Blood flow source proportion prediction method and device, computer device, and storage medium

Abstract

The application relates to a blood flow source proportion prediction method and device, computer equipment and a storage medium. The method comprises the following steps: obtaining a hemodynamic simulation result of a target object after simulated transjugular intrahepatic portosystemic shunt (TIPS) surgery; the hemodynamic simulation result comprises predicted blood flow velocity fields and predicted blood flow pressure fields of each blood vessel branch of a portal vein system, a hepatic artery system and a hepatic vein system; according to the hemodynamic simulation result, the blood flow directions of each blood vessel branch are determined; a scalar transport model provided with a coupling boundary condition is constructed according to the blood flow directions; the coupling boundary condition is determined based on the blood flow concentration mixing relationship among the blood vessel branches; and the scalar transport model is solved to obtain the proportion of a corresponding target blood flow source in a target position of the target object. The method can realize accurate evaluation of the distribution proportion of blood flow of different sources after TIPS surgery.

Description

Technical Field

[0001] This application relates to the field of medical information technology, and in particular to a method, apparatus, computer device, and storage medium for predicting the proportion of blood flow sources. Background Technology

[0002] Transjugular intrahepatic portosystemic shunt (TIPS) is an important means of controlling the portal pressure gradient (PPG) in patients with portal hypertension. However, the portal-systemic shunt established by the TIPS stent after the procedure may allow blood flow, for example from the superior mesenteric vein (rich in intestinal toxins), to bypass the liver and enter the systemic circulation directly, potentially inducing hepatic encephalopathy and thus posing a risk of complications.

[0003] Traditional techniques attempt to simulate and predict the hemodynamic state after TIPS, such as by reconstructing a three-dimensional model of the portal vein or simulating a model connecting the portal vein and hepatic veins to obtain postoperative PPG simulation results. However, such approaches suffer from insufficient modeling accuracy, mismatches between parameter adjustments and measured values, inconsistencies between the model and actual anatomical structures, and limitations in boundary condition settings. Therefore, they are difficult to provide effective quantitative assessment support for the risk of postoperative hepatic encephalopathy and other complications.

[0004] Therefore, existing technologies do not provide an accurate method for assessing the distribution ratio of blood flow from different sources after TIPS. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, device, computer equipment, and storage medium for predicting the proportion of blood flow from different sources after TIPS to address the above-mentioned technical problems.

[0006] Firstly, this application provides a method for predicting the proportion of blood flow sources, the method comprising:

[0007] Obtain the hemodynamic simulation results of the target object after a simulated transjugular intrahepatic portosystemic shunt procedure; the hemodynamic simulation results include the predicted blood flow velocity field and predicted blood flow pressure field of each vascular branch of the portal vein system, hepatic artery system, and hepatic vein system;

[0008] Based on the hemodynamic simulation results, the blood flow direction of each of the blood vessel branches is determined;

[0009] Based on the blood flow direction, a scalar transport model with coupled boundary conditions is constructed; the coupled boundary conditions are determined based on the blood flow concentration mixing relationship between each of the blood vessel branches.

[0010] Solving the scalar transport model yields the proportion of the target blood flow source at the target location of the target object.

[0011] In one embodiment, determining the blood flow direction of each of the blood vessel branches based on the hemodynamic simulation results includes:

[0012] Based on the predicted blood flow velocity field and the predicted blood flow pressure field, the blood flow direction at the exit of each portal vein branch and each hepatic artery branch and the entrance of each hepatic vein branch is determined respectively; the blood flow direction is that the blood flow of the hepatic artery branch flows into the portal vein branch and the hepatic vein branch, the blood flow of the hepatic artery branch merges with the blood flow of the portal vein branch and flows into the hepatic vein branch, or the blood flow of the hepatic artery branch merges with the blood flow of the hepatic vein branch and flows into the portal vein branch.

[0013] In one embodiment, constructing a scalar transport model with coupled boundary conditions based on the blood flow direction includes:

[0014] Based on the target blood flow source, at least one scalar transport model is constructed;

[0015] Based on the blood flow direction, determine the blood flow concentration mixing relationship between the outlet of each portal vein branch and each hepatic artery branch and the inlet of each hepatic vein branch;

[0016] Based on the blood flow concentration mixing relationship, the coupling boundary conditions of the scalar transport model are set.

[0017] In one embodiment, when the blood flow direction of the hepatic artery branch is towards the portal vein branch and the hepatic vein branch, determining the blood flow concentration mixing relationship includes:

[0018] The average blood flow concentration at the exit section of the hepatic artery branch is determined as the blood flow concentration at the exit of the corresponding portal vein branch and the entrance of the hepatic vein branch.

[0019] In one embodiment, when the blood flow direction of the hepatic artery branch merges with the blood flow of the portal vein branch and flows into the hepatic vein branch, determining the blood flow concentration mixing relationship includes:

[0020] The average blood flow concentration after mixing is calculated based on the mass flow rate and average blood flow concentration at the outlet section of the hepatic artery branch and the mass flow rate and average blood flow concentration at the outlet section of the portal vein branch.

[0021] The average blood flow concentration after mixing is determined as the blood flow concentration at the entrance of the hepatic vein branch.

[0022] In one embodiment, when the blood flow direction of the hepatic artery branch merges with the blood flow of the hepatic vein branch and flows into the portal vein branch, determining the blood flow concentration mixing relationship includes:

[0023] The average blood flow concentration after mixing is calculated based on the mass flow rate and average blood flow concentration at the exit section of the hepatic artery branch and the mass flow rate and average blood flow concentration at the inlet section of the hepatic vein branch.

[0024] The average blood flow concentration after mixing is determined as the blood flow concentration at the exit of the portal vein branch.

[0025] In one embodiment, solving the scalar transport model to obtain the proportion of the target blood flow source at the target location of the target object includes:

[0026] With the predicted blood flow velocity field and predicted blood flow pressure field remaining unchanged, the scalar transport model is solved in steady state to obtain the blood flow concentration distribution at each location within the region corresponding to the scalar transport model.

[0027] Based on the blood flow concentration distribution, the blood flow concentration corresponding to the target blood flow source at the target location is determined as the proportion of the corresponding target blood flow source.

[0028] By comparing the proportions at the same target location under different simulated surgical plans, the impact of different stent parameters (such as diameter, length, and placement location) on the degree of shunting of blood from a specific source (such as enteric blood) can be quantitatively assessed, thus providing a quantitative basis for doctors to select optimized surgical plans that reduce the risk of hepatic encephalopathy.

[0029] In one embodiment, obtaining the hemodynamic simulation results of the target object after a simulated transjugular intrahepatic portosystemic shunt includes:

[0030] A preset vascular model and zero-dimensional impedance data of the target object are obtained; the preset vascular model is reconstructed based on the preoperative angiography image of the target object, including the portal vein model, hepatic vein model, and hepatic artery model; the zero-dimensional impedance data is obtained by calibrating the preset vascular model based on the preoperative blood flow status data of the target object.

[0031] Based on the preset vascular model and target stent parameters, construct the target postoperative model;

[0032] Based on the zero-dimensional impedance data and the central arterial pressure of the target object, hemodynamic simulation is performed on the target postoperative model to obtain the hemodynamic simulation results.

[0033] In one embodiment, the process of acquiring the zero-dimensional impedance data includes:

[0034] Based on the preoperative blood flow status data of the target object, the initial boundary conditions of the preset vascular model are determined; the preoperative blood flow status data includes at least portal vein pressure, inferior vena cava pressure, blood flow velocity, and corresponding vascular radius;

[0035] Based on the initial boundary conditions, pressure and flow information are transmitted between the coupling nodes of the portal vein model, hepatic artery model, and hepatic vein model of the preset blood vessel model to perform multi-scale steady-state hemodynamic simulation and obtain initial simulation results.

[0036] Based on the portal vein pressure and the simulated portal vein pressure in the initial simulation results, determine the pressure correction amount;

[0037] Based on the pressure correction amount, the pressure in the portal vein region in the initial simulation results is corrected;

[0038] Based on the pressure and flow rate in the initial simulation results, the portal vein terminal impedance, hepatic vein terminal impedance, hepatic artery impedance, and distal vascular bed impedance are calculated respectively to obtain the zero-dimensional impedance data.

[0039] Secondly, this application provides a blood flow source ratio prediction device, the blood flow source ratio prediction device comprising:

[0040] The simulation result acquisition module is used to acquire the hemodynamic simulation results of the target object after a simulated transjugular intrahepatic portosystemic shunt procedure; the hemodynamic simulation results include the predicted blood flow velocity field and predicted blood flow pressure field of each vascular branch of the portal vein system, hepatic artery system and hepatic vein system;

[0041] The blood flow direction determination module is used to determine the blood flow direction of each of the blood vessel branches based on the hemodynamic simulation results.

[0042] The transport model construction module is used to construct a scalar transport model with coupled boundary conditions based on the blood flow direction; the coupled boundary conditions are determined based on the blood flow concentration mixing relationship between the blood vessel branches.

[0043] The source ratio solution module is used to solve the scalar transport model to obtain the proportion of the target blood flow source at the target location of the target object.

[0044] Thirdly, this application provides a computer device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described above.

[0045] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described above.

[0046] The aforementioned method, apparatus, computer equipment, and storage medium for predicting the proportion of blood flow sources acquire hemodynamic simulation results of the target object after a simulated transjugular intrahepatic portosystemic shunt procedure. These simulation results include predicted blood flow velocity and pressure fields for each vascular branch of the portal venous system, hepatic artery system, and hepatic venous system. Based on these simulation results, the blood flow direction of each vascular branch is determined. A scalar transport model with coupled boundary conditions is constructed based on these blood flow directions. These coupled boundary conditions are determined based on the blood flow concentration mixing relationship between each vascular branch. The scalar transport model is then further processed. The solution process yields the proportion of blood flow originating from the target location within the target object. By utilizing hemodynamic simulation results that incorporate velocity and pressure fields from multiple systems, the limitations of traditional methods that only focus on macroscopic pressure gradients can be overcome, providing a high-fidelity physical field for microscopic blood flow tracking. A scalar transport model with coupled boundary conditions is constructed using the blood flow direction. Based on blood flow concentration, mass conservation and concentration hybrid logic are satisfied at vascular junctions, which can solve the problem that traditional methods cannot distinguish between different sources of blood flow. By solving this model, quantitative source tracing can be achieved on the basis of qualitative simulation, enabling accurate assessment of the proportion of blood flow from different sources after TIPS. Attached Figure Description

[0047] Figure 1 This is a diagram illustrating the application environment of a blood flow source proportion prediction method in one embodiment.

[0048] Figure 2 This is a flowchart illustrating a method for predicting the proportion of blood flow sources in one embodiment;

[0049] Figure 3 This is a flowchart illustrating the blood flow source ratio prediction method in another embodiment;

[0050] Figure 4 This is a schematic diagram of three blood flow directions in one embodiment;

[0051] Figure 5 This is a three-dimensional schematic diagram of a vascular stage model in one embodiment;

[0052] Figure 6 This is a structural block diagram of a blood flow source ratio prediction device in one embodiment;

[0053] Figure 7 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0055] The blood flow source ratio prediction method provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or placed on a cloud or other network server. Terminal 102 can interact with server 104 to obtain hemodynamic simulation results of a target object after a simulated transjugular intrahepatic portosystemic shunt procedure. The hemodynamic simulation results include predicted blood flow velocity and pressure fields for each vascular branch of the portal vein system, hepatic artery system, and hepatic vein system. Based on the hemodynamic simulation results, the blood flow direction of each vascular branch is determined. Based on the blood flow direction, a scalar transport model with coupled boundary conditions is constructed. The coupled boundary conditions are determined based on the blood flow concentration mixing relationship between each vascular branch. The scalar transport model is solved to obtain the proportion of the target blood flow source at the target location of the target object. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, etc. Server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0056] In one embodiment, such as Figure 2 As shown, a method for predicting the proportion of blood flow sources is provided, and this method is applied to... Figure 1 Taking terminal 102 as an example, the method for predicting the proportion of blood flow sources includes:

[0057] Step S100: Obtain the hemodynamic simulation results of the target object after a simulated transjugular intrahepatic portosystemic shunt procedure.

[0058] The target subject referred to in this embodiment can be an individual preparing to undergo transjugular intrahepatic portosystemic shunt (TIPS). Because medical personnel need to ensure the absence of postoperative complications before performing TIPS, simulations are conducted preoperatively based on the different stents used to predict postoperative outcomes. The hemodynamic simulation results after TIPS can be obtained by acquiring a digital anatomical model of the target subject, incorporating changes in vascular structure after stent implantation, for simulation purposes. Furthermore, the digital anatomical model can be a three-dimensional model reconstructed from medical images, or a personalized model modified incorporating pathological features.

[0059] Hemodynamic simulation results can be comprehensive data describing the velocity vector distribution and pressure scalar distribution of blood flow within the portal vein, hepatic artery, and hepatic venous system of the target object under steady-state simulation after TIPS. This data provides initial physical field boundary conditions and driving data for subsequent determination of blood flow direction and construction of a scalar transport model. Furthermore, the hemodynamic simulation results can be obtained through numerical simulation calculations on a three-dimensional geometric model containing the portal vein, hepatic artery, and hepatic venous system using a computational fluid dynamics solver.

