A method and device for assessing the risk of thrombosis in artificial heart valves

Abstract

The present application proposes a method and device for assessing the risk of thrombosis in an artificial heart valve. The method for assessing the risk of thrombosis in an artificial heart valve of the present invention includes: obtaining the physiological scenario after the artificial heart is implanted with an artificial valve through resistance characteristic parameters and elastic expansion characteristic parameters; starting a pulsating pump to output a pulsating flow medium to the artificial heart under the physiological scenario; capturing the motion trajectory of fluorescent particles through a high-speed camera and obtaining the original data of blood flow velocity downstream of the artificial valve; obtaining a fluid mechanics simulation parameter group based on the original data of blood flow velocity; inputting a pre-trained valve thrombosis random forest classifier according to the fluid mechanics simulation parameter group and outputting a thrombosis risk probability map under the physiological scenario. The present invention calculates a fluid mechanics simulation parameter group based on the instantaneous velocity vector field and spatiotemporal distribution information captured by PIV, generates a two-dimensional / three-dimensional thrombosis risk probability map through a random forest classifier and marks the coordinates of high-risk points, thereby achieving accurate prediction of thrombosis probability and thrombosis location.

Description

Technical Field

[0001] The present invention relates to a method and device for assessing the risk of thrombosis in an artificial heart valve, and belongs to the technical field of fluid mechanics experimental data processing. Background Art

[0002] In the practical use of transcatheter pulmonary valve implantation (TPVR), postoperative thrombosis is one of the major complications. To better understand and predict the risk of thrombosis after TPVR, it is very important to quantitatively analyze the changes in the flow field downstream of the pulmonary valve. Particle image velocimetry (PIV) is an optional technical means to measure and analyze process parameters. However, how to combine particle image velocimetry with artificial heart valve scenarios and design methods and calculate data for thrombosis risk assessment remains a blank area in the industry. Therefore, by combining PIV technology with artificial heart valve control systems to predict the points of thrombosis risk, this method of postoperative risk assessment based on experimental data and theoretical analysis is of great reference value for optimizing surgical plans.

[0003] When filling this technological gap, the existing in vitro simulation system can only simulate a static healthy blood flow environment and cannot reproduce the hemodynamic characteristics under various conditions such as pathological and motion conditions, such as hypertension and heart failure. Existing thrombosis prediction models mostly rely on a single blood flow parameter, such as average flow velocity, and lack the coordinated implementation and quantitative analysis of multiple factors to simulate multi-state blood flow environments, resulting in a high misjudgment rate of thrombosis risk points. Summary of the Invention

[0004] In view of the above-mentioned deficiencies in the prior art, the object of the present invention is to provide a method and device for assessing the risk of thrombosis in an artificial heart valve.

[0005] According to an embodiment of the present invention, a first solution is provided: a method for assessing the risk of thrombosis in an artificial heart valve, comprising the following steps:

[0006] installing the artificial valve to be tested in an artificial heart of an extracorporeal pulsatile circulation system, wherein the extracorporeal pulsatile circulation system further comprises a peripheral resistance module and a compliance structure module;

[0007] The peripheral resistance module adjusts the resistance characteristics of the vascular system of the artificial heart and outputs resistance characteristic parameters, the compliance structure module adjusts the elastic expansion capacity of the vascular system of the artificial heart and outputs elastic expansion characteristic parameters, and obtains the physiological scenario after the artificial heart is implanted with an artificial valve through the resistance characteristic parameters and the elastic expansion characteristic parameters, wherein the physiological scenario includes a healthy state, a pathological state, and a motion state;

[0008] Selecting a physiological scenario and starting a pulsating pump in the physiological scenario to output a pulsating flow medium to the artificial heart, wherein the pulsating flow medium includes fluorescent particles, and controlling the pulsating flow medium passing through the artificial valve to circulate along a preset path;

[0009] A dual-pulse laser is started to project a light source onto the flow field downstream of the artificial valve of the artificial heart, and the trajectory of the fluorescent particles is captured by a high-speed camera to obtain the original data of the blood flow velocity downstream of the artificial valve;

[0010] Obtaining a fluid mechanics simulation parameter group based on the original blood flow velocity data, wherein the fluid mechanics simulation parameter group includes wall shear stress, particle residence time, and shear oscillation index;

[0011] The pre-trained valve thrombosis random forest classifier is input according to the fluid dynamics simulation parameter group and outputs the thrombosis risk probability map under the physiological scenario.

[0012] Furthermore, the diversion ratio of the pulsating flow medium is controlled by adjusting the throttle valve opening, so that A% of the pulsating flow medium enters the right atrium through the reflow path and B% of the pulsating flow medium enters the compliance cavity, where A+B=100; the pressure difference ΔP_re between the inlet pressure and the outlet pressure and the blood flow Q of the artificial valve PIV experiment are monitored in real time, and the peripheral resistance value is calculated according to the formula and output as the resistance characteristic parameter: PVR=ΔP_re / Q, and the unit of the peripheral resistance value PVR is dyn·s / cm 5 ; The steps of simulating the physiological scenario of the artificial heart after implantation of the artificial valve based on the peripheral resistance value PVR include: when PVR ≥ the first pressure parameter, the physiological scenario is hypertension pathology or heart failure pathology; when PVR ≤ the second pressure parameter, the physiological scenario is a healthy state.

[0013] Furthermore, the internal pressure of the vascular cavity is adjusted by an air pressure controller to change the cavity volume change rate; the pressure-volume relationship curve is recorded in real time, and the compliance value is calculated according to the formula and output as the elastic expansion characteristic parameter: C=ΔV / ΔP_co, wherein ΔV is the cavity volume change, ΔP_co is the cavity pressure change, and the unit of the compliance value C is mL / mmHg; the steps of simulating the physiological scenario of the artificial heart after implantation of the artificial valve based on the compliance value C include: when C≤the first compliance parameter, the physiological scenario is arteriosclerosis pathology or heart failure pathology; when C≥the second compliance parameter, the physiological scenario is a healthy state.

