WO2017047822A1

WO2017047822A1 - Vascular lesion onset/growth prediction device and method

Info

Publication number: WO2017047822A1
Application number: PCT/JP2016/077757
Authority: WO
Inventors: 高伸八木
Original assignee: イービーエム株式会社
Priority date: 2015-09-18
Filing date: 2016-09-20
Publication date: 2017-03-23
Also published as: JPWO2017047822A1; JP6856937B2

Abstract

[Solution] This device, for predicting the onset and growth of lesions in a blood vessel, acquires shape data of a blood vessel by inputting medical images into the device, and performs numerical fluid analysis; the device further calculates blood flow properties by quantifying blood flow grade, discriminating for parallel flow, confluence, rotation and collision; the device further calculates blood vessel fragility from information about endothelial cell function, and, on the basis of the above, calculates a risk value for the onset or growth of a vascular lesion.

Description

Apparatus and method for predicting vascular lesion onset / growth

The present invention relates to an apparatus for predicting the onset of a vascular lesion and its growth risk by computer simulation.

It is known that the onset of cerebral aneurysms and aortic aneurysms can be caused by stimulation of blood flow. For example, when a cerebral aneurysm is experimentally created, it is empirically known that a cerebral aneurysm occurs on the inner peripheral side of a branch portion of a specific blood vessel. In clinical practice, cerebral aneurysms have been pointed out frequently, and about 60% of them are concentrated in the internal carotid artery-rear traffic artery branch and middle cerebral artery first branch. It is known that cerebral aneurysms occur in such places, but on the other hand, since the onset of cerebral aneurysms increases with aging, there is malignant blood flow that caused the pre-occurrence stage. It is considered that cerebral aneurysm develops due to the synergistic effect of increasing blood vessel vulnerability with aging.

The same applies to coronary stenosis and carotid stenosis, which are representative of arteriosclerosis. In clinical settings, stenosis has been pointed out as a frequent blood vessel or site. It is unlikely that there is no blood flow from the pre-stenosis stage. In other words, it is difficult to predict the onset and growth of stenosis only with the malignancy of the blood flow. Therefore, it was found that it is important to include the fragility of blood vessels in addition to the malignancy of blood flow. That is, when the ratio between the malignancy of blood flow and the fragility of blood vessels exceeds a certain threshold, the risk of developing vascular stenosis increases.

The above also applies to the indexing of blood flow to identify that stenosis once developed due to the occurrence of plaque (hypertrophic lesion in the intima) grows and stabilizes or destabilizes. Here, “stabilization” means so-called stable plaque formation, and indicates that the plaque tissue has been converted to a state where the risk of plaque rupture is low due to fibrosis. On the other hand, “destabilization” is so-called unstable plaque formation, and the plaque tissue is composed of rod-shaped tissue such as macrophages and foam cells, or a part of the tissue is composed of a risk of plaque rupture. It has been converted to a high state.

Special table 2009-518097

The present invention provides an onset / growth risk prediction apparatus for cerebral aneurysms and vascular stenosis based on the above medical insights.

To predict the onset / growth of cerebral aneurysms, it is necessary to quantify the two antagonists of blood flow malignancy and blood vessel vulnerability. In the present invention, the malignancy of blood flow is obtained by computer simulation, and the fragility of blood vessels is obtained by predicting vascular endothelial cell function. Further, it will become clear from the following description that the present invention also provides a standard for the onset, growth, or stabilization / instability conversion of arteriosclerosis.

In order to solve the above-mentioned problem, according to a first main aspect of the present invention, a computer inputs a medical image including an analysis target blood vessel part and information on endothelial cell function, and the computer inputs Blood flow analysis unit that obtains blood vessel shape data from medical data input from the department and performs numerical fluid analysis to obtain blood flow attributes including pressure and velocity fields, and a computer acquired by the blood flow analysis unit A blood flow malignancy calculation unit that determines the blood flow characteristics from the wall shear stress vector based on the blood flow attribute and quantifies the blood flow malignancy, and a computer The blood vessel vulnerability calculation unit that obtains the blood vessel vulnerability from the information, and the computer calculates the vascular lesion from the blood flow malignancy obtained by the blood flow malignancy calculation unit and the blood vessel vulnerability obtained by the blood vessel vulnerability calculation unit. A risk calculation section that calculates a risk value for the disease, or growth, computer, vascular lesions onset and growth prospects device is provided with an output unit for outputting the calculated risk value.