[0060] In this embodiment, the hemodynamic simulation results may include the predicted blood flow velocity field and predicted blood flow pressure field of each vascular branch of the portal vein system, hepatic artery system, and hepatic vein system.

[0061] In this context, vascular branches can be independent ductal units with clearly defined origins, terminations, and diameters within the portal venous system, hepatic artery system, and hepatic venous system. Correspondingly, during simulation, they can serve as basic channels for blood flow transmission and spatial nodes where scalar concentration mixing occurs, thus defining the specific locations for determining blood flow direction and setting boundary conditions.

[0062] Predicting the blood flow velocity field can be achieved by using three-dimensional matrix data that characterizes the displacement vector distribution of blood micro-elements per unit time on the vascular mesh nodes. This data can be used to determine the calculation coefficients of the convection term in the scalar transport equation.

[0063] Predicting the blood flow pressure field can be scalar distribution data representing the intensity of the force exerted by blood on the vessel wall and internal micro-elements at various spatial points within the vascular domain. It can be used to verify the convergence and physiological rationality of hemodynamic simulations and indirectly reflect the distribution of vascular resistance to help determine shunt efficiency.

[0064] Obtaining hemodynamic simulation results of the target object after a simulated transjugular intrahepatic portosystemic shunt procedure can be achieved by performing steady-state simulations on a multi-scale model to calculate the velocity and pressure distributions within the portal vein, hepatic vein, and hepatic artery, thus obtaining velocity and pressure field data for each system. Furthermore, hemodynamic simulation results can be obtained by reading completed transient or steady-state simulation result files from a local database, or by connecting to a high-performance computing cluster in real-time via an interface to obtain online simulation data streams. Alternatively, approximate flow field data can be rapidly reconstructed based on a simplified parametric model, providing a high-fidelity physical environment for blood flow tracking and ensuring that microscopic hybrid calculations are based on an accurate macroscopic flow field.

[0065] Step S200: Determine the blood flow direction of each blood vessel branch based on the hemodynamic simulation results.

[0066] Among them, the blood flow direction can be the macroscopic flow direction of blood in each blood vessel branch determined based on the predicted blood flow velocity field vector characteristics.

[0067] Based on hemodynamic simulation results, the blood flow direction of each vascular branch is determined. For example, this can be achieved by traversing the velocity vectors of all boundary surfaces of the vascular mesh and determining the inflow or outflow state based on the angle between the vectors and the surface normal vectors, thus marking the flow direction attribute of each branch. In other exemplary embodiments, the blood flow direction can also be calculated through binarization based on the dot product sign of the velocity vectors, topological sorting based on streamline tracing techniques, or deduction based on the pressure gradient direction. This clarifies the calculation order and boundary type of scalar transport, avoiding numerical divergence or mixed logic errors caused by unclear flow direction.

[0068] Step S300: Construct a scalar transport model with coupled boundary conditions based on the blood flow direction.

[0069] The scalar transport model can be a mathematical model based on the convection-diffusion equation, used to simulate the migration and mixing process of blood flows from different sources such as the superior mesenteric vein, splenic vein, hepatic artery, and inferior vena cava in a given flow field. For example, the scalar transport model can transform the abstract differences in blood flow sources into calculable concentration values, enabling a quantitative calculation of the distribution ratio of blood from each source after mixing.

[0070] The coupling boundary conditions are determined based on the mixing relationship of blood flow concentrations between different vascular branches. These coupling boundary conditions can be mathematical constraints set at the interface where vascular branches intersect, enforced by the law of conservation of mass and the weighted average rule of component concentrations. They can be dynamically determined based on the blood flow direction of the portal vein, hepatic vein, and hepatic artery branches. By setting coupling boundary conditions, it can be ensured that the scalar concentrations of blood from different sources are correctly mixed according to the flow rate ratio at the confluence, avoiding non-physical jumps or spurious diffusion in numerical calculations. In an exemplary embodiment, the coupling boundary conditions can be derived from the flow rate and scalar concentration values ​​of each upstream branch using a flux balance formula and applied to the model boundary.

[0071] The blood flow concentration mixing relationship can describe the linear superposition of multiple blood flows from different sources at the junction of blood vessels with mass flow weights. For example, it can be expressed as the average blood flow concentration after mixing, calculated based on the mass flow rate and average blood flow concentration at the outlet section of the upstream branch.

[0072] The coupling boundary conditions are set according to the mixing rules of blood flow concentration between various vascular branches. This can be done by identifying all the intersection nodes in the vascular network, calculating the average blood flow concentration at the exit section of the corresponding branch based on the actual blood flow inflow and confluence direction between the portal vein, hepatic artery and hepatic vein, or calculating the average blood flow concentration after mixing according to the formula, and designating it as the blood flow concentration at the inlet of the downstream branch.

[0073] Furthermore, based on the blood flow direction, a scalar transport model with coupled boundary conditions is constructed. This can be done by initializing the scalar transport model, loading the velocity field data obtained from steady-state simulation, configuring the computational domain topology according to the blood flow direction determined between the portal vein, hepatic artery, and branches of the hepatic vein, and then applying the coupled boundary conditions calculated based on each flow direction to the corresponding mesh surface.

[0074] Step S400: Solve the scalar transport model to obtain the proportion of the target blood flow source at the target location of the target object.

[0075] The target location can be a specific spatial region or cross-section within the vascular network designated as an observation point. For example, it could be one or more cross-sections at the TIPS stent site, or a critical site related to the risk of hepatic encephalopathy, such as a systemic circulation inlet. By determining the target location to extract sampling points for the final calculation results, the proportion of various blood components at that location can be determined.

[0076] The target blood flow source can be a specific anatomical blood input source that needs to be independently tracked and quantified in hemodynamic analysis after TIPS. By establishing a dedicated computational channel for each source, decoupled analysis and parallel computation of multiple blood sources in complex vascular networks can be achieved, avoiding proportional distortion caused by interference between signals from different sources. In this embodiment, key entry vessels can be predefined as target blood flow sources based on anatomical knowledge to achieve independent tracking and quantitative evaluation of blood flow from different vessels. For example, depending on the organ, the target blood flow source can include, but is not limited to, one or more of the following: intestinal-derived highly toxic blood flow, spleen-derived metabolic blood flow, hepatic artery-derived oxygenated blood flow, and collateral circulation compensatory blood flow.

[0077] A target blood flow source can refer to a specific anatomical source from which the proportion of blood flow needs to be determined. This proportion can be the percentage of total blood flow volume or mass flow originating from the target blood flow source at a target location. In an exemplary embodiment, the blood flow concentration at the inlet boundary of the target blood flow source can be set to 1, while other inlets can be set to 0 for differentiation. It is understood that in some applications, the proportion of the target blood flow source can be used to quantify the extent to which enterotoxins bypass the liver and enter the systemic circulation, thereby directly relating to the risk level of hepatic encephalopathy.

[0078] Solving the scalar transport model can be exemplified by initiating a numerical iterative algorithm to solve for the distribution of the scalar concentration field under a fixed velocity field background until the residuals converge to a preset accuracy threshold, thereby obtaining the steady-state concentration distribution result. Furthermore, a steady-state iterative solution can be performed using an implicit time-progression scheme to obtain the steady-state concentration distribution result. Additionally, a multigrid acceleration technique can be employed to improve the convergence speed.

[0079] To obtain the proportion of the target blood flow source at the target location of the target object, one can extract the blood concentration at the target location from the steady-state concentration distribution results, thereby obtaining the proportion of the target location from the target blood flow source.

[0080] Furthermore, various transjugular intrahepatic portosystemic shunt (TIPS) procedures can be simulated using different TIPS stent designs, and the concentration ratio of the target blood flow source under different TIPS stent designs can be compared, providing auxiliary reference for medical personnel.

[0081] Taking the risk assessment of hepatic encephalopathy after TIPS surgery as an example, the blood flow source ratio prediction method in this embodiment can be used for a patient with portal hypertension preparing for TIPS surgery. Using preoperative angiographic images, a three-dimensional model including the portal vein, hepatic artery, and hepatic vein is reconstructed, and the geometry after stent implantation is simulated. By performing hemodynamic simulation, detailed velocity and pressure field data within the postoperative portal venous system are obtained. Then, by analyzing the velocity field, the blood flow direction can be determined, including the blood flow from branches such as the superior mesenteric vein and splenic vein converging into the main portal vein and flowing into the hepatic vein through the stent. Subsequently, blood from the intestine can be defined as a high-toxin-concentration tracer, and blood from the spleen as another tracer. The mixed concentration at the portal vein confluence is calculated based on the flow rate of each branch, and this is used as the coupling boundary condition to construct a scalar transport model. After solving the scalar transport model, the tracer concentration value is extracted at the inferior vena cava entrance, thereby determining the proportion of blood from the intestine and spleen at the inferior vena cava entrance. For example, when the blood volume in the superior mesenteric vein reaches 65%, it indicates that most intestinal toxins bypass liver metabolism and directly enter the systemic circulation, potentially posing a risk of hepatic encephalopathy. Thus, the blood flow source ratio prediction method in this embodiment allows medical personnel to simulate postoperative blood flow for various stent parameters (such as diameter, length, and installation location) before surgery, enabling comparison of multiple options without invasive procedures and providing strong data reference for doctors to determine the optimal stent option.

[0082] This embodiment provides a method for predicting the proportion of blood flow sources. It obtains hemodynamic simulation results of a target object after a simulated transjugular intrahepatic portosystemic shunt procedure. The hemodynamic simulation results include predicted blood flow velocity and pressure fields for each vascular branch of the portal venous system, hepatic artery system, and hepatic venous system. Based on the hemodynamic simulation results, the blood flow direction of each vascular branch is determined. Based on the blood flow direction, a scalar transport model with coupled boundary conditions is constructed. These coupled boundary conditions are determined based on the blood flow concentration mixing relationship between the vascular branches. The scalar transport model is then solved to obtain… The proportion of the target blood flow source corresponding to the target location of the target object is determined by using hemodynamic simulation results that include the velocity and pressure fields of multiple systems. This breaks through the limitations of traditional methods that only focus on macroscopic pressure gradients, providing a high-fidelity physical field for microscopic blood flow tracking. A scalar transport model with coupled boundary conditions is constructed using the blood flow direction. Based on the blood flow concentration mixing relationship, mass conservation and concentration mixing logic are satisfied at the vascular junction, which can solve the problem that traditional methods cannot distinguish between different flows from the same source. By solving this model, quantitative source tracing can be achieved on the basis of qualitative simulation, enabling accurate assessment of the proportion of blood flow from different sources after TIPS.

[0083] In one embodiment, the blood flow direction of each vascular branch is determined based on hemodynamic simulation results, including:

[0084] Based on the predicted blood flow velocity field and the predicted blood flow pressure field, the blood flow direction at the exit of each portal vein branch and each hepatic artery branch and the entrance of each hepatic vein branch is determined respectively. The blood flow direction is that the blood flow of the hepatic artery branch flows into the portal vein branch and the hepatic vein branch. The blood flow of the hepatic artery branch merges with the blood flow of the portal vein branch and flows into the hepatic vein branch. In rare special cases caused by abnormal local pressure gradient, the blood flow of the hepatic artery branch may also merge with the blood flow of the hepatic vein branch and flow into the portal vein branch.

[0085] The portal vein branch outlet can be the fluid output interface where each sub-branch of the portal vein system intersects with the main trunk or other blood vessels. In this embodiment, the portal vein branch outlet can be used as a key monitoring point for determining the direction of blood flow on the portal vein side. It can identify topological states such as retrograde flow of hepatic artery blood into the portal vein, normal flow of portal vein blood into the hepatic vein, and retrograde return of hepatic vein blood, thereby determining the method for setting the coupling boundary conditions.

[0086] The hepatic artery branch outlet can be the fluid output interface at the end of each branch of the hepatic artery system, representing the spatial boundary of high-pressure arterial blood flowing out of the branch into the microcirculation or an abnormal shunt path. In this embodiment, the hepatic artery branch outlet can be used as one of the nodes for monitoring whether high-pressure arterial blood is flowing backward or abnormally, determining whether it directly flows into the portal vein, mixes with portal vein blood and flows into the hepatic vein, or mixes with hepatic vein blood and flows into the portal vein, etc., thereby determining the corresponding coupling boundary condition assignment.

[0087] The hepatic vein branch inlet can be a fluid input interface in the hepatic venous system for receiving blood from one or more of the following: the hepatic sinusoidal space, the portal vein, and the hepatic artery. For example, the hepatic vein branch inlet can be a purely hepatic inlet that receives only blood from the hepatic sinusoidal space, a mixed inlet that receives blood shunt from the portal vein, an abnormal inlet that receives direct arterial perfusion, or a special inlet that receives mixed blood flow from the hepatic artery and hepatic vein, etc. This embodiment does not impose any limitations on these aspects.

[0088] Blood flow in the hepatic artery branches can originate from blood pumped by the heart and transported via the hepatic arterial system. Understandably, after TIPS, the blood flow in the hepatic artery branches may exhibit abnormal flow direction due to changes in pressure gradients. For example, there may be pressure competition between the blood flow in the hepatic artery branches and the portal vein branches. Therefore, when the arterial pressure is significantly higher than the portal vein pressure, it may drive blood flow into the portal venous system. This could result in various flow patterns, such as flow into the portal vein and hepatic vein, merging with portal vein blood into the hepatic vein, or merging with hepatic vein blood into the portal vein. Thus, the direction of blood flow in the hepatic artery branches can characterize the presence of hemodynamic disturbances caused by arterial-portal vein shunting, thereby determining whether its outlet concentration should be specified as the inlet concentration of other branches or included in the mixed concentration calculation.

