JP2007244533A - Distortion distribution measuring system, elastic modulus distribution measuring systems, and methods of them - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a distortion distribution measuring system, an elastic modulus distribution measuring system and methods of them which can show high resolution to a micro-object to be measured such as an internal structure etc., of unstable plaque in a blood vessel. <P>SOLUTION: The distortion distribution measuring system 6 has an input part 21, data storage parts 24 and 25, an analysis part 22 and an output part 23, and analyzes distortion distribution 39 from measured data 32 before/after compression of the object 3 to be measured. The analysis part has: a mutual correlation analysis part 26 which calculates a mutual correlation coefficient about positional displacement of the measured data before/after the compression in a desired evaluation area of the measured data; a moving vector setting part 27 for setting a moving vector; a fault vector analysis part 28; a moving vector interpolation/correction analysis part 29; and a distortion distribution analysis part 30 which calculates a spatial variation ratio of consecutive moving vectors. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a strain distribution measurement system that analyzes strain distribution or elastic modulus distribution from ultrasonic echo data or optical interference data before and after compression of a measurement target, and in particular, a strain distribution measurement system that can execute data processing with high resolution. And an elastic modulus distribution measurement system and a method thereof.

Since the measurement of strain distribution or elastic modulus distribution in an object or living tissue is important information for knowing the structure or configuration of the object or living tissue, many studies have been conducted so far.
In recent years, especially in the medical field, images using ultrasonic echoes have been used in the field of non-destructive inspection for cracks due to structural defects inside materials that need to be detected in the manufacturing process of semiconductor elements, etc., and thermal fatigue of metal materials. Diagnosis is used for disease prevention and treatment and is effective. For example, acute coronary syndromes such as myocardial infarction are diseases caused by the rupture of unstable plaques in blood vessels, and the plaques are mainly lesions in which lipids are deposited on the blood vessel wall, which suppresses blood flow. In order to grasp the presence of this plaque, an image diagnosis using an intravascular ultrasonic echo method is performed.

  For example, in Patent Document 1, an ultrasonic beam is used under the name of “ultrasound diagnostic system, strain distribution display method and elastic coefficient distribution display method”, and a subject tissue is divided into a finite number of rectangular parallelepiped elements. A technique is disclosed in which a three-dimensional finite element model is formed and the elastic coefficient distribution information is estimated using the strain distribution information in an elastic equation, assuming that the elastic coefficient, stress, and strain are uniform in each element. In addition, in the estimation, in order to be able to estimate the displacement distribution corresponding to the lateral displacement and to reconstruct the elastic coefficient distribution from only the strain distribution in the ultrasonic beam direction, the correlation calculation means The correlation between the signals before and after compression of the subject tissue is calculated using the envelope signal output from the orthogonal detection means. In order to reduce the amount of calculation, the axial direction in the two-dimensional correlation window and this axial direction are calculated. A technique is disclosed in which a correlation coefficient is calculated for each scanning line that is discrete by a predetermined value in the orthogonal transverse direction.

Patent Document 2 discloses an invention named “Viscoelasticity estimation apparatus and program for soft tissue using ultrasonic waves”. The present invention receives an ultrasonic reflected wave emitted from an ultrasonic probe provided with a moving mechanism, calculates a deformation amount of an object shape from a time change of received data, and performs skin, fat like body tissue It is a technique that makes it possible to estimate elasticity, viscosity, and inertia for each layer even for soft tissues having a hierarchical structure with muscles, bones, and the like.
JP 2004-57652 A JP 2005-144155 A

  However, the conventional techniques described in Patent Document 1 and Patent Document 2 use ultrasonic waves, the resolution is about 100 μm at most, and random elements such as noise in the processing of acquired data. For example, it is difficult to increase the resolution. There has been a problem that the internal structure and instability of the plaque and the presence or absence of minute internal defects of the element cannot be determined in detail.

  The present invention has been made in response to such a conventional situation, and a strain distribution measurement system capable of exhibiting high resolution for a minute measurement object such as an unstable plaque in a blood vessel or an internal structure of a semiconductor element component. And an elastic modulus distribution measuring system and a method thereof.