According to one embodiment of the present invention, the blood flow analysis unit extracts a blood vessel shape from a medical image, generates a calculation grid, and obtains a pressure field and a flow field while considering fluid properties and boundary conditions. Device. In addition, the blood flow analysis unit is configured as a device characterized by inputting a blood vessel shape, blood physical properties, boundary conditions and calculation conditions, and outputting a four-dimensional velocity field and pressure field including time. May be.

According to yet another embodiment, the determination of the blood flow property is performed by obtaining a wall shear stress vector at each position of the blood vessel wall surface of the blood vessel part to be analyzed from the state amount of the blood flow obtained by the blood flow analysis unit, The relative relationship between the direction of the wall shear stress vector at a specific wall position and the direction of the wall shear stress vector at the surrounding wall positions is obtained, and the blood flow characteristics at the wall position are determined from the form, and the determination result is output. It is a device that does.

From the relative angular relationship between the wall shear stress vector τ at a specific wall position and a plurality of wall shear stress vectors at the surrounding wall positions, the blood flow property determination unit calculates the rotation amount rotτ and the divergence divτ that are scalar quantities of the vector field τ. It is determined by comparing these values with the threshold value as the degree of randomness to determine whether it is “parallel”, “confluence”, “rotation”, or “divergence”. When the value is negative or positive outside the threshold range, it is determined to be “rotation”, and when the divergence divτ value of the randomness is a negative value outside the predetermined threshold range, it is determined to be “join”, and the randomness is determined. When the value of the divergence divτ is a positive value outside the predetermined threshold range, it is determined as “collision”, and when both the value of the rotation rotτ and the value of the divergence divτ are within the predetermined threshold value, You may make it discriminate | determine.

According to still another embodiment, the information on the endothelial cell function is obtained by calculating a diameter change from a blood vessel diameter measurement value at rest and at the time of release of blood transfusion.

The risk calculation unit determines whether the blood flow property is “Rotation” or “Merging” and the blood vessel vulnerability is high, and the blood flow property is “collision” and the blood vessel vulnerability is high. If it is determined that there is a risk of onset / growth of a lesion, the apparatus may be used.

In this case, if the vascular region to be analyzed is a cerebral artery, the blood flow characteristics are determined to be “rotation” or “confluence”, and the vascular vulnerability is determined to be high, the onset / growth of arteriosclerosis As a device that judges that there is a risk, and if the blood flow property is judged as “collision” and the vascular vulnerability is judged to be high, it is judged that there is a risk of onset / growth of cerebral aneurysm Also good.

According to yet another embodiment, the apparatus is characterized by determining that there is a risk when the ratio of the malignancy of blood flow to the vascular vulnerability exceeds a threshold value.

According to a second main aspect of the present invention, there is provided a computer-implemented method in which a computer inputs information about a medical image including an analysis target blood vessel region and endothelial cell function; The computer obtains blood vessel shape data from the medical data input in the input process and performs numerical fluid analysis to determine blood flow attributes including pressure and velocity fields, and the computer performs blood flow analysis. The blood flow malignancy calculation step that determines the blood flow malignancy by quantifying the blood flow malignancy from the wall shear stress vector based on the blood flow attributes obtained in the process, and the endothelium input in the input process In the blood vessel vulnerability calculation process to obtain the blood vessel vulnerability from the information about the cell function, the computer performs the blood flow malignancy calculated in the blood flow malignancy calculation process and the blood vessel vulnerability calculation process. A blood vessel lesion onset / growth prediction method comprising a risk calculating step for calculating a risk value for the onset or growth of a vascular lesion from the determined vascular vulnerability and an output step for a computer to output the calculated risk value. is there.

It will become clear from the disclosure of the invention described below that various other problems can be solved.

