Shape-driven deep neural networks for fast acquisition of aortic 3D pressure and velocity flow fields

• Published in: PLoS computational biology Vol. 19; no. 4; p. e1011055
• Main Authors: Pajaziti, Endrit, Montalt-Tordera, Javier, Capelli, Claudio, Sivera, Raphaël, Sauvage, Emilie, Quail, Michael, Schievano, Silvia, Muthurangu, Vivek
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 01.04.2023; Public Library of Science (PLoS)
• Subjects: Aorta; Aortic valve; Artificial neural networks; Biology and Life Sciences; Blood Flow Velocity; Computational fluid dynamics; Computer applications; Computer Simulation; Coronary vessels; Correspondence; Decomposition; Fluid dynamics; Fluid flow; Hemodynamics; Humans; Hydrodynamics; Machine learning; Mathematical models; Mechanical properties; Medicine and Health Sciences; Model testing; Models, Cardiovascular; Neural networks; Neural Networks, Computer; Patients; Physical Sciences; Regression analysis; Regression models; Research and Analysis Methods; Simulation; Standard deviation; Statistical analysis; Three dimensional flow; Velocity; United Kingdom; United States > US
• ISSN: 1553-7358; 1553-734X; 1553-7358
• DOI: 10.1371/journal.pcbi.1011055

Computational fluid dynamics (CFD) can be used to simulate vascular haemodynamics and analyse potential treatment options. CFD has shown to be beneficial in improving patient outcomes. However, the implementation of CFD for routine clinical use is yet to be realised. Barriers for CFD include high computational resources, specialist experience needed for designing simulation set-ups, and long processing times. The aim of this study was to explore the use of machine learning (ML) to replicate conventional aortic CFD with automatic and fast regression models. Data used to train/test the model consisted of 3,000 CFD simulations performed on synthetically generated 3D aortic shapes. These subjects were generated from a statistical shape model (SSM) built on real patient-specific aortas (N = 67). Inference performed on 200 test shapes resulted in average errors of 6.01% ±3.12 SD and 3.99% ±0.93 SD for pressure and velocity, respectively. Our ML-based models performed CFD in ∼0.075 seconds (4,000x faster than the solver). This proof-of-concept study shows that results from conventional vascular CFD can be reproduced using ML at a much faster rate, in an automatic process, and with reasonable accuracy.