SU‐E‐J‐27: Effects of Metal Artifacts of KV and MV CT Images on Structure Delineation and Tissue Electron/Mass Density Calculation

T. he, James Tanyi, Richard Crilly, Wolfram Laub

Research output: Contribution to journalArticle

Abstract

Purpose: To quantitatively evaluate effects of image artifacts of hip prostheses on the accuracy of structure delineation and tissue density calculation on kV and MV CT images. Methods: Five hip prostheses made of stainless steel, titanium and cobalt chrome alloys were positioned inside a water tank and scanned respectively on a Philips CT and a Tomotherapy Hi‐Art unit. Prostheses were positioned to mimic single and bilateral implantations. Rods of tissue materials of lung, water and bone were placed at locations next and distal to metal implants near femoral head, neck and stem of prostheses. kV and MV CT scans were repeated for each placement. On CT images, cross‐sectional outlines of metal implants and tissue rods were delineated. Densities of rod materials were determined and compared to the true values. Results: Metal artifacts were severe on kV CTs and minimal on MV CTs. Cross‐sectional outlines of metal implants and tissue rods on kV CTs were severely distorted by artifacts while those on MV CTs remained clearly identifiable. For kV CTs, deviations of measured tissue density from true value were up to 51.3%, 30.6% and 40.9% respectively for lung, bone and solid water. The magnitude of deviation was generally larger at locations closer to metal implants and greater with bilateral implants than single implant. For MV CTs, deviations of measured density from true value were less than 6% for all three tissue materials either with single or bilateral implants. Magnitude of deviation appeared to be uniform and independent of locations relative to metal implants. Conclusions: High Z metal artifacts on kV CTs can have severe impact on the accuracy of structure delineation and tissue density calculation, while on MV CTs, the impact is substantially less and insignificant. MV CTs should be considered for treatment planning on patients with high Z metal implants.

Original languageEnglish (US)
Pages (from-to)3658
Number of pages1
JournalMedical Physics
Volume39
Issue number6
DOIs
StatePublished - 2012

Fingerprint

Artifacts
Metals
Electrons
Hip Prosthesis
Prostheses and Implants
Water
Bone and Bones
Lung
Femur Neck
Stainless Steel
Cobalt
Titanium

ASJC Scopus subject areas

  • Biophysics
  • Radiology Nuclear Medicine and imaging

Cite this

@article{ba92223a58d04650b0f6b50898fb2706,
title = "SU‐E‐J‐27: Effects of Metal Artifacts of KV and MV CT Images on Structure Delineation and Tissue Electron/Mass Density Calculation",
abstract = "Purpose: To quantitatively evaluate effects of image artifacts of hip prostheses on the accuracy of structure delineation and tissue density calculation on kV and MV CT images. Methods: Five hip prostheses made of stainless steel, titanium and cobalt chrome alloys were positioned inside a water tank and scanned respectively on a Philips CT and a Tomotherapy Hi‐Art unit. Prostheses were positioned to mimic single and bilateral implantations. Rods of tissue materials of lung, water and bone were placed at locations next and distal to metal implants near femoral head, neck and stem of prostheses. kV and MV CT scans were repeated for each placement. On CT images, cross‐sectional outlines of metal implants and tissue rods were delineated. Densities of rod materials were determined and compared to the true values. Results: Metal artifacts were severe on kV CTs and minimal on MV CTs. Cross‐sectional outlines of metal implants and tissue rods on kV CTs were severely distorted by artifacts while those on MV CTs remained clearly identifiable. For kV CTs, deviations of measured tissue density from true value were up to 51.3{\%}, 30.6{\%} and 40.9{\%} respectively for lung, bone and solid water. The magnitude of deviation was generally larger at locations closer to metal implants and greater with bilateral implants than single implant. For MV CTs, deviations of measured density from true value were less than 6{\%} for all three tissue materials either with single or bilateral implants. Magnitude of deviation appeared to be uniform and independent of locations relative to metal implants. Conclusions: High Z metal artifacts on kV CTs can have severe impact on the accuracy of structure delineation and tissue density calculation, while on MV CTs, the impact is substantially less and insignificant. MV CTs should be considered for treatment planning on patients with high Z metal implants.",
author = "T. he and James Tanyi and Richard Crilly and Wolfram Laub",
year = "2012",
doi = "10.1118/1.4734860",
language = "English (US)",
volume = "39",
pages = "3658",
journal = "Medical Physics",
issn = "0094-2405",
publisher = "AAPM - American Association of Physicists in Medicine",
number = "6",

}

TY - JOUR

T1 - SU‐E‐J‐27

T2 - Effects of Metal Artifacts of KV and MV CT Images on Structure Delineation and Tissue Electron/Mass Density Calculation

AU - he, T.