FIG. 1 is a configuration diagram of an endovascular treatment simulation apparatus according to an embodiment of the present invention. FIG. 2 is a diagram conceptually showing functions and processes of the vascular lesion onset / growth prediction apparatus according to an embodiment of the present invention. FIG. 3 is a diagram showing a flow of blood flow analysis in one embodiment of the present invention. FIG. 4 is a schematic diagram illustrating fluid shear stress. FIG. 5 is a schematic diagram illustrating fluid shear stress. FIG. 6 is a diagram showing a global coordinate system related to calculation of wall shear stress. FIG. 7 is a diagram illustrating a local coordinate system related to the calculation of the wall shear stress. FIG. 8 is a diagram graphically showing the shear stress vector superimposed on the three-dimensional shape of the blood vessel. FIG. 9 is a diagram graphically showing the shear stress vector and the pressure superimposed on the three-dimensional shape of the blood vessel. FIG. 10 is a diagram for explaining calculation of randomness according to an embodiment of the present invention. FIG. 11 is an explanatory diagram relating to the interpretation of the degree of randomness in one embodiment of the present invention. FIG. 12 is a diagram representing the relationship between the degree of randomness and various risks as a map in one embodiment of the present invention. FIG. 13 is a visual representation of the relationship between the vascular malignancy calculated in step I, the vascular vulnerability calculated in step II, and the vascular lesion onset / growth risk determined in step III in an embodiment of the present invention. FIG.

Hereinafter, an embodiment of the present invention will be specifically described with reference to the drawings. FIG. 1 shows a configuration diagram of a vascular lesion onset / growth prediction apparatus 10 according to an embodiment of the present invention. In this apparatus 10, a program storage unit 60 and a data storage unit 70 are connected to a bus 50 to which a CPU 20, a memory 30 and an input / output unit 40 are connected.

The program storage unit 60 includes a blood flow analysis unit 11, a blood flow property determination unit 12, a blood flow malignancy degree calculation unit 13, a vasodilation response measurement unit 14, a blood vessel vulnerability calculation unit 15, a risk calculation unit 16, and a result output unit 17. Etc.

The above configuration requirements (blood flow analysis unit 11, blood flow property determination unit 12, blood flow malignancy calculation unit 13, vasodilation response measurement unit 14, vascular vulnerability calculation unit 15, risk calculation unit 16, result output unit 17, etc.) Is actually constituted by computer software stored in a hard disk, and is read out by the CPU, developed on the memory 30 and executed, thereby functioning as each component of the present invention. .

The data storage unit 70 stores 24 calculation condition templates including blood vessel shape information 21, fluid physical properties 22, arterial diameter change information 23, various coefficients and multipliers, and the like.

The contents stored in the program storage unit 60 and the data storage unit 70 can be updated as appropriate by inputting from the outside of the apparatus via the input / output unit 40. Various data stored in this manner can be used for later-described numerical fluid analysis (CFD) and vascular dilation analysis (FMD) as an endothelial cell test. The input / output unit 40 is an interface with various devices and communication means. In addition to a communication line, a display means such as a display, an operation means such as a keyboard and a mouse, and an external storage device (all not shown) are connected. Configured to be possible.

FIG. 2 is a diagram conceptually illustrating the functions and processes of the vascular lesion onset / growth prediction apparatus according to an embodiment of the present invention. The result obtained from the numerical fluid analysis (CFD) and the vasodilator analysis (FMD) as the endothelial cell function test is input.

CFD in Step I is a blood flow simulation, and obtains a velocity field and a pressure field of a blood flow by calculation using a medical image as described later. Here, blood flow malignancy is calculated and output from the velocity field and pressure field of CFD. In this embodiment, the blood flow malignancy is based on the form evaluation of the wall shear stress vector. That is, the shape of the wall shear stress vector is classified into (1) parallel, (2) rotation, (3) merging, and (4) collision (divergence).

In step II, the vascular vulnerability is calculated and output based on the FMD results, such as numerical values for vascular endothelial cell function measured using ultrasound.

In step III, the vascular lesion onset / growth risk calculation unit calculates and outputs a risk related to vascular lesion onset / growth based on the vascular malignancy and vascular vulnerability calculated in steps I and II.