[0014] Furthermore, it also includes the steps of simulating the physiological scenario of the artificial heart after implantation of an artificial valve based on the peripheral resistance value PVR and the compliance value C: synchronously adjusting the throttle valve opening and the internal pressure of the cavity to generate a preset parameter combination: when PVR ≥ the first pressure parameter and C ≤ the first compliance parameter, the physiological scenario is heart failure pathology; when PVR ≤ the second pressure parameter and C ≥ the second compliance parameter, the physiological scenario is a motion state; when PVR ≥ the third pressure parameter and C ≤ the third compliance parameter, the physiological scenario is hypertension combined with arteriosclerosis pathology; according to the preset parameter combination, the pulsating output frequency parameters and the pulsating output flow parameters of the pulsating pump are automatically matched to realize the hemodynamic state of the physiological scenario.

[0015] Furthermore, the vascular system of the artificial heart includes a transparent module, which corresponds to the downstream flow field of the artificial valve, and the transparent module is polycarbonate or optical glass; the transparent module simulates the anatomical morphology of the vascular system, including the right atrial loop structure, the compliance cavity bypass structure, the pulmonary artery structure, and the right ventricular structure; the inner wall surface roughness of the transparent module is ≤0.1μm, and the visible light transmittance of the transparent module is ≥92%.

[0016] Furthermore, the wavelength of the dual-pulse laser is 532nm, the power is 10W, the frame rate of the high-speed camera is 165 frames / s, the resolution is 1920×1200 pixels, the thickness of the light sheet is 1.5mm, and the fluorescent particles evenly distributed in the flow field are excited and the motion trajectory of the fluorescent particles is captured by a high-speed camera and a 35mm focal length lens parallel to the target screen. The concentration of the fluorescent particles is 0.01%.

[0017] Furthermore, the raw blood flow velocity data includes: the instantaneous velocity vector field of the flow field downstream of the artificial valve: the displacement Δx, Δy, Δz and time interval Δt of each fluorescent particle in two-dimensional space / three-dimensional space; the spatiotemporal distribution information of the flow field downstream of the artificial valve: the phase velocity distribution of the systolic period in the cardiac cycle, and the phase velocity distribution of the diastolic period in the cardiac cycle.

[0018] Furthermore, the step of obtaining a fluid dynamics simulation parameter group based on the original blood flow velocity data includes:

[0019] Wall shear stress WSS:

[0020] ,

[0021] μ is the viscosity of the pulsating flow medium, in mPa·s; is the velocity gradient along the wall normal;

[0022] Particle residence time RRT:

[0023] ,

[0024] WSS_mag is the time-averaged wall shear stress amplitude; OSI is the shear oscillation index;

[0025] Shear Oscillation Index OSI:

[0026] ,

[0027] T is a complete cardiac cycle, and τ_ω is the instantaneous wall shear stress vector.

[0028] Furthermore, a random forest classifier was trained based on 300 sets of test data, of which 200 were PIV experimental data of transcatheter pulmonary valve implantation and 100 were PIV experimental data of a healthy control group. The training labels were the presence and absence of thrombus, and the input features were a normalized fluid mechanics parameter group: wall shear stress WSS, particle residence time RRT, and shear oscillation index OSI. In the selected physiological scenario, the wall shear stress WSS, particle residence time RRT, and shear oscillation index OSI were normalized to the [0,1] interval, and the normalized WSS, normalized RRT, and normalized OSI were input into the pre-trained random forest classifier. The pre-trained random forest classifier outputted the thrombosis risk probability value (0-1) of the spatial grid point. When the probability value was ≥0.85, it was marked as a high-risk point. A two-dimensional / three-dimensional thrombosis risk probability map was generated based on the spatial grid data, and the spatial coordinate values ​​and probability values ​​of the high-risk area were marked.

[0029] Furthermore, the step of obtaining a fluid dynamics simulation parameter group based on the raw blood flow velocity data further includes: generating a three-dimensional mesh model of the blood vessel corresponding to the flow field downstream of the artificial valve by 3D scanning based on the anatomical structure of the transparent module, extracting local geometric features of the three-dimensional mesh model of the blood vessel, wherein the local geometric features include the curvature radius R_c and the branch angle θ; mapping the instantaneous velocity vector field of the flow field downstream of the artificial valve to the three-dimensional mesh model of the blood vessel and performing geometric correction on the wall shear stress WSS:

[0030] ,

[0031] Where κ is the dimensionless geometric correction factor; when the branch angle θ ≥ 60°, OSI_new = OSI × 1.3.

[0032] Furthermore, the method further comprises the steps of: real-time monitoring of abnormal blood flow data in the downstream flow field of the artificial valve, wherein the abnormal blood flow state data includes vortex volume ω and mainstream velocity υ, wherein high vortex volume feedback leads to a vortex area where blood is retained, and low velocity feedback leads to a blood flow stagnation area where blood is stagnant; if ω ≥ 100S -1, or υ≤0.1m / s, the pulsating pump feedback adjustment step is started, and the pulsating pump feedback adjustment step includes: adjusting the throttle valve opening to make the split ratio A:B of the pulsating flow medium dynamically match the peripheral resistance value PVR in the range of 800-1200dyn·s / cm 5 between; adjusting the air pressure controller so that the compliance value C ≥ the second compliance parameter; inputting the adjusted fluid mechanics simulation parameter group into the pre-trained valve thrombosis random forest classifier in real time and outputting an updated thrombosis risk probability map under the physiological scenario.

