Performance characteristics of a novel megavoltage cone-beam-computed tomography device

Phys Med Biol. 2012 Feb 7;57(3):N15-24. doi: 10.1088/0031-9155/57/3/N15. Epub 2012 Jan 18.

Abstract

In this work, the image quality of a novel megavoltage cone-beam-computed tomography (CBCT) scanner is compared to three other image-guided radiation therapy devices by analysing images of different-sized quality assurance phantoms. The following devices are compared in terms of image uniformity, signal-to-noise ratio, contrast-to-noise ratio (CNR), electron density to HU conversion, presampling modulation transfer function (MTF(pre)) and combined spatial resolution and noise (Q-factor): (i) the Siemens Artiste kilovoltage (kV) (121 kV) CBCT device, (ii) the Artiste treatment beam line (TBL), 6 MV, (iii) the Tomotherapy (3.5 MV) fan-beam CT and (iv) Siemens' novel approach using a carbon target for a dedicated imaging beam line (IBL), 4.2 MV. Machine settings were selected to produce the same imaging dose for all devices. For a head phantom, IBL scans display CNR values 2.6 ± 0.3 times higher than for the TBL at the same dose level (for a CT-number range of -200 to -60 HU). kV CBCT, on the other hand, displays CNR values 7.9 ± 0.3 times higher than the IBL. There was no significant deviation in spatial resolution between IBL, TBL and Tomotherapy in terms of 50% and 10% MTF(pre). For kV CBCT, the MTF(pre) was significantly higher than those for other devices. In our Q-factor analysis, the IBL (14.6) scores higher than the TBL (7.9) and Tomotherapy (9.7) due to its lower noise level. The linearity of electron density to HU conversion is demonstrated for different-sized phantoms. Employing the IBL instead of the TBL significantly reduces the imaging dose by up to a factor of 5 at a constant image quality level, providing an immediate benefit for the patient.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Cone-Beam Computed Tomography / methods*
  • Diagnostic Imaging / methods
  • Equipment Design
  • Head / diagnostic imaging
  • Humans
  • Image Processing, Computer-Assisted
  • Models, Statistical
  • Pelvis / diagnostic imaging
  • Phantoms, Imaging
  • Radiotherapy / methods*
  • Signal-To-Noise Ratio
  • Tomography Scanners, X-Ray Computed