(Blood flow analysis section)
First, the contents of step I will be described in detail. The blood flow analysis unit 11 acquires a pressure field / flow velocity field based on a medical image input via the input / output unit 40. As this medical image, a vascular tomographic image obtained by MRA (magnetic resonance angiography), CTA (computerized tomography), DSA (digital subtraction angiography), or the like may be used. As shown in FIG. 3, the blood flow analysis unit 11 first receives a medical image (a). Next, a blood vessel shape (surface mesh) is extracted based on the received medical image (b), a calculation grid (volume mesh) is generated (c), while taking into account the fluid physical properties of blood and boundary conditions (wall surface). (D) Furthermore, the flow rate and flow pressure at a plurality of locations in the blood flow are set (e). The fluid physical properties of blood considered at this time are density and viscosity. For the viscosity, a Newtonian fluid model or a non-Newtonian model may be used. Also. The boundary condition is a flow rate designated for each blood vessel, and an actual measurement value, a statistical value, or an estimated value can be used. The calculation condition is a condition used for the simulation, and a pulsatile flow may be used, or a steady flow that can reduce the load on the computing unit may be used. Based on this set flow rate and pressure, the equation is iteratively calculated (f) to obtain the pressure field / velocity field (g). If it solves it, it will become the pressure field and the velocity field in space and time. As described above, blood flow analysis inputs blood vessel shape, blood physical properties, boundary conditions and calculation conditions, and outputs a four-dimensional velocity field and pressure field including time.

(Blood flow property discrimination part)
The blood flow property determination unit 12 is installed with a program that causes a computer to function as the following means. That is, as shown in FIG. 1, the blood flow property determination unit 12 determines the fluid shear stress acting on the blood vessel wall surface by the blood flow and its vector from the pressure field / velocity field of each mesh obtained by the blood flow analysis unit 11. (Hereinafter, simply referred to as “wall shear stress vector”) for each mesh, and a numerical index (disturbance) for determining blood flow properties from the wall shear stress vector calculation unit 121 and the wall shear stress vector. A randomness calculation unit 122 for determining the degree of blood flow, and a determination unit 123 for determining the property of blood flow in each mesh according to the magnitude of the randomness.

(Wall shear stress vector calculation section)
4 and 5 are schematic diagrams showing a method of determining the shear stress vector τ (x, y, z) based on the flow velocity U and the pressure P determined for each mesh in the wall surface shear stress vector calculation unit 121. FIG. is there.

As shown in FIG. 4, the wall shear stress is a viscous force of a fluid acting in a parallel direction with respect to the minute elements forming the blood vessel lumen, and the wall shear stress vector is a vector view of the numerical value. Yes, considering the direction of the force acting on the wall surface. The wall shear stress vector and the pressure are orthogonal to each other, and the pressure is a fluid force acting in the surface normal direction with respect to the center of gravity of the microelement.
In explaining this figure, it is necessary to understand the conversion to the global coordinate system and the local coordinate system. That is, while the pressure P and velocity U used to obtain the shear stress vector are obtained in the global coordinate system as described above, the shear stress acting on the position of the blood vessel wall surface is tangent to the wall surface. It is necessary to convert the pressure and velocity into a local coordinate system based on the blood vessel wall surface in order to obtain the size.

Here, as shown in FIG. 6, the global coordinate system is a single coordinate system for universally indicating the positions of the nodes of the mesh constituting the blood vessel surface and the inside thereof in this system. In the finite element method and the finite volume method, a calculation target is composed of a set of minute elements (triangle, tetrahedron, hexahedron, etc.). Each element has a vertex called a node, and the position information of each element is (X1g, Y1g, Z1g), (X2g, Y2g, Z2g), (X3g, Y3g, Z3g) using the global coordinate system Hold on.

As shown in FIG. 7, the local coordinate system is a local coordinate system defined for each minute triangle element (polygon) constituting the blood vessel surface. Usually, the center of gravity of the minute triangle element is defined as the origin. And the surface normal vector as one axis (Z axis). When the position of each contact point of the microelement is expressed in the local coordinate system, (X1l, Y1 l, Z1 l), (X2 l, Y2 l, Z2 l), (X3 l, Y3 l, Z3 l). The position of the global coordinate system and the position of the local coordinate system can be converted if the position of the center of gravity of the small triangular element and the direction of the surface normal vector are known.

Next, a specific method for determining the wall shear stress will be described. First, the velocity and pressure at each node are acquired in the global coordinate system from the output of the blood flow analysis unit 11 (i-CFD). Next, the triangular element for which the wall shear stress vector is to be obtained is designated. A local coordinate system is set for the triangular element. In the local coordinate system, the position G where the wall shear stress vector is to be calculated is determined (usually, the distance from the wall is made constant for each triangular element, for example, a point entering 0.1 mm from the wall). The flow velocity at this position G is 0 because it is on the wall surface as shown in FIG.