[0089] The blood flow in the portal vein branches can be a low-pressure blood stream rich in metabolic products and potential toxins, originating from abdominal organs such as the intestines and spleen and transported via the portal vein system. In this embodiment, the direction of blood flow in the portal vein branches may be altered by the pressure of the hepatic artery, thereby affecting the distribution ratio of toxins in the systemic circulation. In this embodiment, the direction of blood flow in the portal vein branches can be used to confirm whether toxins bypass liver metabolism and directly enter the systemic circulation, whether they flow backward, or whether they change direction after mixing with arterial blood, thereby accurately calculating the proportion of blood flow originating from different locations.

[0090] Blood flow in the hepatic vein branches can be a mixed flow of blood that has been metabolized by the liver and possibly unmetabolized shunt blood, preparing to flow back into the inferior vena cava. It can also be the source of retrograde flow into the portal vein after mixing with arterial blood under abnormal pressure gradients. In one specific embodiment, the blood flow in the hepatic vein branches can be used as the final assessment target for calculating toxin load.

[0091] For example, based on the predicted blood flow velocity field and the predicted blood flow pressure field, the blood flow direction at the exit of each portal vein branch and each hepatic artery branch, and at the entrance of each hepatic vein branch, can be determined by reading the velocity vector data on the grid surface of a specified anatomical node, calculating the dot product of the vector and the surface normal vector, and combining this with the pressure field difference on both sides of the node to comprehensively determine the macroscopic flow direction attribute of the blood flow to match three specific blood flow directions. In other embodiments, a dual verification method of pressure gradient direction and velocity vector can be combined, or the statistical analysis of the number of times streamlines cross the cross section can be used to achieve accurate identification of local complex blood flow patterns and distinguish between normal antegrade blood flow and pathological retrograde blood flow or eddies.

[0092] Furthermore, determining whether blood flow from a hepatic artery branch flows into a portal vein branch or a hepatic vein branch can be achieved by detecting the velocity vector direction at the exit of the hepatic artery branch. If the vector points into the portal vein branch and a path pointing towards the hepatic vein also exists, then it can be determined that the hepatic artery blood flow flows into the portal vein or the hepatic vein. For example, the average blood flow concentration at the exit section of the hepatic artery branch can be specified as the blood flow concentration at the exit of the portal vein branch and the entrance of the hepatic vein branch.

[0093] To determine whether the blood flow from the hepatic artery branches and the portal vein branches primarily flows into the hepatic vein branches after their confluence, an example can be made by superimposing the inflow fluxes of the hepatic artery and portal vein at the identified confluence point and checking whether the total velocity vector after superposition points towards the entrance of the hepatic vein branch. This confirms whether the mixed blood flow returns along the expected path. Furthermore, the mass flow rate and average blood concentration of the two blood flows can be calculated to determine the mixed concentration.

[0094] Understandably, in rare special cases caused by abnormal local pressure gradients, the blood flow from the hepatic artery branches and the hepatic vein branches may merge and flow into the portal vein branches. Determining whether the blood flow from the hepatic artery branches and the hepatic vein branches merges and flows into the portal vein branches can be achieved by monitoring the flow field characteristics of the area where the hepatic veins and arteries intersect. When the velocity vector points from the hepatic veins or arteries into the interior of the portal vein branches, it can be considered that there is a reverse reflux or short-circuit circulation where the blood flow from the hepatic artery and the hepatic veins merges and flows into the portal vein.

[0095] This embodiment provides a method for predicting the proportion of blood flow sources. By determining the blood flow direction at the exit points of each portal vein branch and each hepatic artery branch, and at the entrance points of each hepatic vein branch, based on the predicted blood flow velocity field and the predicted blood flow pressure field, this method can accurately locate the blood flow vector direction at key anatomical nodes by making in-depth use of microscopic physical quantities. It establishes a clear logical judgment mechanism for complex local hemodynamic phenomena that may occur after TIPS, avoiding the problem of assessment distortion caused by ignoring local reverse blood flow or complex confluence mechanisms. This makes the calculation of the mixing ratio of blood flow from different blood sources with normal hepatic artery blood flow in the systemic circulation more consistent with physiological reality, thereby improving the accuracy and reliability of blood flow source ratio assessment.

[0096] In one embodiment, constructing a scalar transport model with coupled boundary conditions based on the blood flow direction includes:

[0097] Based on the target blood flow source, construct at least one scalar transport model;

[0098] Based on the direction of blood flow, determine the blood concentration mixing relationship between the outlets of each portal vein branch and each hepatic artery branch and the inlet of each hepatic vein branch;

[0099] Based on the blood flow concentration mixing relationship, the coupling boundary conditions of the scalar transport model are set.

[0100] In this embodiment, the number of scalar transport equations can be determined according to the number of blood flow sources to be analyzed. By establishing a dedicated computing channel for each source, decoupled analysis and parallel computing of multi-source blood in complex vascular networks can be achieved, thereby avoiding proportional distortion caused by mutual interference of signals from different sources.

[0101] Furthermore, based on the target blood flow source, at least one scalar transport model is constructed. This can be done by initializing an independent convection-diffusion equation solver instance for each defined target blood flow source, setting the inlet boundary concentration of that specific source to a unit value, and setting the inlet concentration of all other non-target sources to zero, thereby establishing a tracer migration computation domain specific to that source.

[0102] Furthermore, based on the target blood flow source, multiple independent scalar transport models (i.e., solver instances) are constructed. For a given model, the inlet boundary concentration of its corresponding target source is set to 1, while the boundary concentrations of all other source inlets are set to 0. These multiple models can be computed sequentially or simultaneously in parallel in a multi-core or cluster environment without interference. In this way, the migration and mixing process of each specific source of blood can be tracked independently.

[0103] The blood flow concentration mixing relationship can be used to describe the linear superposition relationship of multiple blood flows with different scalar concentrations at a vascular junction, based on the law of conservation of mass and the principle of momentum balance. For example, the average concentration at the downstream section can be equal to the sum of the products of the flow rates of each upstream branch and their corresponding concentrations, divided by the total flow rate; that is, the average blood flow concentration after mixing is calculated by weighting the mass flow rate and the average blood flow concentration. In this embodiment, the blood flow concentration mixing relationship can be based on the determined blood flow direction topology, extracting the mass flow rate and instantaneous concentration values ​​of each upstream branch at the junction, applying a weighted average formula to derive the theoretical mixing concentration value at the downstream interface, and setting corresponding coupling boundary conditions according to the blood flow directions of the portal vein, hepatic vein, and hepatic artery branches.

[0104] Based on the blood flow concentration mixing relationship, the coupling boundary conditions of the scalar transport model are set. For example, the calculated downstream mixing concentration value can be applied to the mesh surface of the corresponding downstream branch inlet as a boundary condition, so that the scalar transport equation automatically satisfies the mixing constraint during the solution process, forming a tight coupling between the upstream and downstream computational domains.

[0105] This embodiment provides a method for predicting the proportion of blood flow sources. By constructing at least one scalar transport model based on the target blood flow source, determining the blood flow concentration mixing relationship between the outlets of each portal vein branch and each hepatic artery branch and the inlet of each hepatic vein branch according to the blood flow direction, and setting the coupling boundary conditions of the scalar transport model based on the blood flow concentration mixing relationship, the abstract target blood flow source can be instantiated into a concrete digital entity. Furthermore, by establishing a matter conservation logic at key nodes of the vascular network, the mixing relationship is transformed into a mandatory constraint term of the scalar transport model, ensuring that the model strictly follows the real physiological mixing law during solution. This effectively avoids the problems of overly simplified boundary condition settings and the inability to simulate the complex mixing process of multi-source blood in traditional techniques. The final calculation result can accurately quantify the specific proportion of blood flow from different sources at a specific location, achieving the technical effect of improving the accuracy and reliability of blood flow source proportion assessment.

[0106] In one embodiment, when the blood flow direction of the hepatic artery branch is towards the portal vein branch and the hepatic vein branch, determining the blood flow concentration mixing relationship includes:

[0107] The average blood flow concentration at the exit section of the hepatic artery branch was determined as the blood flow concentration at the exit of the corresponding portal vein branch and the entrance of the corresponding hepatic vein branch.

[0108] The average blood flow concentration at the exit section of the hepatic artery branch can be a scalar representative value calculated by mass flow weighted integral of the local tracer concentration values ​​of all grid nodes at the exit section of the hepatic artery branch. This value characterizes the overall component composition of the blood flow and simplifies the complex cross-sectional distribution into a single boundary input value.

[0109] Determining the average blood flow concentration at the hepatic artery branch outlet section as the corresponding portal vein branch outlet concentration can be achieved by extracting the calculated average blood flow concentration value from the hepatic artery branch outlet section and directly assigning it to the boundary condition variable of the portal vein branch outlet to match the hepatic artery blood flow direction setting into the portal vein. By setting the average blood flow concentration at the hepatic artery branch outlet section as the corresponding portal vein branch outlet concentration, a rigid concentration transfer channel from the hepatic artery to the portal vein can be established. This ensures that the concentration of the target blood source does not undergo artificial attenuation or distortion when retrograde arterial blood enters the portal venous system, thus truly reflecting the dilution or mixing effect of arterial blood on the portal vein toxin flow.

[0110] Correspondingly, the average blood flow concentration at the exit section of the hepatic artery branch can be determined as the blood flow concentration at the inlet of the corresponding hepatic vein branch. This can be achieved by reusing the average blood flow concentration value at the exit section of the same hepatic artery branch and simultaneously assigning it to the boundary condition variable at the inlet of the hepatic vein branch. This ensures that the arterial blood flowing into the systemic circulation also inherits the same source concentration characteristics, thus completing the concentration coupling on the hepatic vein side.

[0111] This embodiment provides a method for predicting the proportion of blood flow sources. When the blood flow direction of the hepatic artery branch flows to the portal vein branch and the hepatic vein branch, the average blood flow concentration at the exit section of the hepatic artery branch is determined as the blood flow concentration at the exit of the corresponding portal vein branch and the blood flow concentration at the entrance of the corresponding hepatic vein branch. By accurately extracting the average blood flow concentration at the exit section of the hepatic artery branch and simultaneously assigning it as the input concentration at the exit of the downstream portal vein branch and the entrance of the hepatic vein branch, a precise material transport logic of one source and two sinks is established. This avoids the problems of tracer concentration distortion and mass non-conservation caused by the simplification or fragmentation of boundary conditions when dealing with cross-system reverse perfusion and multi-directional shunt. It ensures that the concentration characteristics of blood components originating from the hepatic artery are completely, losslessly and consistently transmitted when they diffuse to different systems. This enables the scalar transport model to realistically reproduce the distribution behavior of high-pressure arterial blood in complex short-circuit paths, thereby improving the accuracy and reliability of blood flow source proportion assessment.

[0112] In one embodiment, when the blood flow direction of the hepatic artery branch is such that it merges with the blood flow of the portal vein branch and flows into the hepatic vein branch, determining the blood flow concentration mixing relationship includes:

[0113] The average blood flow concentration after mixing is calculated based on the mass flow rate and average blood flow concentration at the exit section of the hepatic artery branch and the mass flow rate and average blood flow concentration at the exit section of the portal vein branch.

[0114] The average blood flow concentration after mixing is determined as the blood flow concentration at the entrance of the hepatic vein branch.

[0115] The mass flow rate at the hepatic artery branch outlet section can be defined as the total mass of blood flowing through the hepatic artery branch outlet control surface within a specific time step. It can be calculated using the blood density, velocity field distribution, and geometric area integral of the hepatic artery branch outlet section. This characterizes the mass flux of high-pressure arterial blood injected into the mixing node, ensuring accurate quantification of the impact of high-velocity or large-diameter arterial blood on the mixing concentration. Furthermore, the mass flow rate at the hepatic artery branch outlet section can be obtained by performing an area integral on the dot product of the velocity vector and surface normal vector of each node on the hepatic artery branch outlet grid surface, multiplied by the blood density.

[0116] The mass flow rate at the portal vein branch outlet section can be the total mass of blood flowing through the portal vein branch outlet control surface within a specific time step. It can characterize the material flux of low-pressure blood flow from abdominal organs such as the intestines and spleen into the mixing node. Furthermore, it can reflect the proportion of blood flow rich in metabolites or toxins in the mixing process.

[0117] The average blood flow concentration at the portal vein branch outlet section can be an area-weighted average of scalar concentration values ​​representing a specific blood flow source at the portal vein branch outlet section, characterizing the purity or characteristic intensity of the blood flow before entering the mixing node. In this embodiment, the average blood flow concentration at the portal vein branch outlet section can be obtained by solving the transport equation down to the statistical average concentration value obtained at the outlet section.