[0033] According to an embodiment of the present invention, using the artificial heart valve thrombosis risk assessment method provided in the first embodiment of the present invention, a second embodiment is provided:

[0034] A prosthetic heart valve thrombosis risk assessment system, comprising:

[0035] An artificial heart module, used for installing the artificial valve to be tested in an artificial heart of an extracorporeal pulsatile circulation system, wherein the extracorporeal pulsatile circulation system further comprises a peripheral resistance module and a compliance structure module;

[0036] a physiological scenario simulation module, configured to adjust the resistance characteristics of the artificial heart's vascular system through the peripheral resistance module and output resistance characteristic parameters, adjust the elastic expansion capacity of the artificial heart's vascular system through the compliance structure module and output elastic expansion characteristic parameters, and obtain the physiological scenario of the artificial heart after implantation of an artificial valve through the resistance characteristic parameters and the elastic expansion characteristic parameters, wherein the physiological scenario includes a healthy state, a pathological state, and a motion state;

[0037] A medium output module is used to select a physiological scenario and start a pulsating pump in the physiological scenario to output a pulsating flow medium to the artificial heart, wherein the pulsating flow medium includes fluorescent particles and controls the pulsating flow medium passing through the artificial valve to circulate along a preset path;

[0038] A blood flow data capture module is used to start a dual-pulse laser to project a light source to the flow field downstream of the artificial valve of the artificial heart, capture the trajectory of fluorescent particles through a high-speed camera, and obtain the original data of blood flow velocity downstream of the artificial valve;

[0039] A simulation parameter module, used to obtain a fluid mechanics simulation parameter group based on the original blood flow velocity data, wherein the fluid mechanics simulation parameter group includes wall shear stress, particle residence time, and shear oscillation index;

[0040] The thrombosis probability output module is used to input a pre-trained valve thrombosis random forest classifier according to the fluid dynamics simulation parameter group and output a thrombosis risk probability map under physiological scenarios.

[0041] Compared with the existing technology, the technical solution provided by this application has unique beneficial effects: by dynamically adjusting the parameters of the peripheral resistance value PVR and the compliance value C, a multi-state hemodynamic environment of healthy state, pathological state and motion state is accurately generated; based on the instantaneous velocity vector field and spatiotemporal distribution information captured by PIV, a computational fluid dynamics simulation parameter group is generated, and a two-dimensional / three-dimensional thrombosis risk probability map is generated through a random forest classifier and the coordinates of high-risk points are marked to achieve accurate prediction of thrombosis probability and thrombosis location; through the physiological scene prediction, data collection, and parameter computer machine learning prediction of the artificial heart, an automated evaluation chain is formed without human intervention, breaking through the static prediction limitations of traditional methods and dependence on manual operation, and providing accurate quantitative data support for valve optimization design and postoperative thrombosis prevention. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0043] in:

[0044] Figure 1 A flowchart of a method for assessing thrombosis risk of an artificial heart valve according to an embodiment;

[0045] Figure 2 1 is a structural block diagram of a device for assessing the risk of thrombosis in an artificial heart valve in one embodiment. DETAILED DESCRIPTION

[0046] In order to enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of this application.

[0047] Example 1

[0048] The technical problem solved by this embodiment is that the existing in vitro simulation system can only simulate a static healthy blood flow environment, and cannot reproduce the hemodynamic characteristics under various conditions such as pathological conditions and motion conditions, such as hypertension and heart failure. The existing thrombosis prediction models mostly rely on a single blood flow parameter, such as the average flow velocity, and lack the coordinated implementation and quantitative analysis of the synergistic effects of multiple factors to simulate a multi-state blood flow environment, resulting in a high misjudgment rate of thrombosis risk points.

[0049] To solve the above technical problems, this embodiment provides a method for assessing the risk of thrombosis in an artificial heart valve. Figure 1 As shown, the steps include:

[0050] S101: installing an artificial valve to be tested in an artificial heart of an extracorporeal pulsatile circulation system, wherein the extracorporeal pulsatile circulation system further includes a peripheral resistance module and a compliance structure module;

[0051] S102: adjusting the resistance characteristics of the vascular system of the artificial heart by the peripheral resistance module and outputting a resistance characteristic parameter, adjusting the elastic expansion capacity of the vascular system of the artificial heart by the compliance structure module and outputting an elastic expansion characteristic parameter, and obtaining a physiological scenario after the artificial heart is implanted with an artificial valve by using the resistance characteristic parameter and the elastic expansion characteristic parameter, wherein the physiological scenario includes a healthy state, a pathological state, and a motion state;

[0052] S103: selecting a physiological scenario and starting a pulsating pump in the physiological scenario to output a pulsating flow medium to the artificial heart, wherein the pulsating flow medium includes fluorescent particles, and controlling the pulsating flow medium passing through the artificial valve to circulate along a preset path;

[0053] S104: starting a double-pulse laser to project a light source toward the flow field downstream of the artificial valve of the artificial heart, capturing the trajectory of the fluorescent particles using a high-speed camera and obtaining raw data of the blood flow velocity downstream of the artificial valve;

[0054] S105: Acquire a fluid mechanics simulation parameter group based on the original blood flow velocity data, wherein the fluid mechanics simulation parameter group includes wall shear stress, particle residence time, and shear oscillation index;

[0055] S106: Input a pre-trained valve thrombosis random forest classifier according to the fluid mechanics simulation parameter group and output a thrombosis risk probability map under the physiological scenario.

[0056] By dynamically adjusting the parameters of the peripheral resistance value PVR and the compliance value C, a multi-state hemodynamic environment of healthy state, pathological state and motion state is accurately generated; based on the instantaneous velocity vector field and spatiotemporal distribution information captured by PIV, a computational fluid dynamics simulation parameter group is generated, and a two-dimensional / three-dimensional thrombosis risk probability map is generated through a random forest classifier and the coordinates of high-risk points are marked to achieve accurate prediction of thrombosis probability and thrombosis location; through the physiological scenario prediction, data collection, and parameter computer machine learning prediction of the artificial heart, an automated evaluation chain is formed without human intervention, breaking through the static prediction limitations of traditional methods and dependence on manual operations, and providing accurate quantitative data support for valve optimization design and postoperative thrombosis prevention.

[0057] Example 2

[0058] A method for assessing the risk of thrombosis in a prosthetic heart valve includes the following steps:

[0059] The artificial valve to be tested is installed in an artificial heart of an extracorporeal pulsating circulation system, which also includes a peripheral resistance module and a compliance structure module; the resistance characteristics of the artificial heart's vascular system are adjusted by the peripheral resistance module and resistance characteristic parameters are output; the elastic expansion capacity of the artificial heart's vascular system is adjusted by the compliance structure module and elastic expansion characteristic parameters are output; the physiological scenario after the artificial heart is implanted with the artificial valve is obtained by the resistance characteristic parameters and the elastic expansion characteristic parameters, and the physiological scenario includes a healthy state, a pathological state and a motion state.