If the flow velocity U at a position that is separated from the position G of the wall surface in a normal direction (Z direction of the local coordinate system) by a sufficiently small distance t with respect to the boundary layer thickness of the flow is Ut, Is approximately proportional to the distance n from G,
Un = n · dUt / dZ
It is represented by

The force against the point of the distance n at this speed and the force required to fix the lower surface are equal to each other from the law of action and reaction, and both are proportional to the speed Ut and inversely proportional to the distance Z. Therefore, the force τ per unit area at the point G where the fluid is in contact is as follows.
τ = μ · dUt / dZ

That is, the wall shear stress vector is obtained by calculating the rate of change in the normal direction of the velocity vector parallel to the minute element and multiplying it by the viscosity coefficient of the fluid. There are several methods for calculating the normal direction change rate of the parallel direction velocity vector with respect to the minute element. For example, the speed at each candidate point can be obtained by installing a plurality of candidate points on the Zl axis and interpolating the speed vector from the surrounding speed vector group. In this case, since the distance to the candidate point is different for each ambient velocity vector, interpolation is performed by setting a weight function for the distance. Since the ambient velocity vector is described in the global coordinate system, the velocity component in the plane parallel direction at each candidate point is calculated by coordinate-transforming the interpolated velocity vector into the local coordinate system. Later, when calculating the rate of change in the normal direction, it may be calculated as a primary approximation using one candidate point near the wall, or a polynomial approximation is performed using a plurality of candidate points near the wall, Thereafter, a higher-order differentiation process of mathematical differentiation may be performed.

When this is to be obtained from the velocity U (Xg, Yg, Zg) of the global coordinate system, the velocity Ut at the distance t is decomposed into the local coordinate system (Xl, Yl, Zl), and the respective local coordinate axes. Τ = μ · dUt / dZ can be solved for (Xl, Yl) (Z-axis element is 0) that is parallel to the wall surface.

That is,
τ (Xl) = μ · dUt (Xl) / dZ
τ (Yl) = μ · dUt (Yl) / dZ
Will be calculated.
A vector value τ (Xl, Yl) obtained by combining the local coordinate axes becomes a wall shear stress vector. Accordingly, the wall shear stress vector is a vector having an x-direction component and a y-direction component with respect to the surface within the surface in contact with the blood vessel wall.

FIG. 8 is a diagram in which the shear stress vector along the blood vessel wall thus obtained is superimposed on the three-dimensional shape model.

The force acting on the blood vessel wall acts as a pressure P not only in the direction along the blood vessel wall but also in the direction of collision with the blood vessel wall. This pressure is obtained by applying the pressure at the point G obtained in the global coordinate system to the local coordinate system. It can be obtained as the pressure value in the Zl-axis direction when converted. FIG. 9 shows the pressure values acting on the wall surface in a superimposed manner on FIG. The lighter the color, the higher the pressure.

In this way, the vector calculation unit 121 obtains the wall shear stress and the vector obtained for each polygon.

(Randomness calculator)
Next, the randomness calculation unit 122 obtains the randomness as an index obtained by quantifying the shape of the wall shear stress vector group in each mesh. This randomness is a numerical index indicating the degree of whether or not the wall shear stress vector of a certain mesh is aligned in the same direction as compared with the surrounding wall shear stress vector group. That is, each angle formed between the wall shear stress vector of a mesh for which the degree of randomness is obtained (hereinafter referred to as “target mesh”) and the wall shear stress vector of each mesh adjacent to the target mesh. The degree of randomness is obtained by calculating θ.

FIG. 10 shows the relationship between the shear stress vector in the microelement G (approximate to a point for explanation) used in the system of this embodiment and the shear stress vector in the eight microelements surrounding the element G in a grid pattern. It is a thing. In this example, it is only necessary to extract not the magnitude of the shear stress but only the direction, so that the wall shear stress vector is handled as a unit vector. Strictly speaking, each microelement is in a three-dimensional configuration, but adjacent elements are sufficiently close to each other and are handled in two dimensions. That is, the processing is performed in such a way that each wall shear stress vector is projected onto a two-dimensional plane. FIG. 10 shows a state in which the minute element G and surrounding minute elements are mapped onto a two-dimensional orthogonal coordinate system.