In order to achieve the above object, a strain distribution measurement system according to claim 1 includes an input unit, a data storage unit, an analysis unit, and an output unit, and ultrasonic waves before and after compression of the measurement target. A strain distribution measurement system that analyzes a strain distribution from echo data or optical interference data (hereinafter collectively referred to simply as measurement data), wherein the input unit stores measurement data before and after compression in the data storage unit. The cross-correlation analysis unit for calculating a cross-correlation coefficient related to the positional displacement of the measurement data before and after compression in a desired evaluation region of the measurement data, and the cross-correlation analysis A moving vector setting unit that selects a starting point and an ending point of the position displacement that maximizes the cross-correlation coefficient analyzed in the unit, and sets a moving vector by compression having the starting point and the ending point; An error vector analysis unit that performs first statistical processing on a plurality of movement vectors in the vicinity of the movement vector to remove an error vector from the movement vectors, and a plurality of movements in the vicinity of the movement vector A motion vector interpolation / correction analysis unit that performs a second statistical process on the vector to make the distribution of the motion vector continuous, and a strain distribution analysis unit that calculates a spatial change rate of the continuous motion vector The data storage unit stores output data of at least one analysis executed in the analysis unit, and the output unit stores output data of at least one analysis executed in the analysis unit. It is a means that can be read out from the analysis unit or the data storage unit and output.
In the strain distribution measurement system with the above configuration, the cross-correlation analyzer calculates the cross-correlation coefficient for the position change of the measurement data, selects the start point and end point of the position displacement that maximizes the cross-correlation coefficient, and sets the movement vector. This has the effect of defining a movement vector having the most probable start and end points in the section.
In addition, the error vector analysis unit has an action of removing an error vector that may exist in the movement vector determined using the cross-correlation coefficient, and the movement vector interpolation / correction analysis unit interpolates the movement vector. By performing the process of correction and correction, the movement vector distribution is made continuous.
The strain distribution analysis unit has an effect of analyzing the strain distribution by calculating the spatial change rate of the continuous movement vector.
When optical interference data detected using an OCT (optical coherence tomography) optical system is used as measurement data, internal information (interference signal) of the measurement target is extracted, and the measurement target It has the effect of extracting not only the distribution on the surface of the surface but also the dynamic spatial distribution such as the amount of internal strain and elastic modulus from the internal interference data.

The strain distribution measurement system according to the second aspect of the present invention is the strain distribution measuring system according to the first aspect, wherein the first statistical processing is a standard obtained from a plurality of movement vectors in the vicinity of the movement vector. Processing for determining an error vector when a deviation width of a predetermined coefficient multiple of the standard deviation from the intermediate value exceeds or equals a predetermined width with reference to a statistical analysis value of deviation and intermediate value It is characterized by being.
The strain distribution measuring system having the above configuration has the same operation as that of the first aspect.

The strain distribution measurement system according to claim 3 is the strain distribution measurement system according to claim 1, wherein the first statistical processing includes a set of a plurality of cross-correlation coefficients in the vicinity of the movement vector, This is a process for obtaining a correlation with a set of a plurality of cross-correlation coefficients in the vicinity of other movement vectors adjacent to the movement vector.
The strain distribution measuring system having the above configuration has the same operation as that of the first aspect.

Furthermore, the strain distribution measuring system according to claim 4 is the invention according to any one of claims 1 to 3, wherein the cross-correlation analysis unit, the movement vector setting unit, the error vector analysis unit. And the movement vector interpolation / correction analysis unit has an iteration function, and the evaluation region in which the movement vector continuous by the movement vector interpolation / correction analysis unit exists is divided, and the mutual evaluation is performed again. The correlation analysis unit calculates a cross-correlation coefficient related to the position displacement of the measurement data before and after compression, and the movement vector setting unit determines the start point and end point of the position displacement where the cross-correlation coefficient analyzed by the cross-correlation analysis unit is maximum. Select, set a movement vector by compression having these start and end points, the error vector analysis unit determines and removes the error vector, Motion vector interpolation and correction analyzer is characterized in performing a continuous reduction of the distribution of the movement vectors.
In the strain distribution measurement system configured as described above, the iteration vector is subdivided by the iteration function of the cross-correlation analysis unit, the movement vector setting unit, the error vector analysis unit, and the movement vector interpolation / correction analysis unit, and the repeated movement vector is set. In this case, the error vector removal and the movement vector distribution continuation are performed.

