JPH08289877A - Simulation method for excitation propagation process of tissue and intra-tissue electromagnetic phenomenon diagnostic device - Google Patents

Abstract

PURPOSE: To provide a method capable of easily expressing the anisotropy of conductivity and an excitation transmission speed and performing simulation in a short time in the simulation of the excitation propagation of a heart and a brain, etc. CONSTITUTION: By a heart model constituted by a network, the excitation propagation process is simulated by using a Dijkstra method and a reverse problem for estimating electromagnetic activity inside a tissue from the potential and magnetic field of a body surface is solved.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for simulating an excitement propagation process in the brain (nerve system) or the heart, and a diagnostic device using this method.

[0002]

2. Description of the Related Art Potential distribution (magnetic field distribution) measured on the body surface
), The inverse electrocardiogram problem (inverse electrocardiogram problem), which estimates the electrical phenomena in the heart (intracardiac electromotive force distribution), is an estimation of a two-dimensional distribution to a three-dimensional distribution, and the solution cannot be essentially determined.

Therefore, in the field of the inverse problem of the electrocardiogram having a similar structure, the intracardiac electromotive force is modeled by a single or a plurality of moving dipoles, and the epicardial potential and the time of arrival of excitation of the endocardium and epicardium are estimated. Is being attempted. Furthermore, the number of parameters that can be stably estimated from the body surface electrogram in the subsequent theoretical study is ten and the estimated epicardial potential resolution is 1 cm.
It was shown to be a degree.

In order to obtain a high resolution while maintaining the stability of the solution, from the viewpoint that it is necessary to actively introduce the electrophysiological law condition regarding electromotive force, the order of excitation propagation in the heart is calculated. It has been considered for a long time to simulate with and to use the result as a constraint condition. A cardiac excitation propagation model is constructed in a computer, and an electrocardiogram is derived from the simulation results. The heart model is modified so that the difference from the measured electrocardiogram is minimized, and the lesion site and size are estimated. It used to be done manually, but nowadays it is not impossible due to advances in computational capabilities. It is said that the electrocardiogram is less affected by the conductivity inhomogeneity in the lungs and blood than the electrocardiogram, and it is expected that the use of the magnetocardiogram will make the inverse problem based on the simulation of excitement propagation process a reality.

Conventionally, in the simulation of the excitement propagation process, the myocardium is divided into thousands to tens of thousands of units, and the excitement is propagated between adjacent units at each step, and steps are conducted until the excitement of the entire heart is propagated. A unit division model (cell division model) that repeats is known.

Myocardial fibers run approximately parallel to the ventricular wall, and excitement propagates fast in the direction along the myocardial fibers and slow in the orthogonal direction. This is called anisotropy of excitatory conduction.In the above-mentioned unit division model, an unnatural shape of the excitatory wavefront that depends on the regular arrangement of units is obtained. Expression of sex is not easy algorithmically.

Therefore, a tetrahedral division model has been proposed in which the heart is composed of a large number of tetrahedra, and parameters such as excitatory conduction characteristics are specified at each vertex. In the tetrahedron, linear interpolation is performed so that the characteristics change continuously. The excitation wavefront is divided into triangles, and each vertex is moved at a certain time step to simulate the propagation of the excitation wavefront. By expressing the above parameters with a tensor, it is possible to express anisotropy of excitatory conduction or conductivity. The excitation propagation wavefront is represented by a triangular patch, and the entire excitation propagation process is represented by advancing the apex of the triangle by time steps based on the excitation conduction velocity at that point.

With this method, the anisotropy of the excitation conduction velocity and the conductivity can be expressed in a natural form, but it is difficult to process when the excitation wave front collides, or a part of the triangle is the inner and outer walls of the heart. Since there is a problem in processing when reaching, the triangle must be made fine in order to perform accurate simulation, the amount of calculation becomes enormous, and one excitement propagation process simulation takes about 30 minutes in some cases. .

[0009]

As described above, in the unit division model conventionally proposed as a simulation of excitation propagation of the heart, there is a problem that it is not easy to express anisotropy of excitation conduction velocity or conductivity. there were.

Further, in the tetrahedral segmentation model, the anisotropy of the excitation conduction velocity and the conductivity can be expressed in a natural form, but the amount of calculation becomes enormous and it takes a long time to simulate one excitation propagation. There was a problem of passing.

Further, although it has been pointed out that the conventional method of utilizing the excitement propagation process simulation for improving the stability of the inverse problem solving method has been pointed out, no specific method has been shown. This is because the inverse problem solving method using the simulation of the excitement propagation process has a problem that a potentially enormous calculation time is required and a practical method has not been found.

Therefore, according to the present invention, it is possible to easily express the anisotropy of the excitation conduction velocity and the conductivity in the simulation of the excitation propagation of the tissues such as the heart, the brain and the nervous system, and the simulation can be performed in a short time. An object of the present invention is to provide a method of simulating the excitement propagation process of a tissue and a device for diagnosing an electromagnetic phenomenon in a tissue using this method.

[0013]

The present invention models an organization by a network composed of a plurality of nodes and an edge connecting each of these nodes to a neighboring node, and it is assumed that excitement propagates through the network. Then, the excitement propagation time is set for each side of the network, the excitement arrival time at each node is calculated by the Dijkstra method, and the excitement propagation process of the tissue is simulated.

In another invention, an organization is modeled by a plurality of nodes and a network connected from each node to a neighboring node, and it is assumed that excitement propagates through the network, and for each side of the network. Regarding the simulation means for simulating the excitement propagation process of the tissue by setting the excitement propagation time, calculating the excitement arrival time at each node by the Dijkstra method, and the shape of the tissue, the excitatory conduction characteristic or the electrical property in the simulation Parameter setting means for setting various parameters, electromagnetic field map creating means for creating at least one of a tissue electrogram or a magnetic field map based on the result simulated by the simulation means, and an electrogram created by this electromagnetic field diagram creating means. Or at least one of the magnetic field diagram and the actually measured electrogram or magnetic field. By comparing the figures, it uses an optimization technique that provided and optimization means for optimizing the various parameters set by the parameter setting means.

[0015]

In the invention having such a structure, the excitation wavefront can be calculated in a short time by the Dijkstra method, assuming that the nodes of the network arrive at the same time. In addition, by setting the shape of the network and the time required for the propagation of excitation on each side of the network, the anisotropic property of the excitation conduction velocity and the conductivity is expressed in a natural form.

According to another aspect of the invention, various parameters relating to the shape or electrical properties of the tissue in the simulation are set by the parameter setting means, and the excitement propagation process of the tissue is simulated by the simulation means.

