Modelling pipeline for subject-specific arterial blood flow-A review

• Published in: International journal for numerical methods in biomedical engineering Vol. 27; no. 12; pp. 1868 - 1910
• Main Authors: Sazonov, Igor, Yeo, Si Yong, Bevan, Rhodri L. T., Xie, Xianghua, van Loon, Raoul, Nithiarasu, Perumal
• Format: Journal Article
• Language: English
• Published: Chichester, UK John Wiley & Sons, Ltd 01.12.2011; Wiley
• Subjects: Biological and medical sciences; Blood vessels and receptors; boundary layer; carotid bifurcation; Computational methods in fluid dynamics; Discretization; Exact sciences and technology; Finite element method; Fluid dynamics; Fundamental and applied biological sciences. Psychology; Fundamental areas of phenomenology (including applications); Hemodynamics. Rheology; Image segmentation; Mathematical models; Meshing; Modelling; Physics; pipeline; Pipelines; stenosis; subject-specific modelling; surface and volume meshing; Vertebrates: cardiovascular system; Human; Fluid dynamics; Pipe flow; Image processing; Contour line; Stenosis; Automatic mesh generation; Modeling; Artery; Image method; Blood flow; Pipeline; carotid bifurcation; subject-specific modelling; surface and volume meshing; image segmentation; Circulatory system; Hemodynamics; Mesh generation; Boundary layer
• ISSN: 2040-7939; 2040-7947; 2040-7947
• DOI: 10.1002/cnm.1446

SUMMARY In this paper, a robust and semi‐automatic modelling pipeline for blood flow through subject‐specific arterial geometries is presented. The framework developed consists of image segmentation, domain discretization (meshing) and fluid dynamics. All the three subtopics of the pipeline are explained using an example of flow through a severely stenosed human carotid artery. In the Introduction, the state‐of‐the‐art of both image segmentation and meshing is presented in some detail, and wherever possible the advantages and disadvantages of the existing methods are analysed. Followed by this, the deformable model used for image segmentation is presented. This model is based upon a geometrical potential force (GPF), which is a function of the image. Both the GPF calculation and level set determination are explained. Following the image segmentation method, a semi‐automatic meshing method used in the present study is explained in full detail. All the relevant techniques required to generate a valid domain discretization are presented. These techniques include generating a valid surface mesh, skeletonization, mesh cropping, boundary layer mesh construction and various mesh cosmetic methods that are essential for generating a high‐quality domain discretization. After presenting the mesh generation procedure, how to generate flow boundary conditions for both the inlets and outlets of a geometry is explained in detail. This is followed by a brief note on the flow solver, before studying the blood flow through the carotid artery with a severe stenosis. Copyright © 2011 John Wiley & Sons, Ltd.