JPH10276998A - Simulation system for excitation propagation process - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable to carry out excitation propagation simulation at a high speed, by equipping a calculation means of time required for excitation propagation to preliminarily calculated the time required for excitation propagation and a means to determine time reaching excitation. SOLUTION: In excitation propagation simulation of the cardiac chamber of a heart, a simulation device has a cardiac chamber model constituting means 21 to constitute the cardiac chamber as a geometric model and an excitation propagation network constituting means 22 to constitute the excitation propagation network comprising node points and branch lines against the cardiac chamber model and to calculate and set the time required for excitation propagation between both ends of each branch line. A calculation means 23 for the time required for excitation propagation preliminarily calculates and sets times required for excitation propagation between all the node points, and an initial value setting means 24-1 sets an initial excitation node points and an initial excitation time. Thus a calculation means 24 for the time for reaching excitation propagation calculates each excitation reaching time to all the node points based on the time required for excitation propagation, the initial excitation node point, and the initial excitation time.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an excitement propagation process which simulates the excitement propagation process of the tissue by modeling the tissue as a structure composed of a plurality of points, calculating the excitement propagation arrival time at each point. The present invention relates to a simulation device.

[0002]

2. Description of the Related Art Excitement propagation simulation, which simulates the process of excitement propagation in tissues such as the heart, brain, and nerves, is replaced with the shortest path problem of graph theory, and the Dijkstra method, which is an algorithm for solving the shortest path problem at high speed, is used. Then, a method of performing an excitement propagation simulation is disclosed in
No. 98791. In this method, the excitement is approximated as propagating through a network consisting of nodes and branches (edges) connected from these nodes to nearby nodes. In other words, the excitement propagation of an excitable tissue such as the heart is a branch line connecting multiple nodes to neighboring nodes.
(Excitement Propagation Path) The model is modeled by a network composed of (excitation propagation paths), and the time required for excitation to propagate between both ends is set for each branch of the network.

[0003] Considering the time required for excitement propagation set for each branch line to be the distance of each branch line, an excitement propagation simulation that simulates how the excitement starting from a certain node propagates from a certain node to another node. Can be replaced with the shortest path problem of finding the shortest distance to all nodes.

The above-mentioned Japanese Patent Application No. 7-98791 mentions an application to an inverse problem for estimating an electric phenomenon (structure) inside a tissue from a potential distribution (magnetic field distribution) measured on a tissue surface. , Parameters determined in advance in the model (excitation starting point position, excitation propagation velocity distribution, action potential amplitude distribution, etc.
) Using excitation propagation simulation and potential calculation,
Perform a magnetic field calculation, calculate the body surface potential map and the in vitro magnetic field distribution, and use the optimization method to optimize the parameters that minimize the difference from the actually measured body surface potential map and the in vitro magnetic field distribution An estimation method is shown.

[0005] When the excitation propagation simulation is applied to the inverse problem, the excitation propagation simulation is performed until the difference between the measured body surface potential diagram or the in vitro magnetic field distribution and the calculated body surface potential diagram or the in vitro magnetic field distribution becomes small. Many times (
For example, 100,000 times).

[0006]

In the conventional method, it is necessary to calculate the shortest path problem by the Dijkstra method each time the excitation propagation simulation is performed, and it takes an enormous amount of time to solve the inverse problem. There was a problem.
Also, since the algorithm for calculating the shortest path problem by the Dijkstra method requires a large number of parts to be sequentially advanced, there is a problem that even if a parallel computer is used, it cannot be implemented to use a plurality of processor units. Further, there is a problem that the speedup using the vector computing unit has little effect because the algorithm has few simple repetitive calculations.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a simulation apparatus for an excitement propagation process capable of performing an excitement propagation simulation at a high speed when repeatedly performing an excitement propagation simulation.

[0008]

The invention corresponding to claim 1 is:
An excitement propagation process simulation device that models the tissue as a structure composed of a plurality of points, calculates the excitement propagation arrival time at each point, and simulates the excitement propagation process of the tissue. Excitation propagation time calculation means for obtaining in advance the excitation propagation time required between the points that can be excitable and all points at which excitement can propagate, and all other excitation excursion points from the given initial excitement point Candidate time calculating means for calculating the excitation arrival candidate time of all other points by adding the excitation propagation required time corresponding to the point to the excitation start time; and among the excitation arrival candidate times calculated by the candidate time calculating means, An excitement reaching time determining means for setting the earliest time to the excitement reaching time at that point.

