System for assessing risk of atherosclerotic plaque rupture

Abstract

This application provides a system for assessing the risk of atherosclerotic plaque rupture, comprising: an acquisition module for acquiring a geometric model of the target blood vessel and receiving blood rheological parameters output from a blood rheology detection device used in clinical applications; a base flow field solution module for determining the steady-state flow state of blood within the target blood vessel based on the geometric model and blood rheological parameters, obtaining a base flow field including hemodynamic characteristics around the plaque; a feature extraction module for extracting a first dynamic feature and a second dynamic feature; and an assessment module for assessing the influence trend of non-Newtonian properties of blood on the risk of atherosclerotic plaque rupture. This system, by accurately capturing the actual blood flow characteristics within complex, narrowed blood vessels and employing an adaptive risk discrimination mechanism, can quickly complete individualized assessments on conventional equipment, providing efficient and reliable technical support for clinical diagnosis and treatment.

Description

Technical Field

[0001] This application relates to the field of cardiovascular disease early warning technology, specifically to an assessment system for the risk of atherosclerotic plaque rupture. Background Technology

[0002] Acute myocardial infarction and ischemic stroke are the leading causes of death and disability in humans. Their common pathological basis is the sudden rupture of vulnerable atherosclerotic plaques and the subsequent thrombosis. Accurately identifying the risk of plaque rupture is the core prerequisite for achieving early warning and individualized intervention.

[0003] However, existing plaque rupture risk assessment systems have significant technical flaws. In order to reduce computational costs, traditional hemodynamic simulations tend to simplify blood, which has shear-thinning properties, into a Newtonian fluid. This approach severs the coupling effect between rheological nonlinearity and vascular pathological geometry, resulting in inaccurate estimation of the peak pulsating tensile stress on the plaque fibrous cap.

[0004] Because clinical practice lacks mathematical tools to quantitatively decouple viscous dissipation and inertial transport, it is impossible to determine whether shear thinning in a specific narrow configuration exacerbates stress concentration or induces a protective effect, causing traditional empirical thresholds to completely fail in highly nonnormal flow scenarios.

[0005] Meanwhile, accurately analyzing the aforementioned nonlinear couplings requires computationally expensive full-scale numerical simulations, severely hindering real-time clinical translation. Current early warning indicators are also disconnected from microscopic rheological mechanisms, failing to adaptively adjust their judgment logic according to individual variations in patient rheological parameters, thus failing to meet the needs for personalized and precise early warnings. Summary of the Invention

[0006] To address the shortcomings of the existing technologies, this application provides an assessment system for the risk of atherosclerotic plaque rupture. Through mathematical modeling, it quantitatively reveals the coupling relationship between the blood shear thinning effect and the geometries of stenosis around the plaque, accurately determining whether shear thinning increases or inhibits the risk of plaque rupture, and providing a calculable and adaptive early warning tool for clinical use.

[0007] According to an embodiment of this application, a system for assessing the risk of atherosclerotic plaque rupture is provided, comprising:

[0008] The acquisition module is used to acquire the geometric model of the target blood vessel and receive the blood rheological parameters output by the blood rheology detection device in clinical applications. The geometric model includes at least one narrowing region caused by atherosclerotic plaque. The basic flow field solution module is used to determine the steady-state flow state of blood in the target blood vessel based on the geometric model and the blood rheological parameters, and to obtain the basic flow field containing the hemodynamic characteristics around the plaque; The feature extraction module is used to extract, based on the base flow field, a first dynamic feature characterizing the effect of changes in the non-Newtonian properties of blood on system stability, and a second dynamic feature characterizing the base flow field as most sensitive to physiological disturbances. An evaluation module is used to assess the trend of the influence of the non-Newtonian properties of the blood on the risk of atherosclerotic plaque rupture, based on the degree of correlation between the first kinetic characteristic and the second kinetic characteristic.

[0009] Furthermore, the blood rheological parameters are obtained based on the Carreau-Yasuda constitutive model, which includes a power-law exponent for characterizing the degree of blood shear thinning.

[0010] Furthermore, obtaining the steady-state laminar base flow field containing hemodynamic characteristics around the plaque based on the geometric model and blood rheological parameters includes: solving the steady-state incompressible Navier-Stokes equations in the computational domain of the target blood vessel to obtain the steady-state laminar solution as the base flow field.

[0011] Furthermore, the first dynamic feature is obtained by taking the Fréchet derivative of the blood power law exponent with respect to the linearized dynamic operator of the system in which the base flow field exists. The first dynamic feature is a rheological sensitivity operator used to characterize the sensitivity of the dissipation characteristics of the base flow field to the power law exponent.

[0012] Further, extracting the second dynamic feature includes: linearizing the Navier-Stokes equation at the base flow field to construct a linearized dynamic operator; defining a noise covariance matrix characterizing the physiological fluctuations of the cardiovascular system; solving the Lyapunov equation to obtain a controllable glam matrix; performing eigenvalue decomposition on the controllable glam matrix and extracting the normalized eigenvector corresponding to the largest eigenvalue as the second dynamic feature.

[0013] Furthermore, the stenotic area caused by the atherosclerotic plaque includes the plaque shoulder area and the fibrous cap area.

[0014] Furthermore, the viscosity function formula for the Carreau-Yasuda constitutive model is:

[0015] in, Zero shear viscosity For infinite shear viscosity, For relaxation time, Shear rate, The power-law exponent. This is the Yasuda index, used to control the steepness of the transition zone.

[0016] Furthermore, when the transition parameter a=2 is fixed, the Carreau-Yasuda constitutive model degenerates into a Carreau-type constitutive relation, and the viscosity function simplifies to:

[0017] in, Zero shear viscosity It is a liquidity behavior index;

[0018]

[0019] Let D be the second invariant of the strain rate tensor. For the velocity gradient tensor; when When the coefficient of friction is 2, the model reverts to a Newtonian fluid. when When <2, it characterizes a shear-thinned fluid, and The smaller the value, the stronger the shear thinning effect.

[0020] Furthermore, the solution to the steady-state laminar base flow field includes: Solve the incompressible Navier-Stokes equations within the geometric region Ω of the blood vessel:

[0021] in, For fluid density, For convection operators, For pressure, For gradient operators; Viscosity, For strain rate tensor; The constraints are =0; The equations are solved by numerical simulation or analytical system to obtain the steady-state laminar solution as the base flow field. The base flow field includes the high-shear region at the throat of the patch, the low-shear recirculation region downstream of the patch, and the flow separation region at the shoulder of the patch.

[0022] Furthermore, the construction of the linearized dynamic operator includes: linearizing the Navier-Stokes equations in the weighted kinetic energy inner product space to obtain the linearized dynamic operator, and decomposing the linearized dynamic operator into a symmetric dissipative operator and an antisymmetric convection operator in the weighted kinetic energy inner product space.

