing proton therapy were compared with a digitally reconstructed radiograph (DRR) of planning CT. Two edge images, a reference image and a test image, were extracted by applying a Canny edge detector algorithm to a DRR and a 2D X-ray image, respectively, of each patient before positioning. In a simulation using a humanoid phantom, performed to verify the effectiveness of the proposed method, no registration errorswere observed for given ranges of rotation, pitch, and translation in the x, y, and z directions. For real patients, however, there were discrepancies between the automatic positioning method andmanual positioning by physicians or technicians.Using edged head coronal- and sagittal-viewimages, the average differences in registration between these two methods for the x, y, and z directions were 0.11 cm, 0.09 cm and 0.11 cm, respectively, whereas the maximum discrepancies were 0.34 cm, 0.38 cm, and 0.50 cm, respectively. For rotation and pitch, the average registration errors were 0.95° and 1.00°, respectively, and the maximum errors were 3.6° and 2.3°, respectively. The proposed automatic patient-positioning system based on edge image comparison was relatively accurate for head and neck patients. However, image deformation during treatmentmay render the automatic method less accurate, since the test image many differ significantly from the reference image.

Original languageEnglish
Pages (from-to)322-330
Number of pages9
JournalJournal of Digital Imaging
Volume24
Issue number2
DOIs
Publication statusPublished - 2011 Apr 1
Externally publishedYes

Fingerprint

Patient Positioning
Radiotherapy
Head
Protons
Neck
Proton Therapy
Detectors
Planning
X rays
X-Rays
Physicians
Direction compound

Keywords

  • Automated object detection
  • Digital image processing
  • Radiotherapy

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Radiological and Ultrasound Technology
  • Computer Science Applications
  • Medicine(all)

Cite this

Accuracy of an automatic patient-positioning system based on the correlation of two edge images in radiotherapy. / Yoon, Myonggeun; Cheong, Minho; Kim, Jinsung; Shin, Dong Ho; Park, Sung Yong; Lee, Se Byeong.

In: Journal of Digital Imaging, Vol. 24, No. 2, 01.04.2011, p. 322-330.

Research output: Contribution to journalArticle

Yoon, Myonggeun ; Cheong, Minho ; Kim, Jinsung ; Shin, Dong Ho ; Park, Sung Yong ; Lee, Se Byeong. / Accuracy of an automatic patient-positioning system based on the correlation of two edge images in radiotherapy. In: Journal of Digital Imaging. 2011 ; Vol. 24, No. 2. pp. 322-330.
@article{59cf2d4b21a445e792d856497216efc8,
title = "Accuracy of an automatic patient-positioning system based on the correlation of two edge images in radiotherapy",
abstract = "We have clinically evaluated the accuracy of an automatic patient-positioning system based on the image correlation of two edge images in radiotherapy. Ninetysix head & neck images from eight patients undergoing proton therapy were compared with a digitally reconstructed radiograph (DRR) of planning CT. Two edge images, a reference image and a test image, were extracted by applying a Canny edge detector algorithm to a DRR and a 2D X-ray image, respectively, of each patient before positioning. In a simulation using a humanoid phantom, performed to verify the effectiveness of the proposed method, no registration errorswere observed for given ranges of rotation, pitch, and translation in the x, y, and z directions. For real patients, however, there were discrepancies between the automatic positioning method andmanual positioning by physicians or technicians.Using edged head coronal- and sagittal-viewimages, the average differences in registration between these two methods for the x, y, and z directions were 0.11 cm, 0.09 cm and 0.11 cm, respectively, whereas the maximum discrepancies were 0.34 cm, 0.38 cm, and 0.50 cm, respectively. For rotation and pitch, the average registration errors were 0.95° and 1.00°, respectively, and the maximum errors were 3.6° and 2.3°, respectively. The proposed automatic patient-positioning system based on edge image comparison was relatively accurate for head and neck patients. However, image deformation during treatmentmay render the automatic method less accurate, since the test image many differ significantly from the reference image.",
keywords = "Automated object detection, Digital image processing, Radiotherapy",
author = "Myonggeun Yoon and Minho Cheong and Jinsung Kim and Shin, {Dong Ho} and Park, {Sung Yong} and Lee, {Se Byeong}",
year = "2011",
month = "4",
day = "1",
doi = "10.1007/s10278-009-9269-6",
language = "English",
volume = "24",
pages = "322--330",
journal = "Journal of Digital Imaging",
issn = "0897-1889",
publisher = "Springer New York",
number = "2",