According to a second aspect of the present invention, in the first aspect of the present invention, a potential magnetic field calculating means for calculating a potential on the surface of the tissue or a magnetic field outside the tissue from the result obtained by the excitation arrival time determining means, Parameter setting means for setting parameters for the model and a parameter search means for obtaining an optimum value of the parameter set by the parameter setting means. According to a third aspect of the present invention, there is provided the network according to any one of the first and second aspects, wherein the network is configured by connecting a plurality of nodes and branch lines from each of these nodes to a nearby node. Is modeled by

According to a fourth aspect of the present invention, in the invention according to any one of the first to third aspects, the points that can be the initial excitement point include all points where the excitement can be transmitted.

According to a fifth aspect of the present invention, in the invention according to any one of the first to fourth aspects, the excitement propagation time calculating means calculates the excitement propagation time using the Warshall Floyd method. Is what you do. According to a sixth aspect of the present invention, in the first aspect of the present invention, the excitation propagation time calculating means calculates the excitation propagation time using the Dijkstra method.

According to a seventh aspect of the present invention, there is provided the cell division according to any one of the first to sixth aspects, wherein the tissue is divided into a plurality of units and excitement is propagated between adjacent units. It is modeled by a model. The invention according to claim 8 is the tetrahedron-divided heart model according to any one of claims 1 to 6, wherein the tissue is constituted by a plurality of tetrahedrons, and parameters are set at vertices of the tetrahedron. To model.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a first embodiment of the present invention will be described with reference to FIGS. FIG. 1 is a block diagram showing a circuit configuration of a main part of a simulation device (personal computer) of an excitement propagation process to which the present invention is applied.

Reference numeral 1 denotes a CPU (centr) constituting a control unit main body.
al processing unit). ROM (read only mem) in which program data of processing performed by the CPU 1 is stored.
ory) 2, a RAM (random access memory) in which various memory areas used when the CPU 1 performs processing are formed.
y) 3, I / O ports 4 connected to communication lines, external devices (for example, printers), etc., are connected to the CPU 1 via a system bus 5 respectively. The CPU 1
A storage device interface 7 for controlling data transmission with a storage device 6 such as a floppy disk device or a hard disk device via the system bus 5; a keyboard interface 9 for controlling data transmission with a keyboard 8; It is connected to a display controller 11 that controls a display 10 that displays a magnetocardiogram and various data. The above configuration is used in the following embodiments.

In the first embodiment, an example of an excitation propagation simulation of a ventricle of the heart will be described as an example of an excitable tissue. FIG. 2 is a block diagram showing a functional configuration of a main part of the simulation apparatus of the excitement propagation process according to the first embodiment. This simulation apparatus constitutes a ventricular model constructing means 21 for constructing a ventricle as a geometric model, and an excitement propagation network comprising nodes and branches for the ventricular model constituted by the ventricular model constructing means 21. Excitement propagation network construction means 22 for calculating and setting the required excitation propagation time between both ends of each branch line, and the required excitation propagation time between all (not only adjacent nodes) in this excitation propagation network is calculated in advance.・ Set excitement propagation time required calculation means 2
3, an initial value setting means 24-1 for setting an initial excitement node and an initial excitement time;
Excitement arrival time calculation means 24 which calculates the respective excitement arrival times of all the nodes based on the excitement propagation time set by the above and the initial excitement nodes and the initial excitement times set by the initial value setting means 24-1. It is composed of

The ventricle model construction means 21 constructs a ventricle model in which the shape of the ventricle is represented by a large number of tetrahedrons, as shown in FIG. FIG. 3A shows a model of the right ventricle wall, FIG. 3B shows a model of the left ventricle, and FIG. 3C is a perspective view showing a model of the entire ventricle. . An excitation propagation speed tensor is set at each vertex of the ventricle model. This excitation propagation velocity tensor expresses the anisotropy of excitation propagation, in which excitation propagates rapidly along the direction of the myocardial fiber and propagates slowly in the orthogonal direction. When the excitement starts from one point, the vector t
The excitation propagation velocity v in the direction of is expressed by the following (Equation 1)

[0017]

(Equation 1) Results in an ellipsoidal shape given by This excitation propagation velocity tensor is a second-order symmetric tensor and is represented by a 3 × 3 symmetric matrix.

The excitement propagation network forming means 22
In an actual heart, excitation forms an excitation propagation network which is essentially continuous excitation propagation that propagates in a wavefront shape in the heart parenchyma, and approximates that excitation propagates through a path.

