The reduced-order lumped parameter model (LPM) has great computational efficiency in real-time numerical simulations of haemodynamics but is limited by the accuracy of patient-specific computation. This study proposed a method to achieve the individual LPM modeling with high accuracy to improve the practical clinical applicability of LPM. Clinical data was collected from two medical centres comprising haemodynamic indicators from 323 individuals, including brachial artery pressure waveforms, cardiac output data, and internal carotid artery flow waveforms. The data were expanded to 5000 synthesised cases that all fell within the physiological range of each indicator. LPM of the human blood circulation system was established. A double-path neural network (DPNN) was designed to input the waveforms of each haemodynamic indicator and their key features and then output the individual parameters of the LPM, which was labelled using a conventional optimization algorithm. Clinically collected data from the other 100 cases were used as the test set to verify the accuracy of the individual LPM parameters predicted by DPNN. The results show that DPNN provided good convergence in the training process. In the test set, compared with clinical measurements, the mean differences between each haemodynamic indicator and the estimate calculated by the individual LPM based on the DPNN were about 10 %. Furthermore, DPNN prediction only takes 4 s for 100 cases. The DPNN proposed in this study permits real-time and accurate individualization of LPM's. When facing medical issues involving haemodynamics, it lays the foundation for patient-specific numerical simulation, which may be beneficial for potential clinical application.
Deep Learning , Hemodynamics , Models, Cardiovascular , Humans , Hemodynamics/physiology , Male , Female , Adult
Biodegradable stents can support vessels for an extended period, maintain vascular patency, and progressively degrade once vascular remodeling is completed, thereby reducing the constraints of traditional metal stents. An ideal degradable stent must have good mechanical properties, degradation behavior, and biocompatibility. Zinc has become a new type of biodegradable metal after magnesium and iron, owing to its suitable degradation rate and good biocompatibility. However, zinc's poor strength and ductility make it unsuitable as a vascular stent material. Therefore, this paper reviewed the primary methods for improving the overall properties of zinc. By discussing the mechanical properties, degradation behavior, and biocompatibility of various improvement strategies, we found that alloying is the most common, simple, and effective method to improve mechanical properties. Deformation processing can further improve the mechanical properties by changing the microstructures of zinc alloys. Surface modification is an important means to improve the biological activity, blood compatibility and corrosion resistance of zinc alloys. Meanwhile, structural design can not only improve the mechanical properties of the vascular stents, but also endow the stents with special properties such as negative Poisson 's ratio. Manufacturing zinc alloys with excellent degradation properties, improved mechanical properties and strong biocompatibility and exploring their mechanism of interaction with the human body remain areas for future research.
Biocompatible Materials , Zinc , Humans , Biocompatible Materials/therapeutic use , Absorbable Implants , Alloys , Stents , Magnesium/pharmacology , Magnesium/therapeutic use
Current research on the fatigue properties of degradable zinc alloy stents has not yet considered the issue of the fatigue life changing with material properties during the dynamic degradation process. Therefore, in this paper, we established a fatigue damage algorithm to study the fatigue problem affected by the changing of material properties during the dynamic degradation process of the stent under the action of pulsating cyclic loading. Three models: the dynamic degradation model, the dynamic degradation model under pulsating cyclic loading, and the coupled model of fatigue damage and dynamic degradation, were developed to verify the effect of fatigue damage on stent life. The results show that fatigue damage leads to a deeper degree of inhomogeneous degradation of the stent, which affects the service life of the stent. Fatigue damage is a factor that cannot be ignored. Therefore, when studying the mechanical properties and lifetime of degradable stents, incorporating fatigue damage into the study can help more accurately assess the lifetime of the stents.
