JPH06169895A - Diagnostic system - Google Patents

Abstract

PURPOSE:To identify an abnormal portion by determining the distribution of electrical and magnetic field sources and spatial power from electrical and magnetic field distribution measured on the surface of an object to be inspected, and comparing the obtained results with corresponding distribution in a normal case. CONSTITUTION:A magnetic sensor 1 for detecting magnetic field distribution on the surface of an object to be inspected is provided, together with a sensor 2 for detecting potential or electrical field distribution. Also, a magnetic field source distribution calculation means 3a calculates magnetic field source distribution from the detected magnetic field distribution, A voltage source distribution calculation means 3b calculates voltage source distribution from the detected potential or electrical field distribution. A spatial power distribution calculation means 3c calculates spatial power distribution, on the basis of the inner product of the calculated magnetic field distribution and voltage source distribution. In this case, the electrical field source is a time differential about an electrical dipole, and the voltage source is a physical quantity defined by the product of a current dipole and a resistor. A data processing device 3 is equipped with a comparison means 3e for making a comparison between data 3d in normal state and at least one of the calculated magnetic field source distribution, voltage source distribution source and spatial power distribution, and can identify an abnormal portion as a result of the comparison. According to this construction, a heart can be diagnosed more accurately.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION The present invention finds an electric field, a magnetic field source distribution, and a spatial power distribution from an electric field and a magnetic field distribution measured on the surface of an object to be inspected, and compares the distribution with a normal case to detect an abnormal portion. The present invention relates to a diagnostic device that identifies, diagnoses, and inspects.

[0002]

2. Description of the Related Art Conventionally, as a non-destructive inspection method, an electric potential method for inspecting an inside of an object to be inspected by energizing the object to be inspected and measuring a voltage distribution, There is known a method in which a magnetic field distribution formed when a voltage is applied and a current is applied thereto is measured in a non-contact manner and the presence or absence of a flaw is inspected from the magnetic field distribution.

In the case where the body to be inspected is a living body, a voltage appears on the body surface because nerve activity is carried out by the movement of electric charges, that is, electric current. As examples of measuring and diagnosing this voltage, for example, an electrocardiogram obtained by measuring the voltage from above the heart, an electroencephalogram distribution obtained by measuring the voltage at the head, and the like are known.

[0004]

By the way, in biomedical measurement, for example, the measurement of the electroencephalogram distribution occurs inside because the medium to be measured is a mixture of a dielectric and a resistance, and in particular the skull has a high electrical conductivity. It has been pointed out that it is difficult to capture the phenomenon, and accurate measurement cannot be performed only by measuring the EEG distribution, and similarly, it is pointed out that waveform analysis can be performed only by the electrocardiogram, and it is not possible to know the detailed situation inside the heart. Has been done. As described above, it is difficult to grasp the accurate situation inside the object to be inspected by the conventionally known nondestructive measurement method, and it is not possible to perform highly reliable diagnosis and inspection.

The present invention is intended to solve the above-mentioned problems. The magnetic field distribution, the voltage source distribution, and the power generation distribution inside the object to be inspected are obtained by measuring the magnetic field distribution, the potential or the electric field distribution appearing on the surface. It is an object of the present invention to provide a diagnostic device capable of accurately ascertaining the internal situation and performing highly reliable diagnosis and inspection.

[0006]

As shown in FIG. 1, a diagnostic apparatus according to the present invention comprises a magnetic sensor 1 for detecting a magnetic field distribution on a surface of an object to be inspected, and a sensor 2 for detecting a potential or electric field distribution. Magnetic field source distribution calculating means 3a for calculating a magnetic field source distribution from the detected magnetic field distribution, voltage source distribution calculating means 3b for calculating a voltage source distribution from the detected potential or electric field distribution,
And a data processing device 3 having a spatial power distribution calculating means 3c for calculating a spatial power distribution by taking an inner product of the calculated magnetic field source distribution and voltage source distribution.
Further, the data processing device 3 includes a comparison unit 3e that compares the normal data 3d with at least one of the calculated magnetic field source distribution, voltage source distribution, and spatial power distribution, so that the abnormal portion can be specified by the comparison result. In addition, the calculated magnetic field source distribution, voltage source distribution, spatial power distribution, comparison result, and the like can be output by the output device 4 including the display device 4a, the printer 4b, and the like.