First, a number of representative points (hereinafter referred to as nodes) are provided in the myocardium, and adjacent nodes are connected by branch lines. FIG. 4 illustrates how to select nodes connected by branch lines. Although FIG. 4 is expressed in two dimensions for the sake of explanation, a thick triangle in FIG. 4 corresponds to one tetrahedron shown in FIG. Each branch of a thick triangle (tetrahedron) is N
A new point is set at the equally divided points, and all the points are connected by branches. The initially provided representative point and the newly created point are the nodes of the excitement propagation network, and the first tetrahedron branch and the newly created branch are the branch lines of the excitation propagation network.
It is approximated that the excitement can be propagated only on the branch line of the excitement propagation network created as described above.
By connecting many branch lines to one node, continuous excitation propagation in the myocardium can be accurately approximated.

Next, the time required for excitement propagation between both ends of each branch of the excitement propagation network is all calculated and given to each branch to complete the excitement propagation network. The following equation (Equation 2) is used to calculate the required excitation propagation time r _ij between both ends of the branch line.

[0021]

(Equation 2)

Here, x _i and x _j are position vectors at both ends of the branch line connecting the i-th and j-th nodes.
Is the aforementioned excitation propagation velocity tensor, and s is the integration variable. The integral of (Equation 2) is trapezoidal (Simpson integral formula)
Is used to perform numerical integration to calculate a time r _ij required for excitation propagation between both ends of the branch line.

The required excitation propagation time calculating means 23 calculates the required excitation propagation time a _ij from an arbitrary node to all other nodes based on the aforementioned required excitation propagation time r _ij between both ends of the branch line. Things. By approximating the ventricular excitation propagation with a network, the problem that simulates the excitation propagation assumption can be replaced with the problem that solves the shortest path (shortest path) of the network, using a known algorithm such as the Warshall Floyd method Can be solved efficiently.

Here, an example in which the Warshall Floyd method known as one of the methods for efficiently obtaining the shortest path problem will be described. The excitement propagation time calculation means 23 has N nodes in total, and describes each node as i and j. i and j are the node numbers from 1 to N, and the excitement starts from the node i and the excitement propagation time a _ij required for the excitement to reach the node j is calculated by the next excitement propagation time calculation. Determined by performing processing. FIG. 5 is a diagram showing the flow of the required excitation propagation time calculation processing performed by the CPU 1 (excitation propagation required time calculation means 23).

First, as a process of step 1 (ST1), it is determined whether or not the nodes i and j are connected by a branch line. If it is determined that the node i and the node j are connected by a branch line, the excitation propagation time a _ij =
r _ij, and if it is determined that the nodes i and j are not connected by a branch line, the excitation propagation time a
Set _ij = ∞. Upon completion of the setting of the a _ij, all i, j (= 1,2, ... , N) set in a _ij to determine whether or not it is completed for. If it is determined that the setting of a _ij has not been completed for all i and j, the process returns to the above-described step 1 again.

When it is determined that the setting of _aij has been completed for all i and j, 1 is set to h as an initial value. Next, as the process of step 2 (ST2), a
_It is determined whether or not the value of _ij is less than or equal to the value of a _ih + a _hj (a _ij ≦ a _ih + a _hj ). If it is determined that the value of a _ij is equal to or _smaller than the value of a _ih + a _hj , a _ij is set to a _ij = a _ij, and the value of a _ij is a
_ih + a following not the value of _hj, i.e., the value of a _ij is a _ih
+ A _hj is determined to be larger than a _ij = a _ih + a _hj
Set as When the setting of a _ij is completed, all i,
It is determined whether or not the setting of a _ij has been completed for j (= 1, 2,..., N). Here, a _ij for all i and j
When it is determined that the setting has not been completed, the process returns to the above-described step 2 again.

When it is determined that the setting of _aij has been completed for all i and j, it is determined whether or not h is equal to N. Where h is not equal to N (h is less than N)
When it is determined that h + 1 is set to h + 1, a +1 addition update process is performed, and the process returns to the above-described step 2 again. Further, when it is determined that h is equal to N, the processing for calculating the required time for exciting propagation is ended.

The initial value setting means 24-1 sets the nodes corresponding to the inner walls of the left ventricle and the right ventricle among the nodes of the ventricle model as the initial excitation nodes. Initial excitement nodes are all L
(L ≦ N), and their nodes are P _k (k = 1,...,
L). In the case of the first embodiment, since a large number of nodes of the excitation propagation network are generated based on the nodes of the ventricular model, L is actually a value that is about several times smaller than N. .

