Deep Learning for Computational Hemodynamics: A Brief Review of Recent Advances

Taebi, Amirtahà

doi:10.3390/fluids7060197

Cited by 21 publications

(13 citation statements)

References 93 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A patient-specific simulation investigation, incorporating different boundary conditions, can also deepen the insights into this matter. Furthermore, a study involving a larger population employing machine learning techniques [ 60 , 61 ] can facilitate the exploration of the hypotheses generated in the present study.…”

Section: Resultsmentioning

confidence: 99%

Stress Load and Ascending Aortic Aneurysms: An Observational, Longitudinal, Single-Center Study Using Computational Fluid Dynamics

de Azevedo,

Almeida,

Alvares de Azevedo

et al. 2024

Bioengineering

View full text Add to dashboard Cite

Ascending aortic aneurysm (AAoA) is a silent disease with high mortality; however, the factors associated with a worse prognosis are not completely understood. The objective of this observational, longitudinal, single-center study was to identify the hemodynamic patterns and their influence on AAoA growth using computational fluid dynamics (CFD), focusing on the effects of geometrical variations on aortic hemodynamics. Personalized anatomic models were obtained from angiotomography scans of 30 patients in two different years (with intervals of one to three years between them), of which 16 (53%) showed aneurysm growth (defined as an increase in the ascending aorta volume by 5% or more). Numerically determined velocity and pressure fields were compared with the outcome of aneurysm growth. Through a statistical analysis, hemodynamic characteristics were found to be associated with aneurysm growth: average and maximum high pressure (superior to 100 Pa); average and maximum high wall shear stress (superior to 7 Pa) combined with high pressure (>100 Pa); and stress load over time (maximum pressure multiplied by the time interval between the exams). This study provides insights into a worse prognosis of this serious disease and may collaborate for the expansion of knowledge about mechanobiology in the progression of AAoA.

show abstract

Section: Resultsmentioning

confidence: 99%

Stress Load and Ascending Aortic Aneurysms: An Observational, Longitudinal, Single-Center Study Using Computational Fluid Dynamics

de Azevedo,

Almeida,

Alvares de Azevedo

et al. 2024

Bioengineering

View full text Add to dashboard Cite

show abstract

“…In this study, CFD analysis was performed using the commercial finite element software package ANSYS Workbench 18.2 (Swanson Analysis Systems, Inc., Canonsburg, PA, USA). To calculate flow fields and hemodynamic parameters, blood flow in the iliac veins assumed incompressible Newtonian flow [ 25 ]. Values for blood fluid density and dynamic viscosity (1050 kg/m 3 and 0.003 kg/m/s, respectively) were used according to previous studies [ 32 , 33 , 34 , 35 ].…”

Section: Methodsmentioning

confidence: 99%

“…Hemodynamics, the study of flow changes in the blood using physiology and fluid dynamics, provides quantitative information such as flow resistance, velocity, and pressure that can be used for diagnosing and treating cardiocirculatory system diseases [ 24 , 25 ]. Hemodynamics can be studied using computational fluid dynamics (CFD) [ 26 ], and recent developments in computers and medical imaging software have made the CFD analysis of blood flow both reliable and accessible to clinicians.…”

Section: Introductionmentioning

confidence: 99%

CFD Study of the Effect of the Angle Pattern on Iliac Vein Compression Syndrome

Chen

Fan

et al. 2023

Bioengineering

View full text Add to dashboard Cite

Iliac vein compression syndrome (IVCS, or May–Thurner syndrome) occurs due to the compression of the left common iliac vein between the lumbar spine and right common iliac artery. Because most patients with compression are asymptomatic, the syndrome is difficult to diagnose based on the degree of anatomical compression. In this study, we investigated how the tilt angle of the left common iliac vein affects the flow patterns in the compressed blood vessel using three-dimensional computational fluid dynamic (CFD) simulations to determine the flow fields generated after compression sites. A patient-specific iliac venous CFD model was created to verify the boundary conditions and hemodynamic parameter set in this study. Thirty-one patient-specific CFD models with various iliac venous angles were developed using computed tomography (CT) angiograms. The angles between the right or left common iliac vein and inferior vena cava at the confluence level of the common iliac vein were defined as α1 and α2. Flow fields and vortex locations after compression were calculated and compared according to the tilt angle of the veins. Our results showed that α2 affected the incidence of flow field disturbance. At α2 angles greater than 60 degrees, the incidence rate of blood flow disturbance was 90%. In addition, when α2 and α1 + α2 angles were used as indicators, significant differences in tilt angle were found between veins with laminar, transitional, and turbulent flow (p < 0.05). Using this mathematical simulation, we concluded that the tilt angle of the left common iliac vein can be used as an auxiliary indicator to determine IVCS and its severity, and as a reference for clinical decision making.