[0118] The average blood flow concentration after mixing can be a scalar concentration value representing the new state after the mixing of the two blood flows, calculated based on the law of conservation of mass under the physical assumption of complete merging of hepatic artery and portal vein blood flows. In this embodiment, the average blood flow concentration after mixing can directly depend on the mass flow rate and concentration values ​​of the two upstream branches, and simultaneously serve as the uniquely identifiable value of the blood flow concentration at the inlet of the downstream hepatic vein branch. Accordingly, based on the mass flow rate and average blood flow concentration at the outlet section of the hepatic artery branch, and the mass flow rate and average blood flow concentration at the outlet section of the portal vein branch, the average blood flow concentration after mixing can be calculated by extracting real-time mass flow rate data and corresponding average scalar concentration data at the outlets of the two upstream branches of the hepatic artery and portal vein, applying a mass-weighted averaging algorithm, adding the solute fluxes of the two blood flows, and dividing by the total mass flow rate to obtain a unique mixed concentration value.

[0119] The blood flow concentration at the inlet of the hepatic vein branch can be a scalar concentration boundary condition value set at the beginning of the hepatic vein branch, which characterizes the blood component characteristics before entering the systemic circulation. In this embodiment, the blood flow concentration value at the inlet of the hepatic vein branch can be determined based on the hemodynamic state of the upstream hepatic artery and portal vein.

[0120] Correspondingly, the average blood flow concentration after mixing can be determined as the blood flow concentration at the entrance of the hepatic vein branch by assigning the mixed concentration value calculated in the previous step to all nodes on the boundary grid surface of the hepatic vein branch entrance, thereby using this concentration as the starting condition for scalar transport solution.

[0121] This embodiment provides a method for predicting the proportion of blood flow sources. It calculates the average blood flow concentration after mixing based on the mass flow rate and average blood flow concentration at the exit sections of the hepatic artery and portal vein branches. This mixed average blood flow concentration is then used as the blood flow concentration at the inlet of the hepatic vein branch. By using mass flow rate as a precise weighting factor combined with the average blood flow concentration at each input end to perform mass conservation calculations, the mixed average blood flow concentration is obtained. This effectively avoids calculation biases caused by simple arithmetic averages or empirical estimations. The calculated mixed concentration is assigned as the blood flow concentration at the inlet of the downstream hepatic vein branch. This allows for the construction of high-fidelity dynamic coupling boundary conditions, accurately simulating the actual mixing and dilution process of abnormal and normal hepatic artery blood flow in vivo. It provides physiologically accurate and highly reliable quantitative data support, thereby improving the accuracy and reliability of blood flow source proportion assessment.

[0122] In one embodiment, when the blood flow direction of the hepatic artery branch merges with the blood flow of the hepatic vein branch and flows into the portal vein branch, determining the blood flow concentration mixing relationship includes:

[0123] The average blood flow concentration after mixing is calculated based on the mass flow rate and average blood flow concentration at the exit section of the hepatic artery branch and the mass flow rate and average blood flow concentration at the inlet section of the hepatic vein branch.

[0124] The average blood flow concentration after mixing is determined as the blood flow concentration at the exit of the portal vein branch.

[0125] Specifically, the average blood flow concentration after mixing is calculated based on the mass flow rate and average blood flow concentration at the exit section of the hepatic artery branch and the entrance section of the hepatic vein branch. This can be achieved by reading the real-time mass flow rate data and corresponding average blood flow concentration data at the identified exit points of the hepatic artery and the entrance points of the hepatic vein, substituting them into the mass conservation weighted average formula, and solving for the theoretical concentration value after the two blood flows merge. This reproduces the physical mixing process of high-velocity arterial blood and venous blood at the abnormal confluence point, avoids concentration estimation errors caused by ignoring flow weights, and improves the calculation accuracy of the multi-scale model.

[0126] In this embodiment, the blood flow concentration at the portal vein branch outlet can be a boundary condition value of a specific output interface of the portal vein system, set to be equal to the average blood flow concentration after the hepatic artery and hepatic vein blood flows back into the portal vein in a special scenario. This value characterizes the component properties of abnormally refluxed blood entering the main portal vein. Furthermore, the blood flow concentration at the portal vein branch outlet can be used as the driving boundary of the scalar transport model on the portal vein side, ensuring that the downstream computational domain can correctly inherit the characteristics of the complexly mixed blood upstream. This allows for accurate tracking of the transport behavior of toxins or tracers along the reflux path, enabling multi-scale model simulation of abnormal flow directions.

[0127] The average blood flow concentration after mixing is determined as the blood flow concentration at the portal vein branch outlet. This can be achieved by mapping the value calculated in the previous step and assigning it to the boundary condition variable of the portal vein branch outlet section, overriding the original preset or default value at that location. This makes it a fixed solution or flux input of the scalar transport equation on that boundary, thereby establishing the correct physical connection on the abnormal reverse flow path. This allows the scalar transport model to simulate the real process of blood flowing backward from the arterial and venous mixing zone into the portal vein, solving the complex reverse boundary coupling problem that traditional models cannot handle.

[0128] This embodiment provides a method for predicting the proportion of blood flow sources. By calculating the average blood flow concentration after mixing based on the mass flow rate and average blood flow concentration at the exit section of the hepatic artery branch and the inlet section of the hepatic vein branch, the method determines the average blood flow concentration at the exit of the portal vein branch. By reproducing the physical mixing process of high-velocity arterial blood and venous blood at the abnormal confluence point and ensuring the mass conservation of the tracer in the cross-system flow, and by establishing the correct physical connection on the abnormal reverse flow path, the scalar transport model can simulate the real process of blood flowing backward from the mixing zone into the portal vein. This method can achieve high-precision prediction of the distribution proportion of blood flow from different sources, solve the evaluation bias caused by the inability to handle complex reverse mixing flow, and improve the accuracy and reliability of blood flow source proportion assessment.

[0129] In one embodiment, solving the scalar transport model yields the proportion of the target blood flow source at the target location of the target object, including:

[0130] With the predicted blood flow velocity field and predicted blood flow pressure field remaining unchanged, the scalar transport model is solved in steady state to obtain the blood flow concentration distribution at each location within the region corresponding to the scalar transport model.

[0131] Based on the blood flow concentration distribution, determine the blood flow concentration of the corresponding target blood flow source at the target location, which is used as the proportion of the corresponding target blood flow source.

[0132] Steady-state solutions can be defined as numerical calculations performed on the scalar transport equations, iterating over time until the residuals converge to zero or below a preset threshold, while keeping the flow field driving variables, including velocity and pressure fields, constant. The aim is to obtain a concentration distribution solution that does not change over time. It is understood that steady-state solutions can eliminate transient fluctuations and quickly obtain the final state of blood mixing after equilibrium, thereby significantly reducing computational resource consumption and improving evaluation efficiency. For example, steady-state solutions can employ implicit time-progression schemes or pseudo-time-step methods, updating only the scalar field variables while keeping the convection term coefficients fixed, until the concentration changes in adjacent iterations satisfy the convergence criterion.

[0133] Blood flow concentration distribution can be considered as continuous scalar field data, representing the relative content of blood from different sources, mapped onto the three-dimensional spatial grid nodes of the vascular network after the scalar transport model has been solved. In essence, blood flow concentration distribution can visually demonstrate the migration paths, mixing ranges, and enrichment regions of blood from various sources within the blood vessels, thus providing a comprehensive data view for extracting proportional data at specific locations.

[0134] Furthermore, in this embodiment, the calculated flow velocity and pressure data can be read and set as a constant field. The solver of the fluid dynamics equations can be frozen, and only the scalar transport equation solver can be activated to perform iterative calculations until convergence. This yields the blood concentration distribution at each location within the region corresponding to the scalar transport model, thereby decoupling the flow field calculation from the component transport calculation. This avoids the high cost of repeated fluid dynamics simulations and ensures that the concentration calculation is based on an accurate physical flow field background.

[0135] Based on the blood flow tracer concentration distribution, the relative content of the corresponding target blood flow source in the target location is determined. This can be done by locating the geometric coordinates or cross-sectional markers of the target location in the generated concentration field data, extracting the concentration values ​​of the grid nodes in that area and performing statistical processing to obtain the relative content of the corresponding target blood flow source in the target location, which is then used as the proportion of the corresponding target blood flow source.

[0136] This embodiment provides a method for predicting the proportion of blood flow sources. By employing a decoupled computation strategy, it maintains the high-precision predicted blood flow velocity and pressure fields obtained in the preceding steps while performing steady-state solutions on the scalar transport model. This avoids repeated, time-consuming fluid dynamics iterations, significantly reducing computational costs and improving evaluation efficiency. Simultaneously, by simulating the diffusion and convection of blood from different sources using a fixed real flow field background, it can efficiently obtain the blood flow concentration distribution across the entire region. By directly reading the specific blood flow concentration at the target location, its value is directly mapped to the proportion of the target blood flow source. This transforms the complex multiphase flow mixing problem into a more manageable scalar transport problem. While preserving the real flow velocity and pressure field, it improves computational speed and accurately quantifies the proportion of blood flow from specific sources after TIPS, thereby achieving the technical effect of improving the accuracy and reliability of blood flow source proportion assessment.

[0137] In one embodiment, obtaining the hemodynamic simulation results of the target object after a simulated transjugular intrahepatic portosystemic shunt includes:

[0138] The preset vascular model and zero-dimensional impedance data of the target object are obtained. The preset vascular model is reconstructed based on the preoperative angiography images of the target object, including the portal vein model, hepatic vein model and hepatic artery model. The zero-dimensional impedance data is obtained by calibrating the preset vascular model based on the preoperative blood flow status data of the target object.

[0139] Based on the pre-set vascular model and target stent parameters, construct the target postoperative model;

[0140] Based on zero-dimensional impedance data and the central arterial pressure of the target object, hemodynamic simulation was performed on the postoperative model of the target object to obtain the hemodynamic simulation results.

[0141] The preset vascular model can refer to a three-dimensional anatomical structure model reconstructed based on the preoperative angiography image of the target object. In this embodiment, the preset vascular model is obtained by reconstructing the preoperative angiography image of the target object.

[0142] Angiographic images can be acquired using angiography techniques such as CTA (CT angiography), DSA (digital subtraction angiography), and MRA (magnetic resonance angiography). Through medical image segmentation and surface reconstruction techniques, vascular boundaries are extracted from the preoperative angiographic images to generate a continuous three-dimensional tubular network structure, thus obtaining a pre-defined vascular model. Furthermore, angiographic images can be images showing vascular contrast in the liver region of the target object.

[0143] In this embodiment, the preset vascular model includes a portal vein model, a hepatic vein model, and a hepatic artery model. By including the geometry and topology of at least the portal vein, hepatic vein, and hepatic artery, the individualized spatial configuration of the liver vessels of the target object can be characterized.

[0144] The portal vein model can represent the main portal vein and its tributaries, encompassing a digital model of the luminal structure from the confluence of the superior mesenteric vein and splenic vein to the bifurcation region before entering the liver. This model can be used to track blood inflow and pressure distribution within the portal venous system. The hepatic vein model can represent the main hepatic vein and its major tributaries, connecting the outflow tract within the liver parenchyma to the adjacent inferior vena cava entrance. It serves as a crucial exit path for blood return after portal shunt and participates in pressure gradient calculations. The hepatic artery model can represent the proper hepatic artery and its branches, reflecting the three-dimensional configuration of the hepatic arterial blood supply system. It participates in the calculation of total hepatic blood flow balance and influences the competitive perfusion relationship between the portal vein and the shunt.

[0145] Zero-dimensional impedance data can be a set of lumped parameters reflecting the flow resistance characteristics of various branches in the preoperative vascular system of the target object. It is expressed in impedance values, representing the degree of obstruction to blood flow by vascular segments such as the superior mesenteric vein, splenic vein, portal vein, hepatic artery, hepatic vein, and inferior vena cava. This provides boundary condition inputs that conform to the individual's physiological state for hemodynamic simulation, thereby improving the realism of model predictions. For example, zero-dimensional impedance data may include, but is not limited to, one or more of the following: inlet boundary impedance, branch shunt impedance, and outlet reflux impedance.

[0146] Zero-dimensional impedance data is obtained by calibrating a pre-set vascular model based on the preoperative blood flow status of the target object. For example, the preoperative blood flow status can be blood flow velocity and pressure data obtained by techniques such as preoperative Doppler ultrasound and pressure guidewire measurement, and the zero-dimensional impedance data is obtained by inverse calibration and optimization of the preoperative blood flow status combined with the pre-set vascular model.

[0147] The preset vascular model and zero-dimensional impedance data of the target object obtained in this embodiment can, for example, be obtained by calling a processed individualized model file and parameter set from local storage or a remote server. In an exemplary embodiment, DICOM images can be automatically extracted through a hospital PACS system and a background reconstruction task can be triggered to construct a preset vascular model and obtain zero-dimensional impedance data.

[0148] The target stent can be a stent intended for implantation into a target object. Correspondingly, the target stent parameters can be a set of parameters characterizing the physical properties and spatial configuration of the stent to be implanted. In this embodiment, the target stent parameters can be pre-stored or provided by the user or a host computer. It is understood that the portal vein pressure gradient prediction in this embodiment is applied in the preoperative stage. The target stent parameters correspond to multiple proposed target stent schemes. By simulating the hemodynamic state of a preset vascular model of the target object after the intervention of the target stent, the portal vein pressure gradient is obtained, thereby providing a predictive reference for the target stent before surgery. In an exemplary embodiment, the target stent parameters can be obtained by the clinician selecting a specific model according to the surgical plan and determining the implantation path of the target stent in the anatomical coordinate system through image registration.