The elastic modulus distribution measurement system according to claim 5 has an input unit, a data storage unit, an analysis unit, and an output unit, and analyzes the elastic modulus distribution from measurement data before and after compression of the measurement target. An elastic modulus distribution measuring system, wherein the input unit is a unit capable of inputting measurement data before and after compression and surface pressure data of the measurement target into the data storage unit, and the analysis unit includes the analysis unit A cross-correlation analysis unit that calculates a cross-correlation coefficient related to the positional displacement of the measurement data before and after compression in a desired evaluation region of the measurement data, and a position where the cross-correlation coefficient analyzed by the cross-correlation analysis unit is maximum Select a starting point and an ending point of displacement, set a moving vector by compression with these starting point and ending point, and a plurality of moving vectors in the vicinity of the moving vector An error vector analysis unit that removes an error vector from the movement vectors, and a second statistical process on a plurality of movement vectors in the vicinity of the movement vector. A movement vector interpolation / correction analysis unit for making the distribution of the movement vector continuous, a strain distribution analysis unit for calculating a spatial change rate of the continuous movement vector, and calculation in the surface pressure data and the strain distribution analysis unit An elastic modulus distribution analysis unit that calculates an elastic modulus using the spatial change rate, and the data storage unit stores output data of at least one analysis executed in the analysis unit, and the output unit Is a means capable of reading out and outputting output data of at least one analysis executed by the analysis unit from the analysis unit or the data storage unit.
In the elastic modulus distribution measuring system having the above configuration, in addition to the operation of the strain distribution measuring system according to claim 1, the elastic modulus distribution analyzing unit calculates the spatial change rate calculated by the strain distribution analyzing unit and the surface pressure of the non-measurement target. It has the effect of calculating the elastic modulus using the data.

According to a sixth aspect of the present invention, in the elastic modulus distribution measurement system according to the fifth aspect, the first statistical processing includes a standard deviation obtained from a plurality of movement vectors in the vicinity of the movement vector, and A process of determining an error vector when a deviation width of a predetermined coefficient multiple of the standard deviation from the intermediate value exceeds or equals a predetermined width with reference to a statistical analysis value called an intermediate value It is characterized by this.
The elastic modulus distribution measuring system having the above configuration has the same operation as the elastic modulus distribution measuring system according to claim 5.

  According to a seventh aspect of the present invention, in the elastic modulus distribution measurement system according to the fifth aspect, the first statistical processing includes a set of a plurality of cross-correlation coefficients in the vicinity of the periphery of the movement vector, This is a process for obtaining a correlation with a set of a plurality of cross-correlation coefficients in the vicinity of other movement vectors adjacent to the movement vector.

The elastic modulus distribution measurement system according to claim 8 is the invention according to any one of claims 5 to 7, wherein the cross-correlation analysis unit, the movement vector setting unit, the error vector analysis unit, and the The motion vector interpolation / correction analysis unit has an iteration function, and the cross-correlation analysis is performed again in the evaluation region divided by dividing the evaluation region where the motion vector continuous by the motion vector interpolation / correction analysis unit exists. The unit calculates the cross-correlation coefficient related to the position displacement of the measurement data before and after compression, and the movement vector setting unit selects the start point and end point of the position displacement that maximizes the cross-correlation coefficient analyzed by the cross-correlation analysis unit. Then, a movement vector by compression having these start point and end point is set, and the error vector analysis unit judges and removes the error vector, and the movement vector. Torr interpolation and correction analyzer is characterized in performing a continuous reduction of the distribution of the movement vectors.
In the elastic modulus distribution measurement system having the above-described configuration, as in the strain distribution measurement system according to claim 4, the iteration function of the cross-correlation analysis unit, the movement vector setting unit, the error vector analysis unit, and the movement vector interpolation / correction analysis unit Thus, after the evaluation area is subdivided, it is possible to repeatedly set the movement vector, remove the error vector, and execute the continuation of the distribution of the movement vector.

