Effect of radiation dose and adaptive statistical iterative reconstruction on image quality of pulmonary computed tomography

PurposeThe purpose of this study was to assess the effects of dose and adaptive statistical iterative reconstruction (ASIR) on image quality of pulmonary computed tomography (CT).Materials and methodsInflated and fixed porcine lungs were scanned with a 64-slice CT system at 10, 20, 40 and 400 mAs. Using automatic exposure control, 40 mAs was chosen as standard dose. Scan data were reconstructed with filtered back projection (FBP) and ASIR. Image pairs were obtained by factorial combination of images at a selected level. Using a 21-point scale, three experienced radiologists independently rated differences in quality between adjacently displayed paired images for image noise, image sharpness and conspicuity of tiny nodules. A subjective quality score (SQS) for each image was computed based on Anderson’s functional measurement theory. The standard deviation was recorded as a quantitative noise measurement.ResultsAt all doses examined, SQSs improved with ASIR for all evaluation items. No significant differences were noted between the SQSs for 40%-ASIR images obtained at 20 mAs and those for FBP images at 40 mAs.ConclusionCompared to the FBP algorithm, ASIR for lung CT can enable an approximately 50% dose reduction from the standard dose while preserving visualization of small structures.