Based on the result of this simulation, at least one of the electrogram and the magnetic field diagram is created by the electromagnetic field map creating means. For example, taking the example of the heart as the tissue,
At least one of an electrocardiogram and a magnetocardiogram is created.

The optimizing means compares at least one of the generated electrogram or magnetic field diagram with the actually measured electrogram or magnetic field diagram, and optimizes various parameters by using the optimization method. .

[0019]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A first embodiment of the present invention will be described with reference to FIGS. In addition, in the first embodiment, the present invention is applied to an intra-tissue electromagnetic phenomenon diagnostic apparatus that simulates an excitation propagation process of the heart as a tissue, and the heart model is a tetrahedral segmentation model. Build as a foundation.

1 (a) and 1 (b) are diagrams showing the right ventricle and the left ventricle of the heart model of the tetrahedral segmentation model, and FIG. 2 is a diagram showing the entire ventricle of the heart model. . In this heart model, a ventricle is composed of 501 tetrahedra, and an excitation conduction velocity tensor, a conductivity tensor, an action potential amplitude, etc. are set at each vertex of each tetrahedron.

The excitatory conduction velocity tensor is for expressing the anisotropy of excitatory conduction in which excitatory conduction is fast in the direction along the myocardial fiber and slow in the orthogonal direction.
When excitation starts from one point of the homogeneous myocardium of the excitation conduction velocity tensor S, the excitation conduction velocity v in the direction of the vector t from that point has an ellipsoidal shape given by the following equation (1).

[0022]

[Equation 1]

The T with the upper right of the vector t is the transposed matrix of the vector t. The conductivity tensor is often used to represent the anisotropy of conductivity, and the conductivity tensor is a second-order symmetric tensor, which is expressed as a 3 × 3 symmetric matrix.

In the heart model of this embodiment, a virtual network is assumed in the myocardium of the heart, and it is approximated that excitement propagates only on the side connecting the nodes of the network. Therefore, in the above-described heart model of the tetrahedral model, the vertices of the tetrahedron can be considered as the nodes of the network and the sides of the tetrahedron can be considered as the edges of the network. However, in this case, since the network tension is rough, it is not possible to perform the excitement propagation process simulation with high accuracy.

Therefore, the network is further subdivided as shown in FIG. Note that FIG. 3 is a two-dimensional representation of the network for simplification, and a thick line shows one original tetrahedron. The subdivision is performed by providing a new node that divides each side of the tetrahedron into N equal parts and connecting each vertex of the original tetrahedron to each new node by the new side.

The network of this embodiment is completed by calculating the time required for excitement to propagate between nodes on each side, that is, the excitement propagation required time r for all sides. The excitement propagation required time r between both nodes of the side is calculated by the following equation (2), for example, when the excitement conduction velocity is taken into consideration as a parameter.

[0027]

[Equation 2]

Here, X1 and X2 are position vectors of both nodes, S is the excitation conduction velocity tensor described above, and s is an integration variable, which is an integration with a path connecting the nodes connecting the nodes. The excitement conduction velocity at the side of the network at the site equivalent to Purkinje fibers is set to 250 cm / s here based on physiological knowledge.

FIG. 4 is a block diagram showing a circuit configuration of a main part of a device for diagnosing an electromagnetic phenomenon in a tissue to which the present invention is applied.
1 is a CPU (central process
g unit). A ROM (read only memory) 2 in which program data of processing performed by the CPU 1 is stored, a RAM (random access memory) 3 in which areas of various memories used when the CPU 1 performs processing are formed, a communication line and an external device. I / O connected to (for example, printer)
The port 4 is connected to the CP via the system bus 5, respectively.
It is connected to U1.

Further, the CPU 1 controls the transmission of data to and from the storage device interface 7 for controlling the transmission of data to the storage device 6 such as a floppy disk device or a hard disk device and the keyboard 8 via the system bus 5. It is connected to a keyboard interface 9 and a display controller 11 which controls a display device 10 for displaying various data such as an electrocardiogram and a magnetocardiogram.

In the first embodiment having such a structure,
The excitement propagation process simulation is performed as in the flow of the excitement propagation process simulation process shown in FIG. First,
As the processing of step 1 (ST1), as described above,
Network formation processing is performed to form a network subdivided from the tetrahedral division model according to parameters such as the position, orientation, size, and shape of the heart. Step 2 (S
As a process of T2), for each side of the formed network, the excitation propagation required time between each node is calculated and set by the above-mentioned formula (2), for example.

As the processing of step 3 (ST3), several nodes corresponding to the earliest excitable site of the left ventricle (site that is considered to be the starting point of excitement) are selected, and the excitement start time is used as a physiological finding. Set based on For the right ventricle, select several similar nodes and set the excitation start time. The excitement start time of each of these nodes is 0, for example.
The time is from ms to 15 ms.

As the processing of step 4 (ST4), the excitement arrival times of all other nodes where excitement propagates in the network are calculated using the Dijkstra method. The Dijkstra method is an algorithm for solving the shortest path problem known in graph theory, and is a method for finding the shortest distance between arbitrary two points in a network in which the distance between adjacent nodes is designated. In step 4, the Dijkstra method is used to determine the excitement arrival time of each node of the excitement propagation network by replacing the distance between the two points with the excitement propagation required time.

Calculation of the excitement arrival time using the Dijkstra method is performed as described below. Here, for simplicity, an example of the two-dimensional network shown in FIG. 6 will be described. N (7 in FIG. 6) nodes are connected to ni (i = 1,
, N), and M nodes connecting these nodes (13 in FIG. 6)
For the sides of this book, let the nodes at both ends of each side be ne and nf, and the excitement propagation time thereof be ref.

The nodes selected as the initial excitement nodes are provisionally set such that the given initial excitement time is the excitement start time, and the excitement start times of the other nodes are provisionally set to very large values. deep.

With this as an initial state, the Dijkstra algorithm is executed. First, of all the nodes, the node na having the earliest start time of excitement is selected. Then node n
The excitement start times of all nodes adjacent to a and having an excitement start time later than this node na are calculated. Specifically, if the adjacent excursion start time of the adjacent node nx is
If the excitement arrival time of the excitement start time of a plus the excitement propagation time rax from the node na to the node nx is early (previous), the excitement start time of the node nx is calculated by the new excitement arrival time calculated. Make replacement modifications.

Next, of all the nodes having the excitation start time later than the node na, the node having the earliest excitation start time is selected. This node is nb here. Similarly, the excitement start times of all the nodes adjacent to this node nb and having the excitement start time later than this node nb are corrected by the method described above. By repeating such processing until no nodes having a slow nx are found, the excitement start times of all the nodes can be determined.