The strain distribution measuring method according to claim 9 is the strain distribution measuring method in which the computer analyzes the strain distribution from the measurement data before and after compression of the measurement target while the computer executes each process. A step of storing the measurement data in a readable manner in the data storage unit, and a cross-correlation coefficient related to the positional displacement of the measurement data before and after compression in the desired evaluation region when the cross-correlation analysis unit reads the measurement data from the data storage unit And the movement vector setting unit selects the maximum position displacement start point and end point using the analyzed cross-correlation coefficient as a key, and sets the movement vector by compression having these start point and end point. And an error vector analysis unit performs a first statistical process on a plurality of movement vectors in the vicinity of the movement vector, and the movement A step of removing an error vector from the vector, and a movement vector interpolation / correction analysis unit executes a second statistical process on a plurality of movement vectors in the vicinity of the movement vector, and calculates a distribution of the movement vector. The step of making it continuous, the step of calculating the spatial change rate of the movement vector by which the strain distribution analyzing unit is made continuous, and the step of outputting the distribution of the spatial change rate of the movement vector by the output unit.
In the strain distribution measuring method having the above configuration, the strain distribution measuring system described in claim 1 is regarded as the invention of the method, and has the same operation as the operation of the strain distribution measuring system described in claim 1. .

According to a tenth aspect of the present invention, in the strain distribution measuring method according to the ninth aspect, the first statistical processing includes a standard deviation and an intermediate value obtained from a plurality of movement vectors in the vicinity of the movement vector. This is a process of determining an error vector when a deviation width of a predetermined coefficient multiple of the standard deviation from the intermediate value exceeds or equals a predetermined width with reference to a statistical analysis value called a value It is.
The strain distribution measuring method having the above configuration has the same operation as the strain distribution measuring method according to claim 9.

The strain distribution measuring method according to an eleventh aspect of the present invention is the strain distribution measuring method according to the ninth aspect, wherein the first statistical processing includes a set of a plurality of cross-correlation coefficients in the vicinity of the movement vector, This is a process for obtaining a correlation with a set of a plurality of cross-correlation coefficients in the vicinity of other movement vectors adjacent to the movement vector.
The strain distribution measuring method having the above configuration also has the same operation as the strain distribution measuring method according to claim 9.

The elastic modulus distribution measuring method according to claim 12 is an elastic modulus distribution measuring method in which a new step is added to the strain distribution measuring method according to any one of claims 9 to 11. The step of storing the measurement data before and after compression in the data storage unit so that the input unit can be read includes a step of storing the surface pressure data of the measurement target, and the strain distribution analysis unit is a continuous motion vector After the step of calculating the space change rate, the elastic modulus distribution analyzer has a step of calculating an elastic modulus using the surface pressure data and the space change rate.
In the elastic modulus distribution measuring method having the above-described configuration, the elastic modulus distribution measuring system described in claims 5 to 7 is regarded as a method invention, and the elastic modulus described in claims 5 to 7 is used. It has the same operation as that of the distribution measurement system.

  In the strain distribution measuring system according to the first aspect of the present invention, a cross correlation coefficient related to a change in position of measurement data is calculated, a position displacement is selected, a movement vector is determined, an error vector is removed, and the movement vector By making the distribution continuous, it is possible to eliminate a random component with a high probability and to output a strain distribution with extremely high accuracy.

  Further, the strain distribution measuring system according to the second and third aspects of the present invention is the same as the effect of the first aspect of the present invention.

  In the strain distribution measurement system according to claim 4 of the present invention, the evaluation area is subdivided by the iteration function of the cross-correlation analysis unit, the movement vector setting unit, the error vector analysis unit, and the movement vector interpolation / correction analysis unit and repeated. By executing the calculation, it is possible to further improve the accuracy and resolution of the strain distribution.

  In the elastic modulus distribution measuring system according to the fifth aspect of the present invention, the elastic modulus distribution having a remarkably high accuracy can be output as in the first aspect of the invention.

  In the elastic modulus distribution measuring system according to claims 6 and 7 of the present invention, the same effect as the invention according to claim 5 is obtained.

  In the elastic modulus distribution measurement system according to the eighth aspect of the present invention, as in the strain distribution measurement system according to the fourth aspect, the cross-correlation analysis unit, the movement vector setting unit, the error vector analysis unit, the movement vector interpolation, By subdividing the evaluation area by the iteration function of the correction analysis unit, it is possible to further improve the accuracy and resolution of the strain distribution.

  The strain distribution measuring method according to the ninth to eleventh aspects of the present invention has the same effect as the strain distribution measuring system according to the first to third aspects.

  The elastic modulus distribution measuring method according to the twelfth aspect of the present invention has the same effect as the elastic modulus distribution measuring system according to the fifth aspect.