[0023] Furthermore, the formula for the weighted kinetic energy inner product is:

[0024] in, The computational domain for blood vessels containing plaque regions; For the weight function, For vector dot product, It is a volume infinitesimal element; The linearized dynamic operator decomposition satisfies

[0025]

[0026] in, Here, D is a linearized kinetic operator, and D is a symmetric positive definite dissipation operator representing the viscous dissipation of blood, as shown in the formula:

[0027] C is the antisymmetric convection operator, representing the inertial transport and nonmodal energy transfer of blood flow, and its formula is:

[0028] in, L n express L n Adjoint operators in weighted inner product spaces; The decomposition is performed in a weighted inner product space, which is defined by the kinetic energy norm of the fluid.

[0029] Furthermore, the Lyapunov equation is:

[0030] in, The controllable Graphlam matrix represents the spatial covariance structure of blood flow disturbance in the plaque region; The transpose operator of the linearized dynamics operator; The noise covariance matrix contains the heart rate variability perturbation component and the red blood cell micro-fluctuation perturbation component.

[0031] Furthermore, the noise covariance matrix is ​​decomposed into:

[0032] in, To simulate the inflow fluctuation component caused by heart rate variability; To simulate the microscale erythrocyte agitation component.

[0033] Furthermore, the dominant perturbation direction of the patch instability is a controllable Gauram matrix. The normalized eigenvector v1 corresponding to the largest eigenvalue λ1 represents the flow space mode on the patch surface that is most easily excited by physiological noise, leading to local high shear stress concentration and inducing fiber cap rupture.

[0034] Furthermore, the rheological sensitivity operator The following is obtained by taking the partial derivative of Carreau's viscosity law at the basic strain rate:

[0035] in, Expressed as viscosity parameter The partial derivative; because:

[0036] Furthermore, the convection operator N is independent of n, therefore ; The rheological sensitivity operator describes the full response of the system's dissipation characteristics when the shear thinning intensity changes slightly, including direct viscosity changes and indirect base current corrections; The rheological sensitivity operator has a significant amplitude in the high strain rate region of the narrow throat of the plaque, and the amplitude decays in the downstream low shear reflow region.

[0037] Furthermore, the degree of correlation is measured by an alignment index, which is used to characterize the degree of geometric matching between the second dynamic feature and the principal direction of dissipation increment determined by the first dynamic feature. The alignment index The alignment index is calculated as follows: The image of the second dynamic feature under the action of the rheological sensitivity operator is multiplied by the second dynamic feature itself, and the multiplied product value is normalized to obtain the alignment index; the calculation formula is as follows:

[0038] Where v1 is the dominant eigenvector of the controllable Gauram matrix. For rheologically sensitive operators, ·, · This is the weighted inner product in the flow field space; The alignment index ranges from [-1, 1], and its absolute value reflects the degree of geometric overlap between the optimal forcing direction and the principal direction of dissipation increment.

[0039] Furthermore, a risk mechanism is determined based on the alignment index, and the determination of the risk mechanism includes: When the alignment is greater than a preset first threshold, it is determined to be the Stokes dominant mechanism. At this time, the shear thinning effect reduces the quasi-barrier of plaque rupture and warns that blood shear thinning increases the risk of plaque rupture. When the alignment is less than a preset second threshold, it is determined to be a non-normal misalignment risk mechanism. At this time, the shear thinning effect raises the quasi-barrier of plaque rupture, and the blood shear thinning has a counterintuitive protective effect on plaque rupture. When the alignment index is between the second threshold and the first threshold, it is determined to be a transitional hybrid mechanism; The first threshold is a positive value, and the second threshold is a negative value; the first threshold and the second threshold are empirical values ​​obtained based on historical clinical case data and in vitro experimental calibration.

[0040] Furthermore, the risk mechanism determination also includes defining alignment degree. Where v1 is the dominant eigenvector of the Gram operator and m1 is the dominant eigenvector of the rheologically sensitive operator; under the symmetric limit, the critical alignment is:

[0041] in, The largest eigenvalue of the Gram operator. This is the second largest eigenvalue of the Gram operator; like < A sign reversal occurs, meaning that shear thinning inhibits the risk of plaque rupture; like > This manifests as monotonic instability, meaning that shear thinning promotes the risk of plaque rupture.

[0042] Furthermore, after assessing the trend of the influence of the non-Newtonian properties of the blood on the risk of atherosclerotic plaque rupture, the method further includes generating an atherosclerotic plaque rupture risk warning report, which includes: Risk mechanism type identifier, used to indicate the dominant mechanical mechanism of plaque fibrous cap rupture; The plaque rupture quasi-barrier monotonicity indicator value, determined by the sign and amplitude of the alignment index, is used to characterize the trend direction and sensitivity of plaque stability as a function of the blood power law exponent. The plaque rupture risk contribution score is equal to the alignment index. A positive value indicates that shear thinning increases the risk, while a negative value indicates that shear thinning reduces the risk.

[0043] Furthermore, the system also includes: The risk warning results are corrected by combining the shear stress distribution on the surface of the patch. When the calculated quasi-barrier change trend is inconsistent with the wall shear stress distribution, the alignment diagnosis result shall be used as the priority early warning basis for plaque rupture risk assessment.

[0044] Furthermore, the system also includes: After obtaining the risk warning results, cross-validation was performed by combining them with clinical wall shear stress distribution maps. When the determined risk mechanism is inconsistent with the spatial distribution of high-risk areas of wall shear stress, the non-normal misalignment mechanism diagnosed by the alignment index is used as the priority for early warning, and the rheological-geometric coupling anomaly warning information is marked in the early warning report.

[0045] Furthermore, the system also includes: When changes in the patient’s blood rheological parameters are detected, the alignment index and risk mechanism type determination results are updated, and the updated plaque rupture risk warning information is pushed to the clinical terminal. Furthermore, the target blood vessel is a coronary artery or a carotid artery, and the atherosclerotic plaque is a vulnerable plaque whose rupture leads to acute coronary syndrome or ischemic stroke.

[0046] The application has the following beneficial effects: The atherosclerotic plaque rupture risk assessment system provided in this application accurately captures the actual flow characteristics of blood in complex narrow blood vessels, and can quantitatively reveal the intrinsic relationship between blood rheological properties and plaque lesion geometry. In complex narrowing configurations such as the carotid artery and coronary artery, it can accurately determine whether the non-Newtonian properties of blood exacerbate stress concentration on the plaque surface or induce a protective effect, thereby effectively avoiding false positive or false negative warnings caused by neglecting the coupling relationship between the two in traditional system assessment systems, and providing more reliable reference for clinical practice.