In this embodiment, the form of the wall shear stress vector group is quantified by calculating `` divergence (div) '' and `` rotation (rotation) '' by the vector analysis with respect to the target mesh. A component display at a point G (x, y) obtained by mapping a vector field τ (shear stress vector) of a mesh surrounding a space to the two-dimensional orthogonal coordinate system (x, y) is represented by the following expression.
τ (G) = (τx (x, y), τy (x, y))
It is expressed.

At this time, “scalar field divτ” called “divergence of vector field τ” is defined by the following equation.
divτ = ∂τx / ∂x + ∂τy / ∂y

Similarly, a “scalar field rotτ” called “rotation of vector field τ” is defined by the following equation.
rotτ = ∂τy / ∂x－∂τx / ∂y

(Determination part)
FIG. 11 shows the relationship between the form of the wall shear stress vector group and the values of the “divergence (div)” and “rotation (rot)”. The discriminating unit 123 categorizes the wall shear stress vector group into 1) parallel type, 2) confluence type, 3) rotation type, and 4) collision type.

(Div, rot) = (0, 0) for parallel type, (div, rot) = (negative value, 0) for merge type, (div, rot) = (0, positive or negative value for rotation type ) In the collision type, (div, rot) = (positive value, 0). In the merge type and the collision type, the degree can be quantified according to the increase / decrease of the value of div. In other words, in the case of the confluence type, if the negative value increases in the negative direction, the degree of confluence increases. In the case of the collision type, if the positive value increases in the positive direction, the divergence (collision) occurs. The degree will increase. In the rotation type, positive and negative values appear depending on the rotation direction, but the degree of rotation can be quantified by the magnitude of the absolute value. If the randomness is defined as a vector quantity D = (div, rot), the magnitude can be used as the randomness, and the smaller the randomness, the more the wall shear stress vector of the target mesh, This means that the orientation is aligned with the shear stress vector (parallel type).

Fig. 12 shows the div and rot values mapped. That is, in this figure, the randomness (div, rot) is obtained for a typical example of the shear stress vector. Here, a typical example is an ideal pattern that can be described mathematically, not experimental data. As described above, since the divergence / rotation is calculated using the shear stress vector as a unit vector of size 1, the degree of randomness has already been standardized, which enables comparison between patients. That is, according to this embodiment, the degree of randomness can be obtained as an index that can be evaluated as an absolute value.

In this embodiment, as a degree of randomness, the pressure of the target mesh is combined as a weighting factor so that the damage determination to the blood vessel given when the blood flow collides with the blood vessel wall is performed with higher accuracy. In this embodiment, even when pressure is used, a standardized pressure, that is, a pressure index is used. In this embodiment, a value obtained by dividing each pressure by the average pressure is used as the pressure index after being calculated (multiplied in this example).

As a result, for example, when a collision-type shear stress vector group is formed by collision of blood flow, and when a mainstream flow collides, a local increase in wall pressure can be confirmed. When the secondary flow separated from the main stream collides, the wall pressure does not increase. In such a case, the combination of the shape of the shear stress vector group and the pressure is effective in improving the accuracy, particularly in predicting the thinned portion of the cerebral aneurysm. That is, there are a plurality of ways of indexing the pressure, and a method of superimposing the pressure on the randomness calculated from the shear stress vector may be a multiplication or a power law, or may be a plurality.

That is, the determination unit 123 determines the state of each mesh from the randomness value of each mesh obtained by the randomness calculation unit 122. The wall shear stress vector states here include a “parallel state” that is parallel to the surrounding wall shear stress vector, a “merging state” that extends in a direction approaching the surrounding wall shear stress vector, and a surrounding wall surface. It can be defined as a “rotation state” that rotates with the shear stress vector and a “collision state” in which the direction is radial with respect to the surrounding wall shear stress vector.

(Blood flow malignancy calculator)
The blood flow malignancy calculated by the blood flow malignancy calculation unit 13 is based on the form evaluation of the wall shear stress vector performed by the determination unit 123. That is, the form of the wall shear stress vector is divided into the above states and indexed and classified. It should be noted that a coefficient and a multiplier can be added when designing the index. The coefficient and multiplier may be information derived from the wall shear stress or wall pressure, and are, for example, numerical values of the time instability of the wall shear stress and the degree of unevenness of the wall pressure. Note that, as shown in FIG. 12, the merging state may be divided and identified as “low merging type” or “high merging type” according to the threshold according to the magnitude of the curing risk. The same applies to the rotation state and the collision state. The above is the content of the process I in FIG.