[0149] For example, the target stent parameters may include physical parameters and spatial parameters. The physical parameters may include, but are not limited to, one or more of diameter, length, material stiffness, etc., and the spatial parameters may be, for example, the connection position between the portal vein and the hepatic vein.

[0150] The target postoperative model can be an updated model based on the pre-set vascular model after the intervention of the target stent. It is used to simulate the intrahepatic artificial shunt channel formed after TIPS surgery, its connection with the portal vein, hepatic vein and adjacent inferior vena cava, and the changes in vascular access after stent implantation, so as to realize the numerical extrapolation of the postoperative blood flow status.

[0151] Correspondingly, constructing a target postoperative model can involve implanting an artificial conduit with specific geometric parameters into a pre-defined vascular model and adjusting the local connection topology to generate a new overall structure, thereby obtaining the target postoperative model. In an exemplary embodiment, the origin and termination points between the portal vein and hepatic vein can be selected in the pre-defined vascular model according to the physical and spatial parameters of the planned target stent, and a conduit conforming to the target stent parameters can be inserted. Alternatively, Boolean operations can be used to fuse the stent geometry with the host blood vessel to generate a continuous lumen. Furthermore, the Lagrange particle method can be used to dynamically adjust the contact interface of the conduit wall, thereby achieving a digital reproduction of the vascular access changes after TIPS surgery, enabling virtual testing of the post-intervention state.

[0152] Central arterial pressure (CAP) can be the pressure value measured at the root of the aorta or the proximal large artery in the systemic circulation of the target subject. As the driving force of systemic blood circulation, it affects the baseline pressure level throughout the model. In one exemplary embodiment, CAP is measured by a non-invasive sphygmomanometer of the brachial artery in the upper arm; it can also be measured by other methods. This embodiment does not limit the method of obtaining CAP. Exemplarily, CAP may include, but is not limited to, one or more of the following: systolic CAP, diastolic CAP, and mean CAP.

[0153] Hemodynamic simulation can be a numerical solution process based on the equations of mass and momentum conservation to deduce the flow state of blood in a vascular network, thereby achieving quantitative prediction of key indicators such as pressure, flow velocity, and shear force within the postoperative vascular system. In an exemplary embodiment, a three-dimensional computational fluid dynamics method can be used, with the central arterial pressure serving as the upstream boundary driving condition for the hemodynamic simulation. Given the geometric structure and boundary conditions, the fluid dynamics equations are solved to realize the simulation process. Furthermore, depending on the prediction target, the hemodynamic simulation can employ steady-state simulation, or it can perform transient pulsating flow simulation, coupled fluid-structure interaction (FSI) simulation, etc. For example, to assist in assessing the long-term impact of the target stent on PPG changes after surgery, steady-state simulation can be used to predict the portal vein pressure under steady-state conditions. When it is necessary to assist in assessing the long-term impact of the target stent's own stability on PPG changes, different initial conditions can be set for the inlet, and transient pulsating flow simulation can be performed to obtain the prediction results.

[0154] Hemodynamic simulation of the target postoperative model can be performed based on zero-dimensional impedance data and the central arterial pressure of the target object. This can be achieved by setting the hepatic artery inlet pressure to a constant central arterial pressure, setting the inferior vena cava outlet as the blood flow velocity boundary condition, and using the pressure boundary conditions calculated from the zero-dimensional model for the other inlets. The numerical solver is then used to calculate the pressure and velocity distribution across the entire field, thereby achieving hemodynamic simulation.

[0155] Hemodynamic simulation results can include predicted portal vein pressure gradient values, which are the average pressure difference between the main portal vein and the inferior vena cava outlet output by the hemodynamic simulation. It can be understood that if the actual portal vein pressure gradient value is below 12 mmHg, the risk of varicose vein bleeding can be considered effectively controlled.

[0156] Accordingly, determining the hemodynamic simulation results of the target postoperative model can be achieved by extracting the pressure time series of the mid-portal vein and the inferior vena cava outlet from the simulation results and calculating their mean pressure difference. For example, the gradient can be calculated by averaging the pressure curves of multiple cardiac cycles from the simulation results, or the fundamental component can be extracted from the simulation results based on frequency domain analysis and then restored to the effective pressure difference, thereby providing a reference for output indicators for clinical decision-making by medical personnel.

[0157] This embodiment provides a method for predicting the proportion of blood flow sources. By acquiring a pre-defined vascular model and zero-dimensional impedance data of the target object, a pre-defined vascular model including a portal vein model, a hepatic vein model, and a hepatic artery model is reconstructed based on the target object's preoperative angiography images. The pre-defined vascular model is then calibrated according to the target object's preoperative blood flow state to obtain zero-dimensional impedance data. This allows the pre-defined vascular model to more accurately represent the structure of the target object's hepatic vascular system, addressing the root cause of poor prediction accuracy caused by inaccurate key structures in the model. By constructing a target postoperative model based on the pre-defined vascular model and target stent parameters, the changes in vascular pathways after TIPS surgery can be simulated. Hemodynamic simulation of the target postoperative model is performed based on zero-dimensional impedance data and the target object's central arterial pressure, and the hemodynamic simulation results of the corresponding target postoperative model are determined. This allows for the prediction of the postoperative blood flow redistribution state in the preoperative stage, and the extraction of the average pressure difference between the portal vein and hepatic vein, enabling the prediction results to more accurately reflect the actual postoperative PPG change trend, thus improving the technical effect of hemodynamic simulation accuracy.

[0158] In one embodiment, the process of acquiring zero-dimensional impedance data includes:

[0159] Based on the preoperative blood flow data of the target subject, determine the initial boundary conditions of the preset vascular model; the preoperative blood flow data should include at least portal vein pressure, inferior vena cava pressure, blood flow velocity and corresponding vascular radius;

[0160] Based on the initial boundary conditions, pressure and flow information are transmitted between the coupling nodes of the portal vein model, hepatic artery model, and hepatic vein model in the preset vascular model to perform multi-scale steady-state hemodynamic simulation and obtain the initial simulation results.

[0161] Based on the portal vein pressure and the simulated portal vein pressure in the initial simulation results, the pressure correction amount is determined;

[0162] The pressure in the portal vein region in the initial simulation results is corrected based on the pressure correction amount.

[0163] Based on the pressure and flow rate in the initial simulation results, the portal vein terminal impedance, hepatic vein terminal impedance, hepatic artery impedance, and distal vascular bed impedance were calculated to obtain zero-dimensional impedance data.

[0164] The preoperative blood flow status data can be a set of vascular system operating parameters reflecting the target object's physiological state before surgery, including at least portal vein pressure, inferior vena cava pressure, blood flow velocity, and corresponding vessel radius. This provides realistic initial boundary conditions for the vascular model, ensuring that the simulation input is consistent with the target object's physiological state. For example, the preoperative blood flow status data can be obtained by collecting actual physiological values ​​of one or more key sites such as the portal vein, hepatic vein, hepatic artery, inferior vena cava, splenic vein, and superior mesenteric vein through one or more methods such as Doppler ultrasound, pressure measurement, and CTA imaging.

[0165] Initial boundary conditions can be determined by mapping preoperative blood flow data to the values ​​at the inlet and outlet of a pre-defined vascular model, thereby defining the initial state of the simulation calculation and the initial distribution trend of blood flow within the model. For example, depending on the type of boundary physical quantity, initial boundary conditions can include, but are not limited to, one or more of pressure boundary conditions, velocity boundary conditions, flow rate boundary conditions, and impedance boundary conditions. Further, it can involve assigning measured pressure, velocity, and vessel radius data to the corresponding boundary nodes of the model, and transferring pressure and flow rate information between coupled nodes to obtain the constraints of the computational domain.

[0166] Determining the initial boundary conditions based on the preoperative blood flow data of the target object can be achieved by reading the pressure, velocity, and radius values ​​from the preoperative blood flow data, mapping them to the inlet and outlet nodes of the preset vascular model, and adding the flow rate of the portal vein branch outlet to the flow rate of the hepatic artery branch outlet and assigning the value to the hepatic vein branch inlet to form the initial boundary conditions. This can establish the physical constraints required for simulation calculation and ensure that the model startup state conforms to the patient's preoperative physiological characteristics.

[0167] Multiscale steady-state hemodynamic simulation can be a numerical calculation process that combines macroscopic vascular geometry with microscopic resistance characteristics. Under steady-state assumptions, it solves the blood flow equations and can be used to simulate the stable flow state of blood in a vascular network, providing intermediate data support for impedance parameter calculation. In an exemplary embodiment, multiscale steady-state hemodynamic simulation can transfer pressure and flow information between coupled nodes, add the flow rate at the portal vein branch outlet to the flow rate at the hepatic artery branch outlet and assign the result to the hepatic vein branch inlet, iteratively solving until convergence to obtain a stable flow field distribution.

[0168] The initial simulation results can be the simulation results after reaching a steady state through multi-scale steady-state hemodynamic simulation based on initial boundary conditions. Before pressure correction, the initial simulation results contain pressure and flow distribution data for each node in the model. Alternatively, the initial simulation results can be obtained from the raw data files containing flow and pressure distributions for each branch generated after the multi-scale steady-state hemodynamic simulation process has run. For example, depending on the physical quantities contained in the data, the initial simulation results may include, but are not limited to, one or more of the following: pressure field results, flow field results, and shear force results.

[0169] Transferring pressure and flow information between coupled nodes in the model can be achieved by exchanging data at model interfaces of different scales. This involves assigning the pressure at the inlet of the hepatic vein branch to the outlet of the portal vein branch and the outlet of the hepatic artery branch, thus ensuring the conservation relationship between the macroscopic geometric model and the microscopic resistance model. Furthermore, transferring pressure and flow information between coupled nodes can be achieved by using an explicit coupling strategy to iteratively transfer data or by using an implicit coupling strategy to simultaneously solve a system of equations. This achieves data connectivity across multiple scales, ensuring both blood flow continuity and pressure balance.

[0170] Furthermore, multi-scale steady-state hemodynamic simulations can be performed to obtain initial simulation results. This can involve conducting steady-state coupled computational fluid dynamics simulations on three three-dimensional models, calculating the stable flow field within the vascular network under initial boundary conditions, and outputting the flow rate and pressure distribution of each branch. This can generate uncalibrated simulation baseline data, revealing the differences between the initial assumptions of the model and the actual state, thus providing support for subsequent corrections.

[0171] The pressure correction can be the difference between the measured portal vein pressure and the simulated portal vein pressure in the initial simulation results. Furthermore, determining the pressure correction based on the portal vein pressure and the simulated portal vein pressure can be achieved by extracting the measured blood pressure at any portal vein outlet section and the corresponding pressure value from the initial simulation results, and then calculating the difference between the two.

[0172] Understandably, the pressure correction amount can characterize the deviation between the initial assumptions of the model and the actual state. Using the pressure correction amount can eliminate systematic errors, making the pressure distribution in the simulation model closer to the physiological level of the target object. In this embodiment, the pressure correction amount can be used as a feedback signal to adjust the acquisition of internal model parameters. For example, the pressure correction amount is applied to the initial simulation results, optimizing the accuracy of subsequent impedance calculations by correcting the pressure value in the portal vein region.

[0173] Correcting the pressure in the portal vein region in the initial simulation results can be achieved by calculating the corrected pressure at each inlet and outlet of the portal vein model according to the pressure correction formula, and then updating the pressure distribution. For example, this can be done by uniformly correcting the pressure at all portal vein nodes globally, or by applying a distance-weighted correction to different nodes. This can eliminate system biases, make the simulated pressure field closer to the actual physiological state, and improve the accuracy of impedance calculation.

[0174] Portal vein terminal impedance can be a parameter characterizing the resistance to blood flow in the distal microcirculation of the portal venous system, reflecting the afterload of blood flow into the liver. As part of zero-dimensional impedance data, it can be used to quantify the resistance at the terminal branches of the portal vein. In this embodiment, portal vein terminal impedance can be calculated based on the modified portal vein region pressure and flow data, according to the proportional relationship between portal vein and hepatic vein impedance. For example, portal vein terminal impedance, depending on the vascular branch level, may include, but is not limited to, one or more of the following: left branch terminal impedance, right branch terminal impedance, and main trunk terminal impedance.

[0175] Portal vein distal impedance can be calculated based on the pressure and flow rate from the initial simulation results. For example, the distal impedance value can be calculated using the resistance formula by using the corrected portal vein pressure and the corresponding branch flow rate. Alternatively, accurate parameters reflecting the distal resistance characteristics of the portal vein system can be obtained by using average pressure and average flow rate, or by using peak pressure and peak flow rate.

[0176] Hepatic venous terminal impedance is a parameter characterizing the distal hepatic venous system's resistance to return flow, reflecting the degree of obstruction to blood flow from the liver parenchyma to the inferior vena cava. It can be used to assess the patency of hepatic venous return and influence the calculation results of the portal vein pressure gradient. In this embodiment, hepatic venous terminal impedance can be calculated based on the corrected ratio of hepatic venous outlet pressure to flow rate and the proportional relationship between portal vein and hepatic venous impedance. For example, depending on the draining liver lobe, hepatic venous terminal impedance may include, but is not limited to, one or more of the following: right hepatic venous terminal impedance, middle hepatic venous terminal impedance, and left hepatic venous terminal impedance.