[0047] Meanwhile, because the system can adaptively adjust the judgment logic according to the individualized rheological-geometric matching state of the patient, it fundamentally gets rid of the dependence on a single empirical threshold. It can complete the risk mechanism discrimination without pre-setting a one-size-fits-all threshold, thus significantly reducing the risk of misjudgment caused by individual differences. It is especially suitable for patient groups with large differences in hemodynamic characteristics.

[0048] Furthermore, the system can complete the full risk mechanism diagnosis based solely on the steady-state flow field. This avoids the dependence of traditional full-scale transient numerical simulation on supercomputers and massive computing time, while improving computational efficiency by several orders of magnitude while maintaining an accurate description of the core physical characteristics. This makes it possible to conduct rapid, batch risk assessments of individualized vascular models. The calculations can be completed on regular workstations or even high-performance personal computers, significantly reducing the technical application threshold.

[0049] Ultimately, the evaluation results output by the system can directly serve clinical processes such as preoperative planning, interventional treatment decisions, and postoperative follow-up, helping doctors to gain a more comprehensive understanding of the mechanical state of the patient's plaque within a limited time and providing quantitative basis for developing individualized treatment plans. Therefore, it has good clinical promotion value and industrial application prospects. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other embodiments can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a block diagram of the system structure for assessing the risk of atherosclerotic plaque rupture based on rheological nonlinear characteristics, as described in this application. Figure 2(a) shows the quasi-barrier V. (n) Comparison curves of the two mechanisms as the power law exponent n changes; Figure 2(b) is a graph showing the alignment diagnostic index χ(n) as a function of the power law exponent n; Figure 2(c) shows a comparison of the instantaneous paths of patch perturbation on the projection plane of mode 1-mode 2 under the two mechanisms; Figure 2(d) shows the quasi-barrier derivative dV. A graph showing the variation of / dn with the power law exponent n; Figure 2(e) is a graph showing the Gram condition number κ(Qc) as a function of the power law exponent n; Figure 2 (f1) Quasi-barrier level set and eigenvectors; Figure 2 (f2) Σ' eigenvalues; Figure 2 (f3) shows the aligned and repaired Σ′ 0; Figure 2 (f4) shows the nonnormal transient growth of G(t); Figure 2 (f5) shows the potential barrier curve V along v1. (n); Figure 2 (f6) Alignment Comparison Chart; Figure 3(a) shows the quasi-barrier curves V of Regime II under different nonlinear intensities γ. (n) Comparison chart; Figure 3(b) shows the relative deviation curve of the nonlinear correction relative to the linear barrier; Figure 4(a) shows the quasi-barrier curves V under different lifting coupling intensities. (n) Comparison chart; Figure 4(b) is a graph showing the change in the proportion of non-monotonic regions with the lifting coupling strength; Figure 5(a) shows the convergence curve of relative error as a function of the number of discrete points N; Figure 5(b) is a graph showing the wall time as a function of the number of discrete points N; Figure 6(a) is a bar chart comparing the calculated gMAM values ​​with the closed-form Lyapunov quasi-barrier in four target directions; Figure 6(b) is a schematic diagram of a representative instantaneous path superimposed on a quasi-barrier contour map; Figure 7(a) is a graph showing the variation of det(Σ') with alignment χ. Figure 7(b) shows the quadratic form g. A graph showing how Σ'g changes with alignment χ. Figure 7(c) is a schematic diagram showing the change of the sign of the barrier derivative with the alignment degree χ; Figure 8(a) is a three-dimensional surface plot of χ(n, lift) as a function of the power law exponent n and the lift coupling strength; Figure 8(b) shows the log 10 A three-dimensional surface plot of κ(Qc) as a function of the power-law exponent n and the lifting coupling strength.

[0052] Figure Labels System 100 for assessing the risk of atherosclerotic plaque rupture based on rheological nonlinear characteristics; acquisition module 110; basic flow field solution module 120; feature extraction module 130; evaluation module 140. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be further described in detail below with reference to specific examples and accompanying drawings.

[0054] It should be noted that all uses of "first" and "second" in the embodiments of this application are for the purpose of distinguishing two entities with the same name but different names or different parameters. It can be seen that "first" and "second" are only for the convenience of expression and should not be construed as limiting the embodiments of this application.

[0055] like Figure 1As shown, the atherosclerotic plaque rupture risk assessment system 100 according to this application includes: an acquisition module 110, a basic flow field solution module 120, a feature extraction module 130, and an assessment module 140.

[0056] The acquisition module 110 is used to acquire the geometric model and blood rheological parameters of the target blood vessel. The geometric model includes at least one narrowing region caused by an atherosclerotic plaque.

[0057] Specifically, firstly, a geometric model of the target blood vessel (such as the coronary artery or carotid artery) containing atherosclerotic plaques is obtained. The geometric model includes at least one stenotic region caused by the plaque. At the same time, blood rheological parameters are obtained from the output of a blood rheology detection device used in clinical applications. The blood rheological parameters are obtained based on the Carreau-Yasuda constitutive model.

[0058] To define the rheological constitutive relation of blood, considering the significant shear-thinning properties of blood as it passes through plaque regions in narrow vessels, this embodiment employs the Carreau-Yasuda model to describe the blood viscosity μ as a function of shear rate. The relationship between the changes is modeled as follows:

[0059] in, Zero shear viscosity For infinite shear viscosity, For relaxation time, Shear rate, The power-law exponent. The Yasuda index is used to control the steepness of the transition zone.

[0060] To improve computational efficiency while ensuring physical accuracy, this embodiment fixes the transition parameter a=2 in the analytical calculation. At this point, the model degenerates into a Carreau-type constitutive relation, and the viscosity function simplifies to:

[0061] in, It is a viscosity function. D is the strain rate tensor, and s is the second invariant of the strain rate tensor.

[0062] When n=2, the model reverts to a Newtonian fluid. When n < 2, it represents a shear-thinning fluid, and the smaller n is, the stronger the shear-thinning effect.

[0063] The base flow field solution module 120 is used to determine the steady-state flow state of blood in the target blood vessel based on the geometric model and blood rheological parameters, and obtain the base flow field containing the hemodynamic characteristics around the plaque; Specifically, based on the above constitutive model, the hydrodynamic governing equations for the narrowed vascular region are constructed. The incompressible Navier-Stokes equations are solved within the plaque-containing region Ω:

[0064] in, For fluid density, For convection operators, For pressure, For gradient operators; Viscosity, Let be the strain rate tensor; the constraint condition is: =0.

[0065] Solving this system of equations using numerical simulation or analytical systems yields the steady-state laminar solution (base flow field). This base flow field includes the high-shear region at the throat of the patch's narrow section, the low-shear recirculation region downstream of the patch, and the flow separation region at the patch shoulder, serving as the background field for subsequent linearization analysis.