[0007]

To explain the diagnostic process of the present invention, the coordinate system in two dimensions is a three-dimensional coordinate of (x, y, θ) of the plane position coordinate of x, y and the angle θ of the electromagnetic field source vector.
In three dimensions, the spatial position coordinates of x, y, z, and (x, y, z, θ, φ) of the angle θ from the xy plane and the angle φ from the yz plane of the electromagnetic field source vector It is a five-dimensional coordinate. System equations, u ₁ vector or potential of the n electromagnetic field measured at the local, u _2, ......, if u _n, known vector U is (denotes the transpose superscript ^T) U _{= [u 1, u 2,} ......, u n] written as ^T. If the space in which the electromagnetic field source vector can exist is divided into m areas including the angle, a position that gives a relation between the electromagnetic field source located at an arbitrary position i and a known vector U in this area. If the vector is d _i, it can be written that d _i = [G _1i , G _2i , ..., G _ni ] ^T. Here, the elements G _1i , G _2i , ..., G _{ni of} d _i
Is determined by the Green's function, and if the magnitude of the electromagnetic field source vector at an arbitrary point _i is α _i , the vector U is a linear combination of the vectors d _i (i = 1 to m), so the system equation is Becomes

Next, pattern matching is performed. Generally, the number m of unknowns is greater than the number n of equations, and the solution α _i (i = 1 to m)
Cannot be uniquely determined from the system equations, so assume the following. The magnitude α _i (i = 1 to m) of the electromagnetic field is set as a unit value “1”, and the magnitude is approximated by the degree of spatial concentration including the angle of this unit. The position of the first and most dominant electromagnetic field source is the known vector U
And the pattern of the position vector d _i (i = 1 to m) coincides with each other, and the maximum value is obtained by the following equation. _{^{γ = U T d i / [}} ‖U‖‖d i ‖], i = 1 to m, where the pattern matching index of the maximum value and gamma _h. ‖‖
Indicates the norm of the vector. The position of the second dominant electromagnetic field source is a point where the position h of the first electromagnetic field source is used as a pilot point (that is, α _h = 1) and the maximum pattern matching index γ is obtained by the following equation. γ _j = U ^T (d _h + d _j ) / [‖U‖‖d _h + d
_j ∥], j = 1 to m, j ≠ h where γ _g is the maximum pattern matching index. Similarly, with the points having the maximum value pattern matching indices γ _h and γ _g as pilot points (α _h = 1 and α _g = 1), the remaining distance vector d _k (k = 1 to 1 m, k ≠ h,
Compute the pattern matching index γ _k for k ≠ g). This process is executed m times in total.

Obtained at each step (including angle)
A pattern matching index at m points on the spatial coordinates is added and averaged to obtain a normalized approximate solution, α _i [‖d _i ‖ / ‖U‖] (i = 1 to 1
m) is combined in real space coordinates (xy plane in two dimensions and xyz space in three dimensions) to obtain a normalized approximate electromagnetic field source vector. Here, the fact that the norm ‖d _i ‖ of the distance vector in the normalized approximate solution is a product of α _i (i = 1 to m) is more accurate for the electromagnetic field source closer to the measurement point of the electromagnetic field distribution. Means required to.

In actual calculation, since it takes a lot of calculation time to obtain up to m points, distance vectors are added as long as the pattern matching index taking the maximum value of each step increases from the previous one. A relatively good result can be obtained in a short time by the results obtained by terminating the process until the time when the number decreases.

The spatial power distribution is obtained by taking the inner product of the obtained magnetic field source (current dipole) distribution and voltage source (voltage dipole) distribution. Current dipole distribution, voltage dipole distribution when there is no abnormality,
Spatial power distribution is obtained in advance, and at least one of these is compared with that obtained for the object to be inspected,
Presence / absence of abnormality and identification of abnormal area.