As the Dijkstra method, various variations can be considered on the basis of the method described here, and all these various variations can be applied to the present invention. For example, there are a graph data structure that uses a heap and a graph data structure that uses a two-dimensional array, but it is preferable to select a data structure that is well-balanced in terms of the amount of calculation and the memory storage capacity. When the excitement arrival times of all other nodes are calculated in the process of step 4 described above, this excitement propagation process simulation process is ended.

The electromotive force distribution is calculated for each tetrahedron. If the excitement start time of each vertex of the tetrahedron obtained by the above-described excitement propagation simulation process is beyond the time t currently focused on, it is determined that an electromotive force exists in the tetrahedron. The magnitude of the electromotive force can be calculated from the excitation conduction wavefront, the cross-sectional area of the tetrahedron, the action potential amplitude, and the conductivity tensor. The same electromotive force distribution can be calculated for all the tetrahedra. The above-described calculation of the electromotive force distribution in the heart is the same method as in the conventional tetrahedral divided heart model.

The electrocardiogram shows the electromotive force of each tetrahedron as x, y, z.
The magnetocardiogram can be derived by summing each component, and the magnetocardiogram can be derived by summing the magnetic fields generated by the electromotive force of each tetrahedron outside the body.

As described above, according to the first embodiment, the excitement propagation process is calculated and simulated using the Dijkstra method by the heart model which approximates the excitement propagation path of the myocardium by the network based on the tetrahedral division model. Therefore, the anisotropy of excitement propagation can be expressed in a natural form without causing an unnatural excitation wavefront unlike the unit division model (cell division model). The excitement propagation process simulation can be performed with the same accuracy as. Moreover, compared with the tetrahedral division model, the time required for the simulation (calculation) can be significantly shortened.

A second embodiment of the present invention will be described with reference to FIGS. 7 and 9. In the second embodiment, the electromagnetic waves in the tissue for estimating the electrical phenomenon in the heart from the potential distribution (magnetic field distribution) on the body surface using the simulation of the excitation propagation process in the first embodiment described above. The present invention is applied to a phenomenon diagnostic device. The electromagnetic tissue diagnosing device of the second embodiment is basically the same as the circuit configuration of the main part (see FIG. 4) of the electromagnetic diagnosing device for tissue of the first embodiment described above. Is omitted.

FIG. 7 is a block diagram showing the functional arrangement of the essential parts of this in-tissue electromagnetic phenomenon diagnostic apparatus. The heart model setting unit 21 includes a heart position / direction / shape setting unit 21-1, a stimulation conduction system distribution setting unit 21-2, and an excitation conduction velocity distribution setting unit 21.
-3, a conductivity distribution setting unit 21-4, an action potential amplitude distribution setting unit 21-5 and the like. Each parameter is set in each of these setting sections.

The excitement propagation process simulation unit 22
The excitement propagation process simulation described in the first embodiment is executed based on various parameters set by the heart model setting unit 21. That is, based on the various parameters set by the heart model setting unit 21, the heart model is configured as a network consisting of nodes and sides, the excitement arrival time of each side is set, and the excitement start times of all nodes are set. Calculated using the Dijkstra method.

The potential / magnetic field calculation unit 23 calculates the intracardiac electromotive force distribution for each tetrahedron at the current time t based on the excitation start times of all nodes calculated by the excitation propagation process simulation unit 22. To do. Further, the electromotive force of each tetrahedron is summed up for each of the x, y, and z components to calculate the potential, and the magnetic field generated outside the body by the electromotive force of each tetrahedron is summed to calculate the magnetic field.

The electrocardiogram / magnetocardiogram preparation / display unit 24 prepares an electrocardiogram and a magnetocardiogram based on the electric potential and the magnetic field calculated by the electric potential / magnetic field calculation unit 23, and prepares the electrocardiogram and the magnetocardiogram by the operator. Displayed when there is a display command operation.

The optimization processing section 25 compares the electrocardiogram and magnetocardiogram created by the electrocardiogram / magnetocardiogram creation / display section 24 with the actually measured electrocardiogram and magnetocardiogram, and calculates the difference between them. The heart model setting unit 2 so that
The correction control of the set values of various parameters is performed for 1.

In the second embodiment having such a configuration, the CPU
1 executes the heart shape optimization process shown in FIG. First, as the processing of step 11 (ST11), standard normal parameters for parameters other than position, orientation, and shape of the heart, stimulation conduction system distribution, excitation conduction velocity distribution, conductivity distribution, action potential amplitude distribution Set the fixed value in the example.

As the processing of step 12 (ST12),
An appropriate initial value is set using the parameters of the position, orientation, and shape of the heart as variables. This heart position parameter is
The parameters of the orientation of the heart are set as the values of the coordinates x, y, z, and the relationship of the long axis of the heart to the body axis is set as the values of Euler angles θ, φ, ψ. The parameters of the shape of the heart are the required dimensions D, L, s, t, u, R, C as the geometrical shape as shown in FIG. 9 (a) and FIG. 9 (b).
Is set as the numerical value of. In the second embodiment, an example of a set of 13 parameter groups as the parameters of the shape of the heart will be described, but a parameter group of another set may be used as the parameters of the shape of the heart. is there. As the processing of step 13 (ST13),
Excitation propagation process simulation process described above (see Fig. 5)
)I do.

When this excitement propagation process simulation process is completed, as the process of step 14 (ST14),
An intracardiac electromotive force distribution is calculated from the result of the simulation processing (excitation start times of all nodes), and a potential distribution and a magnetic field distribution are calculated from the intracardiac electromotive force distribution.

As the processing of step 15 (ST15),
An electrocardiogram and a magnetic field map are created from the calculated potential distribution and magnetic field distribution. As the processing of step 16 (ST16), the created electrocardiogram and magnetocardiogram are compared with the actually measured electrocardiogram and magnetocardiogram, and the difference (or index indicating the difference) is calculated.

As the processing of step 17 (ST17),
Based on the calculated difference, it is determined whether or not the setting of the parameters of the position, orientation and shape of the heart is optimal. If it is determined that the setting of the parameters of the position, orientation, and shape of the heart is not optimal, the parameters of the position, orientation, and shape of the heart are corrected so that the difference becomes small in the process of step 18 (ST18). Then, the settings are reset, and the process returns to the above-described step 13 again. Further, when it is determined that the setting of the parameters of the position / direction / shape of the heart is optimum, the heart shape optimization process is ended.

