Localiser-Based Software-Assisted Patient Positioning in Chest CT: A Propensity Score-Matched Study

Background/Objectives: Vertical off-centring in chest CT impairs bowtie-filter beam-hardening correction, degrading noise uniformity and image quality. iCentering derives the patient’s geometric centre from a single posteroanterior localiser and automatically adjusts table height. This study aimed to evaluate whether iCentering reduces off-centring distance (OCD) and improves image quality versus manual positioning in chest CT, and to investigate correlations between OCD and noise characteristics.

Methods

In this retrospective, single-centre study, 192 propensity score-matched pairs from 423 adults undergoing non-contrast chest CT were analysed. OCD, objective image quality (standard deviation (SD), signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), figure of merit (FOM)), noise distribution ratio (NDRA-P), and radiation dose indices were measured. Paired comparisons employed t-tests or Wilcoxon signed-rank tests with 95% confidence intervals (CIs) and effect sizes reported. Directionally stratified Spearman correlation and multivariable regression examined OCD–noise associations.