[0066] The feature extraction module 130 is used to extract, based on the base flow field, a first dynamic feature to characterize the effect of changes in blood non-Newtonian properties on system stability, and a second dynamic feature to characterize the base flow field as most sensitive to physiological disturbances. Specifically, by applying the Freser derivative of the Navier-Stokes operator near the basic current, a linearized dynamic operator is obtained. The linearized perturbation equation is defined as follows:

[0067] in, .

[0068] The weighted kinetic energy inner product is defined, and the mass-weighted kinetic energy inner product is introduced.

[0069]

[0070] Under this inner product, the operator It can be uniquely decomposed into a symmetric positive definite part (dissipative operator D) and an antisymmetric part (convection operator C), that is:

[0071] in, Here, D is a linearized kinetic operator, and D is a symmetric positive definite dissipation operator representing the viscous dissipation of blood, as shown in the formula:

[0072] C is the antisymmetric convection operator, representing the inertial transport and nonmodal energy transfer of blood flow, and its formula is:

[0073] in, L n express L n Adjoint operators in weighted inner product spaces; The decomposition is performed in a weighted inner product space, where the weighted inner product is defined by the fluid's kinetic energy norm.

[0074] The physical meaning of this decomposition is clear: the dissipative component D represents the viscous resistance of blood, while the convective component C represents the inertial transport and nonmodal energy transfer of blood flow. Since C is antisymmetric, it does not generate energy growth or decay under weighted inner products, but is only responsible for the redistribution of energy between different modes (i.e., the nonmodal growth mechanism); while D is positive definite and is responsible for energy dissipation. This decomposition lays the foundation for subsequent identification of nonnormal amplification effects.

[0075] Specifically, to quantify the system response driven by physiological noise (such as heart rate variability and erythrocyte perturbations), the Lyapunov equation is constructed. The steady-state covariance Σ_n must satisfy:

[0076] Where Q is the noise covariance matrix. In this embodiment, the noise covariance is decomposed into two parts, including:

[0077] To simulate inflow fluctuations caused by heart rate variability, To simulate microscale red blood cell agitation.

[0078] Solving the Lyapunov equations above yields the controllable Gauram matrix. .right Eigenvalue decomposition reveals that the eigenvector corresponding to the largest eigenvalue is the dominant perturbation direction (the most easily excited direction). This direction represents the spatial mode of the system that is most sensitive to noise, and is usually localized to a specific geometric location in a narrow blood vessel (such as the larynx of a plaque or the downstream separation region).

[0079] Specifically, if eigenvalue decomposition into Where λ1≥λ2≥...>0, then v1 is the dominant perturbation direction. This direction characterizes the flow space mode on the patch surface that is most easily excited by physiological noise, leading to local high shear stress concentration and inducing fiber cap rupture.

[0080] Specifically, to evaluate the impact of changes in the rheological parameter n on system stability, a rheological sensitivity operator is defined. : in, Expressed as viscosity parameter The partial derivative; because:

[0081] Furthermore, the convection operator N is independent of n, therefore .

[0082] The rheological sensitivity operator describes the full response of the system's dissipative properties to small changes in shear thinning intensity, including direct viscosity changes and indirect base current corrections.

[0083] More specifically, the rheological sensitivity operator is obtained by taking the partial derivative of Carreau's viscosity law at the base flow strain rate, and its calculation formula is as follows:

[0084] in, Let n be the partial derivative of viscosity with respect to the power-law exponent n in the Carreau model. For the basic strain rate tensor, It is a symmetric tensor operator weighted by the background strain rate.

[0085] The rheologically sensitive operator exhibits a significant amplitude in the high-strain-rate region of the patch's narrow throat, with the amplitude attenuating in the downstream low-shear recirculation region. This spatial distribution reveals the physical nature of the rheological-geometric coupling: the influence of the shear-thinning effect is mainly concentrated in the high-strain-rate throat and shoulder regions of the patch, rather than being uniformly distributed throughout the entire flow field. This operator reflects the maximum modulation capability of the shear-thinning effect on the dissipation characteristics near the patch.

[0086] The assessment module 140 is used to assess the trend of the influence of non-Newtonian properties of blood on the risk of atherosclerotic plaque rupture based on the degree of correlation between the first kinetic characteristic and the second kinetic characteristic.

[0087] Specifically, the alignment index is calculated as follows: First, calculate the rheological sensitivity operator. The dominant eigenvector, i.e., the direction of maximum dissipation increment m1, is then used. The dominant perturbation direction v1 is then multiplied by m1:

[0088] Where v1 is the dominant eigenvector of the controllable Gauram matrix. For rheologically sensitive operators, ·, · This is the weighted inner product in the flow field space.

[0089] Furthermore, define the alignment. , where v1 is the dominant eigenvector of the Gram operator and m1 is the dominant eigenvector of the rheological sensitivity operator.

[0090] Under the symmetric limit, the critical alignment is:

[0091] Where λ1 is the largest eigenvalue of the Gram operator, and λ2 is the second largest eigenvalue of the Gram operator.

[0092] like < Sign reversal occurs (i.e., shear thinning inhibits the risk of plaque rupture). like > This manifests as monotonic instability (i.e., shear thinning promotes the risk of plaque rupture).

[0093] Risk assessment rules: based on The sign and numerical range of the symbol, and the risk mechanism for determining hemodynamic status: Firstly, the Stokes-dominant mechanism (rheological instability): if If the alignment is greater than 0 (i.e., the alignment is greater than the preset first threshold), then dV / dn<0 indicates that shear thinning lowers the quasi-barrier, making plaque breakage more likely. In this case, a warning is issued that shear thinning increases the risk of plaque breakage.

[0094] Secondly, non-normal dislocation risk mechanisms (rheologically stable type): if If dV < 0 (i.e., the alignment is less than the preset second threshold), then / dn>0 indicates that shear thinning raises the quasi-barrier, which has a protective effect against plaque breakage. In this case, the warning shear thinning has a counterintuitive protective effect against plaque breakage.

[0095] Third, the transitional hybrid mechanism: if The value between the second and first thresholds indicates a critical state, dominated by higher-order terms, requiring specific numerical analysis.

[0096] The first threshold is a positive value, and the second threshold is a negative value; the first and second thresholds are empirical values ​​obtained based on historical clinical case data and in vitro experimental calibration.

[0097] Based on the above judgment results, it is determined whether the vascular geometry belongs to the Stokes-dominant mechanism or the non-normal-dislocation mechanism.

[0098] Among them, the output alignment index The specific numerical value. If the alignment index... A large negative number indicates that the system is not sensitive to shear thinning; if A value close to zero or positive indicates that minute changes in blood rheological properties can significantly affect the probability of plaque rupture.

