A machine learning approach as a surrogate of finite element analysis–based inverse method to estimate the zero‐pressure geometry of human thoracic aorta

Abstract

Advances in structural finite element analysis (FEA) and medical imaging have made it possible to investigate the in vivo biomechanics of human organs such as blood vessels, for which organ geometries at the zero-pressure level need to be recovered. Although FEA-based inverse methods are available for zero-pressure geometry estimation, these methods typically require iterative computation, which are time-consuming and may be not suitable for time-sensitive clinical applications. In this study, by using machine learning (ML) techniques, we developed an ML model to estimate the zero-pressure geometry of human thoracic aorta given 2 pressurized geometries of the same patient at 2 different blood pressure levels. For the ML model development, a FEA-based method was used to generate a dataset of aorta geometries of 3125 virtual patients. The ML model, which was trained and tested on the dataset, is capable of recovering zero-pressure geometries consistent with those generated by the FEA-based method. Thus, this study demonstrates the feasibility and great potential of using ML techniques as a fast surrogate of FEA-based inverse methods to recover zero-pressure geometries of human organs.

The impact of balloon-expandable transcatheter aortic valve replacement on concomitant mitral regurgitation: a comprehensive computational analysis

“…This assumption, which will be tacked in a future study, could lead to uncertainties on the computed stresses as well as in the tissue compliance [48][49][50]. The pre-stress assumption of the native leaflets might have a small impact on the solution since the transvalvular pressure is very small during TAV deployment.…”

Cancer classification from microarray data for genomic disorder research using optimal discriminant independent component analysis and kernel extreme learning machine

“…The unit covariance with the sum of the marginal Negentrophy maximization is performed via Lagrange, which is shown in Equation (13)…”

“…Depending upon the given D andL M ð Þ, the feature X with its classification accuracy is measured using a function ψ, and k is the constant, which is similar to Equation (13). The learning rule is explained as below.…”

“…Moreover, the significant genes are detected by data mining, supervised and unsupervised machine learning approaches. [10][11][12][13] Because of the small sample size, the number of genes and noise is to affect the above approaches. The lower-dimensional space features are obtained using Discriminant Independent Component and the performance criteria are computed with the Fisher criterion.…”

Intelligent Calibration of Static FEA Computations Based on Terrestrial Laser Scanning Reference

“…The application of the auto encoder deep learning approach indicates a lower reconstruction error, which is compared with the original FEA model. The similar idea can also be applied to estimate, calibrate, and recover the zero-pressure geometry of the patient's thoracic aorta is developed on the basis of the FEA method [30]. The application of different machine learning methods can simplify, predict, and calibrate the complex FEA computations, thus, solving medical problems.…”