[0060] This embodiment reconstructs an extracorporeal pulsatile circulation system that can be installed with an artificial valve to generate pulsatile flow, allowing the use of different artificial valves to test peripheral resistance and compliance values. In particular, a transparent module made of polycarbonate or optical glass is used to simulate the anatomical structure of the vascular system, including a right atrial circuit structure, a compliance chamber bypass structure, a pulmonary artery structure, and a right ventricle structure. The inner wall surface roughness of the transparent module is ≤0.1μm, the visible light transmittance of the transparent module is ≥92%, and the location of the transparent module corresponds to the downstream flow field of the artificial valve. The peripheral resistance module controls the diversion ratio of the pulsatile flow medium by adjusting the throttle valve: A% of the pulsatile flow medium enters the right atrium, tricuspid valve, and right ventricle through the recirculation path and forms a circuit, and B% of the pulsatile flow medium enters the compliance chamber through the medium bypass. The compliance structure module adjusts the internal pressure of the cavity through an air pressure controller to change the cavity volume change ΔV.

[0061] Specifically, the diversion ratio of the pulsating flow medium is controlled by adjusting the throttle valve opening, so that A% of the pulsating flow medium enters the right atrium through the reflow path and B% of the pulsating flow medium enters the compliance chamber, where A+B=100; the pressure difference ΔP_re between the inlet pressure and the outlet pressure and the blood flow Q of the artificial valve PIV experiment are monitored in real time, and the peripheral resistance value is calculated according to the formula and output as the resistance characteristic parameter: PVR=ΔP_re / Q, and the unit of the peripheral resistance value PVR is dyn·s / cm 5 ;

[0062] The peripheral resistance module simulates the resistance characteristics of the vascular system, specifically the amount of resistance encountered during blood flow. Peripheral resistance directly affects cardiac afterload, blood flow velocity, and shear stress distribution. Under healthy conditions, peripheral resistance is normal. Under pathological conditions, peripheral resistance is significantly elevated. This is due to factors such as increased blood flow resistance caused by vascular stenosis or sclerosis, as well as increased resistance caused by peripheral vasoconstriction due to sympathetic nervous system activation. During exercise, peripheral vasodilation reduces resistance.

[0063] The compliance structure module simulates the elastic expansion capacity of the vascular system. In a healthy state, blood vessels are highly compliant, meaning the aorta is elastic and can buffer pulsating pressures. In pathological conditions, vascular wall calcification reduces compliance, resulting in more severe pressure fluctuations. During exercise, when blood flow accelerates, vascular compliance increases, effectively regulating pulsating pressures.

[0064] The steps of simulating the physiological scenario of the artificial heart after implantation of the artificial valve based on the peripheral resistance value PVR include:

[0065] When PVR ≥ the first pressure parameter, the physiological scenario is hypertension pathology or heart failure pathology;

[0066] When PVR≤the second pressure parameter, the physiological scenario is a healthy state.

[0067] Preferably, the first pressure parameter is 1200 dyn·s / cm 5 , the second pressure parameter is 800dyn·s / cm 5 .

[0068] Specifically, the internal pressure of the blood vessel cavity is adjusted by an air pressure controller to change the cavity volume change rate;

[0069] The pressure-volume relationship curve is recorded in real time, and the compliance value is calculated according to the formula and output as the elastic expansion characteristic parameter: C=ΔV / ΔP_co, where ΔV is the change in cavity volume, ΔP_co is the change in cavity pressure, and the unit of the compliance value C is mL / mmHg. The steps of simulating the physiological scenario of the artificial heart after implantation of the artificial valve based on the compliance value C include:

[0070] When C ≤ the first compliance parameter, the physiological scenario is arteriosclerosis pathology or heart failure pathology;

[0071] When C≥the second compliance parameter, the physiological scenario is a healthy state.

[0072] Preferably, the first compliance parameter is 0.8 mL / mmHg, and the second compliance parameter is 1.5 mL / mmHg.

[0073] In a more complex scenario, the method further includes simulating the physiological scenario of an artificial heart after implantation of an artificial valve based on the peripheral resistance value PVR and the compliance value C: synchronously adjusting the throttle valve opening and the internal pressure of the cavity to generate a preset parameter combination:

[0074] When PVR ≥ the first pressure parameter and C ≤ the first compliance parameter, the physiological scenario is heart failure pathology;

[0075] When PVR≤the second pressure parameter and C≥the second compliance parameter, the physiological scenario is a motion state;

[0076] When PVR ≥ the third pressure parameter and C ≤ the third compliance parameter, the physiological scenario is hypertension combined with arteriosclerosis pathology;

[0077] Preferably, the third pressure parameter is 1400 dyn·s / cm 5 , the third compliance parameter is 0.6mL / mmHg.

[0078] The system automatically adjusts the pulsating pump's output frequency (40-120 bpm) and flow rate (2-7 L / min) based on preset parameters to achieve the desired hemodynamic state in physiological scenarios. Simultaneous adjustment of peripheral resistance (PVR) and compliance (C) allows for simulation of a wider range of physiological scenarios.

[0079] A physiological scenario is selected and a pulsating pump is activated under the physiological scenario to output a pulsating flow medium to the artificial heart. The pulsating flow medium includes fluorescent particles, and the pulsating flow medium is controlled to circulate along a preset path through the artificial valve. A dual-pulse laser is activated to project a light sheet onto the flow field downstream of the artificial valve of the artificial heart. The motion trajectory of the fluorescent particles is captured by a high-speed camera, and raw data of blood flow velocity downstream of the artificial valve is obtained.

[0080] Specifically, the dual-pulse laser has a wavelength of 532nm and a power of 10W. The high-speed camera operates at a frame rate of 165 frames per second, with a resolution of 1920×1200 pixels and a light sheet thickness of 1.5mm. The laser excites fluorescent particles evenly distributed in the flow field, and the trajectory of the fluorescent particles is captured parallel to the target screen using a high-speed camera with a 35mm focal length lens. The fluorescent particle concentration is 0.01%. Using these laser parameters, test methods, and fluorescent particle concentration, data collection of transient eddy currents and low-speed stagnation zones is possible.