Hereinafter, a strain distribution measurement system and an elastic modulus distribution measurement system according to the best embodiment of the present invention, and further, a method thereof will be described with reference to FIGS.
FIG. 1 includes an elastic modulus distribution measurement system according to the present embodiment of the present invention, compresses a biological tissue that is an object to be measured, acquires optical interference data obtained before and after that, and acquires elasticity in the biological tissue. It is a conceptual diagram of the tissue elasticity image measurement system for measuring a rate distribution. In the present embodiment, optical interference data having a resolution higher than that of ultrasonic echo data itself is used as measurement data, and a living tissue is used as an object to be measured.
In FIG. 1, the tissue elasticity image measurement system 1 includes a linear driver 12 that applies a compressive force to the object 3 to be measured, and also irradiates the object 3 to be measured with low-coherence near-infrared light. The super luminescent diode 7 is provided as a light source. The low-coherence near-infrared light emitted from the superluminescent diode 7 is split by the beam splitter 8 into one traveling to the objective lens 10 and one traveling to the reference mirror 9. The low-coherence near-infrared light that has traveled to the objective lens 10 is irradiated onto the measurement object 3 and emits backscattered light. The backscattered light is transmitted again through the objective lens 10, travels to the beam splitter 8, and is combined with the reflected light of the low-coherence near-infrared light that travels to the reference mirror 9, and the optical interference data is detected by the photodetector 11. Is detected.
The reference mirror 9 is expressed as Vm / s in FIG. 1, but is scanned at a constant speed. According to the scanning of the reference mirror 9, since the low coherence light source has a wave train corresponding to the coherence length l _c , the interference of the wave train back-scattered at an arbitrary depth position by scanning the reference mirror 9 in the optical axis direction. A signal is obtained.
When the oscillation wavelength spectrum of the low-coherence light source has a Gaussian distribution, the coherence length l _c is expressed by Equation (1), and the spatial resolution in the depth direction is determined. Here, λ _c and Δλ are the center wavelength and full width at half maximum of the light source spectrum of the low coherence light source.

Based on this equation (1), a tomographic image can be calculated by scanning in the direction perpendicular to the optical axis together with depth scanning. The interference signal is obtained in the light condensing region (depth of field) of the low-coherence near-infrared light beam, and the spatial resolution in the direction perpendicular to the optical axis is determined by the beam spot diameter.
As described above, the detection by the OCT (optical coherence tomography) optical system in the present embodiment extracts internal information (interference signal). From the internal interference signal, not only the surface but also internal “distortion” is extracted. It extracts dynamic spatial distributions such as “quantity” and “elastic modulus”.
The optical interference data detected by the photodetector 11 is input from the photodetector 11 to the amplifier 4, further converted into a digital signal via the A / D converter 5, and input to the elastic modulus distribution measurement system 6b.

FIG. 2 shows a system configuration diagram of an elastic modulus distribution measurement system (strain distribution measurement system).
2, the elastic modulus distribution measurement system 6b includes an input unit 21 for inputting optical interference data from the A / D converter 5 shown in FIG. Of course, it is not always necessary to connect the input unit 21 to the A / D converter 5, the optical interference data digitized by the A / D converter 5, or via the amplifier 4 or the A / D converter 5. The raw optical interference data thus obtained is temporarily stored in some medium, and the input unit 21 is connected to this medium, or the raw data is input as the input unit 21 such as a keyboard. Also good.
The optical interference data 32 input from the input unit 21 is stored in the analysis input database 24 or directly input to the analysis unit 22. The optical interference data 32 may be arbitrarily selected from a desired data area and stored.
The analysis unit 22 includes a cross-correlation analysis unit 26, a movement vector setting unit 27, an error vector analysis unit 28, a movement vector interpolation / correction analysis unit 29, a strain analysis unit 30, and an elastic modulus analysis unit 31.
The analysis input database 24 stores general system data 33 including the tissue elasticity image measurement system 1 in addition to the optical interference data 32. For example, the superluminescent diode 7 which is a low coherence light source is stored. The light source wavelength λ _c of the light source spectrum, its full width at half maximum Δλ, and the like correspond to the system data 33, and the coherence length l _{c and the} like can be calculated by the equation (1).
In addition, the analysis input database 24 stores analysis condition data 34 and surface pressure data 35 for calculating the elastic modulus calculated by the elastic modulus analyzing unit 31.
The output obtained by analysis by the analysis unit 22 is stored in the analysis output database 25 and also output to the output unit 23.
In the analysis output database 25, the cross-correlation coefficient data 36, which is the result analyzed by the cross-correlation analysis unit 26, the vector data 37 related to the movement vector set by the movement vector setting unit 27, For example, vector statistical data 38 such as standard deviation, strain distribution data 39 analyzed by the strain analysis unit 30, and elastic modulus distribution data 40 analyzed by the elastic modulus analysis unit 31 are stored. .
Further, the analysis result obtained by the analysis unit 22 is also output to the output unit 23. The output unit 23 includes not only a display such as a CRT (cathode ray tube), a liquid crystal, a plasma display, and an organic EL (electroluminescence), but also an interface capable of transmitting data to other devices and systems.

