Impact of scanning parameters on deep learning coronary artery calcium scoring in non‑gated chest CT.

backgroundTo evaluate the accuracy of deep learning–derived coronary artery calcium scores (DL‑CACS) from non–electrocardiogram (ECG)-gated chest CT against ECG‑gated reference CACS and to determine the impact of CT acquisition parameters on measurement accuracy.

methodsFrom January 2020 to February 2021, 1213 patients at our institution underwent ECG‑gated cardiac CT and non‑gated chest CT within 3 months. An automated pipeline generated DL‑CACS from non‑gated scans. Agreement with ECG‑gated CACS was assessed using Spearman correlation and Bland–Altman analysis; kappa analysis evaluated categorical agreement for the Coronary Artery Calcium Data and Reporting System (CAC‑DRS). Diagnostic performance was evaluated by ROC/AUC. Univariable and multivariable logistic regression quantified associations between acquisition parameters and misclassification.

resultsNon‑gated DL‑CACS correlated strongly with gated CACS across LM, LAD, LCX, RCA and TOTAL scores (ρ ≤ 0.865; all p < 0.001). CAC‑DRS agreement was substantial (κ = 0.641), with accuracy 77.5% and AUC 0.769 (95% CI 0.754–0.785). Collimation width, rotation time, reconstruction slice thickness and reconstruction kernel were independent predictors of misclassification; tube voltage and pitch were not significant. Narrower collimation, slower rotation, thicker slices and smoother kernels were associated with higher odds of misclassification.

conclusionsDL‑based CACS from non–ECG‑gated chest CT shows good agreement with ECG‑gated CACS and useful accuracy, although performance remains sensitive to acquisition settings. In the future, protocol standardisation and algorithm robustness across vendors and parameters may further enhance clinical utility, supporting more consistent performance across sites and facilitating broader adoption in routine practice.