[0081] Specifically, the raw blood flow velocity data includes: the instantaneous velocity vector field of the flow field downstream of the artificial valve: the displacement Δx, Δy, Δz and time interval Δt of each fluorescent particle in two-dimensional space / three-dimensional space; the spatiotemporal distribution information of the flow field downstream of the artificial valve: the phase velocity distribution of the systolic period in the cardiac cycle, and the phase velocity distribution of the diastolic period in the cardiac cycle.

[0082] Obtaining a fluid mechanics simulation parameter group based on the original blood flow velocity data, wherein the fluid mechanics simulation parameter group includes wall shear stress, particle residence time, and shear oscillation index;

[0083] Specifically, the step of obtaining a fluid dynamics simulation parameter group based on the original blood flow velocity data includes:

[0084] Wall shear stress WSS:

[0085] ,

[0086] μ is the viscosity of the pulsating flow medium, in mPa·s; the u value can be used to calculate the particle velocity vector through the particle displacement and time interval, which is It is the velocity gradient along the wall normal; when WSS is less than 0.5Pa, the risk of endothelial cell damage increases, which is the trigger point of thrombosis.

[0087] When calculating the velocity gradient along the wall normal, multiple adjacent velocity points on the wall normal can be selected and the velocity gradient can be calculated by the spatial difference method. .

[0088] For example, the velocity distribution in the area ≤0.5 mm from the wall is extracted, and the velocity gradient curve is fitted to calculate the gradient value.

[0089] Particle residence time RRT:

[0090] ,

[0091] WSS_mag is the time-averaged wall shear stress amplitude; OSI is the shear oscillation index;

[0092] For example, the trajectory of the fluorescent particles is tracked with a frame rate greater than 100 frames, and the residence time of the fluorescent particles in a specific area, such as the valve periphery, is counted. When RRT>5s, it reflects blood flow stagnation and the risk of platelet aggregation is increased.

[0093] Specifically, the instantaneous velocity vector field is calculated through the trajectory of the fluorescent particles, and then the time series data of WSS is calculated. The average value of WSS in a complete cardiac cycle is obtained to obtain WSS_mag. OSI is calculated in the next step. OSI reflects the degree of oscillation of the WSS direction over time. Finally, the particle residence time is derived through public demonstration and key parameters such as platelet aggregation risk are evaluated.

[0094] Shear Oscillation Index OSI:

[0095] ,

[0096] T is a complete cardiac cycle, and τ_ω is the instantaneous wall shear stress vector.

[0097] For example, the directional change of the WSS vector within a complete cardiac cycle is extracted and the oscillation intensity of the direction is calculated by integration. When OSI>0.3, the direction of the shear force is frequently reversed, which easily activates platelets.

[0098] The pre-trained valve thrombosis random forest classifier is input according to the fluid dynamics simulation parameter group and outputs the thrombosis risk probability map under the physiological scenario.

[0099] Specifically, a random forest classifier was trained based on 300 sets of test data, of which 200 were PIV experimental data of transcatheter pulmonary valve implantation and 100 were PIV experimental data of a healthy control group. The training labels were the presence and absence of thrombus, and the input features were a normalized fluid mechanics parameter group: wall shear stress WSS, particle residence time RRT, and shear oscillation index OSI. In the selected physiological scenario, the wall shear stress WSS, particle residence time RRT, and shear oscillation index OSI were normalized to the [0,1] interval, and the normalized WSS, normalized RRT, and normalized OSI were input into the pre-trained random forest classifier. The pre-trained random forest classifier outputted the thrombosis risk probability value (0-1) of the spatial grid point. When the probability value was ≥0.85, it was marked as a high-risk point. A two-dimensional / three-dimensional thrombosis risk probability map was generated based on the spatial grid data, and the spatial coordinate values ​​and probability values ​​of the high-risk area were marked.

[0100] The random forest algorithm is composed of 100 decision trees. Each decision tree randomly selects two features for node splitting. The Niki coefficient is used as the splitting criterion. The final probability value is the average of the prediction results of all decision trees.

[0101] For example, WSS < 0.5 Pa, RRT > 5 s, and OSI > 0.3 can be marked as high-risk points to generate a visual thrombosis risk probability map. The risk map visualization method is to divide the flow field into a spatial grid of 0.5 mm × 0.5 mm, and the probability is mapped into different color areas: red areas with a probability ≥ 0.85 represent high-risk thrombosis areas; yellow areas with a probability between 0.6 and 0.85 represent medium-risk thrombosis areas; and green areas with a probability < 0.6 represent low-risk thrombosis areas.

[0102] Example 3

[0103] The technical problem solved by this embodiment is that the existing model's simulation of vascular anatomical structures is static, such as the static simulation of pulmonary artery branches, pulmonary dilatation chambers, etc., while the real blood flow is dynamically affected by the local aggregate morphology, and the traditional method will increase the analysis error of the flow field.

[0104] This solution further generates a 3D mesh model of the blood vessels corresponding to the flow field downstream of the artificial valve by 3D scanning based on the anatomical structure of the transparent module, and extracts local geometric features of the 3D mesh model of the blood vessels, including the curvature radius R_c and the branch angle θ;

[0105] The instantaneous velocity vector field of the flow field downstream of the artificial valve is mapped to the three-dimensional mesh model of the blood vessel and the wall shear stress WSS is geometrically corrected:

[0106] ,

[0107] Among them, κ is a dimensionless geometric correction factor, which is used to correct the wall shear stress WSS. Its function is to quantify the impact of the local anatomical structure of the blood vessel on the flow field. For example, when the vascular curvature radius R_c is small, the blood flow is prone to form vortices, and the risk of thrombosis increases. Amplifying the WSS value by the κ value can more accurately reflect the shear stress abnormalities in such high-risk areas.

[0108] When the branch angle θ≥60°, OSI_new=OSI×1.3.

[0109] When the branch angle θ is greater than 60°, blood flow is easily separated to form a retention area, and the OSI weight needs to be increased to enhance the thrombosis risk prediction results.

[0110] This solution dynamically couples real-time flow field data with anatomical geometry, further improving the calculation accuracy of parameters and optimizing the accurate prediction of the increased risk of thrombosis caused by local geometric deformation after artificial valve implantation.

