Machine Learning Identification Framework of Hemodynamics of Blood Flow in Patient-Specific Coronary Arteries with Abnormality.

In this study, we put forth a new deep neural network framework to predict flow behavior in a coronary arterial network with different properties in the presence of any abnormality like stenosis. An artificial neural network (ANN) model is trained using synthetic data so that it can predict the pressure and velocity within the arterial network. The data required to train the neural network were obtained from the CFD analysis of several geometries of arteries with specific features in ABAQUS software. The proposed approach precisely predicts the hemodynamic behavior of the blood flow. The average accuracy of the pressure prediction was 98.7%, and the average velocity magnitude accuracy was 93.2%. Our model can also be used to predict fractional flow reserve (FFR), which is one of the main indices to determine the severity of stenosis, and our model predicts this index successfully based on the artery features.

Targeted drug delivery with polydisperse particle transport and deposition in patient-specific upper airway during inhalation and exhalation.

Identifying the deposition pattern of inhaled pharmaceutical aerosols in the human respiratory system and understanding the effective parameters in this process is vital for more efficient drug delivery to this region. This study investigated aerosol deposition in a patient-specific upper respiratory airway and determined the deposition fraction (DF) and pressure drop across the airway. An experimental setup was developed to measure the pressure drop in the same realistic geometry printed from the patient-specific geometry. The unsteady simulations were performed with a flow rate of 15 L/min and different particle diameters ranging from 2 to 30 µm. The results revealed significant flow circulation after the nasal valve in the upper and oropharynx regions, and a maximum local velocity observed in the nasopharynx. Transient cumulative deposition fraction showed that after 2 s of the simulation, all particles deposit or escape the computational domain. About 30 % of the injected large particles (dp ≥ 20 µm) deposited in the first 1 cm away from the nostril and more than 95 % deposited in the nasal airway before entering the oropharynx region. While almost 94 % deposition in trachea was composed of particles smaller than 5 µm. Approximately 20 % of inhaled fine particles (2-5 µm) deposited in the upper airway and the rest deposited in oropharynx, larynx and trachea.

Numerical and Experimental Analysis of Drug Inhalation in Realistic Human Upper Airway Model.

The demand for a more efficient and targeted method for intranasal drug delivery has led to sophisticated device design, delivery methods, and aerosol properties. Due to the complex nasal geometry and measurement limitations, numerical modeling is an appropriate approach to simulate the airflow, aerosol dispersion, and deposition for the initial assessment of novel methodologies for better drug delivery. In this study, a CT-based, 3D-printed model of a realistic nasal airway was reconstructed, and airflow pressure, velocity, turbulent kinetic energy (TKE), and aerosol deposition patterns were simultaneously investigated. Different inhalation flowrates (5, 10, 15, 30, and 45 L/min) and aerosol sizes (1, 1.5, 2.5, 3, 6, 15, and 30 µm) were simulated using laminar and SST viscous models, with the results compared and verified by experimental data. The results revealed that from the vestibule to the nasopharynx, the pressure drop was negligible for flow rates of 5, 10, and 15 L/min, while for flow rates of 30 and 40 L/min, a considerable pressure drop was observed by approximately 14 and 10%, respectively. However, from the nasopharynx and trachea, this reduction was approximately 70%. The aerosol deposition fraction alongside the nasal cavities and upper airway showed a significant difference in pattern, dependent on particle size. More than 90% of the initiated particles were deposited in the anterior region, while just under 20% of the injected ultrafine particles were deposited in this area. The turbulent and laminar models showed slightly different values for the deposition fraction and efficiency of drug delivery for ultrafine particles (about 5%); however, the deposition pattern for ultrafine particles was very different.

Numerical study of the effect of hemodynamic variables on LDL concentration through the single layer of the Left Anterior Descending coronary artery (LAD) under the heart pulse.