Boundary condition settings are key risk factors for the accuracy of noninvasive quantification of fractional flow reserve (FFR) based on computed tomography angiography (i.e., FFRCT). However, transient numerical simulation-based FFRCT often ignores the three-dimensional (3D) model of coronary artery and clinical statistics of hyperemia state set by boundary conditions, resulting in insufficient computational accuracy and high computational cost. Therefore, it is necessary to develop the custom function that combines the 3D model of the coronary artery and clinical statistics of hyperemia state for boundary condition setting, to accurately and quickly quantify FFRCT under steady-state numerical simulations. The 3D model of the coronary artery was reconstructed by patient computed tomography angiography (CTA), and coronary resting flow was determined from the volume and diameter of the 3D model. Then, we developed the custom function that took into account the interaction of stenotic resistance, microcirculation resistance, inlet aortic pressure, and clinical statistics of resting to hyperemia state due to the effect of adenosine on boundary condition settings, to accurately and rapidly identify coronary blood flow for quantification of FFRCT calculation (FFRU). We tested the diagnostic accuracy of FFRU calculation by comparing it with the existing methods (CTA, coronary angiography (QCA), and diameter-flow method for calculating FFR (FFRD)) based on invasive FFR of 86 vessels in 73 patients. The average computational time for FFRU calculation was greatly reduced from 1-4 h for transient numerical simulations to 5 min per simulation, which was 2-fold less than the FFRD method. According to the results of the Bland-Altman analysis, the consistency between FFRU and invasive FFR of 86 vessels was better than that of FFRD. The area under the receiver operating characteristic curve (AUC) for CTA, QCA, FFRD and FFRU at the lesion level were 0.62 (95% CI: 0.51-0.74), 0.67 (95% CI: 0.56-0.79), 0.85 (95% CI: 0.76-0.94), and 0.93 (95% CI: 0.87-0.98), respectively. At the patient level, the AUC was 0.61 (95% CI: 0.48-0.74) for CTA, 0.65 (95% CI: 0.53-0.77) for QCA, 0.83 (95% CI: 0.74-0.92) for FFRD, and 0.92 (95% CI: 0.89-0.96) for FFRU. The proposed novel method might accurately and rapidly identify coronary blood flow, significantly improve the accuracy of FFRCT calculation, and support its wide application as a diagnostic indicator in clinical practice.
The Special Issue entitled "Biomechanical Study and Analysis for Cardiovascular/Skeletal Materials and Devices" addresses biological functional materials and devices relevant to cardiovascular diseases and orthopedic conditions [...].
Hemodynamic parameters are of great significance in the clinical diagnosis and treatment of cardiovascular diseases. However, noninvasive, real-time and accurate acquisition of hemodynamics remains a challenge for current invasive detection and simulation algorithms. Here, we integrate computational fluid dynamics with our customized analysis framework based on a multi-attribute point cloud dataset and physics-informed neural networks (PINNs)-aided deep learning modules. This combination is implemented by our workflow that generates flow field datasets within two types of patient personalized models - aorta with fine coronary branches and abdominal aorta. Deep learning modules with or without an antecedent hierarchical structure model the flow field development and complete the mapping from spatial and temporal dimensions to 4D hemodynamics. 88,000 cases on 4 randomized partitions in 16 controlled trials reveal the hemodynamic landscape of spatio-temporal anisotropy within two types of personalized models, which demonstrates the effectiveness of PINN in predicting the space-time behavior of flow fields and gives the optimal deep learning framework for different blood vessels in terms of balancing the training cost and accuracy dimensions. The proposed framework shows intentional performance in computational cost, accuracy and visualization compared to currently prevalent methods, and has the potential for generalization to model flow fields and corresponding clinical metrics within vessels at different locations. We expect our framework to push the 4D hemodynamic predictions to the real-time level, and in statistically significant fashion, applicable to morphologically variable vessels.