Hereinafter, an analysis method in the analysis unit 22 will be described with reference to FIGS. FIG. 3 is a flowchart showing the steps of the elastic modulus distribution measuring method (strain distribution measuring method) according to the embodiment of the present invention. In addition, in order to clarify the relationship with the hardware that executes each process, the same components as those in the system configuration diagram disclosed in FIG.
First, the cross correlation analysis unit 26 sets the initial correlation analysis resolution as shown in step S1 of FIG. For this setting, the initial correlation analysis resolution stored in advance as part of the analysis condition data 34 in the analysis input database 24 may be automatically read and set, or the cross-correlation analysis unit 26 inputs the output to the output unit 23. A message prompting input from the unit 21 may be displayed, and the cross-correlation analyzing unit 26 may read out from the input unit 21 after waiting for an input from the input unit 21 by the system user.
The initial correlation analysis resolution will be described in the resolution enhancement in step S6, which will be described later. However, it is possible to perform analysis with higher accuracy by executing iteration that gradually increases the resolution of analysis. The initial value is used to input a low resolution in advance.

Next, as shown in step S <b> 2 in FIG. 3, the cross correlation analysis unit 26 reads the optical interference data 32 using the optical interference data 32 input from the input unit 21 or from the analysis input database 24.
The optical interference data 32 includes a speckle pattern in a predetermined inspection region before and after compression by the linear driver 12 of the measurement object 3 shown in FIG. By analyzing the cross-correlation with respect to the pattern (step S2), the displacement before and after compression with respect to the measurement object 3 is obtained, and thereby the movement vector is analyzed (step S3).
The cross-correlation analysis unit 26 evaluates the local speckle pattern similarity using the correlation value R _{i, j} represented by the equation (2), as shown in step S2.

f (X _i , Z _j ) and g (X _i , Z _j ) are the center position (i, j) set in the optical interference data 32 before and after compression, and the speckle in the M _x × M _z pixel inspection region・ It is a pattern.
Within the search area (N _x × N _z pixels), the inspection area is translated, and correlation values R _{i, j} (ΔX, ΔZ), which are cross-correlation coefficients, are calculated for each movement amount. ΔX and ΔZ are respective movement amounts in the X-axis direction and the Z-axis direction shown in FIG.
As a result, as shown in the equation (3), the movement amount U _{i, j} giving the maximum correlation value (an arrow as a vector notation is attached to the upper part of U) is transferred to the movement vector setting unit 27 in step S3. As shown, it can be determined as a movement vector before and after compression.

As shown in the equation (2), f (a bar as an average value is added to the upper part of f) and g (a bar as an average value is attached to the upper part of g) are expressed as f ( X _i , Z _j ) and g (X _i , Z _j ) represent the average value in the inspection region, and the deviation from the average value is used and the correlation value is normalized.