show abstract

“…Given the powerful feature-extraction capabilities in multidomain regression and pattern recognition, both machine learning (ML) and deep learning (DL) methods have shown successful applications in various fields, such as physiological signal diagnosis, medical image separation, smart medical care, etc ( LeCun et al, 2015 ; Li et al, 2019 ; Mittal et al, 2019 ; Noorbakhsh-Sabet et al, 2019 ; Bhandary et al, 2020 ; Li et al, 2021b ; Wang et al, 2022 ). The ML and DL-based methodology is also considered as an alternative to the CFD method for blood flow analysis ( Taebi, 2022 ) because it is of high potential to implement the mapping of anatomic geometries and CFD-driven flow fields, which enables accomplishing fast and accurate hemodynamic prediction for clinical applications. Recently, the ML/DL models have been verified capable of predicting the reduced-order simulation results in a computationally inexpensive way when merely employing some limited flow information, i.e., the velocities and pressures at the centerline or cross-section of a vessel ( Itu et al, 2016 ; Sklet, 2018 ; Sarabian et al, 2021 ).…”

Section: Introductionmentioning

confidence: 99%

Deep learning-based hemodynamic prediction of carotid artery stenosis before and after surgical treatments

Wang

et al. 2023

Front. Physiol.

View full text Add to dashboard Cite

Hemodynamic prediction of carotid artery stenosis (CAS) is of great clinical significance in the diagnosis, prevention, and treatment prognosis of ischemic strokes. While computational fluid dynamics (CFD) is recognized as a useful tool, it shows a crucial issue that the high computational costs are usually required for real-time simulations of complex blood flows. Given the powerful feature-extraction capabilities, the deep learning (DL) methodology has a high potential to implement the mapping of anatomic geometries and CFD-driven flow fields, which enables accomplishing fast and accurate hemodynamic prediction for clinical applications. Based on a brain/neck CT angiography database of 280 subjects, image based three-dimensional CFD models of CAS were constructed through blood vessel extraction, computational domain meshing and setting of the pulsatile flow boundary conditions; a series of CFD simulations were undertaken. A DL strategy was proposed and accomplished in terms of point cloud datasets and a DL network with dual sampling-analysis channels. This enables multimode mapping to construct the image-based geometries of CAS while predicting CFD-based hemodynamics based on training and testing datasets. The CFD simulation was validated with the mass flow rates at two outlets reasonably agreed with the published results. Comprehensive analysis and error evaluation revealed that the DL strategy enables uncovering the association between transient blood flow characteristics and artery cavity geometric information before and after surgical treatments of CAS. Compared with other methods, our DL-based model trained with more clinical data can reduce the computational cost by 7,200 times, while still demonstrating good accuracy (error<12.5%) and flow visualization in predicting the two hemodynamic parameters. In addition, the DL-based predictions were in good agreement with CFD simulations in terms of mean velocity in the stenotic region for both the preoperative and postoperative datasets. This study points to the capability and significance of the DL-based fast and accurate hemodynamic prediction of preoperative and postoperative CAS. For accomplishing real-time monitoring of surgical treatments, further improvements in the prediction accuracy and flexibility may be conducted by utilizing larger datasets with specific real surgical events such as stent intervention, adopting personalized boundary conditions, and optimizing the DL network.

show abstract

Deep Learning for Computational Hemodynamics: A Brief Review of Recent Advances

Cited by 21 publications

References 93 publications

Stress Load and Ascending Aortic Aneurysms: An Observational, Longitudinal, Single-Center Study Using Computational Fluid Dynamics

Stress Load and Ascending Aortic Aneurysms: An Observational, Longitudinal, Single-Center Study Using Computational Fluid Dynamics

CFD Study of the Effect of the Angle Pattern on Iliac Vein Compression Syndrome

Deep learning-based hemodynamic prediction of carotid artery stenosis before and after surgical treatments

Contact Info

Product

Resources

About