The excitement arrival time calculating means 24 calculates that k is 1
Takes a value of L from, p _k has a value of from 1 N, when the excitation start time of the initial excitement node P _k is given as t _pk, excitement arrival time t _i for all nodes, including also p _k , By performing the next excitement arrival time calculation process. FIG.
FIG. 7 is a diagram showing a flow of an excitement arrival time calculation process performed by the CPU 1 (excitation arrival time calculation means 24).

First, as a process of step 3 (ST3), it is determined whether or not the node i has been set as an initial excitation node by the initial value setting means 24-1. If it is determined that the node i is set as the initial excitement node, the excitement arrival time t _i = t _{pk at} that node i is set. If it is determined that the node i is not set as the initial excitement node, the excitement arrival time t Set _i = ∞. Upon completion of the setting of the t _i, the setting of the t _i for all i to determine whether or not it is completed. Here, if it is determined that the setting of t _i has not been completed for all i, the process returns to the above-described step 3 again.

If it is determined that t _i has been set for all _i , the value of t _i is equal to or less than the value of t _pk + a _pki (t _i ≦ t _pk + a) in step 4 (ST4).
_pki ). If it is determined that the value of t _i is equal to or less than the value of t _pk + a _pki , t _i = t _i is set, and
The value of t _i is not less than the value of t _pk + a _pki , ie,
If it is determined that the value of t _i is greater than the value of t _pk + a _pki ,
Set t _i = t _pk + a _pki . Upon completion of the setting of the t _i, all i, the k (all i, with respect to the combination of k), setting the t _i is termination decision. If it is determined that the setting of t _i has not been completed for all i and k, the process returns to step 4 again. Also, for all i and k, t
_When it is determined that the setting of _i has been completed, the excitement arrival time calculation processing is terminated.

As described above, according to the first embodiment, the excitation propagation time required calculating means 23 is provided, and the excitation propagation time r _ij between both ends of the branch line constituted by the excitation propagation network forming means is provided. The required excitement propagation time a _ij from any node to all the other nodes is calculated in advance, and the excitement arrival time calculation means 24 calculates the excitement arrival time based on this a _ij , whereby the excitement arrival time is calculated. Means 24
Can greatly reduce the calculation time. Therefore, when performing the excitement propagation simulation multiple times,
Since only the excitement arrival time calculation means 24 needs to calculate, the time required for the entire excitement propagation simulation can be reduced.

Further, the same result as that of the method of determining the time of arrival of excitement by the Dijkstra method can be obtained. When the Dijkstra method is used, a calculation amount proportional to approximately N ² is required, and the algorithm is complicated and the innermost In the first embodiment, since L is a fraction of N, the calculation content in the loop of the first embodiment is complicated. The calculation of the excitement arrival time of the form can be executed with a calculation amount of a fraction or less. Therefore, in the case of calculating a large number of times at which the initial excitation nodes and the excitation start times of the initial excitation nodes differ from each other for the same excitation propagation network, the time required for the calculation can be greatly reduced.

Further, the algorithm for calculating the time of arrival of excitement according to the first embodiment can divide the entire processing into N or L pieces and calculate independently, so that when a parallel computer is used, the processing can be efficiently distributed. Can be done. Further, for the same reason, the vector computing unit can be used efficiently. Therefore, in addition to shortening the calculation time due to the reduction in the amount of calculation, it is possible to further reduce the time required for calculation when using a parallel computer or a vector operation unit. Even when a parallel computer and a vector computing unit are not used, since the processing is a simple repetitive operation, it is possible to implement the high-speed means of the computer such as pipeline processing, cache memory, registers, etc. so as to operate efficiently. Therefore, even when a parallel computer or a vector computing unit is not used, the time required for the calculation can be further reduced more than the reduction in the calculation time due to the reduction in the amount of calculation.

A second embodiment of the present invention will be described with reference to FIG. An example in which the above-described excitation propagation simulation method according to the first embodiment is applied to an inverse problem of estimating an initial excitation time distribution of an intraventricular wall from a magnetocardiogram to be measured.

FIG. 7 is a block diagram showing a functional configuration of a main part of the simulation apparatus according to the second embodiment.
This simulation apparatus forms a ventricle model forming means 31 for forming a ventricle as a geometric model, and an excitation propagation network comprising nodes and branches for the ventricle model formed by the ventricular model forming means 31. Excitement propagation network composing means 32 for calculating and setting the required excitement propagation time between both ends of each branch line, and pre-calculates the required excitement propagation time between all (not only between adjacent nodes) in this excitement propagation network A set excitation propagation time calculation unit 33, a parameter search unit 34 described later, and a parameter search for a magnetocardiogram to be subjected to inverse problem analysis (for example, a magnetocardiogram actually obtained as a result of measurement, a task magnetocardiogram). Displaying the magnetocardiogram measurement result setting means 35, which is output to the means 34, and the estimated initial excitation site and the excitation start time. And a constant result displaying means 36..

