Abstract
Localizer orientation can significantly impact the automatic tube-current modulation (ATCM) in computed tomography (CT). The purpose of this study was to introduce a methodology for assessing ATCM consistency and to compare the outcomes in image noise and radiation dose between a single anterior-posterior (AP) localizer and a combination of lateral and AP (LAT + AP) localizers. A total of 299 prospective patients referred for a routine CT thorax examination were randomly assigned to a single AP or LAT + AP localizer. Measurements of image noise and patient size were performed in the thorax and liver, and the corresponding effective milliampere-second (mAseff) at these positions were collected. Performance analysis was conducted using the normalized root mean squared error (nRMSE) of mAseff and image noise relative to patient size. A smaller nRMSE indicates more consistent inter-patient variance in image noise and mAseff for a given size. The LAT + AP group had a statistically significant lower nRMSE for mAseff and image noise in the thorax region, and for mAseff (but not image noise) in the liver region. A post-hoc finding indicated that a subgroup of female patients with laterally protruding breasts was overexposed (+57% radiation dose) in the thorax region when using the AP localizer. Using LAT + AP localizer resulted in more consistent image noise levels and radiation doses compared to single AP localizer. The effect was small, except for the subgroup of females, where radiation doses were higher in the AP group. The findings are specific to the CT system used in this study and may not be generalizable to other CT models or software versions. Nonetheless, the proposed methodology provides a valuable approach for evaluating ATCM consistency.