[0012]

The present invention will be described in more detail below.
In order to facilitate understanding, a method of measuring a magnetic field distribution and obtaining a current distribution will be described. As an analytical model, as shown in FIG. 2A, when the current is evenly distributed at the current density J in a conductor having an arbitrary cross-sectional shape, the cross-sectional shape of the conductor is obtained from the measurement of the magnetic field distribution around the conductor. Considering an example, assuming that the current density J flows in the cross-sectional area Δs, the current i on the cross-sectional area Δs is given by i = ΔsJ. Therefore, as shown in FIG.
The two-dimensional cross section of FIG. 2A is divided into minute surfaces of Δs, and
As shown in (c), consider an analytical model in which a current i flows through each minute surface and is distributed in the shape of a conductor cross section. At this time, if the distribution of the current i in FIG.
The cross-sectional shape of the conductor is required. Since all the currents in FIG. 2 (c) have the same magnitude, when this analytical model is applied to the problem of non-uniform current distribution, the areas with high current density have large shapes and the areas with low current density have small shapes. As a result, the obtained analysis result represents not the cross-sectional shape of the conductor but the current distribution on the cross-section of the conductor.

Next, let us consider how to obtain the distribution of the current i shown in FIG. 2 by measuring the magnetic field distribution. As shown in FIG. 3, the current i located at the distance of x ₁₁ from the measurement surface ab
Measurement points 1 by ₁ and the magnetic field H ₁₁ in the 2 and H ₁₂ (perpendicular magnetic field component in the measurement plane a-b) is, H ₁₁ = i _{1 [y 11 / {2π (x 11} 2 + y 11 2)} ] ... (1a) H ₁₂ = i ₁ [y ₁₂ / {2π (x ₁₁ ² + y ₁₂ ² )}] (1b) When formula (1a) and formula (1b) are written together, U = i ₁ X ...... (2a) U = [H ₁₁ , H ₁₂ ] ^T (the upper subscript ^T is transposed) …… (2b) X ₁ ＝ 1 / (2π) [y ₁₁ / (x ₁₁ ² + y ₁₁ ² ), y ₁₂ / (x ₁₁ ² + y ₁₂ ² )] ^T (2c). Therefore, assuming that the number of measurement points is n and the number of input points of the current that is the magnetic field source is m, equation (2a) is Becomes Here, U and X _k are column vectors of order n.
Further, the elements of U are H ₁₁ and H _{12 in the} equation (2b),
The element of X _k corresponds to the element on the right side of the expression (2c).

Now, letting C _{nm be} the coefficient matrix of n rows and m columns in which the elements of the matrix consist of the column vector X _k , equation (3) can also be expressed in the following form. U = C _nm I (4) where I is a column vector of order m. That, I = _{[i 1, i 2, ......,} i m ] is given by ^T ...... (5).

The equation (3) or (4) is a system equation of the problem (inverse problem) of obtaining the current distribution from the magnetic field distribution measured on the surface of the object to be inspected, and generally the measurement point (known point or expression). N is the number of current input points (unknown) m
Since it is less, and n <m (6), it is impossible to uniquely obtain the current vector I. Therefore, in the inverse problem, some constraint condition should be found to find a unique solution, or all solutions satisfying the system equation without constraint condition should be obtained, and only physically meaningful solutions should be adopted. Some kind of ingenuity is required.

Next, a method of solving the system equation in the present invention will be described. (A) Single-input estimation Norm of vector U in equation (3) And the vector U is normalized to U ′ = (1 / u _N ) U (8) using this norm u _N. Here, u _{j on the} right side of the equation (7) is the vector U
The element of the j-th row of is shown.

Next, the norm x _kN is similarly set for the vector X _{k at the} current input point k. Then, the vector X _k is normalized. That is, X _k ′ = (1 / x _kN ) X _k (10) Here, x _{jk on the} right side of the equation (9) is a vector X _k.
The element of the j-th row of is shown. The vector U'of equation (8) and (1
The angle (vector matching) between the vectors X _k ′ in the expression 0) is evaluated by the following Cauchy-Schwarz relational expression. Here, γ _k has a value of −1 ≦ γ _k ≦ 1 (12). When | γ _k | = 1, the vectors X _k ′ and U
The angle becomes zero between ′, which means that the patterns match perfectly. Thus the current input point given as _k = 1~m | γ k | if the input point k = p which maximizes _{obtained, u N U'= i p x} pN X'p ...... (13 ) it is because, the current i _p is determined as _{i p = u N / x pN} ...... (14). In the case of obtaining a single input, the magnitude of the current can be uniquely obtained in an average sense for the input position regardless of the distance or the angle.