As in the first embodiment, the ventricle model structuring means 31 represents the ventricle shape by a large number of tetrahedrons, and each node of the ventricle model has the excitation propagation velocity tensor of the first embodiment. In addition, the conductivity tensor and the action potential amplitude are set. The conductivity tensor is for expressing the anisotropy of conductivity, and is often used for expressing the conductivity anisotropy of a substance having a crystal structure. The conductivity tensor is also a second-order symmetric tensor like the excitation propagation speed tensor, and is represented by a 3 × 3 symmetric matrix.

The excitement propagation network composing means 32 and the excitement propagation required time calculating means 33 are provided by the first
Since the configuration is the same as that of the embodiment, description thereof is omitted here.

The parameter search means 34 includes an initial parameter setting means 34-1 for setting an initial excitement node and an initial excitement time, and an excitement propagation time set by the excitement propagation time calculation means 33 and the initial parameter setting. An excitation arrival time calculating means 34-2 for calculating the respective excitation arrival times of all the nodes based on the initial excitation nodes and the initial excitation times set by the means 34-1; and the conduction set by the ventricular model construction means. A magnetic field on the surface of the heart is calculated based on the rate tensor, the action potential amplitude, and the respective excitation arrival times of all the nodes calculated by the excitation arrival time calculating means 34-2 to create a magnetocardiogram (estimated magnetocardiogram). Magnetic field calculation means 34-3, the subject magnetocardiogram from the magnetocardiogram measurement result setting means 35 and the magnetic field calculation means 34-3
A convergence determining means 34-4 for comparing the estimated magnetocardiogram with the estimated magnetocardiogram and evaluating the difference, and reducing the difference between the task magnetocardiogram and the estimated magnetocardiogram evaluated by the convergence determining means 34-4. Parameters set by the initial parameter setting means 34-1 (here, initial excitement nodes and initial excitement times)
And parameter changing means 34-5 for changing the parameter.

The excitement arrival time calculation processing performed by the excitation arrival time calculation means 34-2 includes the excitation arrival time calculation processing (see FIG. 6) performed by the excitation arrival time calculation means 24 of the first embodiment described above. This is the same, and the description is omitted here. The initial parameter setting unit 34-1 and the convergence determination unit 34-4 included in the parameter search unit 34,
A specific execution method of the parameter changing unit 34-5 is as follows.
It depends on known optimization algorithms used for parameter search. These execution methods may be determined according to a known optimization algorithm.

In the second embodiment having such a configuration, the process of calculating the required time of excitement propagation between all nodes by the required time of excitement propagation calculation means 33 is performed only once.
Optimization processing is performed in the parameter search means 34. That is, the initial value is set by the initial parameter setting unit 34-1 by the parameter changing unit 34-5 until the difference between the subject magnetocardiogram evaluated by the convergence determining unit 34-4 and the estimated magnetocardiogram falls within a preset range. The parameter setting is changed, and with this change, calculation processing is performed by the excitation arrival time calculation means 34-2 and the magnetic field calculation means 34-3, and the target magnetocardiogram is estimated again by the convergence determination means 34-4. The difference from the magnetocardiogram is evaluated, and the above processing is repeated. When the difference between the subject magnetocardiogram evaluated by the convergence determining means 34-4 and the estimated magnetocardiogram falls within a preset range, the parameters set by the initial parameter setting means (the initial excitation node and the initial excitation Time) is the solution to the inverse problem. A diagram showing the solution of the inverse problem and the state of the excitement propagation based on the solution is displayed by the estimation result display means 36. That is, the excitement arrival time calculation process of the excitement arrival time calculation means 34-2 is a process that is repeatedly performed along with the optimization.

As described above, according to the second embodiment, similarly to the above-described first embodiment, the calculation time in the excitement arrival time calculation means 34-2 can be greatly reduced. In addition, when solving the inverse problem, the excitement arrival time calculation process of the excitation arrival time calculation means 34-2 is repeated many times, so that the above-described effect of shortening the calculation time becomes more remarkable. Therefore, the effects of the first embodiment described above can also be obtained in the second embodiment.