Heart attack is one of the most common causes of death in the world. Coronary artery disease is the most recognized cause of heart attack whose onset and progression have been attributed to low-density lipoprotein (LDL) passing through the wall of the artery. In this paper, hemodynamic variables as well as the concentration of LDL through the coronary porous artery at the Left Anterior Descending coronary artery (LAD), and its first diagonal branch (D1) under the heart motion investigated using computational simulation. The geometry that has been studied in this paper is the first bifurcation of Left Anterior Descending (LAD) that has been placed on a perimeter of hypothetical sphere representative of the heart geometry. Sinusoidal variations of sphere radii, simulated pulsating movement of the heart. Blood has been considered as a Newtonian and incompressible flow with pulsatile flow rate and real physiological profile. The plasma filtration boundary condition used over the walls in order to simulate the concentration of LDL to a one-layer artery wall. Variations in the concentration of LDL on the artery wall and its relation to oscillation on shear stress on the artery wall under different conditions are presented. Moreover, the effects of the pulsating inlet flow and dynamic movement of the artery are explored. The results declared that minimum shear stress and maximum LDL concentration take place at the bifurcation and on the myocardial wall which is in complete agreement with clinical studies. Furthermore, it has been shown that the heart pulse has a slight effect on the average time of concentration (0.1% increase); however, by analyzing all time steps, one could observe that the maximum concentration rises in some time steps; where this increases the possibility of LDL presence and helps them diffuse inside the artery wall.

Atheroprone sites of coronary artery bifurcation: Effect of heart motion on hemodynamics-dependent monocytes deposition.

Atherosclerosis as a common cardiovascular disease is a result of both adverse hemodynamics conditions and monocyte deposition within coronary arteries. It is known that the adhesion of monocytes on the arterial wall and their interaction with the vascular surface are one of the main parameters in the initiation and progression of atherosclerosis. In this work, hemodynamic parameters and monocyte deposition have been investigated in a 3D computational model of the Left Anterior Descending coronary artery (LAD) and its first diagonal branch (D1) under the heart motion. A one-way Lagrangian approach is performed to trace the monocyte particles under different blood flow regimes and heart motion conditions. The hemodynamic results show that the myocardial wall, and also the flow divider wall can be candidates for atheroprone sites. The dynamic movement and pulsatile inlet changed the flow rate between branches about 21% compared to the static case and steady inlet. On the other hand, the calculation of monocytes' depositional behavior illustrates that they settle down downstream the LAD-D1 bifurcation and on the myocardial wall. The deposition rate is closely associated with the inlet type and changing the steady inlet to the sinusoidal and real physiologic profile showed a 150% increase in the deposition rate. These results ensure that the myocardial wall and LAD-D1 bifurcation are the desirable locations for atherosclerosis. These results are in good agreement with the clinical observations.

The effect of the second excitation frequency mode under different conditions on the fluid streaming and microparticles acoustophoresis with the aim of separating biological cells.

BACKGROUND AND OBJECTIVE: In this study, the effect of the second excitation frequency mode under different conditions on the fluid streaming and its microparticles displacement is investigated. METHODS: For this purpose, some variable parameters such as the particle diameter, microchannel aspect ratio, and applied frequency modes have been selected to study. The resulted acoustic streaming was scrutinized to understand the physics of the problem under different geometrical and input conditions. Finally, the effect of the increasing the microparticle size and aspect ratio of the microchannel, simultaneously, has been evaluated. RESULTS: The results demonstrated that increasing the microparticle size accelerates the displacement of the microparticles. On the other hand, changing the aspect ratio affects the formation of the microparticle distribution and it also changes the velocity of the microparticles due to the gradient of the second-order pressure. CONCLUSIONS: The obtained results have wide applications in the military, medical, petrochemical, and other related studies.

Computational investigation of stenosis in curvature of coronary artery within both dynamic and static models.

BACKGROUND AND OBJECTIVE: Blood flow variation during cardiac cycle is the main mechanism of atherosclerotic development which is dependent on. METHODS: The present work mainly tends to investigate stenosis effect in dynamic curvature of coronary artery. This paper presents numerical investigations on wall shear stress profiles in three-dimensional pulsatile flow through curved stenotic coronary arteries for both static and dynamic model. In order to do so, three-dimensional models related to the curved arteries with two degrees of stenosis (30% and 50%). RESULTS: Lower amount of wall shear stress is found near the inner wall of artery distal to the plaque region (stenosis) and in both percentages of stenosis the maximum wall shear stress will accrue in the middle of the stenosis; however it is much more in the higher rate of stenosis. CONCLUSIONS: A chaotic wall shear stress region is also observed downstream of stenosis in the severe stenosis case. Finally it concluded that the arterial wall motion affects the wall shear stress and the plaque formation site.