[0111] Example 4

[0112] The technical problem solved by this embodiment is that the existing prediction method simulates physiological scenarios through preset parameters. Therefore, it is fundamentally impossible to achieve dynamic adjustment of flow field data. The deviation between traditional simple simulations and the actual physiological state is one of the reasons why the accuracy of simulation results needs to be improved.

[0113] In order to solve this technical problem, this embodiment further provides the following steps:

[0114] Real-time monitoring of abnormal blood flow data in the flow field downstream of the artificial valve, wherein the abnormal blood flow state data includes vorticity ω and mainstream velocity υ, wherein high vorticity feedback leads to a vortex area where blood is retained, and low flow velocity feedback leads to a stagnant blood area where blood is stagnant;

[0115] If ω≥100S -1 , or υ≤0.1m / s, the pulsating pump feedback adjustment step is started, and the pulsating pump feedback adjustment step includes: adjusting the throttle valve opening to make the split ratio A:B of the pulsating flow medium dynamically match the peripheral resistance value PVR in the range of 800-1200dyn·s / cm 5 between; adjusting the air pressure controller so that the compliance value C ≥ the second compliance parameter;

[0116] The adjusted fluid dynamics simulation parameter group is input into the pre-trained valve thrombosis random forest classifier in real time and an updated thrombosis risk probability map under physiological scenarios is output.

[0117] This dynamic control method achieves risk reassessment through automatic system adjustment when the flow field is abnormal. It integrates fluid mechanics monitoring and hardware control systems, and improves the system's ability to respond to real physiological conditions.

[0118] Example 5

[0119] The technical problem specifically solved by this embodiment is that the solutions of Examples 1-4 are transparent modular vascular systems with fixed anatomical morphology, while the vascular anatomical structures of the objects that actually need to be simulated, such as the degree of pulmonary artery dilatation, the angle of pulmonary artery branches, the morphology of the right atrium, etc., are significantly different in the vascular anatomical structures of different objects. Simulating only the hemodynamic characteristics still lacks the influence of personalized data of the vascular anatomical structure on the simulation results. However, when the existing vascular system with fixed anatomical morphology is not accurate enough for the thrombosis risk assessment results, the anatomical morphology will not be further considered. After Examples 1-4 more accurately solve the thrombosis risk assessment, it is still necessary to further improve the accuracy of thrombosis risk assessment through personalized simulation of the vascular system.

[0120] Based on the extracorporeal pulsatile circulation system after adding an artificial valve to the artificial heart of Examples 1-4, this example further adds personalized vascular module customization:

[0121] The target's heart structure and large blood vessel structure are acquired through clinical MRI imaging scans. The heart structure includes the right atrium and right ventricle, and the large blood vessel structure includes pulmonary artery parameters. Three-dimensional contour data of the vascular lumen is acquired through medical imaging processing. An individualized vascular model is generated based on the three-dimensional contour data of the vascular lumen. The format of the individualized vascular model file is STL format.

[0122] A personalized transparent vascular module is obtained using a 3D printing device. The transparent material of the personalized transparent vascular module is compatible with the transparent module of the original artificial heart's vascular system, and can be made of polycarbonate with a light transmittance of ≥92%. The inner wall surface roughness of the personalized transparent vascular module is ≤0.1μm. The 3D contour data of the personalized transparent vascular module includes pulmonary artery size, pulmonary artery angle, right atrial circuit curvature, etc.

[0123] The personalized transparent vascular module is replaced with the original transparent module of the artificial heart, and the personalized transparent vascular module, the throttle valve of the peripheral resistance module, and the air pressure controller of the compliance structure module constitute a personalized extracorporeal circulation system;

[0124] The physiological scenario after the artificial heart is implanted with an artificial valve is obtained through resistance characteristic parameters and elastic expansion characteristic parameters, the physiological scenario is selected and the pulsating pump is started under the physiological scenario to output pulsating flow medium to the artificial heart, a fluid dynamics simulation parameter group is obtained according to the original data of blood flow velocity, the fluid dynamics simulation parameter group is corrected according to individualized parameter calibration and a real hemodynamic simulation parameter group is obtained, the individualized parameter calibration step includes: adjusting the throttle valve opening and the air pressure controller according to the peripheral resistance characteristics and elastic expansion capacity in the resting state, adjusting and outputting the corrected resistance characteristic parameters and the corrected elastic expansion force characteristic parameters, and synchronously adjusting the output frequency of the pulsating pump to a resting heart rate of 60-100 times / minute and the flow parameter to a cardiac output of 2-7L / minute;

[0125] The real hemodynamic simulation parameter group includes personalized wall shear stress, personalized particle retention time, and personalized shear oscillation index. The personalized wall shear stress, personalized particle retention time, and personalized shear oscillation index are input into a pre-trained random forest classifier to output a specific thrombosis risk probability map for the specific anatomical structure of a specific target user, and high-risk areas are marked using the specific thrombosis risk probability map.

[0126] By reproducing the patient's true anatomy through a personalized vascular module, the assessment bias caused by the mismatch between the standard module and the patient's actual vascular morphology is resolved. The accuracy of thrombosis risk prediction can be improved by 20%-30%. The output specific risk map can guide physicians in selecting the most suitable valve model. For example, patients with valvular disease and localized pulmonary artery dilatation can be given priority for a short-stem valve to reduce the incidence of postoperative thrombosis. Simply replacing the transparent module and calibrating the parameters is unnecessary, eliminating the need to modify the core measurement and assessment processes of the original solution, making it easy to upgrade and apply to existing experimental platforms.

[0127] Specifically, the calculation formulas for personalized wall shear stress, personalized particle retention time, and personalized shear oscillation index are consistent with those in Examples 1-4. While the protocols in Examples 1-4 used standard transparent module blood flow velocity data for calculation, this protocol utilizes personalized transparent module blood flow velocity technology to account for anatomical differences that lead to different flow velocity distributions, while the parameter calculation logic remains consistent.