A third embodiment of the present invention will be described with reference to FIGS. In the first and second embodiments described above, the excitation propagation time required calculating means 2
In the third embodiment, an algorithm based on the Warshall Floyd method in the graph theory was constructed for the calculation of the required excitation propagation time in the third and third embodiments. A description will be given of a graph theory constructed by an algorithm based on the Dijkstra method in graph theory. Note that, in the third embodiment, the main part functional configuration of either the first embodiment or the second embodiment described above may be used, and the description of the main part functional configuration will be omitted.

FIG. 8 is a diagram showing the flow of the required time for the required propagation of the excitement to be performed by the means for calculating the required time of the required propagation of the excitement. First, 1 is set to a variable k, and L initial excitation nodes Pk are defined. Therefore, k takes a value from 1 to L. next,
In the process of step 5 (ST5), the excitation start time t _pk of the node Pk is set to 0, _tpk is set to 0, and the other nodes i (i takes a value from 1 to N. Here, i is Pk.
Not included. ∞ to t _i the excitement start time t _i of) as ∞
Set. Next, as a process of step 6 (ST6),
The Dijkstra method described later (see Japanese Patent Application No. 7-98791)
Performing a t _i calculation process due to).

When the t _i calculation processing by the Dijkstra method is completed, each t _i calculated by the t _i calculation processing by the Dijkstra method is set to a _pki . Next, it is determined whether or not the variable k is equal to L. Here, if it is determined that the variable k is not equal to L (k is smaller than L), the variable k is updated by adding +1 to the variable k, and the process returns to the step 5 again. When it is determined that the variable k is equal to L, the processing for calculating the time required for the propagation of the excitement is terminated.

FIG. 9 is a diagram showing a flow of an example of the t _i calculation process by the Dijkstra method in step 6 in the excitement propagation time calculation process performed by the excitement propagation time calculation means. First, 1 is set to the variable i, and step 7 (ST7)
Is determined whether or not the node Pk and the node i are connected by a branch line. Here, if it is determined that the node Pk and the node i are not connected by a branch line, 興奮 is set to the excitation arrival time t _i of the node i, and the processing shifts to step 8 (ST8) to be described later. It has become. If it is determined that the node Pk and the node i are connected by a branch line, _rpki is set to the excitation arrival time t _i of the node i, and the process _proceeds to the next step 8. ing.

The process of step 8 determines whether or not the variable i is equal to the total number of nodes N (N-1 if Pk = N). If it is determined that the variable i is not equal to N (i is smaller than N), the variable i is subjected to addition and update processing of +1 and the processing returns to the above-mentioned step 7 again. If it is determined that the variable i is equal to N, 2 is set to the variable m, and as a process in step 9 (ST9), the numerical value m set in the variable m is set to the m-th node among all the nodes. Attention node x where the excitement arrival time is earlier
Set as That is, the excited arrival time of the noted node x is t _x . When there are a plurality of nodes having the same excited arrival time, priority is given to the nodes having the smallest node number in order.

Next, 1 is set to a variable i, and step 1 is executed.
As a process of 0 (ST10), it is determined whether or not the target node x and the node i are connected by a branch line. Here, if it is determined that the target node x and the node i are not connected by a branch line, step 11 (ST11
).

When it is determined that the node of interest x and the node i are connected by a branch line, the excitation arrival time t _x of the node of interest x and the branch line connecting the node of interest x and the node i are determined. arrival time obtained by adding the excitement propagation time required r _xi between both ends (
t _x + r _xi ) is set at the node i, the excitement attainment time t
Determine if it is earlier than _i . Now, it is determined that the calculated arrival time (t _x + r _xi) is not earlier than the excitation time of arrival t _i that are set (simultaneously or slow) Step 1
1 is performed. In addition, if the calculated arrival time (t _x + r _xi) is determined to be earlier than the excitement arrival time t _i that has been set, the arrival time of calculation of the excitement arrival time t _i that is set to the node i (t _x + r _xi ) Is performed (update setting) and the next step 1
Shift to the processing of 1.

The process in step 11 determines whether the variable i is equal to the total number N of nodes. Here, if it is determined that i is not equal to (less than) N, the addition and update processing of +1 is performed on the variable i, and the process returns to the above-described step 10 again. When it is determined that i is equal to N, it is determined whether or not the variable m is equal to the total number N of the nodes.
Here, if it is determined that m is not equal to (less than) N, the process of adding +1 to the variable m is performed, and the process returns to the above-described step 9 again. If it is determined that m is equal to N, the processing for calculating t _i by the Dijkstra method is terminated, and the processing for calculating the time required for the excitation propagation is performed again.
(FIG. 8). Note that the t _i calculation process by the Dijkstra method is an example, and another algorithm may be used.