[0177] The hepatic vein terminal impedance can be calculated based on the pressure and flow rate data from the initial simulation results. For example, the pressure and flow rate data at the hepatic vein outlet can be extracted, and their ratio can be calculated as the terminal impedance. Furthermore, the resistance to hepatic vein return can be quantified by calculating the impedance of each hepatic vein branch individually or by combining the total hepatic vein impedance, thereby improving the description of the outlet boundary of the zero-dimensional impedance data.

[0178] Hepatic artery impedance is a parameter characterizing the blood flow resistance of the hepatic arterial system. It can be used in whole-liver blood flow balance calculations to compensate for perfusion differences caused by portal vein blood flow variations, thus reflecting the impact of arterial blood supply on intrahepatic hemodynamics. In this embodiment, hepatic artery impedance can be calculated based on hepatic artery inlet pressure and flow data. For example, depending on the blood supply area, hepatic artery impedance may include, but is not limited to, one or more of the following: left lobe artery impedance, right lobe artery impedance, and whole-liver artery impedance.

[0179] Hepatic artery impedance is calculated based on the pressure and flow rate from the initial simulation results. For example, arterial system resistance can be calculated based on the hepatic artery inlet pressure and the flow rate into the liver parenchyma. Furthermore, arterial blood supply resistance parameters can be supplemented by calculations based on systolic parameters, diastolic parameters, etc., thereby ensuring the integrity of the dual blood supply model.

[0180] Distal vascular bed impedance can be a parameter characterizing the overall resistance to blood flow in the microcirculatory tissue at the terminal end of a vascular network. It can encompass capillaries, venous return segments, etc., and serve as an equivalent substitute for boundary conditions, simplifying the modeling of complex microcirculatory structures. For example, distal vascular bed impedance can be derived from the ratio of overall pressure drop to total flow. Furthermore, depending on organ affiliation, distal vascular bed impedance can include, but is not limited to, one or more of the following: distal spleen impedance, distal mesenteric impedance, and distal renal impedance.

[0181] Accordingly, calculating the distal vascular bed impedance based on the pressure and flow rate in the initial simulation results can involve combining the overall circulatory pressure drop and total flow rate to derive the distal impedance of the inferior vena cava, the distal impedance of the superior mesenteric vein, and the distal impedance of the splenic vein. Furthermore, by estimating the total peripheral resistance and calculating based on organ-specific blood flow distribution ratios, the complex distal structures can be simplified, and stable boundary resistance parameters can be provided.

[0182] This embodiment provides a method for predicting the proportion of blood flow sources. By acquiring preoperative blood flow status data including portal vein pressure, inferior vena cava pressure, blood flow velocity, and corresponding vessel radius, initial boundary conditions consistent with the physiological condition of the target object can be established. Based on the geometric model reconstructed from the patient's CTA image, multi-scale steady-state hemodynamic simulation is performed to obtain initial simulation results. The measured portal vein pressure is compared with the simulated portal vein pressure to calculate the pressure correction amount. The pressure in the portal vein region in the initial simulation results is corrected, which can eliminate the systematic error between the initial assumptions of the model and the real physiological environment. Subsequently, based on the corrected pressure and flow information, portal vein impedance, hepatic vein impedance, hepatic artery impedance, distal inferior vena cava impedance, distal superior mesenteric vein impedance, and distal splenic vein impedance are calculated respectively to generate zero-dimensional impedance data that accurately quantifies the resistance characteristics of different vascular branches and distal bed. This enables closed-loop calibration from measured data to model parameters, allowing subsequent simulations to more realistically simulate the blood flow resistance distribution of a specific object, thereby improving the accuracy of hemodynamic simulation.

[0183] In some embodiments, determining the initial boundary conditions of the preset vascular model based on the preoperative blood flow status data of the target object further includes: calculating the inlet flow of the main portal vein and the main hepatic artery based on the blood flow velocity and the vascular radius; using the inlet flow or blood flow velocity of the main portal vein and the main hepatic artery as the corresponding inlet boundary conditions; and using the inferior vena cava pressure as the outlet boundary condition.

[0184] In some embodiments, based on the pressure and flow rate in the initial simulation results, the portal vein terminal impedance, hepatic vein terminal impedance, hepatic artery impedance, and distal vascular bed impedance are calculated to obtain zero-dimensional impedance data, including: obtaining the portal hypertension type of the target object; if the portal hypertension type is presinusoidal portal hypertension, then setting the ratio of portal vein terminal impedance to hepatic vein terminal impedance as a first ratio value; if the portal hypertension type is sinusoidal portal hypertension, then setting the ratio of portal vein terminal impedance to hepatic vein terminal impedance as a second ratio value; calculating the portal vein terminal impedance and hepatic vein terminal impedance based on the ratio of portal vein terminal impedance to hepatic vein terminal impedance, and the pressure and flow rate in the initial simulation results; and determining the hepatic artery impedance and distal vascular bed impedance based on the initial simulation results, portal vein terminal impedance, and hepatic vein terminal impedance.

[0185] The first and second ratio values ​​can be fixed values ​​preset based on clinical statistical data, or they can be dynamically determined in the following ways: obtaining the ratio of the hepatic vein pressure gradient (HVPG) to the portal vein pressure gradient of the target object; and determining the ratio of the portal vein terminal impedance to the hepatic vein terminal impedance based on the ratio and the preset mapping relationship.

[0186] In some embodiments, the target stent parameters include stent diameter, length, and connection location; constructing the target postoperative model based on the preset vascular model and the target stent parameters includes: constructing a stent channel model between the portal vein model and the hepatic vein model based on the target stent parameters; preprocessing the portal vein model, hepatic vein model, and hepatic artery model connected based on the stent channel model to obtain the target postoperative model; the preprocessing includes at least one of smoothing, truncation, and mesh generation.

[0187] In some embodiments, the hemodynamic simulation of the target postoperative model is performed based on zero-dimensional impedance data and the central arterial pressure of the target object. Determining the hemodynamic simulation result of the corresponding target postoperative model further includes: setting pressure boundaries for multiple inlets in the target postoperative model based on zero-dimensional impedance data and central arterial pressure; the inlets include the hepatic artery inlet, the superior mesenteric vein inlet, the splenic vein inlet, and the inferior vena cava inlet; based on zero-dimensional impedance data, pressure and flow information are transmitted between the coupling nodes of the portal vein model, the hepatic artery model, and the hepatic vein model of the target postoperative model to perform multi-scale steady-state hemodynamic simulation until convergence is achieved, and the target simulation result is obtained; the hemodynamic simulation result is determined based on the target simulation result.

[0188] In some embodiments, the process of obtaining the preset vascular model further includes: obtaining preoperative angiography images of the target object; extracting the luminal structures of the portal vein, hepatic vein, and hepatic artery from the preoperative angiography images according to a medical image segmentation algorithm; constructing portal vein models, hepatic vein models, and hepatic artery models respectively based on the luminal structures; and preprocessing the portal vein models, hepatic vein models, and hepatic artery models to obtain the preset vascular model; the preprocessing includes at least one of smoothing, truncation, and mesh generation.

[0189] In some embodiments, the portal vein model, hepatic vein model, and hepatic artery model are locally connected at multiple coupling nodes through a zero-dimensional model, wherein: the inlet flow of the hepatic vein model is the sum of the corresponding outlet flow of the portal vein model and the corresponding outlet flow of the hepatic artery model; the corresponding outlet pressure of the portal vein model, the corresponding outlet pressure of the hepatic artery model, and the corresponding inlet pressure of the hepatic vein model are equal.

[0190] To more clearly illustrate the technical solution of this application, a detailed embodiment is also provided.

[0191] Currently, patients with portal hypertension can choose transjugular intrahepatic portosystemic shunt (TIPS) to improve clinically risky symptoms such as gastrointestinal bleeding. During TIPS, the stent diameter and placement significantly impact the degree of portal pressure relief and the probability of complications. For example, a larger stent results in a lower portal pressure gradient and a lower probability of gastrointestinal bleeding, but may increase the risk of hepatic encephalopathy later in the patient's life. According to the latest Baveno VII guidelines, the portal pressure gradient (PPG) should be measured before and after TIPS treatment, and reducing the absolute PPG to <12 mmHg is associated with near-complete prevention of portal hypertension-related bleeding and is the primary goal for hemodynamic success. However, due to a lack of measurable data, quantitative methods for assessing hepatic encephalopathy are currently lacking.

[0192] Hepatic encephalopathy (HEEE) refers to brain dysfunction occurring in patients with severe liver disease. It results from the accumulation of toxic substances in the bloodstream that should normally be metabolized and removed by the liver, which then enter the brain. Patients with HEEE may experience confusion, disorientation, somnolence, and changes in personality, behavior, and mood. Generally, these toxic substances originate from the small intestine and are normal breakdown products of proteins. Normally, toxins absorbed into the bloodstream by the small intestine are cleared by the liver as they pass through. However, after transhepatic portosystemic shunt (TIPS), toxins from the small intestine may completely bypass the liver via the TIPS stent between the portal venous system and systemic circulation, which is believed to be associated with an increased risk of HEEE.

[0193] In current clinical treatment, due to the lack of effective measurement methods, it is impossible to predict the potential risks caused by different surgical procedures. Even the measurement of postoperative portal vein pressure gradient (PPG) can only be obtained after puncture and catheter placement. Therefore, some scholars have proposed using computational fluid dynamics simulation to predict the effects of different surgical procedures in advance, in order to guide the selection of treatment plans.

[0194] In current clinical treatment, due to the lack of effective measurement methods, it is impossible to predict the potential risks caused by different surgical procedures. Even the measurement of postoperative portal vein pressure gradient (PPG) can only be obtained after puncture and catheter placement. Therefore, some scholars have proposed using computational fluid dynamics simulation to predict the effects of different surgical procedures in advance, in order to guide the selection of treatment plans.

[0195] Currently, hemodynamic simulations based on computational fluid dynamics for patients with portal hypertension typically predict the postoperative portal pressure gradient (PPG), but there is no method to evaluate the source of postoperative blood flow, especially blood flow from the superior mesenteric vein.

[0196] Understandably, traditional techniques present certain difficulties in conducting coupled simulations based on multiple three-dimensional models, while simplified simulations, such as those considering only the portal vein or both the portal and hepatic veins, cannot accurately obtain the pressure and flow status of blood flow.

[0197] This embodiment reconstructs the portal vein, hepatic vein, and hepatic artery, and couples the three computational domains using a zero-dimensional model to solve the flow field and pressure field. At the same time, it defines the coupling boundary of blood components based on the flow exchange of the three computational domains, thereby realizing the computational analysis of blood flow sources.

[0198] In one embodiment, a method for predicting the proportion of blood flow origin is provided. This method includes: geometrically reconstructing the portal vein, hepatic vein, and hepatic artery based on the patient's CTA images; coupling the three computational domains using a zero-dimensional model; calculating and calibrating the parameters of the zero-dimensional model using clinically measured preoperative portal vein pressure gradient (PPG) values; adding a TIPS shunt stent model to the geometric model according to the TIPS surgical protocol; updating boundary conditions to predict the postoperative model; and obtaining the flow and pressure distribution of the three-dimensional model. After predicting the blood flow and pressure distribution characteristics of the three-dimensional model, the flow field of the computational domain is fixed, a scalar transport model is activated, and the blood flow origin is calculated by coupling relevant boundaries (mainly referring to the portal vein outlet, hepatic artery outlet, and hepatic vein inlet). This yields the proportion of blood shunted from the superior mesenteric vein by the TIPS stent. Medical personnel can then use the shunting proportion data under different protocols as a reference to assess the risk of complications (hepatic encephalopathy).

[0199] Specifically, such as Figure 3 As shown, the method for predicting the proportion of blood flow sources includes:

[0200] Models of portal vein, hepatic vein, and hepatic artery reconstructed based on patient CTA images.

[0201] The zero-dimensional model parameters were calibrated based on the patient's preoperative PPG values.

[0202] Depending on the specific virtual surgical plan, a TIPS stent is added to the patient's preoperative geometric model.

[0203] First, a steady-state simulation was performed on the multi-scale model to calculate the velocity and pressure distribution within the portal vein, hepatic vein, and hepatic artery.

[0204] The scalar transport model is used to calculate blood flow from the superior mesenteric vein, splenic vein, hepatic artery, and inferior vena cava. The number of scalar transport equations depends on the number of blood flow sources to be analyzed. For example, if only blood flow from the superior mesenteric vein is to be analyzed, one scalar transport equation is sufficient. If blood flow from the superior mesenteric vein and splenic vein is to be analyzed, two scalar transport equations are to be used, and so on.

[0205] The flow field within the computational domain is obtained by fixing the multi-scale simulation calculation, and scalar transport equations are added as needed.

[0206] Furthermore, the scalar transport equation takes the form of:

[0207]

[0208] Where ρ is the density of blood flow, u i φ represents the velocity of blood flow (i represents the x, y, and z directions). k This represents the blood flow concentration from a specific inlet, where 1 indicates all blood flow originates from that inlet, and 0 indicates no blood flow originates from that inlet. i Let i be the coordinate system of the computational domain (i represents the x, y, and z directions). The diffusion coefficient can be set to 0 when solving for the source of blood flow. As the source term, its value can be set to 0 when solving for the source of blood flow.