The problem of estimating the optimal parameters of the position, orientation and shape of the heart by correcting and resetting the parameters of the position, orientation and shape of the heart is a generally known nonlinear optimization method. Newton's method, conjugate gradient method, steepest descent method, simplex method, simulated annealing method, etc. are currently proposed as efficient algorithms that can be solved by.

As described above, according to the second embodiment, the same effect as that of the first embodiment can be obtained, and further, since there is the instability of the estimation due to the individual difference in the conventional heart shape, it can be used as an inverse problem solving method. Although a practical method has not been found, it is possible to obtain optimum values for each parameter of the heart position, orientation, and shape by estimating the heart position, orientation, and shape, and the inverse problem A solution can be realized.

A third embodiment of the present invention will be described with reference to FIG. The second embodiment described above estimates the position, orientation, and shape of the heart, whereas the third embodiment differs in that the parameters of the stimulation conduction system distribution are estimated. Therefore, the circuit configuration of the main part (see FIG. 4) and the basic functional configuration of the apparatus for diagnosing the electromagnetic phenomenon in tissue according to the third embodiment (see FIG. 4)
(See FIG. 7) is basically the same as the intra-tissue electromagnetic phenomenon diagnostic apparatus of the above-described embodiment, and therefore its description is omitted here.

The CPU 1 executes the stimulation conduction system distribution optimization process shown in FIG. First, step 21 (ST21
) Processing, parameters other than the stimulation conduction system distribution parameters, heart position / direction / shape, excitation conduction velocity distribution, conductivity distribution, and action potential amplitude distribution parameters are set to standard fixed values for normal cases. To do.

As the processing of step 22 (ST22),
An appropriate initial value is set using the parameters of the stimulus conduction system distribution as variables. The parameters of the stimulus conduction system distribution are set as the distribution of Purkinje fibers, the initial excitation position, and the initial excitation time. The distribution of Purkinje fibers is represented by a table that enumerates the edges (or nodes) corresponding to the positions of Purkinje fibers among all the edges (or nodes) in the model network. The initial excitement position and the initial excitement time are similarly represented by a table. In the excitement propagation process simulation process described later, a special calculation process is performed to calculate the excitement propagation required time of the side (or the side between the nodes) corresponding to the table.

As the processing of step 23 (ST23),
Excitation propagation process simulation process described above (see Fig. 5)
)I do. When this excitement propagation process simulation process is completed, as a process of step 24 (ST24), an intracardiac electromotive force distribution is calculated from the result of the simulation process (excitation start time of all nodes), and from this intracardiac electromotive force distribution, Calculate potential distribution and magnetic field distribution.

As the processing of step 25 (ST25),
An electrocardiogram and a magnetic field map are created from the calculated potential distribution and magnetic field distribution. As the process of step 26 (ST26), the created electrocardiogram and magnetocardiogram are compared with the actually measured electrocardiogram and magnetocardiogram, and the difference (or index indicating the difference) is calculated.

As the processing of step 27 (ST27),
Based on the calculated difference, it is determined whether the parameter setting of the stimulus conduction system distribution is optimum. Here, if it is determined that the parameter setting of the stimulus conduction system distribution is not optimal, the parameter of the stimulus conduction system distribution is corrected and reset so that the difference becomes small as the process of step 28 (ST28). The processing is returned to the above-mentioned step 23 again. When it is determined that the parameter setting of the stimulus conduction system distribution is optimum, this stimulus conduction system distribution optimization process is ended.

As a method for correcting and resetting the parameters of the stimulus conduction system distribution, a combination optimization method can be applied. As an algorithm of the combination optimization method, a simulated annealing method and a genetic algorithm are currently proposed.

As described above, according to the third embodiment, the same effect as that of the first embodiment can be obtained, and further, since the conventional stimulus conduction system distribution is unstable in estimation due to individual difference, the inverse problem is caused. Although a practical method has not been found as a solution, the optimum value of each parameter of the stimulus conduction system distribution can be obtained by using the stimulus conduction system distribution as an estimation target,
An inverse problem solving method can be realized.

A fourth embodiment of the present invention will be described with reference to FIG. In the fourth embodiment, the excitation conduction velocity distribution is estimated. Therefore, regarding the circuit configuration (see FIG. 4) and the basic functional configuration (see FIG. 7) of the essential part of the tissue electromagnetic phenomenon diagnosing device of the fourth embodiment, the tissue electromagnetic phenomenon diagnosing device of the above-described embodiment is used. Because they are basically the same,
The description is omitted here.

The CPU 1 executes the excitatory conduction velocity distribution optimization process shown in FIG. First, step 31 (ST3
As the processing of 1), the fixed values of standard normal cases are set for each parameter other than the excitation conduction velocity distribution parameter, heart position / direction / shape, stimulation conduction system distribution, conductivity distribution, action potential amplitude distribution parameter. Set.

As the processing of step 32 (ST32),
An appropriate initial value is set with the parameter of the excitatory conduction velocity distribution as a variable. When the excitation conduction velocity distribution is displayed by the symmetrical 6 components of the excitation conduction velocity tensor set at all the vertices in the tetrahedral model, the number of parameters of this excitation conduction velocity distribution is the number of vertices × 6. If the directions of the three axes of the ellipsoid showing the excitatory conduction characteristic are fixed for each vertex and only the excitatory conduction velocity in each axial direction is the estimation target, the number of parameters is the number of vertices × 3. Further, assuming that the ellipsoid is a spheroid, only the excitation conduction velocities in the major axis direction and the minor axis direction are targets for estimation, and therefore the number of parameters is the number of vertices × 2. Furthermore, if the ratio of the major axis to the minor axis of the ellipsoid is fixed for each vertex, the number of parameters can be reduced to the number of vertices × 1.

The smaller the number of parameters, the more stable the excitation conduction velocity distribution can be estimated, but conversely the approximation accuracy will deteriorate. Therefore, it is preferable to select an appropriate number of parameters in consideration of the balance between estimation stability and approximation accuracy.

As the processing of step 33 (ST33),
Excitation propagation process simulation process described above (see Fig. 5)
)I do. When this excitement propagation process simulation process is completed, as a process of step 34 (ST34), an intracardiac electromotive force distribution is calculated from the result of the simulation process (excitation start time of all nodes), and from this intracardiac electromotive force distribution, Calculate potential distribution and magnetic field distribution.

As the processing of step 35 (ST35),
An electrocardiogram and a magnetic field map are created from the calculated potential distribution and magnetic field distribution. As the processing of step 36 (ST36), the created electrocardiogram and magnetocardiogram are compared with the actually measured electrocardiogram and magnetocardiogram, and the difference (or index indicating the difference) is calculated.

