Model based iterative reconstruction IMR gives possibility to evaluate thinner slice thicknesses than conventional iterative reconstruction iDose(4) - a phantom study

• Computed tomography (CT) is one of the most important modalities in a radiological department, which produces images with high diagnostic confidence, but in some cases contributes to a high radiation dose to the patient. The radiation dose can be reduced by the use of advanced image reconstruction algorithms. This study was done on a Philips Brilliance iCT with iterative reconstruction iDose(4) and model-based iterative reconstruction IMR. The purpose was to investigate the effect on the image quality with thin slice images reconstructed with IMR, compared to standard slice thickness reconstructed with iDose(4). Objective measurements of noise and contrast-to-noise ratio were performed using an image quality phantom, an anthropomorphic phantom and clinical cases. Subjective evaluations of low-contrast resolution were performed by observers using an image quality phantom. IMR gives strong noise reduction and enhanced low-contrast and thereby enable selection of thinner slice thickness. Objective evaluation of image noise shows that thin slices reconstructed with IMR provides lower noise than thicker slice images reconstructed with iDose(4). With IMR the slice thickness is of less importance for the noise. With thinner slices the partial volume artefacts becomes less pronounced. In conclusion, we have shown that IMR enables reduction of the slice thickness and at the same time maintain or even reduce the noise level compared to iDose(4) reconstruction with standard slice thickness. This will subsequently result in an improvement of image quality for images reconstructed with IMR.

Subject headings

MEDICIN OCH HÄLSOVETENSKAP  -- Klinisk medicin -- Radiologi och bildbehandling (hsv//swe)

MEDICAL AND HEALTH SCIENCES  -- Clinical Medicine -- Radiology, Nuclear Medicine and Medical Imaging (hsv//eng)

Keyword

computed tomography

iterative reconstruction

image quality

Publication and Content Type

kon (subject category)

ref (subject category)