Scholarlink

Background <p>Accurate quantification of aortic valve calcification (AVC) on contrast-enhanced computed tomography angiography (CTA) is pivotal for planning surgical and transcatheter aortic valve replacement. The optimal Hounsfield unit (HU) threshold for calcification detection on contrast-enhanced images remains unresolved, and every prior validation study has relied on non-contrast Agatston scoring—itself an imaging estimate—as the reference standard. This study validated two widely used fixed HU thresholds (450 HU and 850 HU) and a self-configuring nnU-Net deep learning model against ex vivo gravimetric calcium weight as an absolute physical ground truth.</p> Methods <p>Four hundred patients were included in a retrospective cohort study with a pre-specified temporal validation split: 300 with CT-confirmed AVC and 100 with normal aortic valves. Fifty chronologically later AVC patients who underwent elective open surgical aortic valve replacement (SAVR) within seven days of clinically indicated pre-operative contrast-enhanced CTA formed the locked surgical validation cohort; their excised native leaflets underwent standardised high-temperature ashing (550&#xa0;°C, 12&#xa0;h) and analytical weighing (precision 0.1&#xa0;mg) to obtain gravimetric calcium mass. The remaining 350 cases served exclusively for nnU-Net development (280 training / 70 internal validation). CT-derived calcium mass-equivalent estimates were quantified on the validation cohort and compared with gravimetric weight using Pearson and Spearman correlation and Bland-Altman analysis.</p> Results <p>The nnU-Net achieved the strongest observed correlation with gravimetric weight (Pearson <i>r</i> = 0.967; bias + 6.2&#xa0;mg; RMSE 13.7&#xa0;mg), significantly outperforming the 450 HU threshold for correlation (<i>r</i> = 0.864; bias + 36.2&#xa0;mg; RMSE 42.1&#xa0;mg; Steiger <i>p</i> &lt; 0.001) and showing a non-significant trend toward stronger correlation than 850 HU (<i>r</i> = 0.929; bias + 17.5&#xa0;mg; RMSE 23.9&#xa0;mg; Steiger <i>p</i> = 0.085). Compared with 850 HU, nnU-Net provided lower bias and RMSE, although the difference in Pearson r did not reach statistical significance. The 450 HU method exhibited significant proportional bias (<i>p</i> = 0.024), whereas neither 850 HU nor nnU-Net did. The nnU-Net achieved a mean Dice coefficient of 0.873 and intersection-over-union of 0.812.</p> Conclusions <p>Against physically weighed calcium, nnU-Net deep learning segmentation provided the most favourable overall performance profile on contrast-enhanced CTA, with the lowest bias and RMSE and the strongest observed correlation. The improvement in Pearson correlation over 850 HU represented a non-significant trend, whereas the error and agreement metrics favoured nnU-Net. Among fixed thresholds, 850 HU substantially outperformed 450 HU, offering direct physical-rather than surrogate imaging-evidence to support 850 HU as the preferred fixed threshold in standard contrast-enhanced protocols.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Validation of aortic valve calcification quantification on contrast-enhanced computed tomography against ex vivo gravimetric analysis: comparison of fixed Hounsfield unit thresholds and deep learning segmentation

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月20日

查看全文
  • Daisong Jiang,
  • Wanting Zhang,
  • Lulu Liu,
  • Geng Li,
  • Xiaoke Shang,
  • Nianguo Dong,
  • Ye Yang,
  • Jian Yang,
  • Haibo Zhang,
  • Qing Zhou,
  • Chao Jian,
  • Yuan Zhao,
  • Buqing Ni,
  • Yongfeng Shao,
  • Jian Liu,
  • Zhong Wu

摘要

Background

Accurate quantification of aortic valve calcification (AVC) on contrast-enhanced computed tomography angiography (CTA) is pivotal for planning surgical and transcatheter aortic valve replacement. The optimal Hounsfield unit (HU) threshold for calcification detection on contrast-enhanced images remains unresolved, and every prior validation study has relied on non-contrast Agatston scoring—itself an imaging estimate—as the reference standard. This study validated two widely used fixed HU thresholds (450 HU and 850 HU) and a self-configuring nnU-Net deep learning model against ex vivo gravimetric calcium weight as an absolute physical ground truth.

Methods

Four hundred patients were included in a retrospective cohort study with a pre-specified temporal validation split: 300 with CT-confirmed AVC and 100 with normal aortic valves. Fifty chronologically later AVC patients who underwent elective open surgical aortic valve replacement (SAVR) within seven days of clinically indicated pre-operative contrast-enhanced CTA formed the locked surgical validation cohort; their excised native leaflets underwent standardised high-temperature ashing (550 °C, 12 h) and analytical weighing (precision 0.1 mg) to obtain gravimetric calcium mass. The remaining 350 cases served exclusively for nnU-Net development (280 training / 70 internal validation). CT-derived calcium mass-equivalent estimates were quantified on the validation cohort and compared with gravimetric weight using Pearson and Spearman correlation and Bland-Altman analysis.

Results

The nnU-Net achieved the strongest observed correlation with gravimetric weight (Pearson r = 0.967; bias + 6.2 mg; RMSE 13.7 mg), significantly outperforming the 450 HU threshold for correlation (r = 0.864; bias + 36.2 mg; RMSE 42.1 mg; Steiger p < 0.001) and showing a non-significant trend toward stronger correlation than 850 HU (r = 0.929; bias + 17.5 mg; RMSE 23.9 mg; Steiger p = 0.085). Compared with 850 HU, nnU-Net provided lower bias and RMSE, although the difference in Pearson r did not reach statistical significance. The 450 HU method exhibited significant proportional bias (p = 0.024), whereas neither 850 HU nor nnU-Net did. The nnU-Net achieved a mean Dice coefficient of 0.873 and intersection-over-union of 0.812.

Conclusions

Against physically weighed calcium, nnU-Net deep learning segmentation provided the most favourable overall performance profile on contrast-enhanced CTA, with the lowest bias and RMSE and the strongest observed correlation. The improvement in Pearson correlation over 850 HU represented a non-significant trend, whereas the error and agreement metrics favoured nnU-Net. Among fixed thresholds, 850 HU substantially outperformed 450 HU, offering direct physical-rather than surrogate imaging-evidence to support 850 HU as the preferred fixed threshold in standard contrast-enhanced protocols.