Hemodynamics , Neural Networks, Computer , Humans , Aorta , Algorithms , Computer Simulation
Introduction: Hemodynamic diagnosis indexes (HDIs) can comprehensively evaluate the health status of the cardiovascular system (CVS), particularly for people older than 50 years and prone to cardiovascular disease (CVDs). However, the accuracy of non-invasive detection remains unsatisfactory. We propose a non-invasive HDIs model based on the non-linear pulse wave theory (NonPWT) applied to four limbs. Methods: This algorithm establishes mathematical models, including pulse wave velocity and pressure information of the brachial and ankle arteries, pressure gradient, and blood flow. Blood flow is key to calculating HDIs. Herein, we derive blood flow equation for different times of the cardiac cycle considering the four different distributions of blood pressure and pulse wave of four limbs, then obtain the average blood flow in a cardiac cycle, and finally calculate the HDIs. Results: The results of the blood flow calculations reveal that the average blood flow in the upper extremity arteries is 10.78 ml/s (clinically: 2.5-12.67 ml/s), and the blood flow in the lower extremity arteries is higher than that in the upper extremity. To verify model accuracy, the consistency between the clinical and calculated values is verified with no statistically significant differences (p < 0.05). Model IV or higher-order fitting is the closest. To verify the model generalizability, considering the risk factors of cardiovascular diseases, the HDIs are recalculated using model IV, and thus, consistency is verified (p < 0.05 and Bland-Altman plot). Conclusion: We conclude our proposed algorithmic model based on NonPWT can facilitate the non-invasive hemodynamic diagnosis with simpler operational procedures and reduced medical costs.
Percutaneous coronary intervention with stent implantation is one of the most commonly used approaches to treat coronary artery stenosis. Stent malapposition (SM) can increase the incidence of stent thrombosis, but the quantitative association between SM distance and stent thrombosis is poorly clarified. The objective of this study is to determine the biomechanical reaction mechanisms underlying stent thrombosis induced by SM and to quantify the effect of different SM severity grades on thrombosis. The thrombus simulation was performed in a continuous model based on the diffusion-convection response of blood substance transport. Simulated models included well-apposed stents and malapposed stents with various severities where the detachment distances ranged from 0 to 400 µm. The abnormal shear stress induced by SM was considered a critical contributor affecting stent thrombosis, which was dependent on changing SM distances in the simulation. The results illustrate that the proportion of thrombus volume was 1.88% at a SM distance of 75 µm (mild), 3.46% at 150 µm, and 3.93% at 400 µm (severe), but that a slight drop (3.18%) appeared at the detachment distance of 225 µm (intermediate). The results indicate that when the SM distance was less than 150 µm, the thrombus rose notably as the gap distance increased, whereas the progression of thrombogenicity weakened when it exceeded 150 µm. Therefore, more attention should be paid when SM is present at a gap distance of 150 µm. Moreover, when the SM length of stents are the same, thrombus tends to accumulate downstream towards the distal end of the stent as the SM distance increases.
Traditional inert materials used in internal fixation have caused many complications and generally require removal with secondary surgeries. Biodegradable materials, such as magnesium (Mg)-, iron (Fe)- and zinc (Zn)-based alloys, open up a new pathway to address those issues. During the last decades, Mg-based alloys have attracted much attention by researchers. However, the issues with an over-fast degradation rate and release of hydrogen still need to be overcome. Zn alloys have comparable mechanical properties with traditional metal materials, e.g., titanium (Ti), and have a moderate degradation rate, potentially serving as a good candidate for internal fixation materials, especially at load-bearing sites of the skeleton. Emerging Zn-based alloys and composites have been developed in recent years and in vitro and in vivo studies have been performed to explore their biodegradability, mechanical property, and biocompatibility in order to move towards the ultimate goal of clinical application in fracture fixation. This article seeks to offer a review of related research progress on Zn-based biodegradable materials, which may provide a useful reference for future studies on Zn-based biodegradable materials targeting applications in orthopedic internal fixation.
A new protein foaming-consolidation method for preparing porous zinc was developed using three proteins (egg white protein (EWP), bovine bone collagen protein (BBCP), and fish bone collagen protein (FBCP)) as both consolidating and foaming agents. The preparation route utilized powder mixing and sintering processing, which could be divided into three steps: slurry preparation, low-temperature foaming, and high-temperature sintering. The morphological characteristics of the pore structures revealed that the porous zinc had an interconnected open-cell structure. Compared to the porous zinc prepared with EWP or BBCP, the porous zinc prepared with FBCP possessed the largest average pore size and the highest compressive properties. The porosity of the porous zinc increased with the stirring time, the content of protein and sucrose, and higher sintering temperatures. Moreover, a compression test and immersion test were performed to investigate the stress-strain behavior and corrosion properties of the resulting porous zinc. A fluctuated stress plateau could be found due to the brittle fracture of the porous cells. The porous zinc prepared with FBCP showed the highest compressive strength and elastic modulus. The corrosion rate of the porous zinc obtained through an immersion test in vitro using simulated bodily fluids on the thirty-second day was close to 0.02 mm/year. The corresponding corrosion mechanism of porous zinc was also discussed.