[0099] Furthermore, after assessing the influence trend of the non-Newtonian properties of the blood on the risk of atherosclerotic plaque rupture, the method further includes generating an atherosclerotic plaque rupture risk warning report. This report includes: a risk mechanism type identifier, used to indicate the dominant mechanical mechanism of plaque fibrous cap rupture; a plaque rupture quasi-barrier monotonicity indicator value, determined by the sign and amplitude of the alignment index, used to characterize the trend direction and sensitivity of plaque stability as a function of the power law exponent; and a plaque rupture risk contribution score, the value of which is equal to the alignment index, with positive values ​​indicating increased risk due to shear thinning and negative values ​​indicating decreased risk due to shear thinning.

[0100] When the Stokes-dominant mechanism is identified, clinicians should be aware of the risk of plaque rupture caused by abnormal blood viscosity (such as anemia or polycythemia) and intervention measures are recommended. When the non-normal misalignment risk mechanism is identified, it suggests that the shear thinning effect has a counterintuitive protective effect on plaque rupture, and the applicability of traditional wall shear stress indices should be reassessed.

[0101] Furthermore, this embodiment also includes: correcting the risk warning results by combining the wall shear stress distribution on the patch surface; when the calculated quasi-barrier change trend is inconsistent with the wall shear stress distribution, the alignment diagnosis result is used as the priority warning basis for patch rupture risk assessment.

[0102] Furthermore, after obtaining the risk warning results, cross-validation is performed by combining the clinical wall shear stress distribution map; when the determined risk mechanism is inconsistent with the spatial distribution of the high-risk area of ​​wall shear stress, the non-normal misalignment mechanism diagnosed by the alignment index is used as the priority warning basis, and the rheological-geometric coupling abnormality warning information is marked in the warning report.

[0103] In addition, when changes in a patient's blood rheological parameters are detected, the alignment index and risk mechanism type determination results are updated, and the updated warning information is pushed to the clinical terminal.

[0104] To verify the effectiveness of the above system, this embodiment constructs a three-modal Galerkin simplified model to simulate the flow in the narrow shear layer around the patch, with the specific settings as follows: The three modes with the most significant energy growth in a narrow flow field are selected as the basis Φ3=span{φ1, φ2, φ3}, and the Galerkin hypothesis is applied. : φ1 (flow stripe / slow mode): mainly concentrated in the low shear recirculation region downstream of the patch, which is the slowest evolving and most amplified mode (slow mode).

[0105] φ2 (wall normal disturbance / fast mode): concentrated in the narrow throat and shoulder of the patch, where the average shear and lift coupling is strongest.

[0106] φ3 (spreading modulation): provides a three-dimensional saturation path.

[0107] Projected system equations: Projecting the original equations yields a simplified three-degree-of-freedom system:

[0108] Where x is the magnitude vector of the three modes, It is its derivative with respect to time; where It is a diagonal matrix, representing the modal damping ratio; It is a strictly upper triangular matrix, representing the nonnormal lifting linkage effect.

[0109] To clearly demonstrate the two distinct physical mechanisms of rheological monotonicity, an experiment was designed with Mechanism I (Stokes-dominated) as the control group and Mechanism II (non-normal-dominated) as the experimental group for comparison: Mechanism I: Damping Matrix Settings

[0110] Flow matrix (No convection coupling n>0); The results show that dV is monotonically increasing. / dn>0, shear thinning lowers the potential barrier.

[0111] Mechanism II: Damping Matrix Settings

[0112] The results show that: non-monotonic / sign-inverted, dV / dn<0 (in some intervals), shear thinning may raise the potential barrier.

[0113] As shown in Figure 2(a), the V of mechanism I The (n) curve shows a monotonically increasing trend.dV / dn≈0.28>0, which verifies that in pure viscous / low Reynolds number flow, increasing n increases the potential barrier, while shear thinning decreases the potential barrier, which is consistent with conventional intuition.

[0114] And mechanism II's V The curve (n) shows a distinct non-monotonic region near n≈1.6 (shaded area in Figure 2(a)), indicating a sign reversal. At n=1.5, the potential barrier along the dominant direction is about 21% higher than at n=2.0, demonstrating that in nonnormal flow, shear thinning actually increases the barrier to plaque breakup, playing a protective role, which is completely opposite to the results of mechanism I.

[0115] As shown in Figure 2(b), the alignment diagnostic index χ(n) is significantly lower in mechanism II than in mechanism I (χ≈0.08), revealing the mathematical root of sign inversion, namely the severe misalignment between the dominant eigenvector v1 of the Gram operator and the dominant eigenvector m1 of the rheological sensitivity operator (χ=| v1, m1 |²≈0.08).

[0116] As shown in Figure 2(c), the comparison of the instantaneous path on the projection plane of mode 1-mode 2 under the two mechanisms shows that the instantaneous path of mechanism I is approximately straight (gradient dynamics dominates), while the instantaneous path of mechanism II is significantly curved. Mode 1 (slow mode in the downstream backflow region of the patch) is excited first, and then mode 2 (fast mode in the throat of the patch) is amplified through the nonnormal lifting effect. This is a typical feature of energy transfer in nonnormal flow.

[0117] As shown in Figure 2(d), the quasi-barrier derivative dV The variation of / dn with the power law exponent shows that the derivative of mechanism I is positive and stable (approximately +0.28), while the derivative of mechanism II crosses zero at n≈1.6, changes from positive to negative, and is negative in the interval [1.4, 1.9].

[0118] As shown in Figure 2(e), the Gram condition number κ(Qc) varies with the power law exponent, showing that κ≈2 for mechanism I (condition number close to 1, system close to normal), and κ for mechanism II spans the order of 10²~10³ (vertical axis is logarithmic coordinate), indicating that the system has a strong non-normal energy amplification capability - which is the dynamic root of the sign inversion.

[0119] As shown in Figures 2(f1) to 2(f2), the misalignment (χ) between v1 and m1 is the root cause of the inversion, and non-normality (G_max) is the key catalytic condition for triggering the inversion; both are indispensable. This perfectly supports the core claim of the alignment criterion proposed in this invention: χ is a calculable diagnostic index for assessing how the risk of plaque rupture changes with shear thinning.

[0120] In the numerical verification of the normal misalignment mechanism, Figure 2 (f1) shows the spatial distribution of the quasi-barrier level set and the dominant eigenvectors v1 (red), v2 (gray), and m1 (purple) of the Gram matrix, as well as the dominant direction of the dissipative increment operator M. It is evident that there is a significant angular misalignment (χ≈0.116) between v1 and m1, which is the root cause of the sign inversion. This misalignment results in the covariance derivative matrix Σ′ exhibiting one positive and one negative eigenvalue, as shown in Figure 2 (f2), indicating that the system's response to the power-law exponent n is direction-dependent. The covariance in the v1 direction increases with increasing n (v1... Σ′v1≈+1.099>0), which is precisely the mathematical signal for sign reversal.