(Vascular expansion response measurement unit)
The vasodilation reaction analysis unit 14 for examining the endothelial cell function processes the result of measuring the change in the vascular diameter of the brachial artery using ultrasound. Specifically, the blood vessel diameter at rest and the change in the brachial artery blood vessel diameter after a predetermined time of blood transfusion are measured in real time by an ultrasonic image, and the result is output to the blood vessel vulnerability calculation unit 15. .

(Vessel fragility calculator)
The evaluation object in vascular vulnerability is a vascular endothelial cell. It is known that vascular endothelial cells regulate physiological functions by sensing wall shear stress of blood flow. This degree of expansion is known to represent endothelial cell function. In view of this, the vascular vulnerability calculation unit 15 in the present embodiment calculates the vascular vulnerability by measuring the change in the diameter of the artery after release of blood transfusion in real time as described above. Specifically, the vascular vulnerability degree% FMD is calculated by the following formula.
% FMD = (Dmax－Drest) / Drest

Here, Drest indicates the diameter of the blood vessel at rest, and Dmax indicates the maximum blood vessel diameter after the blood is released. The stronger the degree of vasodilation, the more the endothelial cell function is maintained. The above is the detail of the process II.

(Risk calculation department)
In step III, the vascular lesion onset / growth risk calculator calculates the vascular lesion onset / growth risk based on the blood flow malignancy and vascular vulnerability calculated in steps I and II. The risk may be output as having a risk when the ratio between the blood flow malignancy and the vascular vulnerability exceeds a threshold. For example, in the case of an intracranial cerebral artery, the risk of developing a vascular lesion / growth risk is determined and digitized based on the following. That is, when it is determined that the blood flow malignancy is a rotation type or a confluence type and the vascular vulnerability is a high value, it is determined that there is an onset / growth risk of arteriosclerosis, and the risk value is output. Further, when the blood flow malignancy in the cerebral blood vessel is a collision type and the vascular fragility is determined to be high, it is determined that there is a risk of onset / growth of cerebral aneurysm, and the risk value is output. The result output unit 17 visualizes the result, transmits it to the input / output unit 40, and displays it on a display (not shown) or the like.

In addition, the configuration of each part of the apparatus according to the present invention is not limited to the illustrated configuration example, and various modifications can be made as long as substantially the same operation is achieved.

Since the present invention predicts and outputs the onset and growth risk of vascular lesions by computer simulation, including those related to devices, methods, etc., it does not fall under the so-called medical practice or treatment method and is highly industrial With the availability of