(B) Estimation of multiple inputs (b-1) Uniform current distribution Let us consider a case where a current is evenly distributed in the cross section of a conductor as shown in Fig. 2. This condition is static. This is true when the conductor is homogeneous in the magnetic field system.As a concrete example, it is assumed that a current i having a magnitude equal to the current input points k = 1 and k = 3 is flowing. , U is iX ₁ + iX ₃ (15), the norm u _N in equation (15) is Becomes Therefore the normalized vector U'is Therefore, the normalized vector U'is irrelevant to the magnitude of the input current i. This means that the pattern matching method that finds only the input position regardless of the magnitude of the input current is effective for the problem of finding a plurality of current input points with a uniform current distribution. Therefore, the vector X of each current input point
_{For k} (k = 1 to m), the norm x _kN of the equation (9) is obtained, and X _k is normalized by the equation (10) to obtain X _k ′. Thus assessing the angle between the obtained vectors X _k 'and U'and (11) in Cauchy-Schwarz relationship of expression. As a result, the most dominant (| γ _k | is closest to 1)
It is assumed that a different current input point p is obtained. Next, the vector of the other remaining current input points is added to this current input point p. That, X _p and ^(k), as a vector obtained by adding the vector X _k to a vector X _p, determining the ^{_{X p (k) = X p}} + X k (k = 1~m, k ≠ p) ...... (18) . Here, X _p is the vector of the first current input point p, and the vector of the remaining current input points of X _k . Norm of vector X _p ^(k ) in equation (18) And the vector X _p ^(k) is normalized as X _p ^(k) '= (1 / x _pN ^(k) ) X _p ^(k) (20). In equation (20), X _p ^(k) ′ and norm u
The pattern matching between the vectors U'normalized by _N is evaluated by the Cauchy-Schwarz relational expression (11). As a result, the current input point at which | γ _k | (k ≠ p) becomes the maximum becomes the second current input point. In the same manner, a current input point at which | γ _k | becomes maximum successively is added to obtain a plurality of current input points. Therefore, if there are multiple current input points,
It is possible to sequentially find the input current position by adding the vector by the current input point by the pattern matching method. However, when the input current is added in this way, if the current distributions are similar, the field patterns that they give to the measurement points will be the same pattern, and even if the shape is obtained as a result, the size is unique. You have to be careful that you are not asked to. This is also clear from the fact that a conductor having a circular cross section gives the same pattern to the periphery regardless of the radius.

[B-2] Inhomogeneous current distribution Due to the case where the conductor is not a uniform medium and the skin effect (instantaneous current distribution and magnetic field distribution is assumed in a system that changes with time), an uneven current in the conductor cross section Is distributed. When the current matching point is obtained by applying the above-described pattern matching method assuming uniform current distribution to this problem, the magnitude of the current density is expressed by the concentration degree of the current input point. That is, the part where the current input points are concentrated has a high current density,
The dispersed portion is represented as having a low current density. In this way, the macroscopic distribution of current can be estimated.

(C) Three-dimensional problem Three-dimensional current components include those in the x-axis direction as shown in FIG. 4 (a) and those in the y-axis direction as shown in FIG. 4 (b). Some include. Now, considering the magnetic fields H _na and H _nb in the normal direction to the yz plane at the measurement point on the y axis, both magnetic fields have the same current direction in the z direction perpendicular to the xy plane. Sometimes, their size is maximum. Therefore, H _na = (i / 2π) {ΔY / (ΔX ² + ΔY ² )} sinα (21a) H _nb = (i / 2π) {ΔY / (ΔX ² + ΔY ² )} cosβ (21b) Can write These means that when a two-dimensional model is locally applied to a three-dimensional problem, the result obtained is to extract the two-dimensional model with the most suitable component as the main component.
Therefore, the electric distribution of the three-dimensional problem can be evaluated by decomposing the three-dimensional space into two-dimensional models in directions orthogonal to each other and combining the results obtained by applying the two-dimensional model to each part.