[0121] As a reverse control, when the direction of M is manually rotated to be completely aligned with v1, both eigenvalues ​​of Σ′ become negative, as shown in Figure 2(f3), confirming that misalignment is a necessary condition for reversal. Simultaneously, the system exhibits a significant transient energy amplification effect, with the nonnormal transient growth G(t) reaching a peak of G_max≈1.38 at t≈0.64, as shown in Figure 2(f4), demonstrating that the system possesses sufficient nonnormality to drive reversal. Under the combined effect of the above mechanisms, the potential barrier curve along the v1 direction shows that the potential barrier at the power-law exponent n=1.5 (shear-thinning) is approximately 21% higher than that at n=2.0 (Newtonian fluid), as shown in Figure 2(f5). This indicates that shear-thinning in this direction actually inhibits the risk of plaque breakup, completely contrary to the prediction of the Stokes-dominated mechanism.

[0122] Figure 2 (f6) further shows that, due to the existence of nonnormality, the effective critical alignment χ_c is significantly increased, making χ≈0.1164 greater than the symmetry limit χ. c ≈0.0706, yet it is still sufficient to trigger sign inversion—this reveals the key catalytic role of nonnormality in inversion, that is, the stronger the nonnormality, the higher the effective critical value, and the easier it is for sign inversion to occur.

[0123] The above six sub-graphs together form a complete argument loop: geometric misalignment → mathematical indeterminacy → nonnormal amplification → sign reversal, progressively verifying the effectiveness of the alignment criterion proposed in this invention from principle to phenomenon.

[0124] Based on the multidimensional diagnoses in Figures 2(a) to 2(f6), in mechanism II, χ ≈ 0.08–0.116, the system is near a critical state. Due to the presence of nonnormality, the effective critical value is higher than the symmetry limit χ_c^sym ≈ 0.071, thus reversal still occurs. The significant bending of the instanton path in Figure 2(c) and the high condition number in Figure 2(e) together provide direct evidence for the existence of nonnormality.

[0125] Furthermore, to demonstrate the robustness of the system in this application, as shown in Figure 3(a), Mechanism 2: Informal system, quasi-barrier curve V under different nonlinear intensities γ (γ=0, 0.1, 0.3, 0.6) (n) indicates that the non-monotonic structure exists stably under the four conditions of γ=0 (purely linear), γ=0.1, γ=0.3, and γ=0.6, and the non-monotonic region is always clearly distinguishable, indicating that the system of this application is not sensitive to changes in nonlinear intensity and has high qualitative robustness.

[0126] As shown in Figure 3(b), the relative deviation of the nonlinear correction relative to the linear barrier (γ=0) shows that even under the strong nonlinear condition of γ=0.6, the relative deviation is still less than 10%, which confirms that the near-critical approximation based on linearization analysis in this application has quantitative accuracy under medium threshold amplitude.

[0127] As shown in Figure 4(a), the quasi-barrier curves V under different lifting coupling intensities (0, 0.5, 1.0, 1.5, 2.0, 2.5) are shown. (n) shows that when the uplift intensity is 0, the system degenerates into mechanism one, namely the Stokes dominant mechanism, and the curve increases monotonically. As the uplift intensity increases, the barrier curve gradually flattens and develops a non-monotonic region. The non-monotonicity increases with the increase of nonnormality.

[0128] As shown in Figure 4(b), the non-monotonic region (i.e., dV) The percentage of the region where / dn<0 occupies the range of n values ​​as the coupling strength increases shows that there is a critical nonnormality threshold (about 0.8). Below this threshold, there is no non-monotonic region (monotonic stability). Above this threshold, the proportion of non-monotonic region increases rapidly and tends to saturate, clearly depicting the phase transition process from monotonic to non-monotonic.

[0129] This finding has important clinical implications: the non-monotonic protective effect of shear thinning should only be considered when the non-normality caused by plaque geometry exceeds a certain level; in vessels with weaker non-normality, the traditional monotonic cognition still applies.

[0130] As shown in Figure 5(a), the relative error of gMAM (geometric minimum action method) decreases to O(N) as the number of discrete points N increases. - ²) On the order of magnitude. As shown in Figure 5(b), the wall computation time increases polynomially with increasing N, indicating that the gMAM system is feasible and scalable in practical computation.

[0131] As shown in Figure 6(a), the comparison between the calculated gMAM value and the closed-form Lyapunov quasi-barrier in the four target directions shows that the two are similar in 10... -6The accuracy of the gMAM numerical calculation is consistent within the specified range. As shown in Figure 6(b), the representative instantaneous path is superimposed on the quasi-barrier contour map, which intuitively shows the geometric shape of the optimal escape path in phase space.

[0132] As shown in Figure 7(a), the relationship between det(Σ') and χ indicates that uncertainty occurs below the critical χ, verifying the theoretical prediction of Theorem 7. As shown in Figure 7(b), the quadratic form g... The sign of Σ'g changes with the change of χ at the critical χ (g=v1), further confirming the precise mathematical relationship between sign reversal and alignment. As shown in Figure 7(c), the change of the sign of the barrier derivative with χ clearly marks the boundary between the monotonic and non-monotonic regions. The purple dashed line marks the position of the corrected counterexample (χ≈0.116), which is located in the non-monotonic region, consistent with the conclusion of Figure 2(f).

[0133] As shown in Figure 8(a), the surface curve of χ(n, lift) as a function of the power-law exponent n and the lift coupling strength reveals the distribution law of alignment in the parameter space—χ decreases with increasing lift strength and increases slightly with increasing n. As shown in Figure 8(b), the condition number of the controllable Galam matrix is ​​taken as the base-10 logarithm: log 10 κ(Qc) spans four orders of magnitude, and Figure 8(b) reflects the huge range of variation in the intensity of nonnormality, further validating the effectiveness of the condition number κ as a quantitative indicator of nonnormality.

[0134] Furthermore, in order to calculate the quasi-barrier V (n) The geometric minimum action method gMAM is used to minimize the Freidlin-Wentzell action functional. The calculation process is as follows: Path discretization: γ:[0,1]→ d Discretize into N = 40 to 60 points.

[0135] For linear dynamics or weakly nonlinear case The geometric action is defined as:

[0136] Where Φ is the geometric action functional; γ' is the path... γ The derivative with respect to parameter s; f(γ) is the deterministic drift term (linear or weakly nonlinear) of the system at the path point γ; γ is the path variable (distinct from the nonlinear coefficients, which are already represented as parameters in the expression of f(x)). Discretize the path as... N =40 With 60 points, fixed endpoint constraints are applied, and the discretization action is minimized using the L-BFGS algorithm (Limited-memory Broyden-Fletcher-Goldfarb-Shanno algorithm) with periodic arc length reparameterization.