AU - Tanyi, James

AU - Crilly, Richard

AU - Laub, Wolfram

PY - 2012

Y1 - 2012

N2 - Purpose: To quantitatively evaluate effects of image artifacts of hip prostheses on the accuracy of structure delineation and tissue density calculation on kV and MV CT images. Methods: Five hip prostheses made of stainless steel, titanium and cobalt chrome alloys were positioned inside a water tank and scanned respectively on a Philips CT and a Tomotherapy Hi‐Art unit. Prostheses were positioned to mimic single and bilateral implantations. Rods of tissue materials of lung, water and bone were placed at locations next and distal to metal implants near femoral head, neck and stem of prostheses. kV and MV CT scans were repeated for each placement. On CT images, cross‐sectional outlines of metal implants and tissue rods were delineated. Densities of rod materials were determined and compared to the true values. Results: Metal artifacts were severe on kV CTs and minimal on MV CTs. Cross‐sectional outlines of metal implants and tissue rods on kV CTs were severely distorted by artifacts while those on MV CTs remained clearly identifiable. For kV CTs, deviations of measured tissue density from true value were up to 51.3%, 30.6% and 40.9% respectively for lung, bone and solid water. The magnitude of deviation was generally larger at locations closer to metal implants and greater with bilateral implants than single implant. For MV CTs, deviations of measured density from true value were less than 6% for all three tissue materials either with single or bilateral implants. Magnitude of deviation appeared to be uniform and independent of locations relative to metal implants. Conclusions: High Z metal artifacts on kV CTs can have severe impact on the accuracy of structure delineation and tissue density calculation, while on MV CTs, the impact is substantially less and insignificant. MV CTs should be considered for treatment planning on patients with high Z metal implants.

AB - Purpose: To quantitatively evaluate effects of image artifacts of hip prostheses on the accuracy of structure delineation and tissue density calculation on kV and MV CT images. Methods: Five hip prostheses made of stainless steel, titanium and cobalt chrome alloys were positioned inside a water tank and scanned respectively on a Philips CT and a Tomotherapy Hi‐Art unit. Prostheses were positioned to mimic single and bilateral implantations. Rods of tissue materials of lung, water and bone were placed at locations next and distal to metal implants near femoral head, neck and stem of prostheses. kV and MV CT scans were repeated for each placement. On CT images, cross‐sectional outlines of metal implants and tissue rods were delineated. Densities of rod materials were determined and compared to the true values. Results: Metal artifacts were severe on kV CTs and minimal on MV CTs. Cross‐sectional outlines of metal implants and tissue rods on kV CTs were severely distorted by artifacts while those on MV CTs remained clearly identifiable. For kV CTs, deviations of measured tissue density from true value were up to 51.3%, 30.6% and 40.9% respectively for lung, bone and solid water. The magnitude of deviation was generally larger at locations closer to metal implants and greater with bilateral implants than single implant. For MV CTs, deviations of measured density from true value were less than 6% for all three tissue materials either with single or bilateral implants. Magnitude of deviation appeared to be uniform and independent of locations relative to metal implants. Conclusions: High Z metal artifacts on kV CTs can have severe impact on the accuracy of structure delineation and tissue density calculation, while on MV CTs, the impact is substantially less and insignificant. MV CTs should be considered for treatment planning on patients with high Z metal implants.

UR - http://www.scopus.com/inward/record.url?scp=85024801017&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85024801017&partnerID=8YFLogxK

U2 - 10.1118/1.4734860

DO - 10.1118/1.4734860

M3 - Article

AN - SCOPUS:85024801017

VL - 39

SP - 3658

JO - Medical Physics

JF - Medical Physics

SN - 0094-2405

IS - 6

ER -