}

TY - JOUR

T1 - Accuracy of an automatic patient-positioning system based on the correlation of two edge images in radiotherapy

AU - Yoon, Myonggeun

AU - Cheong, Minho

AU - Kim, Jinsung

AU - Shin, Dong Ho

AU - Park, Sung Yong

AU - Lee, Se Byeong

PY - 2011/4/1

Y1 - 2011/4/1

N2 - We have clinically evaluated the accuracy of an automatic patient-positioning system based on the image correlation of two edge images in radiotherapy. Ninetysix head & neck images from eight patients undergoing proton therapy were compared with a digitally reconstructed radiograph (DRR) of planning CT. Two edge images, a reference image and a test image, were extracted by applying a Canny edge detector algorithm to a DRR and a 2D X-ray image, respectively, of each patient before positioning. In a simulation using a humanoid phantom, performed to verify the effectiveness of the proposed method, no registration errorswere observed for given ranges of rotation, pitch, and translation in the x, y, and z directions. For real patients, however, there were discrepancies between the automatic positioning method andmanual positioning by physicians or technicians.Using edged head coronal- and sagittal-viewimages, the average differences in registration between these two methods for the x, y, and z directions were 0.11 cm, 0.09 cm and 0.11 cm, respectively, whereas the maximum discrepancies were 0.34 cm, 0.38 cm, and 0.50 cm, respectively. For rotation and pitch, the average registration errors were 0.95° and 1.00°, respectively, and the maximum errors were 3.6° and 2.3°, respectively. The proposed automatic patient-positioning system based on edge image comparison was relatively accurate for head and neck patients. However, image deformation during treatmentmay render the automatic method less accurate, since the test image many differ significantly from the reference image.

AB - We have clinically evaluated the accuracy of an automatic patient-positioning system based on the image correlation of two edge images in radiotherapy. Ninetysix head & neck images from eight patients undergoing proton therapy were compared with a digitally reconstructed radiograph (DRR) of planning CT. Two edge images, a reference image and a test image, were extracted by applying a Canny edge detector algorithm to a DRR and a 2D X-ray image, respectively, of each patient before positioning. In a simulation using a humanoid phantom, performed to verify the effectiveness of the proposed method, no registration errorswere observed for given ranges of rotation, pitch, and translation in the x, y, and z directions. For real patients, however, there were discrepancies between the automatic positioning method andmanual positioning by physicians or technicians.Using edged head coronal- and sagittal-viewimages, the average differences in registration between these two methods for the x, y, and z directions were 0.11 cm, 0.09 cm and 0.11 cm, respectively, whereas the maximum discrepancies were 0.34 cm, 0.38 cm, and 0.50 cm, respectively. For rotation and pitch, the average registration errors were 0.95° and 1.00°, respectively, and the maximum errors were 3.6° and 2.3°, respectively. The proposed automatic patient-positioning system based on edge image comparison was relatively accurate for head and neck patients. However, image deformation during treatmentmay render the automatic method less accurate, since the test image many differ significantly from the reference image.

KW - Automated object detection

KW - Digital image processing

KW - Radiotherapy

UR - http://www.scopus.com/inward/record.url?scp=79960095517&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79960095517&partnerID=8YFLogxK

U2 - 10.1007/s10278-009-9269-6

DO - 10.1007/s10278-009-9269-6

M3 - Article

VL - 24

SP - 322

EP - 330

JO - Journal of Digital Imaging

JF - Journal of Digital Imaging

SN - 0897-1889

IS - 2

ER -