Effect of patient centering on patient dose and image noise in chest CT

Abstract

Objective: The objective of our study was to evaluate the effect of vertical centering on dose and image noise in chest MDCT of different-sized patients using anthropomorphic phantoms and retrospectively studying examinations of clinical patients.

Materials and methods: Three different anthropomorphic phantoms were scanned using different vertical centering (offset ± 6 cm) and were assessed with radiation dose-monitoring software. The effect of vertical positioning on the radiation dose was studied using the volume CT dose index, dose-length product, and size-specific dose estimates for different-sized phantoms. Image noise was determined from CT number histograms. Vertical positioning for chest CT examinations of 112 patients ranging from neonates to adults were retrospectively assessed.

Results: Radiation doses were highest when using the posteroanterior scout image for automatic exposure control (AEC) and when phantoms were set in the lowest table position, and radiation doses were lowest when phantoms were set in the uppermost table position. For the adult phantom, relative doses increased by 38% in the lowest table position and decreased by 23% in the highest table position. Similarly, doses for pediatric 5-year-old and newborn phantoms were 21% and 12% higher in the lowest table position and 12% and 8% lower in the highest table position, respectively. The effect decreased when a lateral scout image was used for AEC. The relative noise was lowest when the phantoms were properly centered and increased with vertical offset. In clinical patients, we observed offset with a median value varying from 25 to 35 mm below the isocenter.

Conclusion: Regardless of patient size, most patients in this study were positioned too low, which negatively affected both patient dose and image noise. Miscentering was more pronounced in smaller pediatric patients.

Keywords: chest CT; image quality; off-centering; optimization; size-specific dose estimate.