Next, the magnetic field source (current dipole) and voltage source (voltage dipole) are obtained by measuring the magnetic field distribution, potential or electric field distribution on the surface of the object to be inspected, and the spatial power distribution is obtained from this. Will be described. The current dipole is a time derivative of the electric dipole, and the voltage dipole is a physical quantity defined by the product of the current dipole and the resistance.

Most of the problems of the electromagnetic field system are attributed to solving the Poisson type equation shown in the following equation. λ∇ ² ψ = −σ (23) Here, λ, ψ and σ represent the parameter of the medium, the potential or field, and the density of the source, respectively. Here, Green function G = 1 / (4π | r |)
Is used, the following equation is obtained as the integral system of the equation (23). ψ = ∫ [σ / (4π | r | λ)] dv (24) When current flows in the conductor, λ and φ in the equation (23)
And σ are conductivity k [S / m] and potential φ, respectively.
[V], time derivative of charge density −∂ρ / ∂t [C / m
³ S)]. According to the equation (24), the potential φ becomes φ = −∫ [(∂ρ / ∂t) / 4π | r | k)] dv (25). When the equation (25) is discretized into a minute space, Becomes Here, each term of Formula (26) is as follows.

U = [φ _1, φ _2, ... φ _n ] ^T α _i = − [(∂ρ _i / ∂t) / ΔV _i = P _Ui, i = 1 to m d _i = [1 / ( 4π)] [(n _i · a _1a, / r _1i ), (n _i · a _2a, / r _2i ) ... (n _i · a _na, / r _ni )] n: number of measurement points m: number of divisions : V = ΔV ₁ + ΔV ₂ + ... + ΔV _m (27) α _i (= P _vi [Vm]) n _i , a _ij and r _ij (i = 1 to m, j) in the equation (27). = 1 to n) are a voltage dipole, a unit vector in the direction of the voltage dipole, a unit vector in the direction of the measurement point from the source point, and a distance between the source point and the measurement point, respectively. Further, the magnetic field vector E on the conductor surface is obtained by taking the gradient of the scalar potential φ (E = − ▽ φ).

Next, the system equation of the magnetic field system becomes similar to the equation (26) according to the Biot-Ssvart law. Here, the respective parameters are U = [H ₁ , H ₂ , ... H _n ] ^T α _i = J _i ΔV ₁ = P _{1 i ,} i = _{1 to m} d _i = [1 / (4π)] [ _{(n i × a 1i / r} 1i 2), a _{(n i × a 2i / r} 2i 2) ··· (n i × a ni / r ni 2) ] ^T ...... (28).

In equation (28), J _i ΔV ₁ (= P _1i
[Am], i = 1 to m) is a current dipole, J _i is a current density in the minute deposition ΔV ₁ , H ₁ , H ₂ , ... H _n is a measured magnetic field, and other parameters are expressed by the equation (27). Similar to the one inside. The important thing to note here is
In the voltage dipole system, the scalar quantity is measured, and its basis vector d _i (i = 1 to m) is composed of the inner product, while in the current dipole system, the vector quantity is measured,
The basis vector d _i (i = 1 to m) is composed of outer products. Furthermore, since it is generally a measurement region or a finite surface region and the estimation target region is deposition, the following relational expression holds between the number of measurement points n and the number of divisions m in both systems.

M >> n (29) In the equation (26), the vector u is the vector d _i (i = 1 to m).
It is shown that a linear combination of is represented. Therefore (2
6) is Can be transformed. Therefore, the solution of the first group of equation (30) is as follows.

[U / | U |] ^T [d ₁ / | d ₁ |], [U / | U |] ^T [d ₂ / | d ₂ |], ... [U / | U |] ^T [D _h / | d _h |], [U / | U |] ^T [d _m / | d _m |] (31a) If γ _h = [U / | U |] ^T [d _h / | d _{If h} |] takes a maximum value, the solution of the second group is as follows with d _h as a pilot.