[0137] Hyperplane The optimal escape point on is This equation holds exactly in the case of purely linear dynamics, and also has the following properties in the case of weak nonlinearity. The precision of the order.

[0138] in, g Let be the normal vector of the threshold hyperplane. h Σ is the threshold constant. n For controllable personality Ram matrix; x The optimal escape point is the phase space location that the system is most likely to traverse when escaping from the steady state to the threshold hyperplane.

[0139] Optimization solution: The quasi-Newton method optimizer is used to optimize the path by reparameterizing the arc length in order to find the path with minimum action.

[0140] For the plaque rupture threshold hyperplane g, x =b, the optimal escape point is reached. Calculation, where It is a controllable personality Ram matrix.

[0141] To verify the computational accuracy of this system and to visually demonstrate the flow field evolution process under the two risk mechanisms, Figures 2(c), 5(a), 5(b), 6(a), and 6(b) present the corresponding computational results obtained based on the Generalized Minimalist Autonomous Inference (gMAM) system.

[0142] The atherosclerotic plaque rupture risk assessment system provided in this application accurately captures the actual flow characteristics of blood in complex and narrow blood vessels, and can quantitatively reveal the intrinsic relationship between rheological properties and plaque geometry. This allows for accurate determination of whether shear thinning effects exacerbate or alleviate plaque rupture risk in key clinical sites such as the carotid artery and coronary artery, effectively avoiding the misjudgment problems of traditional systems. Furthermore, this system can adaptively adjust its judgment logic without relying on empirical thresholds, and only requires steady-state flow field calculations to complete the diagnosis, significantly reducing reliance on supercomputers and computation time. This enables rapid batch assessment of individualized blood vessels on conventional equipment, and the output assessment results can directly serve preoperative planning, interventional decisions, and postoperative follow-up, providing practical, efficient, and accurate technical support for clinical practice.

[0143] The above description represents the preferred embodiments of this application. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles described in this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A system for assessing the risk of atherosclerotic plaque rupture, characterized in that, include: The acquisition module is used to acquire the geometric model of the target blood vessel and receive the blood rheological parameters output by the blood rheology detection device in clinical applications. The geometric model includes at least one narrowing region caused by atherosclerotic plaque. The basic flow field solution module is used to determine the steady-state flow state of blood in the target blood vessel based on the geometric model and the blood rheological parameters, and to obtain the basic flow field containing the hemodynamic characteristics around the plaque; The feature extraction module is used to extract, based on the base flow field, a first dynamic feature characterizing the effect of changes in the non-Newtonian properties of blood on system stability, and a second dynamic feature characterizing the base flow field as most sensitive to physiological disturbances. An evaluation module is used to assess the trend of the influence of the non-Newtonian properties of the blood on the risk of atherosclerotic plaque rupture, based on the degree of correlation between the first kinetic characteristic and the second kinetic characteristic.

2. The system for assessing the risk of atherosclerotic plaque rupture according to claim 1, characterized in that, The blood rheological parameters were obtained based on the Carreau-Yasuda constitutive model, which includes a power-law exponent to characterize the degree of blood shear thinning.

3. The system for assessing the risk of atherosclerotic plaque rupture according to claim 1, characterized in that, The process of obtaining the steady-state laminar base flow field, which includes hemodynamic characteristics around the plaque, based on the geometric model and blood rheological parameters, includes: Solve the steady-state incompressible Navier-Stokes equations within the computational domain of the target blood vessel to obtain the steady-state laminar solution as the base flow field.

4. The system for assessing the risk of atherosclerotic plaque rupture according to claim 1, characterized in that, The first dynamic characteristic is obtained by taking the Fréchet derivative of the blood power law exponent with respect to the linearized dynamic operator of the system in which the base flow field exists. The first dynamic characteristic is a rheological sensitivity operator used to characterize the sensitivity of the dissipation characteristics of the base flow field to the power law exponent.

5. The system for assessing the risk of atherosclerotic plaque rupture according to claim 1, characterized in that, Extracting the second dynamic feature includes: The Navier-Stokes equations are linearized at the base flow field to construct a linearized dynamic operator; a noise covariance matrix characterizing the physiological fluctuations of the cardiovascular system is defined. The controllable Gauram matrix is ​​obtained by solving the Lyapunov equation; the controllable Gauram matrix is ​​subjected to eigenvalue decomposition, and the normalized eigenvector corresponding to the largest eigenvalue is extracted as the second dynamic feature.

6. The system for assessing the risk of atherosclerotic plaque rupture according to claim 1, characterized in that, The narrowed area caused by the atherosclerotic plaque includes the plaque shoulder area and the fibrous cap area.

7. The system for assessing the risk of atherosclerotic plaque rupture according to claim 2, characterized in that, The viscosity function formula for the Carreau-Yasuda constitutive model is as follows: in, Zero shear viscosity For infinite shear viscosity, For relaxation time, Shear rate, The power-law exponent. This is the Yasuda index, used to control the steepness of the transition zone.

8. The system for assessing the risk of atherosclerotic plaque rupture according to claim 2, characterized in that, When the transition parameter a=2 is fixed, the Carreau-Yasuda constitutive model degenerates into a Carreau-type constitutive relation, and the viscosity function simplifies to: in, Zero shear viscosity It is a liquidity behavior index; Let D be the second invariant of the strain rate tensor. For the velocity gradient tensor; when When the coefficient of friction is 2, the model reverts to a Newtonian fluid. when When <2, it characterizes a shear-thinned fluid, and The smaller the value, the stronger the shear thinning effect.

9. The system for assessing the risk of atherosclerotic plaque rupture according to claim 3, characterized in that, The solution to the steady-state laminar base flow field includes: Solve the incompressible Navier-Stokes equations within the geometric region Ω of the blood vessel: in, For fluid density, For convection operators, For pressure, For gradient operators; Viscosity, For strain rate tensor; =0 is a constraint condition; The equations are solved by numerical simulation or analytical system to obtain the steady-state laminar solution as the base flow field. The base flow field includes the high-shear region at the throat of the patch, the low-shear recirculation region downstream of the patch, and the flow separation region at the shoulder of the patch.

10. The system for assessing the risk of atherosclerotic plaque rupture according to claim 5, characterized in that, The construction of the linearized dynamic operator includes: linearizing the Navier-Stokes equations in a weighted kinetic energy inner product space to obtain the linearized dynamic operator, and decomposing the linearized dynamic operator into a symmetric dissipative operator and an antisymmetric convection operator in the weighted kinetic energy inner product space.