[0128] Example 6

[0129] This embodiment provides a system for assessing the risk of thrombosis in an artificial heart valve. Figure 2 Shown, including:

[0130] An artificial heart module 100 is used to install an artificial valve to be tested in an artificial heart of an extracorporeal pulsatile circulation system, wherein the extracorporeal pulsatile circulation system further includes a peripheral resistance module and a compliance structure module;

[0131] a physiological scenario simulation module 200 for adjusting the resistance characteristics of the artificial heart's vascular system through the peripheral resistance module and outputting resistance characteristic parameters, adjusting the elastic expansion capacity of the artificial heart's vascular system through the compliance structure module and outputting elastic expansion characteristic parameters, and obtaining a physiological scenario after the artificial heart is implanted with an artificial valve through the resistance characteristic parameters and the elastic expansion characteristic parameters, wherein the physiological scenario includes a healthy state, a pathological state, and a motion state;

[0132] The medium output module 300 is used to select a physiological scenario and activate a pulsating pump to output a pulsating flow medium to the artificial heart under the physiological scenario. The pulsating flow medium includes fluorescent particles and controls the pulsating flow medium to circulate along a preset path through the artificial valve.

[0133] The blood flow data capture module 400 is used to start a dual-pulse laser to project a light source to the flow field downstream of the artificial valve of the artificial heart, capture the trajectory of the fluorescent particles through a high-speed camera, and obtain the original data of the blood flow velocity downstream of the artificial valve;

[0134] A simulation parameter module 500 is used to obtain a fluid dynamics simulation parameter group based on the original blood flow velocity data, wherein the fluid dynamics simulation parameter group includes wall shear stress, particle residence time, and shear oscillation index;

[0135] The thrombosis probability output module 600 is used to input a pre-trained valve thrombosis random forest classifier according to the fluid dynamics simulation parameter group and output a thrombosis risk probability map under physiological scenarios.

[0136] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When executed, the program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media used in the various embodiments provided herein may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0137] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0138] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A method for assessing the risk of thrombosis in an artificial heart valve, characterized in that: The steps include: installing the artificial valve to be tested in an artificial heart of an extracorporeal pulsatile circulation system, wherein the extracorporeal pulsatile circulation system further comprises a peripheral resistance module and a compliance structure module; The peripheral resistance module adjusts the resistance characteristics of the vascular system of the artificial heart and outputs resistance characteristic parameters, the compliance structure module adjusts the elastic expansion capacity of the vascular system of the artificial heart and outputs elastic expansion characteristic parameters, and obtains the physiological scenario after the artificial heart is implanted with an artificial valve through the resistance characteristic parameters and the elastic expansion characteristic parameters, wherein the physiological scenario includes a healthy state, a pathological state, and a motion state; Selecting a physiological scenario and starting a pulsating pump in the physiological scenario to output a pulsating flow medium to the artificial heart, wherein the pulsating flow medium includes fluorescent particles, and controlling the pulsating flow medium passing through the artificial valve to circulate along a preset path; A dual-pulse laser is started to project a light source onto the flow field downstream of the artificial valve of the artificial heart, and the trajectory of the fluorescent particles is captured by a high-speed camera to obtain the original data of the blood flow velocity downstream of the artificial valve; Obtaining a fluid mechanics simulation parameter group based on the original blood flow velocity data, wherein the fluid mechanics simulation parameter group includes wall shear stress, particle residence time, and shear oscillation index; The pre-trained valve thrombosis random forest classifier is input according to the fluid dynamics simulation parameter group and outputs the thrombosis risk probability map under the physiological scenario; The diversion ratio of the pulsating flow medium is controlled by adjusting the opening of the throttle valve, so that A% of the pulsating flow medium enters the right atrium through the return path and B% of the pulsating flow medium enters the compliance cavity, where A+B=100; The pressure difference ΔP_re between the inlet and outlet pressures of the artificial valve PIV experiment and the blood flow Q are monitored in real time. The peripheral resistance value is calculated according to the formula and output as the resistance characteristic parameter: PVR=ΔP_re / Q. The unit of the peripheral resistance value PVR is ; The steps of simulating the physiological scenario of the artificial heart after implantation of the artificial valve based on the peripheral resistance value PVR include: When PVR ≥ the first pressure parameter, the physiological scenario is hypertension pathology or heart failure pathology; When PVR≤the second pressure parameter, the physiological scenario is a healthy state.

2. The method for assessing the risk of thrombosis in an artificial heart valve according to claim 1, wherein: The internal pressure of the blood vessel cavity is adjusted by the air pressure controller to change the cavity volume change rate; The pressure-volume relationship curve is recorded in real time, and the compliance value is calculated according to the formula and output as the elastic expansion characteristic parameter: C = ΔV / ΔP_co, where ΔV is the change in cavity volume, ΔP_co is the change in cavity pressure, and the unit of the compliance value C is mL / mmHg; The steps of simulating the physiological scenario of the artificial heart after implantation of the artificial valve based on the compliance value C include: When C ≤ the first compliance parameter, the physiological scenario is arteriosclerosis pathology or heart failure pathology; When C≥the second compliance parameter, the physiological scenario is a healthy state.

3. The method for assessing the risk of thrombosis in an artificial heart valve according to claim 1 or 2, wherein: The method further includes the steps of simulating a physiological scenario of an artificial heart after implantation of an artificial valve based on the peripheral resistance value PVR and the compliance value C: Synchronously adjust the throttle valve opening and the internal pressure of the cavity to generate a preset parameter combination: When PVR ≥ the first pressure parameter and C ≤ the first compliance parameter, the physiological scenario is heart failure pathology; When PVR≤the second pressure parameter and C≥the second compliance parameter, the physiological scenario is a motion state; When PVR ≥ the third pressure parameter and C ≤ the third compliance parameter, the physiological scenario is hypertension combined with arteriosclerosis pathology; The pulsating output frequency parameters and pulsating output flow parameters of the pulsating pump are automatically matched according to the preset parameter combination to achieve the hemodynamic state of the physiological scenario.

4. The method for assessing the risk of thrombosis in an artificial heart valve according to claim 1, wherein: The vascular system of the artificial heart includes a transparent module, which corresponds to the downstream flow field of the artificial valve and is made of polycarbonate or optical glass; The transparent module simulates the anatomical morphology of the vascular system, including the right atrial circuit structure, the compliance cavity bypass structure, the pulmonary artery structure, and the right ventricular structure; The surface roughness of the inner wall of the transparent module is ≤0.1 μm, and the visible light transmittance of the transparent module is ≥92%.