As described above, all combinations of the time required for the propagation of the excitation from all the initial excitation nodes to all the other nodes are calculated and set in advance. The initial excitement nodes of the third embodiment are excitement arrival time calculating means 24, 34-2.
May be selected so as to include all of the initial excitation nodes that may be specified in. For example, most simply, all nodes can be selected as P _k = k.

As described above, according to the third embodiment, even if the algorithm based on the Dijkstra method is used in advance in the excitation propagation time calculating means 23 and 33, all the initial excitation nodes are set in advance from all other excitation nodes ( Excitation propagation time up to (including all nodes not connected by branch lines) can be calculated and set, so that the same effect as in the above-described first and second embodiments can be obtained. Can be obtained.

In the first, second and third embodiments described above, the ventricle model constituted by the ventricle model constructing means 21 and 31 is described by a method of approximating the ventricle model by the network by the excitement propagation network constructing means 22 and 32. However, the present invention is not limited to this. The myocardium is divided into thousands to tens of thousands of units, and excitement is propagated between adjacent units at each step until excitement is propagated to the entire heart. A cell model that repeats steps and a heart are composed of many tetrahedrons, parameters such as excitation propagation characteristics are specified at each vertex, and linear interpolation is performed so that the characteristics change continuously within the tetrahedron, and the excitation wavefront is divided into triangles Alternatively, the excitation propagation may be approximated by a tetrahedral division model in which each vertex is moved at a certain time step. In this case, the calculation of the excitement propagation time can be performed in substantially the same manner as in the third embodiment. At this time, the node numbers i and P _k are cell numbers, t _i is the excitation arrival time of the i-th cell, and a _{pk, i} is P _k
What is necessary is just to replace it with the time required for the excitement propagation from the i-th cell to the i-th cell. However, in the t _i calculation process by the Dijkstra method, an excitement propagation simulation is performed using a cell division model, and the required excitement propagation time from the P _kth cell to the i th cell is stored in t _i .

In the first, second, and third embodiments described above, the simulation of the excitation propagation of the heart and the inverse problem solving method using the simulation have been described. However, the present invention is not limited to this. In addition, the present invention can be applied to a simulation method of excitation propagation of nerve tissue and an inverse problem solving method using the same. That is, it can be applied to the problem of estimating the activity in the brain from the measured magnetoencephalogram and the brain.

For example, a description will be given of a brain excitation propagation simulation. First, regarding the configuration of the excitation propagation network, nodes are provided at a plurality of points in the brain. Although a site where a node is provided may be random in the cerebral cortex, providing a node at a site defined anatomically or physiologically can realize a more accurate excitation propagation simulation.

For example, nodes are set based on the Bradmann's brain map shown in FIGS. 10A and 10B and the cerebral cortex areas obtained by subclassification by researchers thereafter. FIG. 10 shows “K. Brodmann: Vergleichende Lokalisat
ionslehre der Groshirnrinde, Barth, Leipzig 1909 ”
FIG. 10 (a) is a diagram showing the cytological cortical area on the outer surface of the cerebral hemisphere, and FIG. 10 (b)
FIG. 2 is a diagram showing a cytological cortical area on the inner surface of the cerebral hemisphere.

In the setting of the nodes, it is possible to finely set a region where the function localization is fine and clear, such as the motor area, within a known range. For example, FIG. 11 shows the four Brodmann fields of FIG. 10 finely set to 4-1 to 4-20. Of course, it may be further subdivided. In addition, nodes are similarly set for the basal ganglia, brain stem, and cerebellum, and these nodes are connected by branch lines to express neural projection.

At the time of connection of the branch lines, it is assumed that the direction of the excitement propagation is given, and the propagation is not performed in the opposite direction. That is, assuming that the time required for transmission of the excitement from the node i to the node j is r _ij as in the first embodiment, the node j to the node i
It is assumed that no excitement is transmitted to the user, so _rji = ∞. However,
If it is known that nerve projections exist in the opposite direction physiologically, the known excitation propagation time required is _rji . Further, in each block of all territories, each node connects branch lines by a combination of all conceivable nodes.

When calculating the time required for excitement propagation between the nodes, the sum of the time required for nerve fiber conduction and the time required for synaptic excitation propagation is given. In the calculation of the required excitement propagation time, when the required excitement propagation time is calculated using the Dijkstra method or the Warshall Floyd method, the excitement propagation time is not changed at all from the first, second, and third embodiments. The required time can be calculated.