Most of the studies on the finite element analysis (FEA) of biodegradable vascular stents (BVSs) during the degradation process have limited the accuracy of the simulation results due to the application of the uniform degradation model. This paper aims to establish an FEA model for the non-uniform degradation of BVSs by considering factors such as the dynamic changes of the corrosion properties and material properties of the element, as well as the pitting corrosion and stress corrosion. The results revealed that adjusting the corrosion rate according to the number of exposed surfaces of the element and reducing the stress threshold according to the corrosion status accelerates the degradation time of BVSs by 26% and 25%, respectively, compared with the uniform degradation model. The addition of the pitting model reduces the service life of the BVSs by up to 12%. The effective support of the stent to the vessel could reach at least 60% of the treatment effect before the vessel collapsed. These data indicate that the proposed non-uniform degradation model of BVSs with multiple factors produces different phenomena compared with the commonly used models and make the numerical simulation results more consistent with the real degradation scenario.
Background: Although aortic valve reconstruction has become an alternative treatment for aortic valve disease, the design of the geometric parameters of the reconstructed leaflet still mainly depends on the experience of doctors. The present study investigates the effects of the height of the leaflets on the performance and biomechanical states of the reconstructed aortic valve. Methods: This numerical study was carried out using the finite element approach and the lattice Boltzmann method. The dynamic and biomechanical characteristics of the leaflets were evaluated by using the finite element approach, while the blood flow in the aortic sinus was evaluated by applying the lattice Boltzmann method. Three types of leaflets with different heights were designed. Then the dynamic characteristics, stress distribution, and effective orifice area (EOA) of the aortic valve and flow pattern were calculated as the indicators. Results: The results demonstrated that the height of the leaflets could indeed regulate the performance and the biomechanical states of the aortic valve. The rapid valve opening times of the 3 types of leaflets gradually reduced along with the decrease of the height ratio (HR_0.8: 120 ms vs. HR_1.0: 68 ms vs. HR_1.2: 31 ms), while the rapid valve closing times (RVCTs) of the 3 types of leaflets were similar to each other (approximately 75 ms). Moreover, the radial displacement of the leaflet at the fully open time increased along with the decrease of the HR of the leaflets (HR_0.8: 8 mm vs. HR_1.0: 6 mm vs. HR_1.2: 4 mm). In addition, the stress level of the leaflets also increased with the increase of the height of the leaflets (max stress, HR_0.8: 0.5 MPa, vs. HR_1.0: 1.1 MPa, vs. HR_1.2: 1.8 MPa). Similarly, the low velocity region near the ascending aortic wall and the wall shear stress (WSS) level on the ventricular side of the leaflets also increased along with the increase of the HR of the leaflets. Conclusions: In short, the height of the leaflets mainly affects the opening performance of the reconstructed aortic leaflets. The HR of the reconstructed leaflets for adults should be less than 1.0 to balance the opening and closing performance of aortic leaflets.
Mineralized collagen is the basic unit in hierarchically organized natural bone with different structures. Polyacrylic acid (PAA) and periodic fluid shear stress (FSS) are the most common chemical and physical means to induce intrafibrillar mineralization. In the present study, non-mineralized collagen, extrafibrillar mineralized (EM) collagen, intrafibrillar mineralized (IM) collagen, and hierarchical intrafibrillar mineralized (HIM) collagen induced by PAA and FSS were prepared, respectively. The physical and chemical properties of these mineralized collagens with different microstructures were systematically investigated afterwards. Transmission electron microscopy (TEM) and scanning electron microscopy (SEM) showed that mineralized collagen with different microstructures was prepared successfully. The pore density of the mineralized collagen scaffold is higher under the action of periodic FSS. Fourier transform infrared spectroscopy (FTIR) analysis showed the formation of the hydroxyapatite (HA) crystal. A significant improvement in the pore density, hydrophilicity, enzymatic stability, and thermal stability of the mineralized collagen indicated that the IM collagen under the action of periodic FSS was beneficial for maintaining collagen activity. HIM collagen fibers, which are prepared under the co-action of periodic FSS and sodium tripolyphosphate (TPP), may pave the way for new bone substitute material applications.