5. The method for assessing the risk of thrombosis in an artificial heart valve according to claim 1, wherein: The wavelength of the dual-pulse laser is 532nm, the power is 10W, the frame rate of the high-speed camera is 165 frames / s, the resolution is 1920×1200 pixels, the thickness of the light sheet is 1.5mm, and the fluorescent particles evenly distributed in the flow field are excited. The motion trajectory of the fluorescent particles is captured by a high-speed camera and a 35mm focal length lens parallel to the target screen. The concentration of the fluorescent particles is 0.01%.

6. The method for assessing the risk of thrombosis in an artificial heart valve according to claim 1, wherein: The blood flow velocity raw data includes: The instantaneous velocity vector field of the flow field downstream of the artificial valve: the displacement Δx, Δy, Δz and time interval Δt of each fluorescent particle in two-dimensional space / three-dimensional space; The spatiotemporal distribution information of the flow field downstream of the artificial valve: the phase velocity distribution during the systolic period of the cardiac cycle and the phase velocity distribution during the diastolic period of the cardiac cycle.

7. The method for assessing the risk of thrombosis in an artificial heart valve according to claim 6, wherein: The step of obtaining a fluid dynamics simulation parameter group according to the original blood flow velocity data comprises: Wall shear stress WSS: ， μ is the viscosity of the pulsating flow medium, in mPa·s; is the velocity gradient along the wall normal; Particle residence time RRT: ， WSS_mag is the time-averaged wall shear stress amplitude; OSI is the shear oscillation index; Shear Oscillation Index OSI: ， T is a complete cardiac cycle, and τ_ω is the instantaneous wall shear stress vector.

8. The method for assessing the risk of thrombosis in an artificial heart valve according to claim 6, wherein: A random forest classifier was trained based on 300 test data sets, including 200 PIV experimental data from transcatheter pulmonary valve implantation (TPV) and 100 PIV experimental data from a healthy control group. The training labels were the presence and absence of thrombus, and the input features were the normalized fluid dynamics parameter set: wall shear stress (WSS), particle residence time (RRT), and shear oscillation index (OSI). In the selected physiological scenario, the wall shear stress WSS, particle residence time RRT, and shear oscillation index OSI are normalized to the range [0,1], and the normalized WSS, normalized RRT, and normalized OSI are input into the pre-trained random forest classifier; The pre-trained random forest classifier outputs the thrombosis risk probability value (0-1) of the spatial grid point. When the probability value is ≥0.85, it is marked as a high-risk point. Based on the spatial grid data, a two-dimensional / three-dimensional thrombosis risk probability map is generated, and the spatial coordinate values ​​and probability values ​​of the high-risk area are marked.

9. The method for assessing the risk of thrombosis in an artificial heart valve according to claim 4, wherein: The step of obtaining a fluid dynamics simulation parameter group according to the original blood flow velocity data further includes: Based on the anatomical structure of the transparent module, a 3D mesh model of the blood vessels corresponding to the flow field downstream of the artificial valve is generated through 3D scanning, and local geometric features of the 3D mesh model of the blood vessels are extracted, including the curvature radius R_c and the branch angle θ; The instantaneous velocity vector field of the flow field downstream of the artificial valve is mapped to the three-dimensional mesh model of the blood vessel and the wall shear stress WSS is geometrically corrected: ， Where, κ is the dimensionless geometric correction factor; When the branch angle θ≥60°, OSI_new=OSI×1.

3.

10. The method for assessing the risk of thrombosis in an artificial heart valve according to claim 1 or 2, characterized in that: Also includes the steps: Real-time monitoring of abnormal blood flow data in the flow field downstream of the artificial valve, wherein the abnormal blood flow data includes vorticity ω and mainstream velocity υ, wherein high vorticity feedback leads to a vortex area where blood is retained, and low flow velocity feedback leads to a stagnant blood area where blood is stagnant; If ω≥100S -1 , or υ≤0.1m / s, the pulsating pump feedback adjustment step is started, and the pulsating pump feedback adjustment step includes: adjusting the throttle valve opening to make the diversion ratio A:B of the pulsating flow medium dynamically match the peripheral resistance value PVR in the range of 800-1200 between; adjusting the air pressure controller so that the compliance value C ≥ the second compliance parameter; The adjusted fluid dynamics simulation parameter group is input into the pre-trained valve thrombosis random forest classifier in real time and an updated thrombosis risk probability map under physiological scenarios is output.

11. A prosthetic heart valve thrombosis risk assessment system, characterized in that: A method for assessing thrombosis risk of an artificial heart valve according to any one of claims 1 to 10, the system comprising: An artificial heart module, used for installing the artificial valve to be tested in an artificial heart of an extracorporeal pulsatile circulation system, wherein the extracorporeal pulsatile circulation system further comprises a peripheral resistance module and a compliance structure module; a physiological scenario simulation module, configured to adjust the resistance characteristics of the artificial heart's vascular system through the peripheral resistance module and output resistance characteristic parameters, adjust the elastic expansion capacity of the artificial heart's vascular system through the compliance structure module and output elastic expansion characteristic parameters, and obtain the physiological scenario of the artificial heart after implantation of an artificial valve through the resistance characteristic parameters and the elastic expansion characteristic parameters, wherein the physiological scenario includes a healthy state, a pathological state, and a motion state; A medium output module is used to select a physiological scenario and start a pulsating pump in the physiological scenario to output a pulsating flow medium to the artificial heart, wherein the pulsating flow medium includes fluorescent particles and controls the pulsating flow medium passing through the artificial valve to circulate along a preset path; A blood flow data capture module is used to start a dual-pulse laser to project a light source to the flow field downstream of the artificial valve of the artificial heart, capture the trajectory of fluorescent particles through a high-speed camera, and obtain the original data of blood flow velocity downstream of the artificial valve; A simulation parameter module, used to obtain a fluid mechanics simulation parameter group based on the original blood flow velocity data, wherein the fluid mechanics simulation parameter group includes wall shear stress, particle residence time, and shear oscillation index; The thrombosis probability output module is used to input a pre-trained valve thrombosis random forest classifier according to the fluid dynamics simulation parameter group and output a thrombosis risk probability map under physiological scenarios.

Images