As the processing of step 37 (ST37),
Based on the calculated difference, it is determined whether the parameter setting of the excitation conduction velocity distribution is optimum. If it is determined that the parameter setting of the excitatory conduction velocity distribution is not optimal, the parameter of the excitatory conduction velocity distribution is corrected / reset so as to reduce the difference as the process of step 38 (ST38). The process returns to the above-mentioned step 33 again. When it is determined that the parameter setting of the excitation conduction velocity distribution is optimum, the excitation conduction velocity distribution optimizing process is ended.

As a method of correcting and resetting the parameters of the excitation conduction velocity distribution, the parameters of the excitation conduction velocity distribution are a plurality of continuous quantities, and this parameter and the measurement object quantity have a non-linear relationship. The non-linear optimization method described in the second embodiment can be applied.

As described above, according to the fourth embodiment, it is possible to obtain the same effect as that of the first embodiment described above, and further, by using the excitation conduction velocity distribution as an estimation target, each of the excitation conduction velocity distributions can be obtained. The optimum values of the parameters can be obtained, and the inverse problem solving method can be realized.

A fifth embodiment of the present invention will be described with reference to FIG. The fifth embodiment is directed to estimation of action potential amplitude distribution. Therefore, regarding the circuit configuration (see FIG. 4) and the basic functional configuration (see FIG. 7) of the essential parts of the tissue electromagnetic diagnosing device of the fifth embodiment, the tissue electromagnetic phenomenon diagnosing device of the above-described embodiment is used. Because they are basically the same,
The description is omitted here.

The CPU 1 executes the action potential amplitude distribution optimizing process shown in FIG. First, step 41 (ST4
As the processing of 1), the fixed values of standard normal cases are applied to the parameters other than the action potential amplitude distribution parameters, the position / direction / shape of the heart, the stimulation conduction system distribution, the excitation conduction velocity distribution, and the conductivity distribution parameters. Set.

As the processing of step 42 (ST42),
Excitation propagation process simulation process described above (see Fig. 5)
)I do. When this excitement propagation process simulation process is completed, R is set as the process of step 43 (ST43).
0 is set to the storage area i formed in AM3.

Next, in step 44 (ST44), it is determined whether the numerical value i set in the storage area i is equal to the total number N of vertices in the tetrahedral model. If it is determined that the numerical value i set in the storage area i is not equal to the total number N (i is less than N), the numerical value i set in the storage area i is changed as the process of step 45 (ST45). Then, +1 addition processing is performed, and step 4
6 (ST46), a unit action potential amplitude is set only at the vertex of the i-th registered tetrahedral model.

As the processing of step 47 (ST47),
Based on the result of the simulation process in step 42 (excitation start time of all nodes (vertices)),
An intracardiac electromotive force distribution is calculated in a state in which the action potential amplitude of that unit is set, and an electric potential distribution and a magnetic field distribution are calculated from this intracardiac electromotive force distribution.

As the processing of step 48 (ST48),
An electrocardiogram and a magnetic field map are created from the calculated potential distribution and magnetic field distribution. As the processing of step 49 (ST49), a vector bi in which M potentials (potentials of each electrode) or magnetic fields (values of each detection coil) are vertically arranged from the created electrocardiogram and magnetocardiogram is created, and this is created as a potential bi.・ Magnetic field operator B
= (B1, b2, ..., bN). Therefore, the operator B is an M × N matrix. When the setting of this bi is completed, the process returns to the above-described step 44.

When it is judged in the above-mentioned step 44 that the numerical value i set in the storage area i is equal to the total number N of vertices, the completed operator B = (b1, From b2, ..., bN), an equation for deriving the potential / magnetic field distribution from the action potential amplitude distribution is created. A vector f in which N action potential amplitude variables are vertically arranged.
And a vector b in which M electric potentials or magnetic fields derived from this vector f are vertically arranged, the above equation is given by
It is created as (3). That is, the vector f shows the action potential amplitude distribution, and the vector b shows the measured body surface potential distribution or magnetic field distribution.

[0079]

(Equation 3)

As the processing of step 51 (ST51),
An estimation matrix G (f = Gb is obtained by the operation processing known in the linear optimization method for the created equation.
) Is calculated. As a method of calculating the estimation matrix G, four examples thereof are shown by equations (4) to (7).

[0081]

[Equation 4]

In equation (4), with the upper right corner of the matrix B +
Is the Moore-Penrose generalized inverse matrix of operator B.
In the equation (5), t in the upper right of the matrix B is a transposed matrix of the operator B, λ is a regularization parameter, and the matrix I is a matrix whose diagonal elements are 1 and all other elements are 0.

In equation (6), the matrix P is a penalty matrix. In equation (7), the lower right Q of matrix B
Is obtained by the equation (9) when the singular value (eigenvalue) decomposition is performed on the operator B as shown in the equation (8). Note that R is called the rank of the matrix B, and Q <R.

As the processing of step 52 (ST52),
An equation f = Gb is created from the obtained estimation matrix G,
The optimized vector f is obtained, and the action potential amplitude distribution optimizing process is ended. As described above, according to the fifth embodiment, it is possible to obtain the same effect as that of the first embodiment described above, and further to obtain the optimum action potential amplitude distribution by using the action potential amplitude distribution as the estimation target. And the inverse problem solving method can be realized. further,
In the fifth embodiment, the linear optimization method can be applied by ignoring the relationship with the excitatory conduction velocity distribution and independently setting the action potential amplitude distribution as the estimation target. Compared to the optimization method and the combination optimization method,
The time required for the optimization process can be greatly reduced.

The optimization process for estimating the action potential amplitude distribution in the fifth embodiment can also be applied to the conductivity distribution optimization process. FIG. 13 shows the flow of the optimization process of the conductivity distribution. Therefore, the description of the conductivity distribution optimization processing is almost the same as the action potential amplitude distribution optimization processing, and therefore the description thereof is omitted. Note that g is a vector in which n conductivity variables are vertically arranged.