[U / | U |] ^T [d _h + d ₁ ) / | d _h + d ₁ |], [U / | U |] ^T [d _h + d ₂ ) / | d _h + d ₂ |], ... , 1,…, [U / | U |] ^T [d _h + d _m ) / | d _h + d _m |] (31b) The same process is repeated until the inner product, that is, the peak value of γ is obtained. As a general solution of the equation (26), α ₁ [| d ₁ | / | U |] ≈ [U / | U |] ^T {[d ₁ / | d ₁ |] + [d _h + d ₁ ) / | d _h + D ₁ |] + ……} α ₂ [| d ₂ | / | U |] ≈ [U / | U |] ^T {[d ₂ / | d ₂ |] + [d _h + d ₂ ) / | d _h + D ₂ |] + ……} ………………………… α _h [| d _h | / | U |] ≈ {[U / | U |] ^T {[d _h / | d _h |] + 1 + 1 + ……} ………………………… α _m [| d _m | / | U |] ≈ [U / | U |] ^T {[d _m / | d _m |] + [d _h + d _m ) / | d _h + d _m |] + ...} (32) is obtained.

In this way, the current dipole distribution is obtained from the magnetic field, and the voltage dipole distribution is obtained from the potential or the electric field, and the spatial power distribution is obtained by obtaining the inner product of the independent results.

Next, as a first verification example to which the present invention is applied, the result of estimating the voltage from the 6 × 6 = 36 voltage measurement points on the upper surface of the cube by the voltage dipole method is shown in FIG. ), A magnetic field perpendicular to 6 × 6 = 36 planes is measured at a plane 5 mm away from the upper surface of the cubic body, and the current distribution is estimated by the current dipole method. The result is shown in FIG. 5B, voltage dipole and current dipole. Figure 5 (c) shows the result of the spatial power distribution obtained by taking the inner product between the offspring.
And the correct current distribution is shown in FIG. 5 (d).

As a second verification example, 6 × 6 on the upper surface of the cubic body is used.
Fig. 6 (a) shows the results of estimating the voltage from the = 36 voltage measurement points using the voltage dipole method, and measures the magnetic field perpendicular to the 6 x 6 = 36 planes 5 mm away from the upper surface of the cube. 6 (b) shows the result of estimating the current distribution by the current dipole method, and FIG. 6 (c) shows the result of obtaining the spatial power distribution obtained by taking the inner product between the voltage dipole and the current dipole. The correct current distribution is shown in FIG.

As a third verification example, 6 × 6 on the upper surface of the cube is used.
= 36 results from estimating the voltage from 36 voltage measurement points using the voltage dipole method. Measure the magnetic field perpendicular to the 6 x 6 = 36 planes on the surface 5 mm away from the upper surface of the cube. Fig. 7 (b) shows the result of estimating the current distribution by the current dipole method, and Fig. 7 (c) shows the result of obtaining the spatial power distribution obtained by taking the inner product between the voltage dipole and the current dipole. The correct current distributions are shown in FIG.

From the results of FIGS. 5 to 7, it can be seen that the voltage dipole method shows a voltage distribution that spreads three-dimensionally, and the current dipole method shows a current distribution that spreads symmetrically in two dimensions. The difference between the two is considered as follows. The voltage dipole method uses a scalar quantity as a known number, and its basis vector is composed of an inner product. Furthermore, the voltage dipole method is essentially the source current − (∂ρ _i / ∂t) ΔV _{i in} equation (27).
It is established assuming (i = 1 to m). Also, the result is not a true current distribution but a divergent voltage distribution. These facts reflect the characteristic that the solution has a three-dimensional spread. On the other hand, the current dipole method uses a vector quantity as a known number, and its basis vector is composed of outer products. Therefore, the solution is a current distribution with a rotational characteristic.

From the above, when the current dipole method is used to calculate from the vertical component of the measured magnetic field, the current distribution that runs parallel to the measurement surface is obtained, and when the voltage dipole method is used to calculate from the potential on the measurement surface, the divergence is obtained. It is considered that this is effective for the three-dimensionally spreading current distribution. In this way, a solution pattern with some ambiguity can be obtained from both the voltage dipole method and the current dipole method.
Comparing (c) and FIG. 5 (d), FIG. 6 (c) and FIG. 6 (d), and FIG. 7 (c) and FIG. 7 (d), respectively, a power distribution near the correct current distribution is obtained. It turns out that is well concentrated. In particular, it can be seen that the closer to the measurement surface, the more accurate. Therefore, FIG. 5 (c), FIG. 6 (c), and FIG. 7 (c)
It can be seen that the most reliable current distribution can be obtained from the spatial power distribution of.