[0209] Set appropriate boundary conditions for entry points from different sources.

[0210] Furthermore, in order to achieve the coupling of the boundary conditions of the portal vein, hepatic vein, and hepatic artery during the calculation and analysis of blood flow sources, specific boundary conditions should be set, and the setting method should be determined according to the blood flow direction of the branches of the portal vein, hepatic vein, and hepatic artery.

[0211] The coupling boundary for blood flow concentration is set according to the blood flow direction of the portal vein, hepatic vein, and hepatic artery.

[0212] For each group of portal vein, hepatic vein, and hepatic artery branches, the blood flow direction calculated by the multi-scale model may fall into three categories, such as... Figure 4 As shown, the blood flow from the hepatic artery into the portal vein and the hepatic vein, respectively; the blood flow from the hepatic artery merges with the blood flow from the portal vein and flows into the hepatic vein; and the blood flow from the hepatic artery merges with the blood flow from the hepatic vein and flows into the portal vein. Among these, R... SMV R represents the distal impedance of the superior mesenteric vein. SV For the distal impedance of the splenic vein, R IVCThe impedance of the inferior vena cava is given by PV_out_1, the outlet of the first branch of the portal vein is given by HV_in_1, the inlet of the first branch of the hepatic vein is given by HA_out_1, and the outlet of the first branch of the hepatic artery is given by HA_out_1. The flow rate at the outlet of the first branch of the portal vein is Q. PV_1 The pressure is P PV_1 The outflow rate at the first branch of the hepatic artery is Q. HA_1 The pressure is P HA_1 The inflow rate at the first branch of the hepatic vein is Q. HV_1 The pressure is P HV_1 R PV_1 R is the impedance of the first exit branch of the portal vein. HV_1 The impedance of the first inlet branch of the hepatic vein, R HA_1 The impedance of the first exit branch of the hepatic artery.

[0213] When analyzing the proportion of blood flow from the superior mesenteric vein at any point within the computational domain, the blood flow concentration φ at the superior mesenteric vein inlet boundary is used. k Set to 1, the blood flow concentration φ at the inlet of the proper hepatic artery, the inlet of the inferior vena cava, and the inlet of the splenic vein. k Set to 0.

[0214] like Figure 5 As shown, the portal vein, hepatic vein, and hepatic artery are each truncated to generate three main inlets / outlets. The inlets / outlets of the three branches correspond one-to-one: the outlet of the first branch of the portal vein is coupled to the outlet of the first branch of the hepatic artery and the inlet of the first branch of the hepatic vein; the outlet of the second branch of the portal vein is coupled to the outlet of the second branch of the hepatic artery and the inlet of the second branch of the hepatic vein; and the outlet of the third branch of the portal vein is coupled to the outlet of the third branch of the hepatic artery and the inlet of the third branch of the hepatic vein. In subsequent steps, zero-dimensional models will be established between these three sets of coupled inlets / outlets to couple their flow and pressure. Wherein, IVC_out is the inferior vena cava outlet, IVC_in is the inferior vena cava inlet, SMV_in is the superior mesenteric vein inlet, SV_in is the splenic vein inlet, PV_out_1 is the outlet of the first branch of the portal vein, PV_out_2 is the outlet of the second branch of the portal vein, PV_out_3 is the outlet of the third branch of the portal vein, HV_in_1 is the inlet of the first branch of the hepatic vein, HV_in_2 is the inlet of the second branch of the hepatic vein, HV_in_3 is the inlet of the third branch of the hepatic vein, HA_in is the inlet of the proper hepatic artery, HA_out_1 is the outlet of the first branch of the hepatic artery, HA_out_2 is the outlet of the second branch of the hepatic artery, and HA_out_3 is the outlet of the third branch of the hepatic artery.

[0215] The locations for portal vein branch outlets, hepatic vein branch inlets, and hepatic artery branch outlets are determined based on their respective blood flow directions.

[0216] When blood flow from the hepatic artery enters the portal vein and hepatic vein, the average blood flow concentration φ at the exit section of the hepatic artery branch is... k The blood flow concentration specified as the exit point of the portal vein branch and the entrance point of the hepatic artery branch;

[0217] When the blood flow from the hepatic artery merges with the blood flow from the portal vein and flows into the hepatic vein, the mass flow rate and mean blood concentration at the exit sections of the hepatic artery branches and the portal vein branches are calculated, and the mean blood concentration after mixing is also calculated. The blood flow concentration at the entrance of the hepatic vein branch is specified and calculated using the following formula:

[0218]

[0219] In the formula, It is the calculated blood flow concentration at the entrance of the i-th hepatic vein branch. It is the mass flow rate at the outlet of the i-th hepatic artery branch. It is the average blood flow concentration at the outlet of the i-th hepatic artery branch. It is the mass flow rate at the outlet of the i-th portal vein branch. It is the average blood flow concentration at the outlet of the i-th portal vein branch.

[0220] When the blood flow from the hepatic artery and hepatic vein merges and flows into the portal vein, the mass flow rate and mean blood flow concentration at the exit section of the hepatic artery branch and the inlet section of the hepatic vein branch are calculated, and the mean blood flow concentration after mixing is also calculated. , is specified as the blood flow concentration at the exit of the portal vein branch.

[0221]

[0222] in, It is the calculated blood flow concentration at the outlet of the i-th portal vein branch. It is the mass flow rate at the outlet of the i-th hepatic artery branch. It is the average blood flow concentration at the outlet of the i-th hepatic artery branch. It is the mass flow rate at the inlet of the i-th hepatic vein branch. It is the average blood flow concentration at the entrance of the i-th hepatic vein branch.

[0223] Through the above coupling, the portal vein branch outlet, hepatic artery branch outlet, and hepatic vein branch inlet can be coupled together when calculating the blood flow source, thereby accurately calculating the proportion of blood flow source at different locations.

[0224] Similarly, when analyzing the proportion of blood flow from the splenic vein at any point within the computational domain, the blood flow concentration φ at the splenic vein inlet boundary is used. k Set to 1, the blood flow concentration φ at the superior mesenteric vein inlet, proper hepatic artery inlet, and inferior vena cava inlet.k Set to 0. The coupling method for the portal vein branch outlet, hepatic artery branch outlet, and hepatic vein branch inlet is the same as described above.

[0225] Similarly, when analyzing the proportion of blood flow from the hepatic artery at any point within the computational domain, the blood flow concentration φ at the boundary of the proper hepatic artery inlet is used. k Set to 1, the blood flow concentration φ at the superior mesenteric vein inlet, splenic vein inlet, and inferior vena cava inlet. k Set to 0. The coupling method for the portal vein branch outlet, hepatic artery branch outlet, and hepatic vein branch inlet is the same as described above.

[0226] Similarly, when analyzing the proportion of blood flow from the inferior vena cava at any point within the computational domain, the blood flow concentration φ at the inferior vena cava inlet boundary is used. k Set to 1, the blood flow concentration φ at the superior mesenteric vein inlet, splenic vein, and proper hepatic artery inlet. k Set to 0. The coupling method for the portal vein branch outlet, hepatic artery branch outlet, and hepatic vein branch inlet is the same as described above.

[0227] Calculate the average blood flow concentration from the superior mesenteric vein in the cross-section of the TIPS shunt stent under different virtual surgical procedures.

[0228] When performing the above analysis, a corresponding number of scalar transport equations can be added according to the blood source to be analyzed. After setting the boundary conditions as described above, multiple scalar transport equations can be solved steadily simultaneously to obtain the proportion analysis of different blood flow sources. Since the scalar transport model used in the simulation is faster than the multiphase flow model, and it does not require solving the velocity and pressure fields of the computational domain simultaneously, it is also suitable for comparing the effects of TIPS stent placement and size in multiple virtual surgical schemes.

[0229] After completing the blood flow source calculation and analysis, a cross-section can be established at the TIPS stent site, and the average blood flow concentration from different vessels (superior mesenteric vein, splenic vein, and hepatic artery) at this cross-section can be calculated. Considering that the risk of hepatic encephalopathy is related to the direct flow of blood from the superior mesenteric vein into the inferior vena cava via the TIPS catheter, medical personnel can use this value to compare the average blood flow concentration from the superior mesenteric vein across the cross-section of the TIPS stent under different surgical procedures. This helps determine the surgical procedure with a lower risk of postoperative hepatic encephalopathy (corresponding to the average blood flow concentration from the superior mesenteric vein). (Lower surgical options).

[0230] This embodiment provides a method for predicting the proportion of blood flow sources. It obtains hemodynamic simulation results of a target object after a simulated transjugular intrahepatic portosystemic shunt procedure. The hemodynamic simulation results include predicted blood flow velocity and pressure fields for each vascular branch of the portal venous system, hepatic artery system, and hepatic venous system. Based on the hemodynamic simulation results, the blood flow direction of each vascular branch is determined. Based on the blood flow direction, a scalar transport model with coupled boundary conditions is constructed. These coupled boundary conditions are determined based on the blood flow concentration mixing relationship between each vascular branch. The scalar transport model is solved to obtain the target blood flow source at the target location of the target object. The proportion of blood flow from different sources can be determined by using hemodynamic simulation results that include the velocity and pressure fields of multiple systems. This breaks through the limitations of traditional methods that only focus on macroscopic pressure gradients, providing a high-fidelity physical field for microscopic blood flow tracking. A scalar transport model with coupled boundary conditions is constructed using the blood flow direction. Based on the blood flow concentration mixing relationship, mass conservation and concentration mixing logic are satisfied at the vascular junction, which can solve the problem that traditional methods cannot distinguish between different sources of blood flow. By solving this model, quantitative source tracing can be achieved on the basis of qualitative simulation, enabling accurate assessment of the proportion of blood flow from different sources after TIPS. By determining the outlet of each portal vein branch and each hepatic artery branch based on the predicted blood flow velocity field and predicted blood flow pressure field, the proportion of blood flow from different sources can be determined. The blood flow direction at the entrance of each hepatic vein branch allows for precise localization of blood flow vector directions at key anatomical nodes using microscopic physical quantities. A clear logical judgment mechanism is established for complex local hemodynamic phenomena that may occur after TIPS, avoiding assessment distortion caused by neglecting local reverse blood flow or complex confluence mechanisms. This makes the calculation of the mixing ratio of blood flow from different sources with normal hepatic artery blood flow in the systemic circulation more consistent with physiological reality, achieving the technical effect of improving the accuracy and reliability of blood flow source ratio assessment. By constructing at least one scalar transport model based on the target blood flow source, and determining the exit points of each portal vein branch and each hepatic artery branch, as well as the direction of each hepatic vein branch, based on the blood flow direction. By understanding the blood flow concentration mixing relationship between inlets and setting coupled boundary conditions for the scalar transport model based on this relationship, the abstract target blood flow source can be instantiated into a concrete digital entity. Furthermore, by establishing a matter conservation logic at key nodes of the vascular network, the mixing relationship is transformed into a mandatory constraint term for the scalar transport model. This ensures that the model strictly adheres to the real physiological mixing laws during the solution process. This effectively avoids the problems of overly simplified boundary conditions and the inability to simulate complex mixing processes of multi-source blood in traditional techniques. The final calculation results can accurately quantify the specific proportion of blood flow from different sources at a specific location, thus improving the accuracy and reliability of blood flow source proportion assessment.By employing a decoupled computational strategy, while maintaining the high-precision predicted blood flow velocity and pressure fields obtained in the preceding stages, steady-state solutions are performed on the scalar transport model. This avoids repeated, time-consuming hydrodynamic iterations, significantly reducing computational costs and improving evaluation efficiency. Simultaneously, by simulating the diffusion and convection of blood from different sources using a fixed, realistic flow field background, the blood flow concentration distribution across the entire region can be efficiently obtained. Furthermore, by directly reading the specific blood flow concentration at the target location and mapping its value directly to the proportion of the target blood flow source, the complex multiphase flow mixing problem is transformed into a more manageable scalar transport problem. This approach improves computational speed while preserving the realistic flow velocity and pressure fields, accurately quantifying the proportion of blood flow from specific sources after TIPS surgery, thereby enhancing the accuracy and reliability of blood flow source proportion assessment.

[0231] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0232] Based on the same inventive concept, this application also provides an apparatus for implementing the method described above. The solution provided by this apparatus is similar to the solution described in the above method; therefore, specific limitations in one or more apparatus embodiments provided below can be found in the limitations of the method described above, and will not be repeated here.

[0233] In one embodiment, such as Figure 6 As shown, this application provides a blood flow source ratio prediction device, which includes:

[0234] The simulation result acquisition module 100 is used to acquire the hemodynamic simulation results of the target object after a simulated transjugular intrahepatic portosystemic shunt procedure. The hemodynamic simulation results include the predicted blood flow velocity field and predicted blood flow pressure field of each vascular branch of the portal venous system, hepatic artery system, and hepatic venous system.

[0235] The blood flow direction determination module 200 is used to determine the blood flow direction of each blood vessel branch based on the hemodynamic simulation results.

[0236] The transport model construction module 300 is used to construct a scalar transport model with coupled boundary conditions based on the blood flow direction. The coupled boundary conditions are determined based on the blood flow concentration mixing relationship between the various vascular branches.

[0237] The source ratio solution module 400 is used to solve the scalar transport model to obtain the proportion of the target blood flow source at the target location of the target object.