BACKGROUND AND OBJECTIVES: The bicuspid aortic valve (BAV) is a major risk factor for the progression of aortic dilation (AD) because of the induced abnormal blood flow environment in aorta. The differences in the development of AD induced by BAV phenotypes remains unclear. Therefore, the objective of this study was to assess the potential locations of AD induced by different phenotypes of BAV. The different effects of opening orifice area and leaflet orientation on ascending aortic hemodynamics in Type-1 BAV was investigated by means of numerical simulation. METHODS: Finite element dynamic analysis was performed on tricuspid aortic valve (TAV) and BAV models to simulate the motion of the leaflets and obtain the geometrical characteristics of AV at peak systole as a reference, which were used for aortic models. Then, four sets of aortic fluid models were designed according to the leaflet fusion types [TAV; BAV (left-right-coronary cusp fusion, LR; right-non-coronary cusp fusion, RN; left-non-coronary cusp fusion, LN)], and the computational fluid dynamics method was applied to compare the hemodynamic differences within the aorta at peak systole. RESULTS: The maximum opening area of BAV was significantly reduced, resulting in alterations in aortic hemodynamics compared with TAV. The velocity streamlines were essentially parallel to the aortic wall in TAV. The average pressure and wall shear stress in aorta tend to be stable. In contrary, the eccentricity of BAV orifice jet resulted in high-velocity flow directed toward the ascending aorta (AA) wall and aortic arch for LR and LN; RN features an asymmetrical velocity distribution toward the outer bend of the middle AA, and eccentric flow tends to impact the distal AA. As the flow angle is associated with distinct flow impingement locations, different degrees of WSS and pressure concentration occur along the aortic wall from the AA to the aortic arch in three BAV types. CONCLUSIONS: The BAV morphotype affects the aortic hemodynamics, and the abnormal blood flow associated with BAV may play a role in AD. The different BAV phenotypes determine the direction of blood flow jet and change the expression of dilation. LR is likely to cause dilation of the tubular AA; RN results in dilation of the middle AA to proximal aortic arch; and LN causes an increased incidence of the tubular AA and the proximal aortic arch.
Bicuspid Aortic Valve Disease , Heart Valve Diseases , Aortic Valve/physiology , Dilatation , Heart Valve Diseases/complications , Hemodynamics/physiology , Humans , Phenotype
OBJECTIVE: Pulse wave has been considered as a message carrier in the cardiovascular system (CVS), capable of inferring CVS conditions while diagnosing cardiovascular diseases (CVDs). Clarification and prediction of cardiovascular function by means of powerful feature-abstraction capability of machine learning method based on pulse wave is of great clinical significance in health monitoring and CVDs diagnosis, which remains poorly studied. METHODS: Here we propose a machine learning (ML)-based strategy aiming to achieve a fast and accurate prediction of three cardiovascular function parameters based on a 412-subject database of pulse waves. We proposed and optimized an ML-based model with multi-layered, fully connected network while building up two high-quality pulse wave datasets comprising a healthy-subject group and a CVD-subject group to predict arterial compliance (AC), total peripheral resistance (TPR), and stroke volume (SV), which are essential messengers in monitoring CVS conditions. RESULTS: Our ML model is validated through consistency analysis of the ML-predicted three cardiovascular function parameters with clinical measurements and is proven through error analysis to have capability of achieving a high-accurate prediction on TPR and SV for both healthy-subject group (accuracy: 85.3%, 86.9%) and CVD-subject group (accuracy: 88.3%, 89.2%). DISCUSSION: The independent sample t-test proved that our subject groups could represent the typical physiological characteristics of the corresponding population. While we have more subjects in our datasets rather than previous studies after strict data screening, the proposed ML-based strategy needs to be further improved to achieve a disease-specific prediction of heart failure and other CVDs through training with larger datasets and clinical measurements. CONCLUSION: Our study points to the feasibility and potential of the pulse wave-based prediction of physiological and pathological CVS conditions in clinical application.
