.png){kind=icon}
.png){kind=icon}
Return
Improving aortic calcium assessment in contrast-enhanced CT: exploring Hounsfield Unit thresholds for cardiovascular scoring
.png){kind=icon}
PRE
AI
P
C
K
V
J
E
S
T
K
T
K
L
DOI：10.1007/s11547-026-02223-y.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Cardiovascular disease is the leading global cause of death, and vascular calcification is a key marker of atherosclerotic burden. This study aimed to determine optimal Hounsfield Unit (HU) thresholds for quantifying aortic calcification on contrast-enhanced CT. This retrospective single-center study included 1125 patients (344 women, 781 men) who underwent triphasic CT angiography. Fixed (100–900 HU) and dynamic thresholds were applied to arterial and venous phases and compared with non-contrast Agatston scores. Correlation, linear regression and Bland–Altman analyses were used to assess agreement and predictive performance. In the venous phase, the 300 HU threshold demonstrated the strongest correlation with the Agatston score (thoracic aorta Spearman’s rank correlation coefficient ρ = 0.93; abdominal aorta: ρ = 0.92; all p < 0.001) and the highest predictive accuracy (R2 = 0.66 and 0.68). In the arterial phase, higher thresholds were required, with optimal performance at 900 HU for the thoracic aorta (R2 = 0.51, p < 0.001) and 800 HU for the abdominal aorta (R2 = 0.45, p < 0.001). Dynamic thresholding showed lower R2 values than optimal fixed thresholds. Bland-Altman analysis for the venous 300 HU threshold showed low bias at low calcium burden but widening limits of agreement and increasing underestimation at higher Agatston scores. A fixed threshold of 300 HU in the venous phase may represent a pragmatic surrogate marker for estimating aortic calcification on contrast-enhanced CT when non-contrast imaging is unavailable.
Keywords:
Atherosclerosis
Computed tomography
Agatston score
Volumetric
Body composition
Blood vessel
Journal
IF:
4.8
Papers:
206
Citations:
0
Organization
Cited Papers