[0238] In some embodiments, the blood flow direction determination module 200 is further configured to:

[0239] Based on the predicted blood flow velocity field and the predicted blood flow pressure field, the blood flow direction at the exit of each portal vein branch and each hepatic artery branch and the entrance of each hepatic vein branch is determined respectively; the blood flow direction is that the blood flow of the hepatic artery branch flows into the portal vein branch and the hepatic vein branch, the blood flow of the hepatic artery branch merges with the blood flow of the portal vein branch and flows into the hepatic vein branch, or the blood flow of the hepatic artery branch merges with the blood flow of the hepatic vein branch and flows into the portal vein branch.

[0240] In some embodiments, the transport model building module 300 is further configured to:

[0241] Based on the target blood flow source, at least one scalar transport model is constructed;

[0242] Based on the blood flow direction, determine the blood flow concentration mixing relationship between the outlet of each portal vein branch and each hepatic artery branch and the inlet of each hepatic vein branch;

[0243] Based on the blood flow concentration mixing relationship, the coupling boundary conditions of the scalar transport model are set.

[0244] In some embodiments, when the blood flow direction of the hepatic artery branch is towards the portal vein branch and the hepatic vein branch, determining the blood flow concentration mixing relationship includes:

[0245] The average blood flow concentration at the exit section of the hepatic artery branch is determined as the blood flow concentration at the exit of the corresponding portal vein branch and the entrance of the hepatic vein branch.

[0246] In some embodiments, when the blood flow direction of the hepatic artery branch is such that it merges with the blood flow of the portal vein branch and flows into the hepatic vein branch, determining the blood flow concentration mixing relationship includes:

[0247] The average blood flow concentration after mixing is calculated based on the mass flow rate and average blood flow concentration at the outlet section of the hepatic artery branch and the mass flow rate and average blood flow concentration at the outlet section of the portal vein branch.

[0248] The average blood flow concentration after mixing is determined as the blood flow concentration at the entrance of the hepatic vein branch.

[0249] In some embodiments, when the blood flow direction of the hepatic artery branch merges with the blood flow of the hepatic vein branch and flows into the portal vein branch, determining the blood flow concentration mixing relationship includes:

[0250] The average blood flow concentration after mixing is calculated based on the mass flow rate and average blood flow concentration at the exit section of the hepatic artery branch and the mass flow rate and average blood flow concentration at the inlet section of the hepatic vein branch.

[0251] The average blood flow concentration after mixing is determined as the blood flow concentration at the exit of the portal vein branch.

[0252] In some embodiments, the source scaling module 400 is further configured to:

[0253] With the predicted blood flow velocity field and predicted blood flow pressure field remaining unchanged, the scalar transport model is solved in steady state to obtain the blood flow concentration distribution at each location within the region corresponding to the scalar transport model.

[0254] Based on the blood flow concentration distribution, the blood flow concentration corresponding to the target blood flow source at the target location is determined as the proportion of the corresponding target blood flow source.

[0255] By comparing the proportions at the same target location under different simulated surgical plans, the impact of different stent parameters (such as diameter, length, and placement location) on the degree of shunting of blood from a specific source (such as enteric blood) can be quantitatively assessed, thus providing a quantitative basis for doctors to select optimized surgical plans that reduce the risk of hepatic encephalopathy.

[0256] In some embodiments, the simulation result acquisition module 100 is further configured to:

[0257] The hemodynamic simulation results of the target object after a simulated transjugular intrahepatic portosystemic shunt procedure include:

[0258] A preset vascular model and zero-dimensional impedance data of the target object are obtained; the preset vascular model is reconstructed based on the preoperative angiography image of the target object, including the portal vein model, hepatic vein model, and hepatic artery model; the zero-dimensional impedance data is obtained by calibrating the preset vascular model based on the preoperative blood flow status data of the target object.

[0259] Based on the preset vascular model and target stent parameters, construct the target postoperative model;

[0260] Based on the zero-dimensional impedance data and the central arterial pressure of the target object, hemodynamic simulation is performed on the target postoperative model to obtain the hemodynamic simulation results.

[0261] In some embodiments, the simulation result acquisition module 100 is further configured to:

[0262] Based on the preoperative blood flow status data of the target object, the initial boundary conditions of the preset vascular model are determined; the preoperative blood flow status data includes at least portal vein pressure, inferior vena cava pressure, blood flow velocity, and corresponding vascular radius;

[0263] Based on the initial boundary conditions, pressure and flow information are transmitted between the coupling nodes of the portal vein model, hepatic artery model, and hepatic vein model of the preset blood vessel model to perform multi-scale steady-state hemodynamic simulation and obtain initial simulation results.

[0264] Based on the portal vein pressure and the simulated portal vein pressure in the initial simulation results, determine the pressure correction amount;

[0265] Based on the pressure correction amount, the pressure in the portal vein region in the initial simulation results is corrected;

[0266] Based on the pressure and flow rate in the initial simulation results, the portal vein terminal impedance, hepatic vein terminal impedance, hepatic artery impedance, and distal vascular bed impedance are calculated respectively to obtain the zero-dimensional impedance data.

[0267] Each module in the aforementioned blood flow source ratio prediction device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0268] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 7As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a method for predicting the proportion of blood flow sources. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0269] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0270] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the blood flow source ratio prediction method of any of the above embodiments:

[0271] Obtain the hemodynamic simulation results of the target object after a simulated transjugular intrahepatic portosystemic shunt procedure; the hemodynamic simulation results include the predicted blood flow velocity field and predicted blood flow pressure field of each vascular branch of the portal vein system, hepatic artery system, and hepatic vein system;

[0272] Based on the hemodynamic simulation results, the blood flow direction of each of the blood vessel branches is determined;

[0273] Based on the blood flow direction, a scalar transport model with coupled boundary conditions is constructed; the coupled boundary conditions are determined based on the blood flow concentration mixing relationship between each of the blood vessel branches.

[0274] Solving the scalar transport model yields the proportion of the target blood flow source at the target location of the target object.

[0275] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the blood flow source ratio prediction method of any of the above embodiments:

[0276] Obtain the hemodynamic simulation results of the target object after a simulated transjugular intrahepatic portosystemic shunt procedure; the hemodynamic simulation results include the predicted blood flow velocity field and predicted blood flow pressure field of each vascular branch of the portal vein system, hepatic artery system, and hepatic vein system;

[0277] Based on the hemodynamic simulation results, the blood flow direction of each of the blood vessel branches is determined;

[0278] Based on the blood flow direction, a scalar transport model with coupled boundary conditions is constructed; the coupled boundary conditions are determined based on the blood flow concentration mixing relationship between each of the blood vessel branches.

[0279] Solving the scalar transport model yields the proportion of the target blood flow source at the target location of the target object.

[0280] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0281] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0282] The technical features of the above embodiments can be combined in any way. For the sake of brevity, 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.

[0283] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for predicting the proportion of blood flow sources, characterized in that, The method for predicting the proportion of blood flow sources includes: Obtain the hemodynamic simulation results of the target object after a simulated transjugular intrahepatic portosystemic shunt procedure; the hemodynamic simulation results include the predicted blood flow velocity field and predicted blood flow pressure field of each vascular branch of the portal vein system, hepatic artery system, and hepatic vein system; Based on the hemodynamic simulation results, the blood flow direction of each of the blood vessel branches is determined; Based on the blood flow direction, a scalar transport model with coupled boundary conditions is constructed; the coupled boundary conditions are determined based on the blood flow concentration mixing relationship between each of the blood vessel branches. Solving the scalar transport model yields the proportion of the target blood flow source at the target location of the target object.

2. The method for predicting the proportion of blood flow sources according to claim 1, characterized in that, The step of determining the blood flow direction of each blood vessel branch based on the hemodynamic simulation results includes: Based on the predicted blood flow velocity field and the predicted blood flow pressure field, the blood flow direction at the exit of each portal vein branch and each hepatic artery branch and the entrance of each hepatic vein branch is determined respectively; the blood flow direction is that the blood flow of the hepatic artery branch flows into the portal vein branch and the hepatic vein branch, the blood flow of the hepatic artery branch merges with the blood flow of the portal vein branch and flows into the hepatic vein branch, or the blood flow of the hepatic artery branch merges with the blood flow of the hepatic vein branch and flows into the portal vein branch.

3. The method for predicting the proportion of blood flow sources according to claim 1 or claim 2, characterized in that, The construction of a scalar transport model with coupled boundary conditions based on the blood flow direction includes: Based on the target blood flow source, at least one scalar transport model is constructed; Based on the blood flow direction, determine the blood flow concentration mixing relationship between the outlet of each portal vein branch and each hepatic artery branch and the inlet of each hepatic vein branch; Based on the blood flow concentration mixing relationship, the coupling boundary conditions of the scalar transport model are set.

4. The method for predicting the proportion of blood flow sources according to claim 3, characterized in that, When the blood flow direction of the hepatic artery branches is towards the portal vein branches and the hepatic vein branches, determining the blood flow concentration mixing relationship includes: The average blood flow concentration at the exit section of the hepatic artery branch is determined as the blood flow concentration at the exit of the corresponding portal vein branch and the entrance of the hepatic vein branch.

5. The method for predicting the proportion of blood flow sources according to claim 3, characterized in that, When the blood flow direction of the hepatic artery branch merges with the blood flow of the portal vein branch and flows into the hepatic vein branch, determining the blood flow concentration mixing relationship includes: The average blood flow concentration after mixing is calculated based on the mass flow rate and average blood flow concentration at the outlet section of the hepatic artery branch and the mass flow rate and average blood flow concentration at the outlet section of the portal vein branch. The average blood flow concentration after mixing is determined as the blood flow concentration at the entrance of the hepatic vein branch.

6. The method for predicting the proportion of blood flow sources according to claim 3, characterized in that, When the blood flow direction of the hepatic artery branch merges with the blood flow of the hepatic vein branch and flows into the portal vein branch, determining the blood flow concentration mixing relationship includes: The average blood flow concentration after mixing is calculated based on the mass flow rate and average blood flow concentration at the exit section of the hepatic artery branch and the mass flow rate and average blood flow concentration at the inlet section of the hepatic vein branch. The average blood flow concentration after mixing is determined as the blood flow concentration at the exit of the portal vein branch.

7. The method for predicting the proportion of blood flow sources according to claim 1, characterized in that, Solving the scalar transport model to obtain the proportion of the target blood flow source at the target location of the target object includes: With the predicted blood flow velocity field and predicted blood flow pressure field remaining unchanged, the scalar transport model is solved in steady state to obtain the blood flow concentration distribution at each location within the region corresponding to the scalar transport model. Based on the blood flow concentration distribution, the blood flow concentration corresponding to the target blood flow source at the target location is determined as the proportion of the corresponding target blood flow source.

8. The method for predicting the proportion of blood flow sources according to claim 1, characterized in that, The hemodynamic simulation results of the target object after a simulated transjugular intrahepatic portosystemic shunt procedure include: A preset vascular model and zero-dimensional impedance data of the target object are obtained; the preset vascular model is reconstructed based on the preoperative angiography image of the target object, including the portal vein model, hepatic vein model, and hepatic artery model; the zero-dimensional impedance data is obtained by calibrating the preset vascular model based on the preoperative blood flow status data of the target object. Based on the preset vascular model and target stent parameters, construct the target postoperative model; Based on the zero-dimensional impedance data and the central arterial pressure of the target object, hemodynamic simulation is performed on the target postoperative model to obtain the hemodynamic simulation results.

9. The method for predicting the proportion of blood flow sources according to claim 8, characterized in that, The process of acquiring the zero-dimensional impedance data includes: Based on the preoperative blood flow status data of the target object, the initial boundary conditions of the preset vascular model are determined; the preoperative blood flow status data includes at least portal vein pressure, inferior vena cava pressure, blood flow velocity, and corresponding vascular radius; Based on the initial boundary conditions, pressure and flow information are transmitted between the coupling nodes of the portal vein model, hepatic artery model, and hepatic vein model of the preset blood vessel model to perform multi-scale steady-state hemodynamic simulation and obtain initial simulation results. Based on the portal vein pressure and the simulated portal vein pressure in the initial simulation results, determine the pressure correction amount; Based on the pressure correction amount, the pressure in the portal vein region in the initial simulation results is corrected; Based on the pressure and flow rate in the initial simulation results, the portal vein terminal impedance, hepatic vein terminal impedance, hepatic artery impedance, and distal vascular bed impedance are calculated respectively to obtain the zero-dimensional impedance data.

10. A blood flow source ratio prediction device, characterized in that, The blood flow source ratio prediction device includes: The simulation result acquisition module is used to acquire the hemodynamic simulation results of the target object after a simulated transjugular intrahepatic portosystemic shunt procedure; the hemodynamic simulation results include the predicted blood flow velocity field and predicted blood flow pressure field of each vascular branch of the portal vein system, hepatic artery system and hepatic vein system; The blood flow direction determination module is used to determine the blood flow direction of each of the blood vessel branches based on the hemodynamic simulation results. The transport model construction module is used to construct a scalar transport model with coupled boundary conditions based on the blood flow direction; the coupled boundary conditions are determined based on the blood flow concentration mixing relationship between the blood vessel branches. The source ratio solution module is used to solve the scalar transport model to obtain the proportion of the target blood flow source at the target location of the target object.

11. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 9.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 9.

Images