Abstract：
OBJECTIVE: Three-dimensional (3D) modelling of aortic leaflets remains difficult due to insufficient resolution of medical imaging. We aimed to model the coaptation and load-bearing surfaces of the aortic leaflets and adapt this workflow to aid in the design of aortic valve neocuspidizations.
METHODS: Geometric morphometrics, using landmarks and semilandmarks, was applied to the geometric determinants of the aortic leaflets from computed tomography, followed by an isogeometric analysis using Non-Uniform Rational Basis Splines (NURBS). Ten aortic valve models were generated, measuring determinants of leaflet geometry defined as 3D NURBS curves, and leaflet coaptation and load-bearing surfaces were defined as 3D NURBS surfaces. Neocuspidizations were obtained by either shifting the upper central coaptation landmark towards the sinotubular junction or using parametric neo-landmarks placed on a centerline drawn between the centroid of the aortic root base and centroid of a circle circumscribing the three upper commissural landmarks.
RESULTS: The ratio of the leaflet free margin length to the geometric height was 1.83, whereas the ratio of the commissural coaptation height to the central coaptation height was 1.93. The median coaptation surface was 137 mm2 (IQR 58) and the median load-bearing surface was 203 mm2 (60) per leaflet. Neocuspidization multiplied the central coaptation height by 3.7 and the coaptation surfaces by 1.97 and 1.92 using the native coaptation axis and centroid coaptation axis, respectively.
CONCLUSIONS: Geometric Morphometrics reliably defined the coaptation and load-bearing surfaces of aortic leaflets, enabling an experimental 3D design for the in silico neocuspidization of aortic valves.
METHODS: Geometric morphometrics, using landmarks and semilandmarks, was applied to the geometric determinants of the aortic leaflets from computed tomography, followed by an isogeometric analysis using Non-Uniform Rational Basis Splines (NURBS). Ten aortic valve models were generated, measuring determinants of leaflet geometry defined as 3D NURBS curves, and leaflet coaptation and load-bearing surfaces were defined as 3D NURBS surfaces. Neocuspidizations were obtained by either shifting the upper central coaptation landmark towards the sinotubular junction or using parametric neo-landmarks placed on a centerline drawn between the centroid of the aortic root base and centroid of a circle circumscribing the three upper commissural landmarks.
RESULTS: The ratio of the leaflet free margin length to the geometric height was 1.83, whereas the ratio of the commissural coaptation height to the central coaptation height was 1.93. The median coaptation surface was 137 mm2 (IQR 58) and the median load-bearing surface was 203 mm2 (60) per leaflet. Neocuspidization multiplied the central coaptation height by 3.7 and the coaptation surfaces by 1.97 and 1.92 using the native coaptation axis and centroid coaptation axis, respectively.
CONCLUSIONS: Geometric Morphometrics reliably defined the coaptation and load-bearing surfaces of aortic leaflets, enabling an experimental 3D design for the in silico neocuspidization of aortic valves.
摘要：
目的：由于医学成像分辨率不足，主动脉小叶的三维（3D）建模仍然很困难。我们旨在对主动脉瓣叶的接合和承重表面进行建模，并调整此工作流程以帮助设计主动脉瓣新胸瓣。
方法：几何形态计量学，使用地标和半地标，应用于计算机断层扫描的主动脉瓣叶的几何决定因素，然后使用非均匀有理基准样条(NURBS)进行等几何分析。生成了十个主动脉瓣模型，定义为3DNURBS曲线的小叶几何形状的测量决定因素，小叶接合和承重表面被定义为3DNURBS表面。通过将上中央接合标志移向窦管交界处或使用参数新标志放置在主动脉根基部的质心和围绕三个上连合标志的圆的质心之间的中心线上，可以获得新的穿孔。
结果：小叶自由边缘长度与几何高度之比为1.83，而连合高度与中央接合高度之比为1.93。每个小叶的中位接合表面为137mm2（IQR58），中位承载表面为203mm2（60）。使用天然对合轴和质心对合轴，新对合将中心对合高度乘以3.7，将对合表面乘以1.97和1.92，分别。
结论：几何形态测量可靠地定义了主动脉瓣叶的接合和承重表面，为主动脉瓣的计算机新穿孔化进行实验性3D设计。
方法：几何形态计量学，使用地标和半地标，应用于计算机断层扫描的主动脉瓣叶的几何决定因素，然后使用非均匀有理基准样条(NURBS)进行等几何分析。生成了十个主动脉瓣模型，定义为3DNURBS曲线的小叶几何形状的测量决定因素，小叶接合和承重表面被定义为3DNURBS表面。通过将上中央接合标志移向窦管交界处或使用参数新标志放置在主动脉根基部的质心和围绕三个上连合标志的圆的质心之间的中心线上，可以获得新的穿孔。
结果：小叶自由边缘长度与几何高度之比为1.83，而连合高度与中央接合高度之比为1.93。每个小叶的中位接合表面为137mm2（IQR58），中位承载表面为203mm2（60）。使用天然对合轴和质心对合轴，新对合将中心对合高度乘以3.7，将对合表面乘以1.97和1.92，分别。
结论：几何形态测量可靠地定义了主动脉瓣叶的接合和承重表面，为主动脉瓣的计算机新穿孔化进行实验性3D设计。