11. The system for assessing the risk of atherosclerotic plaque rupture according to claim 10, characterized in that, The formula for the weighted kinetic energy inner product is: in, The computational domain for blood vessels containing plaque regions; For the weight function, For vector dot product, It is a volume infinitesimal element; The linearized dynamic operator decomposition satisfies in, Here, D is a linearized kinetic operator, and D is a symmetric positive definite dissipation operator representing the viscous dissipation of blood, as shown in the formula: C is the antisymmetric convection operator, representing the inertial transport and nonmodal energy transfer of blood flow, and its formula is: in, L n express L n Adjoint operators in weighted inner product spaces; The decomposition is performed in a weighted inner product space, which is defined by the kinetic energy norm of the fluid.

12. The system for assessing the risk of atherosclerotic plaque rupture according to claim 5, characterized in that, The Lyapunov equation is: in, The controllable Graphlam matrix represents the spatial covariance structure of blood flow disturbance in the plaque region; The transpose operator of the linearized dynamics operator; The noise covariance matrix contains the heart rate variability perturbation component and the red blood cell micro-fluctuation perturbation component.

13. The system for assessing the risk of atherosclerotic plaque rupture according to claim 12, characterized in that, The noise covariance matrix is ​​decomposed into: in, To simulate the inflow fluctuation component caused by heart rate variability; To simulate the microscale erythrocyte agitation component.

14. The system for assessing the risk of atherosclerotic plaque rupture according to claim 5, characterized in that, The dominant perturbation direction of the patch instability is the controllable Gauram matrix. The normalized eigenvector v1 corresponding to the largest eigenvalue λ1 represents the flow space mode on the patch surface that is most easily excited by physiological noise, leading to local high shear stress concentration and inducing fiber cap rupture.

15. The system for assessing the risk of atherosclerotic plaque rupture according to claim 4, characterized in that, The rheological sensitivity operator The following is obtained by taking the partial derivative of Carreau's viscosity law at the basic strain rate: in, Expressed as viscosity parameter The partial derivative; because: Furthermore, the convection operator N is independent of n, therefore ; The rheological sensitivity operator describes the full response of the system's dissipation characteristics when the shear thinning intensity changes slightly, including direct viscosity changes and indirect base current corrections; The rheological sensitivity operator has a significant amplitude in the high strain rate region of the narrow throat of the plaque, and the amplitude decays in the downstream low shear reflow region.

16. The system for assessing the risk of atherosclerotic plaque rupture according to claim 4, characterized in that, The degree of correlation is measured by an alignment index, which is used to characterize the degree of geometric matching between the second dynamic feature and the principal direction of dissipation increment determined by the first dynamic feature. The alignment index The alignment index is calculated as follows: the image of the second dynamic feature under the action of the rheological sensitivity operator is multiplied by the second dynamic feature itself, and the inner product value is normalized to obtain the alignment index. The calculation formula is as follows: Where v1 is the dominant eigenvector of the controllable Gauram matrix. For rheologically sensitive operators, ·, · This is the weighted inner product in the flow field space; The alignment index ranges from [-1, 1], and its absolute value reflects the degree of geometric overlap between the optimal forcing direction and the principal direction of dissipation increment.

17. The system for assessing the risk of atherosclerotic plaque rupture according to claim 16, characterized in that, The risk mechanism is determined based on the alignment index, and the determination of the risk mechanism includes: When the alignment is greater than a preset first threshold, it is determined to be the Stokes dominant mechanism. At this time, the shear thinning effect reduces the quasi-barrier of plaque rupture and warns that blood shear thinning increases the risk of plaque rupture. When the alignment is less than a preset second threshold, it is determined to be a non-normal misalignment risk mechanism. At this time, the shear thinning effect raises the quasi-barrier of plaque rupture, and the blood shear thinning has a counterintuitive protective effect on plaque rupture. When the alignment index is between the second threshold and the first threshold, it is determined to be a transitional hybrid mechanism; The first threshold is a positive value, and the second threshold is a negative value; the first threshold and the second threshold are empirical values ​​obtained based on historical clinical case data and in vitro experimental calibration.

18. The system for assessing the risk of atherosclerotic plaque rupture according to claim 17, characterized in that, The risk mechanism assessment also includes defining alignment. Where v1 is the dominant eigenvector of the Gram operator and m1 is the dominant eigenvector of the rheologically sensitive operator; under the symmetric limit, the critical alignment is: in, The largest eigenvalue of the Gram operator. This is the second largest eigenvalue of the Gram operator; like < A sign reversal occurs, meaning that shear thinning inhibits the risk of plaque rupture; like > This manifests as monotonic instability, meaning that shear thinning promotes the risk of plaque rupture.

19. The system for assessing the risk of atherosclerotic plaque rupture according to claim 16, characterized in that, After assessing the trend of the influence of the non-Newtonian properties of the blood on the risk of atherosclerotic plaque rupture, the method further includes generating an atherosclerotic plaque rupture risk warning report, which includes: Risk mechanism type identifier, used to indicate the dominant mechanical mechanism of plaque fibrous cap rupture; The plaque rupture quasi-barrier monotonicity indicator value, determined by the sign and amplitude of the alignment index, is used to characterize the trend direction and sensitivity of plaque stability as a function of the blood power law exponent. The plaque fracture risk contribution score is equal to the alignment index. A positive value indicates that shear thinning increases the risk, while a negative value indicates that shear thinning reduces the risk.

20. The system for assessing the risk of atherosclerotic plaque rupture according to claim 19, characterized in that, Also includes: The risk warning results are corrected by combining the shear stress distribution on the surface of the patch. When the calculated quasi-barrier change trend is inconsistent with the wall shear stress distribution, the alignment diagnosis result shall be used as the priority early warning basis for plaque rupture risk assessment.

21. The system for assessing the risk of atherosclerotic plaque rupture according to claim 20, characterized in that, Also includes: After obtaining the risk warning results, cross-validation was performed by combining them with clinical wall shear stress distribution maps. When the determined risk mechanism is inconsistent with the spatial distribution of high-risk areas of wall shear stress, the non-normal misalignment mechanism diagnosed by the alignment index is used as the priority for early warning, and the rheological-geometric coupling anomaly warning information is marked in the early warning report.

22. The system for assessing the risk of atherosclerotic plaque rupture according to claim 16, characterized in that, Also includes: When changes in the patient's blood rheological parameters are detected, the alignment index and risk mechanism type determination results are updated, and the updated plaque rupture risk warning information is pushed to the clinical terminal.

23. The system for assessing the risk of atherosclerotic plaque rupture according to claim 1, characterized in that, The target blood vessel is a coronary artery or a carotid artery, and the atherosclerotic plaque is a vulnerable plaque whose rupture leads to acute coronary syndrome or ischemic stroke.

Images