The second to fifth embodiments described above can be combined. That is, first, the electrocardiogram and the magnetocardiogram are measured at rest, and the heart shape optimization process of the second embodiment is performed on the resting electrocardiogram and magnetocardiogram to optimize the parameters of the position, orientation, and shape of the heart. Find the value. Next, the physical or mental stress is applied to measure the electrocardiogram and magnetocardiogram under load, and for the electrocardiogram and magnetocardiogram under load, the obtained position, orientation, and shape parameters of the heart are calculated. Using the optimum value as a fixed value,
The stimulation conduction system distribution optimization processing of the third embodiment, the excitation conduction velocity distribution optimization processing of the fourth embodiment, or the action potential amplitude distribution optimization processing (conductivity distribution optimization processing) of the fifth embodiment is performed, Stimulation conduction system distribution, excitation conduction velocity distribution or action potential amplitude distribution (conductivity distribution) can be estimated.

The heart shape optimization process of the second embodiment is performed on the resting electrocardiogram and magnetocardiogram, and the stimulation conduction system distribution optimization process of the third embodiment is performed to determine the position of the heart. The optimum values of the orientation and shape parameters and the optimum values of the stimulation conduction system distribution parameters are obtained, and the obtained optimum values are set as fixed values, and the excitation conduction of the fourth embodiment is performed with respect to the electrocardiogram and the magnetocardiogram under load. Velocity distribution optimization processing or 5th
The action potential amplitude distribution optimization process (conductivity distribution optimization process) of the embodiment can be performed to estimate the excitatory conduction velocity distribution or the action potential amplitude distribution (conductivity distribution).

Contrary to the above, the optimum value of the parameter of the excitation conduction velocity distribution or action potential amplitude distribution (conductivity distribution) is obtained from the electrocardiogram and magnetocardiogram under load, and this optimum value is fixed. Alternatively, the position / direction / shape of the heart or the stimulus conduction system distribution can be estimated from the electrocardiogram and magnetocardiogram at rest.

As described above, first, the optimum values of the respective parameters are obtained by using the electrocardiogram and the magnetocardiogram at rest and the electrocardiogram and the magnetocardiogram under load, and the obtained optimum parameters are used as fixed values. By estimating the parameter, a highly reliable value can be used as a fixed value, and more accurate optimization processing can be performed.

A sixth embodiment of the present invention is shown in FIGS. 14 and 15.
Will be described with reference to. In the above-described second to fifth embodiments, the optimization process for estimating one kind of parameters (the parameters of the position, orientation, and shape of the heart, the parameters of the stimulation conduction distribution, etc.) has been described. On the other hand, in the sixth embodiment, an optimization process for estimating a combination of a plurality of types of parameters will be described. In addition, regarding the circuit configuration (see FIG. 4) and the basic functional configuration (see FIG. 7) of the essential parts of the electromagnetic tissue diagnostic apparatus for tissue of the sixth embodiment, the electromagnetic phenomenon diagnostic apparatus for tissue of the embodiment described above is used. Since they are basically the same, the description thereof is omitted here.

The CPU 1 executes the plural kinds of parameter optimization processing shown in FIG. First, step 61 (ST
61) As a process, a standard normal example is used for the parameters of the position / orientation / shape of the heart, the parameters of the stimulation conduction system distribution, the parameters other than the parameters of the action potential amplitude distribution, the parameters of the excitation conduction velocity distribution and the conductivity distribution. Set a fixed value for.

As the processing of step 62 (ST62),
An appropriate initial value is set using the parameters of the stimulus conduction system distribution as variables. As the processing of step 63 (ST63),
An initial value registered using the parameters of the position, orientation, and shape of the heart as variables or an appropriate initial value is set when there is no registered initial value.

As the processing of step 64 (ST64),
Excitation propagation process simulation process described above (see Fig. 5)
)I do. When this excitement propagation process simulation process is completed, as the process of step 65 (ST65), the action potential amplitude distribution estimation process, which will be described later, is completed. When this action potential amplitude estimation process is completed, the action is executed as step 66 (ST66). A vector f in which N action potential amplitudes obtained by the potential amplitude estimation process are vertically arranged is registered as an initial value.

As the processing of step 67 (ST67),
The position / orientation of the heart based on the difference between the vector b in which M potentials or magnetic fields calculated in the action potential amplitude estimation process are vertically arranged and the potential / magnetic field obtained from the actually measured electrocardiogram and magnetocardiogram. -Determine whether or not the shape parameter settings are optimal.

If it is determined that the setting of the parameters of the position, orientation and shape of the heart is not optimal, step 68 is executed.
As the process of (ST68), the parameters of the position, orientation, and shape of the heart are corrected and reset using the non-linear optimization method so that the difference becomes small, and the process returns to the process of step 64 described above. It has become.

If it is determined that the parameters of the position, orientation and shape of the heart are optimal, step 69 (S
As a process of T69), the parameters of the position, orientation and shape of the heart set at this time are registered as initial values.

Next, as the processing of step 70 (ST70), it is determined whether or not the setting of the parameters of the stimulus conduction system distribution is optimum based on the above difference. Here, when it is determined that the setting of the parameters of the stimulus conduction system distribution is not optimal, as a process of step (ST71), the optimization method of combining the parameters of the stimulus conduction system distribution is used so that the above difference becomes small. To correct and reconfigure, and again to step 63 above.
The process returns to (ST63). Further, when it is determined that the parameter setting of the stimulus conduction system distribution is optimum, the plural kinds of parameter optimization process is ended.

FIG. 15 is a diagram showing a flow of the action potential amplitude distribution estimation processing described above performed by the CPU 1. In addition,
This action potential amplitude distribution estimation processing is basically step 4 of the action potential amplitude distribution optimization processing (see FIG. 12) described above.
It is composed of the processing from 3 to step 52.

First, as the processing of step 81 (ST81), 0 is set in the storage area i formed in the RAM 3. Next, as the processing of step 82 (ST82), it is determined whether the numerical value i set in the storage area i is equal to the total number N of vertices in the tetrahedral model.

Here, the numerical value i set in the storage area i
Is not equal to the total number N (i is less than N), as a process of step 83 (ST83), an addition process of +1 is performed on the numerical value i set in the storage area i, and then step 84 ( As the processing of ST84), the unit action potential amplitude is set only at the vertex in the tetrahedron model registered as the i-th.

As the processing of step 85 (ST85),
Based on the result of the simulation processing in step 64 (excitation start time of all nodes (vertices)),
An intracardiac electromotive force distribution is calculated in a state in which the action potential amplitude of that unit is set, and an electric potential distribution and a magnetic field distribution are calculated from this intracardiac electromotive force distribution.

As the processing of step 86 (ST86),
An electrocardiogram and a magnetic field map are created from the calculated potential distribution and magnetic field distribution. As the processing of step 87 (ST87), a vector bi in which M potentials (potentials of each electrode) or magnetic fields (values of each detection coil) are vertically arranged is created from the created electrocardiogram and magnetocardiogram, and this is created as a potential bi.・ Magnetic field operator B
= (B1, b2, ..., bN). Therefore, the operator B is an M × N matrix. When the setting of this bi is completed, the process returns to the above-mentioned step 82.