Next, an example in which the present invention is applied to diagnose a heart disease will be described with reference to FIGS. FIG. 8 (a),
8 (b) and 8 (c) are voltage distributions estimated from the electrocardiogram of a healthy subject using the voltage dipole method, current distributions estimated from the magnetocardiogram of the same healthy subject using current dipoles, respectively.
The spatial power distribution is the inner product of the two.

On the other hand, FIG. 9 (a), FIG. 9 (b) and FIG.
(C) is a voltage distribution estimated from the electrocardiogram of a person with each disease (Pulmonary Stenosis) using the voltage dipole method, a current distribution estimated from the magnetocardiogram of the same patient using the current dipole method, and both of them. It is the spatial power distribution that is the inner product.

8 (a), 8 (b), 9 (a) and 9 (b), the voltage and current dipole method has a pattern matching index γ for increasing the contrast when evaluating the result. Although it is necessary to display a certain value or more, the spatial power distribution diagrams of FIGS. 8 (c) and 9 (c) can omit such work, and can determine a normal heart and a diseased heart. Differences can be clarified. Thus, by combining the electrocardiogram and the magnetocardiogram to obtain the voltage distribution, the current distribution, and the spatial power distribution, more accurate diagnosis of the heart becomes possible.

[0038]

As described above, according to the present invention, the voltage distribution and the current distribution can be obtained with certain ambiguity by the voltage dipole method and the current dipole method, and the inner product of these can be obtained from the voltage distribution and the current distribution. The spatial power distribution can be obtained, and it is possible to make a reliable diagnosis by comprehensively looking at these.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a diagnostic device of the present invention.

FIG. 2 is a diagram showing an analysis model.

FIG. 3 is a diagram illustrating a method of obtaining a current distribution from a magnetic field measurement.

FIG. 4 is a diagram illustrating a three-dimensional problem.

FIG. 5 shows a voltage distribution, a current distribution in the first verification example,
It is a figure which shows a spatial power distribution and a correct current distribution.

FIG. 6 shows a voltage distribution and a current distribution in the second verification example.
It is a figure which shows a spatial power distribution and a correct current distribution.

FIG. 7 shows a voltage distribution, a current distribution in a third verification example,
It is a figure which shows a spatial power distribution and a correct current distribution.

FIG. 8 is a diagram showing a voltage distribution, a current distribution, and a spatial power distribution of the heart of a healthy person.

FIG. 9 is a diagram showing a voltage distribution, a current distribution, and a spatial power distribution of a heart of a diseased person.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Magnetic sensor 1, 2 ... Sensor which detects electric potential or electric field distribution, 3 ... Data processing device, 3a ... Magnetic field source distribution calculation means, 3b ... Voltage source distribution calculation means, 3c ... Spatial power distribution calculation means, 3d ... Normal Time data 3d, 3e ... Comparison means,
3 ... Output device, 3a ... Display device, 3b ... Printer.

Claims (3)

[Claims] 1. A detection means for detecting a magnetic field distribution and a potential or electric field distribution on a surface of an object to be inspected, and a magnetic field source distribution calculation means for calculating a magnetic field source distribution from the detected magnetic field distribution,
A voltage source distribution calculating means for calculating a voltage source distribution from the detected potential or electric field distribution, and a spatial power distribution calculating means for calculating a spatial power distribution by taking an inner product of the calculated magnetic field source distribution and voltage source distribution are provided. Diagnostic device. 2. The apparatus according to claim 1, further comprising comparing a result obtained in a normal state with respect to at least one of a magnetic field source distribution, a voltage source distribution, and a spatial power distribution with a measured and calculated result. And a means for identifying an abnormal part based on a comparison result. 3. The device according to claim 1, wherein
A diagnostic device further comprising an output means for outputting a calculation result or a comparison result.