Efficiency of a predictive surrogate model for hemodynamic predictions of blood flow in an idealized carotid artery stenosis DOI
Gang He, Zhang Li,

Li-Cai Zhao

и другие.

Physics of Fluids, Год журнала: 2024, Номер 36(12)

Опубликована: Дек. 1, 2024

This study presents a predictive surrogate model (PSM) for predicting hemodynamic variables in idealized carotid artery stenosis. The PSM integrates long short-term memory (LSTM) and proper orthogonal decomposition (POD) techniques. model's accuracy is evaluated two different stenosis conditions the For simulation of full-order stenosis, fluid–structure interaction (FSI) solver employed to between blood vessel wall. Casson used estimate viscosity non-Newtonian flow. These are selected accurately capture hemodynamics across various conditions. examines pressure, wall shear stress (WSS), velocity components, oscillatory index (OSI) variables. reconstruction error reduced order calculated based on chosen number POD modes. It noteworthy that OSI significantly higher than other components derivatives (i.e., WSS) both LSTM under conditions, showing promising results despite inherent complexities physiological situations. While effectively predicts WSS indices with reliable scales, exhibits slightly larger errors.

Язык: Английский

A predictive surrogate model of blood haemodynamics for patient-specific carotid artery stenosis DOI
M. Barzegar Gerdroodbary, Sajad Salavatidezfouli

Journal of The Royal Society Interface, Год журнала: 2025, Номер 22(224)

Опубликована: Март 1, 2025

In this study, the haemodynamic factors inside patient-specific carotid artery with stenosis are evaluated via a predictive surrogate model. The technique of proper orthogonal decomposition (POD) is used for reducing order main model and consequently, long short-term memory employed prediction blood flow parameters, i.e. velocity pressure along stenosis. efficiency proposed machine learning has been in arteries with/without Besides, reconstruction error analysis performed different POD mode numbers. Our results demonstrate that value at stages cardiac cycle great impact on method estimation haemodynamics. presence intensifies complexity flow, magnitude errors increased when exists artery.

Язык: Английский

Процитировано

3

A predictive surrogate model based on linear and nonlinear solution manifold reduction in cardiovascular FSI: A comparative study DOI Creative Commons
M. Barzegar Gerdroodbary, Sajad Salavatidezfouli

Computers in Biology and Medicine, Год журнала: 2025, Номер 189, С. 109959 - 109959

Опубликована: Март 5, 2025

This study investigates the fluid-structure interaction (FSI) simulation of abdominal aorta, with a particular focus on hemodynamic alterations induced by aneurysmal deformations. The behavior within aorta is highly dependent geometric characteristics aneurysm, necessitating use patient-specific models to ensure accurate predictions. primary objective this research enhance predictive capability flow and structural indices in complex FSI biomechanical setting under varying physiological conditions, namely rest exercise states. paper presents comparative analysis between two distinct yet promising surrogate models: Proper Orthogonal Decomposition coupled Long Short-Term Memory (POD + LSTM) Convolutional Neural Network combined (CNN LSTM). methodology, model selection, performance are discussed detail, providing insights into efficacy limitations each approach context personalized cardiovascular simulations.

Язык: Английский

Процитировано

2

Linear Surrogate Modelling for Predicting Hemodynamic in Carotid Artery Stenosis During Exercise Conditions DOI
Feng Wang,

Wensheng Shi,

Haibin Zhang

и другие.

Chinese Journal of Physics, Год журнала: 2025, Номер unknown

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

0

Estimation of the fuel mixing of annular extruded fuel multi-jets in cavity flame holder at the supersonic combustion chamber via predictive surrogate model DOI

Dechen Wei,

Yuanyuan Jiao, Yukun Fan

и другие.

Engineering Analysis with Boundary Elements, Год журнала: 2024, Номер 163, С. 369 - 377

Опубликована: Март 27, 2024

Язык: Английский

Процитировано

1

Computational Modeling Approach to Profile Hemodynamical Behavior in a Healthy Aorta DOI Creative Commons
Ahmed M. Al‐Jumaily, Mohammad Al‐Rawi, Djelloul Belkacemi

и другие.

Bioengineering, Год журнала: 2024, Номер 11(9), С. 914 - 914

Опубликована: Сен. 12, 2024

Cardiovascular diseases (CVD) remain the leading cause of mortality among older adults. Early detection is critical as prognosis for advanced-stage CVD often poor. Consequently, non-invasive diagnostic tools that can assess hemodynamic function, particularly aorta, are essential. Computational fluid dynamics (CFD) has emerged a promising method simulating cardiovascular efficiently and cost-effectively, using increasingly accessible computational resources. This study developed CFD model to aorta geometry tetrahedral polyhedral meshes. A healthy was modeled with mesh sizes ranging from 0.2 1 mm. Key parameters, including blood pressure waveform, difference, wall shear stress (WSS), associated parameters like relative residence time (RRT), oscillatory index (OSI), endothelial cell activation potential (ECAP) were evaluated. The performance simulations, focusing on accuracy processing time, assessed determine clinical viability. demonstrated clinically acceptable results, achieving over 95% while reducing simulation by up 54%. entire process, image construction post-processing completed in under 120 min. Both types (tetrahedral polyhedral) provided reliable outputs analysis. provides novel demonstration impact type obtaining accurate data, quickly efficiently, simulations aortic assessments. beneficial routine check-ups, offering improved diagnostics populations limited healthcare access or higher disease risk.

Язык: Английский

Процитировано

0

Efficiency of a predictive surrogate model for hemodynamic predictions of blood flow in an idealized carotid artery stenosis DOI
Gang He, Zhang Li,

Li-Cai Zhao

и другие.

Physics of Fluids, Год журнала: 2024, Номер 36(12)

Опубликована: Дек. 1, 2024

This study presents a predictive surrogate model (PSM) for predicting hemodynamic variables in idealized carotid artery stenosis. The PSM integrates long short-term memory (LSTM) and proper orthogonal decomposition (POD) techniques. model's accuracy is evaluated two different stenosis conditions the For simulation of full-order stenosis, fluid–structure interaction (FSI) solver employed to between blood vessel wall. Casson used estimate viscosity non-Newtonian flow. These are selected accurately capture hemodynamics across various conditions. examines pressure, wall shear stress (WSS), velocity components, oscillatory index (OSI) variables. reconstruction error reduced order calculated based on chosen number POD modes. It noteworthy that OSI significantly higher than other components derivatives (i.e., WSS) both LSTM under conditions, showing promising results despite inherent complexities physiological situations. While effectively predicts WSS indices with reliable scales, exhibits slightly larger errors.

Язык: Английский

Процитировано

0