When it is judged in the process of step 82 that the numerical value i set in the storage area i is equal to the total number N of vertices, the completed operator B = (b1, From b2, ..., bN, the equation (Equation (3
) Reference) is created.

As the processing of step 89 (ST89),
An estimation matrix G (f = Gb is obtained by the operation processing known in the linear optimization method for the created equation.
) Is calculated.

As the processing of step 90 (ST90),
An equation f = Gb is created from the obtained estimation matrix G,
The optimized vector f is obtained, and the action potential amplitude distribution estimation processing is ended. Like this
According to the embodiment, it is possible to obtain the same effect as that of the first embodiment described above, and further, by estimating a plurality of types of parameters in combination, it is possible to obtain individual differences such as the position, orientation, and shape of the heart. One parameter and another parameter can be estimated at the same time, and the destabilization of the solution of the inverse problem due to individual differences can be significantly reduced.

In the first to sixth embodiments described above, a tissue for solving an inverse problem of estimating an electromagnetic phenomenon in the heart from an electrocardiogram or a magnetocardiogram using a simulation of an excitation propagation process in the heart. Although the apparatus for diagnosing an internal electromagnetic phenomenon has been described, the present invention is not limited to the heart as a target tissue, but can be applied to, for example, brain and nervous system tissue.

In the brain model in the simulation of nerve excitement propagation process of the brain, nodes (representative points) are set at anatomically and physiologically defined sites. For example, for the cerebral cortex, it is set based on the cerebral cortex volume field by Brodmann and the subsequent subclassification. For areas where functional localization is more clear, such as the motor cortex, it is better to make detailed settings within the known range. Nodes (representative points) are similarly set for the basal ganglia, interbrain, and cerebellum. Edges connect these nodes to represent neural projections.

The direction of excitatory conduction is defined on each side, and the opposite direction is assumed not to be conducted. The time required for excitatory conduction between nodes is the time required for conducting nerve fibers plus the time required for excitatory conduction at synapses.

The brain model configured in this way (effective graph
), Set the initial excitement point, and use the Dijkstra method known in graph theory to determine the excitement start time (excitation arrival time) of all other nodes. It is assumed that the magnetic field is generated from each node and that it is generated outside the skull with a preset waveform from the obtained excitation start time.

At this time, not all the nerve conduction paths are always effective, and the conduction paths which are considered not to be related to the activity of the brain for a certain period are ignored or a sufficiently large value is set for the excitation conduction time. I'll do it.

The Dijkstra method for the brain model is almost the same as the method described in the first embodiment described above. However, as a node adjacent to a certain node na, among nodes directly connected by an edge, those The only difference is that only the nodes whose conduction direction is defined to be conducted to the nodes of are considered.

Using the simulation of the brain's excitement propagation process described above, the neural activity in the brain can be estimated from the comparison with the actually measured cerebral magnetic field. That is, first consider a time series of a series of magnetic field distributions corresponding to the activity of the brain during a certain period. An example is an auditory evoked magnetic field in which the response to an auditory stimulus is added and averaged in synchronization with the stimulus, but other evoked magnetic fields or spontaneous magnetic fields due to complicated combinations can be considered. The magnetic field distributions obtained from the above simulations and these measured magnetic field distributions (
(Time series) of the initial excitation point, the waveform of the magnetic field generated from each node, the time required for excitatory conduction at the synapse, and the time required for excitatory conduction at the nerve conduction path are effective parameters. Correct the set of conduction paths, etc. and estimate (optimize) these values. Also for this optimization, various generally known optimization algorithms can be used.

[0113]

As described above in detail, according to the present invention,
It is easy to express the anisotropy of excitatory conduction velocity and conductivity in the simulation of excitement propagation in tissues such as heart, brain and nervous system, and the simulation can be performed in a short time. Due to
Provided is a method for simulating a tissue excitement propagation process capable of solving an inverse problem of estimating electromagnetic activity in a tissue from an actually measured potential or magnetic field on the body surface, and an apparatus for diagnosing an electromagnetic phenomenon in a tissue using this method. .

[Brief description of drawings]

FIG. 1 is a diagram showing a right ventricle and a left ventricle of a heart model of a tetrahedral division model, which is a basis of a heart model constructed by an excitation propagation process simulation used in an apparatus for diagnosing electromagnetic phenomenon in tissue in one embodiment of the present invention. The figure which shows a ventricle.

FIG. 2 is a diagram showing an entire ventricle of a heart model of the tetrahedral segmentation model of the embodiment.

FIG. 3 is a diagram for explaining subdivision of constructing a heart model by a network from the heart model of the tetrahedral segmentation model of the same embodiment.

FIG. 4 is a block diagram showing a circuit configuration of a main part of the tissue electromagnetic phenomenon diagnosing device of the embodiment.

FIG. 5 is a diagram showing a flow of excitement propagation process simulation processing performed in the tissue electromagnetic phenomenon diagnosing device of the embodiment.

FIG. 6 is a diagram for explaining calculation of excitement arrival time by the Dijkstra method used in the excitement propagation process simulation process performed in the tissue electromagnetic phenomenon diagnostic apparatus of the embodiment.

FIG. 7 is a block diagram showing a functional configuration of essential parts of the tissue electromagnetic phenomenon diagnosing device of the embodiment.

FIG. 8 is a diagram showing a flow of heart shape optimization processing performed by the intra-tissue electromagnetic phenomenon diagnostic apparatus of the embodiment.

FIG. 9 is a view showing parameters of the position, orientation, and shape of the heart in the heart shape optimization process performed by the intra-tissue electromagnetic phenomenon diagnostic apparatus of the embodiment.

FIG. 10 is a diagram showing a flow of stimulation conduction system distribution optimization processing performed by the intra-tissue electromagnetic phenomenon diagnostic apparatus of the embodiment.

FIG. 11 is a diagram showing the flow of an excitation conduction velocity distribution optimization process performed by the intra-tissue electromagnetic phenomenon diagnostic apparatus of the embodiment.

FIG. 12 is a diagram showing a flow of action potential amplitude distribution optimizing processing performed by the intra-tissue electromagnetic phenomenon diagnosing device of the embodiment.

FIG. 13 is a diagram showing a flow of a conductivity optimizing process performed in the tissue electromagnetic phenomenon diagnosing apparatus of the embodiment.