[0060]

As described in detail above, according to the present invention,
When the excitement propagation simulation is repeatedly performed, it is possible to provide an excitement propagation process simulation apparatus capable of performing the excitement propagation simulation at high speed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a circuit configuration of a main part of a simulation device for an excitement propagation process according to a first embodiment of the present invention.

FIG. 2 is an exemplary block diagram showing a functional configuration of a main part of the simulation device of the excitement propagation process according to the embodiment;

FIG. 3 is a diagram showing a right ventricle wall, a left ventricle, and the entire ventricle of a heart model of a tetrahedron division model which is a basic heart model constituted by the simulation apparatus of the excitement propagation process of the embodiment.

FIG. 4 is an exemplary view for explaining a subdivision for constructing a heart model by a network from a heart model of a tetrahedral division model configured by the simulation apparatus of the excitation propagation process according to the embodiment;

FIG. 5 is a diagram showing a flow of an excitement propagation time calculation process performed by an excitement propagation time calculation unit of the simulation apparatus of the excitement propagation process according to the embodiment;

FIG. 6 is a diagram showing a flow of an excitement arrival time calculation process performed by an excitement arrival time calculation unit of the simulation apparatus of the excitement propagation process according to the embodiment;

FIG. 7 is a block diagram showing a functional configuration of a main part of a simulation device for an excitement propagation process according to a second embodiment of the present invention.

FIG. 8 is a diagram showing a flow of an excitement propagation time calculation process performed by an excitement propagation time calculation unit of the simulation apparatus of the excitement propagation process according to the third embodiment of the present invention.

FIG. 9 is a diagram showing a flow of an example of t _i calculation processing by the Dijkstra method during the excitation propagation time calculation processing performed by the excitation propagation time calculation means of the simulation apparatus of the excitation propagation process of the embodiment.

FIG. 10 is a diagram showing a Bradmann brain map.

FIG. 11 is a diagram showing the subdivision of four cytostructural cortical fields in the Brodmann brain map.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... CPU, 21, 31 ... Ventricle model construction means, 22, 32 ... Excitement propagation network construction means, 23, 33 ... Excitement propagation required time calculation means, 24, 34-2 ... Excitement arrival time calculation means, 34 ... Parameter search means.

Claims (8)

[Claims] 1. Modeling an organization as a structure composed of a plurality of points, calculating an excitation propagation arrival time at each point,
An excitement propagation process simulation apparatus for simulating an excitement propagation process of the tissue, wherein an excitement propagation time required between a point that can be an initial excitement point among all the points and all points where excitement can propagate is obtained in advance. An excitement propagation required time calculating means to be added, and the excitement propagation required time corresponding to all the other points from the given initial excited point is added to the excitement start time to calculate all the other points. A candidate time calculating means for calculating an excitement reaching candidate time; and an excitement reaching time determining means for setting the earliest time among the excitement reaching candidate times calculated by the candidate time calculating means to an excitement reaching time at that point. An apparatus for simulating an excitement propagation process, characterized in that: 2. A potential magnetic field calculating means for calculating a potential on the surface of a tissue or a magnetic field outside the tissue from a result obtained by the means for determining the time of excitation arrival, and a parameter setting means for setting parameters for the tissue model. The apparatus for simulating an excitement propagation process according to claim 1, further comprising: parameter search means for obtaining an optimum value of the parameter set by the parameter setting means. 3. The model according to claim 1, wherein the tissue is modeled by a network constituted by connecting a plurality of nodes and a branch line from each of these nodes to a nearby node. A simulation device for an excitement propagation process according to claim 1. 4. The excitement propagation simulation apparatus according to claim 1, wherein the points that can be the initial excitement point include all points where excitement can be propagated. 5. The excitement propagation according to claim 1, wherein said excitement propagation time calculation means calculates the excitement propagation time using a Warshall Floyd method. Simulation device. 6. The excitement propagation simulation apparatus according to claim 1, wherein said excitement propagation time calculation means calculates the excitement propagation time using Dijkstra's algorithm. . 7. The model according to claim 1, wherein the tissue is divided into a plurality of units and is modeled by a cell division model that defines that the excitement propagates between adjacent units. The described excitation propagation simulation apparatus. 8. The tissue is constituted by a plurality of tetrahedrons,
The excitation propagation simulation apparatus according to any one of claims 1 to 6, wherein the model is modeled by a tetrahedron-divided heart model that sets parameters at the vertices of the tetrahedron.