Cardiovascular Diseases , Pulse Wave Analysis , Cardiovascular Diseases/diagnosis , Heart Rate , Humans , Machine Learning , Pulse Wave Analysis/methods , Stroke Volume
Mineralized collagen is a natural organic-inorganic composite. The combination of organic collagen and inorganic apatite to form different nanostructures is the key to producing bone substitutes with biomechanical properties that are as identical to normal bone as possible. However, the formation of apatite with different nanostructures during collagen mineralization is unexplored. Here, pyrophosphate (Pyro-P), as an important hydrolysate of adenosine triphosphate in the body, was introduced to prepare mineralized collagen under the regulation of alkaline phosphatase (ALP) and orthophosphate (Ortho-P). Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) results showed that mineralized collagen, which combined with different crystallinities and multilayered structured apatite, was successfully prepared. A combination of ion chromatography (IC), Fourier transform infrared (FTIR) spectroscopy, circular dichroism (CD), and thermogravimetry (TG) analyses revealed the crucial role of Ortho-P in the formation of multilayered flower-shaped apatite with different crystallinities and in the maintenance of mineralization balance. Mineralization balance is of great significance for maintaining normal bone morphology during bone regeneration. Overall, our results provide a promising method to produce new bone substitute materials for the repair of large bone defects and a deeper insight into the mechanisms of biomineralization.
Alkaline Phosphatase , Apatites , Bone and Bones , Collagen , Microscopy, Electron, Scanning , Phosphates , Spectroscopy, Fourier Transform Infrared
BACKGROUND: Inadequate scaffolding performance hinders the clinical application of the biodegradable zinc alloy stents. OBJECTIVE: In this study we propose a novel stent with the tenon-and-mortise structure to improve its scaffolding performance. METHODS: 3D models of stents were established in Pro/E. Based on the biodegradable zinc alloy material and two numerical simulation experiments were performed in ABAQUS. Firstly, the novel stent could be compressed to a small-closed ring by a crimp shell and can form a tenon-and-mortise structure after expanded by a balloon. Finally, 0.35 MPa was applied to the crimp shell to test the scaffolding performance of the novel stent and meanwhile compare it with an ordinary stent. RESULTS: Results showed that the novel stent decreased the recoiling ratio by 70.7% compared with the ordinary stent, indicating the novel structure improved the scaffolding performance of the biodegradable zinc alloy stent. CONCLUSION: This study proposes a novel design that is expected to improve the scaffolding performance of biodegradable stents.
Alloys , Zinc , Finite Element Analysis , Humans , Prosthesis Design , Stents
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.
The interventional treatment of cerebral aneurysm requires hemodynamics to provide proper guidance. Computational fluid dynamics (CFD) is gradually used in calculating cerebral aneurysm hemodynamics before and after flow-diverting (FD) stent placement. However, the complex operation (such as the construction and placement simulation of fully resolved or porous-medium FD stent) and high computational cost of CFD hinder its application. To solve these problems, we applied aneurysm hemodynamics point cloud data sets and a deep learning network with double input and sampling channels. The flexible point cloud format can represent the geometry and flow distribution of different aneurysms before and after FD stent (represented by porous medium layer) placement with high resolution. The proposed network can directly analyze the relationship between aneurysm geometry and internal hemodynamics, to further realize the flow field prediction and avoid the complex operation of CFD. Statistical analysis shows that the prediction results of hemodynamics by our deep learning method are consistent with the CFD method (error function <13%), but the calculation time is significantly reduced 1,800 times. This study develops a novel deep learning method that can accurately predict the hemodynamics of different cerebral aneurysms before and after FD stent placement with low computational cost and simple operation processes.