Claims

An input unit for inputting information about a medical image including a vascular region to be analyzed and an endothelial cell function;
A computer that obtains blood vessel shape data from medical data input from the input unit and performs numerical fluid analysis to obtain a blood flow attribute including a pressure field and a velocity field;
A computer determines a blood flow property from a wall shear stress vector based on a blood flow attribute acquired by the blood flow analysis unit, and calculates a blood flow malignancy by quantifying the blood flow malignancy,
A blood vessel vulnerability calculating unit for obtaining a blood vessel vulnerability from information on the endothelial cell function input from the input unit;
A risk calculation unit that calculates a risk value for the onset or growth of a vascular lesion from a blood flow malignancy obtained by the blood flow malignancy calculation unit and a blood vessel vulnerability obtained by the blood vessel vulnerability calculation unit When,
A blood vessel lesion onset / growth prediction apparatus, wherein the computer has an output unit for outputting the calculated risk value.
The apparatus of claim 1.
The blood flow analysis unit is an apparatus that extracts a blood vessel shape from the medical image, generates a calculation grid, and obtains a pressure field and a flow field while considering fluid physical properties and boundary conditions.
The apparatus of claim 1.
The blood flow analysis unit receives a blood vessel shape, blood physical properties, boundary conditions and calculation conditions, and outputs a four-dimensional velocity field and pressure field including time.
The apparatus of claim 1.
The blood flow property is determined by determining a wall shear stress vector at each position of the blood vessel wall surface of the blood vessel part to be analyzed from the state amount of blood flow obtained by the blood flow analysis unit, and determining the wall surface at a specific wall surface position. A device that obtains the relative relationship between the direction of the shear stress vector and the direction of the wall surface shear stress vector at the surrounding wall surface position, determines the property of the blood flow at the wall surface position from the form, and outputs the determination result.
The apparatus of claim 4.
The blood flow characteristics are discriminated by indexing into “parallel”, “rotation”, “confluence”, and “collision”.
The apparatus of claim 5.
The blood flow property determination unit
From the relative angular relationship between the wall shear stress vector τ at the specific wall position and a plurality of wall shear stress vectors at the surrounding wall positions, the rotation rotτ and the divergence divτ, which are scalar quantities of the vector field τ, are obtained, and their values are obtained. By comparing with the threshold as the degree of randomness, it is determined whether it is in the above "parallel", "confluence", "rotation", "divergence",
When the value of the rotation rotτ of the randomness is a negative value or a positive value outside a predetermined threshold range, it is determined as “rotation”,
When the value of the divergence divτ of the messiness is a negative value outside a predetermined threshold range, it is determined as “join”,
When the value of the divergence divτ of the randomness is a positive value outside a predetermined threshold range, it is determined as “collision”,
The apparatus is characterized in that it is determined as “parallel” when both the value of the rotation rotτ and the value of the divergence divτ are within a predetermined threshold.
The apparatus of claim 1.
The apparatus is characterized in that the information on the endothelial cell function is obtained by calculating a diameter change from a blood vessel diameter measurement value at rest and at the time of release of blood transfusion.
The apparatus of claim 1.
The risk calculation unit determines that the blood flow property is “rotation” or “confluence” and the blood vessel vulnerability level is high, and the blood flow property is “collision”. A device that determines that there is a risk of onset / growth of lesions when the degree is determined to be high.
The apparatus of claim 8.
If the blood vessel part to be analyzed is a cerebral artery, the blood flow property determination is “rotation”, “confluence”, and the vascular vulnerability is determined to be high, the risk of onset / growth of arteriosclerosis An apparatus that determines that there is a risk of onset / growth of a cerebral aneurysm when it is determined that the blood flow property is “collision” and the vascular vulnerability is determined to be high.
The apparatus of claim 1.
An apparatus for determining that there is a risk when a ratio between the blood flow malignancy and the vascular vulnerability exceeds a threshold value.
A method performed by a computer,
An input step in which a computer inputs information about a medical image including an analysis target blood vessel region and endothelial cell function;
A computer obtains blood vessel shape data from the medical data input in the input step and executes a numerical fluid analysis to obtain a blood flow attribute including a pressure field and a velocity field;
A blood flow malignancy calculation step for determining a blood flow malignancy by quantifying the blood flow malignancy by determining a property of the blood flow from the wall shear stress vector based on the blood flow attribute obtained in the blood flow analysis step;
A blood vessel vulnerability calculating step for obtaining a blood vessel vulnerability from information on the endothelial cell function input in the input step;
Risk calculation in which a computer calculates a risk value for the onset or growth of a vascular lesion from the blood flow malignancy obtained in the blood flow malignancy calculation step and the blood vessel vulnerability obtained in the blood vessel vulnerability calculation step Process,
A method for predicting vascular lesion onset / growth, wherein the computer has an output step of outputting the calculated risk value.
A software program executed by a computer, the following steps:
An input step in which a computer inputs information about a medical image including an analysis target blood vessel region and endothelial cell function;
A computer obtains blood vessel shape data from the medical data input in the input step and executes a numerical fluid analysis to obtain a blood flow attribute including a pressure field and a velocity field;
A blood flow malignancy calculation step for determining a blood flow malignancy by quantifying the blood flow malignancy by determining a property of the blood flow from the wall shear stress vector based on the blood flow attribute obtained in the blood flow analysis step;
A blood vessel vulnerability calculating step for obtaining a blood vessel vulnerability from information on the endothelial cell function input in the input step;
Risk calculation in which a computer calculates a risk value for the onset or growth of a vascular lesion from the blood flow malignancy obtained in the blood flow malignancy calculation step and the blood vessel vulnerability obtained in the blood vessel vulnerability calculation step Process,
A software program comprising: an instruction for causing a computer to execute an output step of outputting the calculated risk value.