[Dose optimization for multislice computed tomography protocols of the midface]

  • Martin Lorenzen
  • U Wedegärtner
  • C Weber
  • Ute Lockemann
  • G Adam
  • J Lorenzen

Abstract

PURPOSE: To optimize multislice computed tomography (MSCT) protocols of the midface for dose reduction and adequate image quality. MATERIALS AND METHODS: MSCT (Somatom Volume Zoom, Siemens) of the midface was performed on 3 cadavers within 24 hours of death with successive reduction of the tube current, applying 150, 100, 70 and 30 mAs at 120 kV as well as 40 and 21 mAs at 80 kV. At 120 kV, a pitch of 0.875 and collimation of 4x1 mm were used, and at 80 kV, a pitch of 0.7 and collimation of 2x0.5 mm. Images were reconstructed in transverse and coronal orientation. Qualitative image analysis was separately performed by two radiologists using a five-point scale (1 = excellent; 5 = poor) applying the following parameters: image quality, demarcation and sharpness of lamellar bone, overall image quality, and image noise (1 = minor; 5 = strong). The effective body dose [mSv] and organ dose [mSv] of the ocular lens (using the dosimetry system "WINdose") were calculated, and the interobserver agreement (kappa coefficient) was determined. RESULTS: For the evaluation of the lamellar bone, adequate sharpness, demarcation and image quality was demonstrated at 120 kV/30 mAs, and for the overall image quality and noise, 120 kV/40 mAs was acceptable. With regard to image quality, the effective body dose could be reduced from 1.89 mSv to 0.34 mSv and the organ dose of the ocular lens from 27.2 mSv to 4.8 mSv. Interobserver agreement was moderate (kappa = 0.39). CONCLUSION: Adequate image quality was achieved for MSCT protocols of the midface with 30 mAs at 120 kV, resulting in a dose reduction of 70 % in comparison to standard protocols.

Bibliographical data

Original languageGerman
Article number2
ISSN1438-9029
Publication statusPublished - 2005
pubmed 15666236