FIG. 14 is a diagram showing the flow of a multi-type parameter optimization process performed by the tissue electromagnetic phenomenon diagnosing device of the embodiment.

FIG. 15 is a diagram showing the flow of action potential amplitude distribution estimation processing performed by the intra-tissue electromagnetic phenomenon diagnostic apparatus of the embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... CPU, 21 ... Heart model setting part, 22 ... Excitation propagation process simulation part, 23 ... Potential / magnetic field calculation part, 24 ... Electrocardiogram / magnetocardiogram preparation display part, 25 ... Optimization processing part.

Claims (15)

[Claims] 1. A Dijkstra method in which an organization is modeled by a network constituted by connecting a plurality of nodes and an edge from each of these nodes to a neighboring node and setting an excitement propagation time for each edge of the network. A method for simulating the excitement propagation process of an organization, wherein the excitement arrival time at each node is calculated to simulate the excitement propagation process of the organization. 2. An organization is modeled by a plurality of nodes and a network connected from each node to a neighboring node, an excitement propagation time is set for each side of the network, and an excitement arrival time at each node is set by the Dijkstra method. A simulation means for simulating the excitement propagation process of the tissue, and a parameter setting means for setting various parameters relating to the shape of the tissue, the excitatory conduction characteristic or the electrical property in the simulation by the simulation means, Electromagnetic field map creating means for creating at least one of the electrogram or magnetic field map of the tissue based on the result simulated by the simulation means, and actual measurement with at least one of the electrogram or magnetic field map created by this electromagnetic field map creating means Compared with the electrogram or magnetic field diagram, the difference between An apparatus for diagnosing an electromagnetic phenomenon in tissue, comprising: an optimizing means for optimizing various parameters set by the parameter setting means by using an optimizing method so as to be as small as possible. 3. The intra-tissue electromagnetic phenomenon diagnosing device according to claim 2, wherein parameters other than the parameters relating to the shape of the tissue set by the parameter setting means are fixed, and the optimizing means relates to the shape of the tissue. An apparatus for diagnosing electromagnetic phenomenon in tissue, characterized by optimizing parameters. 4. The intra-tissue electromagnetic phenomenon diagnosing device according to claim 2, wherein parameters other than the parameters of the stimulation conduction system distribution set by the parameter setting means are fixed, and the optimizing means sets the stimulation transmission system. An apparatus for diagnosing electromagnetic phenomenon in tissue, characterized by optimizing distribution parameters. 5. The apparatus for diagnosing electromagnetic phenomenon in tissue according to claim 2, wherein parameters other than those of the action potential amplitude distribution set by the previous parameter setting means are fixed, and the optimizing means sets the action potential amplitude. An apparatus for diagnosing electromagnetic phenomenon in tissue, characterized by optimizing distribution parameters. 6. The apparatus for diagnosing electromagnetic phenomenon in tissue according to claim 2, wherein parameters other than the parameters of the conductivity distribution set by the parameter setting means are fixed, and the optimizing means sets the conductivity distribution. An apparatus for diagnosing electromagnetic phenomenon in tissue, characterized by optimizing parameters. 7. The intra-tissue electromagnetic phenomenon diagnosing device according to claim 2, wherein the parameter setting means fixes a parameter other than a predetermined parameter to at least one of the electrogram and the magnetic field diagram of the tissue in the first state. Then, the optimizing means optimizes the predetermined parameter, and the value of the optimized predetermined parameter is set to the parameter setting means with respect to at least one of the electrogram or the magnetic field diagram of the tissue in the second state. Fixed by, the optimization means,
A device for diagnosing electromagnetic phenomenon in tissue, which is characterized by optimizing parameters other than predetermined parameters. 8. The intra-tissue electromagnetic phenomenon diagnosing device according to claim 7, wherein at least one of a potential diagram and a magnetic field diagram of the tissue at rest is other than the parameter relating to the shape of the tissue set by the parameter setting means. With the parameter fixed, the optimizing means optimizes the parameter relating to the shape of the tissue, and with respect to at least one of the electrogram or the magnetic field diagram of the tissue under load, the value of the parameter relating to the optimized tissue shape. Is fixed by the parameter setting means, and the optimizing means optimizes parameters other than the parameters relating to the shape of the tissue. 9. The intra-tissue electromagnetic phenomenon diagnosing device according to claim 7, wherein a parameter relating to the shape of the tissue or stimulation conduction set by the setting means with respect to at least one of the electrogram and the magnetic field diagram of the tissue at rest. By fixing parameters other than system distribution parameters, the optimizing means optimizes parameters relating to the shape of the tissue or parameters of the stimulus conduction system distribution, and with respect to at least one of the electrogram or magnetic field diagram of the tissue under load. , Fixing the value of the parameter related to the shape of the optimized tissue and the parameter of the stimulation conduction system distribution by the parameter setting means,
The apparatus for diagnosing electromagnetic phenomenon in tissue, wherein the optimizing means optimizes parameters other than the parameters relating to the shape of the tissue and the parameters of the stimulation conduction system distribution. 10. The apparatus for diagnosing electromagnetic phenomenon in tissue according to claim 2, wherein parameters other than a plurality of parameters among various parameters set by the parameter setting means are estimated and fixed, and the optimizing means sets the plurality of parameters. An apparatus for diagnosing electromagnetic phenomenon in tissue, which is characterized by simultaneously optimizing parameters. 11. The apparatus for diagnosing electromagnetic phenomenon in tissue according to claim 10, wherein a plurality of parameters simultaneously optimized by the optimizing means are parameters of excitation conduction velocity distribution, tissue shape parameters or stimulation conduction system distribution. And an electromagnetic phenomenon diagnosing device in a tissue. 12. The apparatus for diagnosing electromagnetic phenomenon in tissue according to claim 10, wherein a plurality of parameters simultaneously optimized by the optimizing means are parameters of action potential amplitude distribution, tissue shape parameters or stimulation conduction system distribution. And an electromagnetic phenomenon diagnosing device in a tissue. 13. The intra-tissue electromagnetic phenomenon diagnosing device according to claim 10, wherein a plurality of parameters simultaneously optimized by the optimizing means include a conductivity distribution parameter, a tissue shape parameter, and a stimulation conduction system distribution. A device for diagnosing an electromagnetic phenomenon in a tissue, which is one of the parameters. 14. The method of simulating the excitement propagation process of tissue according to claim 1, wherein the tissue is a heart, brain, or nervous system tissue. 15. The apparatus for diagnosing electromagnetic phenomenon in tissue according to claim 2, wherein the tissue is heart, brain or nervous system tissue.