Fig. 1 | Scientific Reports

Fig. 1

From: Reduced order computational fluid dynamic simulations in the thoracic aorta are associated with disease recorded in a medical biobank

Fig. 1

Overview of the study design. Thoracic segmentations of the aorta were performed on > 3,000 CT scans using a nnU-Net. Meshes (visualized as “Ao Model”) were constructed from the segmentations using The Vascular Modeling Toolkit (VMTK) 1.4.0 (http://www.vmtk.org/). After segmentation and meshing, aortic anatomic parameters were calculated including the cross-sectional area of the vessel A(x) and the length L of the centerline from the aortic valve to the inferior thoracic vertebrae. Simulations were performed using A(x) and L as an input to the ROM, while resistance and flow conditions were permitted to vary over a physiologic range defined by the literature. Hemodynamic and anatomic traits were associated with diseases codes in the Penn Medicine Biobank (PMBB) obtained from the electronic health record (EHR), adjusting for the age and sex of the patient.

Back to article page