Here, a specific example regarding the analysis of the correlation value and the setting of the movement vector will be described with reference to FIG.
FIG. 4 is a conceptual diagram for explaining the cross-correlation analysis performed in the cross-correlation analysis unit 26. In FIG. 4, the vertical axis indicates the Z direction shown in the vicinity of the measurement target object 3 in FIG. 1, and the horizontal axis also indicates the X direction. A region indicated by a dotted line with a reference symbol N is a search region (5 × 5 pixels), and a region indicated by a solid line with a reference symbol M is an inspection region (3 × 3 pixels). The portion indicated by the symbol P with a cross in the center is the center position, and the coordinates are indicated by (6, 6). When this inspection area is translated in the exploration area, there are eight ways to move, up and down, left and right, diagonally upper right, upper left, lower right, and lower left, and the correlation value can be calculated for each movement amount. Is possible. The start point and end point of this parallel movement are taken as the start point and end point of the displacement of the measurement object 3 before and after compression. The coordinates other than the center position included in the inspection area are circled, but when the center position is translated in eight directions, the speckle pattern at the center position and each position This means the magnitude of the correlation value calculated with the speckle pattern.
In FIG. 4, since the correlation value of the speckle pattern obtained by translating to the upper right is the largest, it is considered that there is a displacement from the center position to the coordinate indicated by the sign Q in the upper right direction. It is possible to select the position (6, 6) as the start point and the upper right coordinate (7, 7) as the end point.
As described above, the cross-correlation analysis unit 26 calculates the correlation value of the speckle pattern while translating the inspection region within the exploration region.
Then, the movement vector setting unit 27 sets the movement vector 45 using the point giving the maximum value among the correlation values obtained by the cross correlation analysis unit 26 as the end point of displacement with respect to the center position of the inspection region. . In FIG. 3, this is the step shown as the movement vector analysis in step S3.
The correlation values R _{i, j} analyzed by the cross correlation analysis unit 26 in this way are stored as cross correlation coefficient data 36 in the analysis output database 25 by the cross correlation analysis unit 26 and further output to the output unit 23. You may make it do.
The vector data 37 relating to the movement vector 45 set by the movement vector setting unit 27 is also stored in the analysis output database 25 by the movement vector setting unit 27. You may transmit and you may output to the output part 23.
In FIG. 4, an inspection area and an exploration area divided into rectangular meshes are conceptualized by adopting an orthogonal coordinate system based on the X-Z axis, but the invention is not particularly limited to the orthogonal coordinate system, and various A coordinate system may be used.

When the movement vectors have been set for the center positions of all the inspection areas as described above, a set of movement vectors as shown in FIG. 5 is considered. Each of the arrows shown in FIG. 5 is a movement vector having a start point and an end point, and the end point is a result of analyzing the correlation value around the start point by the cross-correlation analysis unit 26 as described above. The coordinates having the highest correlation value are selected and set as the end point.
As shown in FIG. 5, the movement vectors 45 set in this way are often not all directed in the same direction. This is because an error vector is included in the movement vector. This error vector is based on the fact that the speckle pattern contains less noise and less information on the speckle pattern itself.
In FIG. 5, it appears that what is considered to be an error vector whose direction is different from that of other movement vectors, as indicated by the symbol K, is included.
Therefore, in the present invention, in order to eliminate the error vector, the error vector analysis unit 28 is provided in the analysis unit 22.
The error vector analysis unit 28 eliminates an error vector as indicated by a symbol K in FIG. 5, but has the configuration shown in FIG. 6 in detail.

FIG. 6 is a system configuration diagram of the error vector analysis unit. In FIG. 6, a standard deviation analyzing unit 46 and an intermediate value analyzing unit 47 for analyzing the movement vector are determined in order to determine whether the error vector is an error vector, and it is determined whether the error vector is an error vector based on the analysis result. In addition, an error vector determination unit 48 that excludes the error vector data from the vector data 37 and a movement vector error correlation analysis unit 49 that excludes the error vector independently are provided. In the flowchart shown in FIG. 3, it is a process shown as step S4.
The standard deviation analysis unit 46, the intermediate value analysis unit 47, the error vector determination unit 48, and the analysis unit of the movement vector error correlation analysis unit 49 are the standard deviation analysis unit 46, the intermediate value analysis unit 47, and the error vector determination unit 48. It is necessary for the set, but it does not always require the moving vector error correlation analyzing unit 49, and one set of the standard deviation analyzing unit 46, the intermediate value analyzing unit 47 and the error vector determining unit 48 and the moving vector error correlation analyzing unit It is sufficient that at least one of 49 is provided. Moreover, even if it is a thing other than these, what is necessary is just the analysis part which can analyze the movement vector which remove | deviates from the predetermined reference | standard from several movement vectors as an error vector, and can be excluded, and the analysis method does not ask | require It is. The predetermined standard is input and stored in the analysis input database 24 as a part of the analysis condition data 34 via the input unit 21, or an error vector analysis unit 28 inputs the input from the input unit 21. It is good to read and use.

Next, a specific analysis by the standard deviation analysis unit 46, the intermediate value analysis unit 47, and the error vector determination unit 48 will be described.
In the standard deviation analysis unit 46, first, data related to the movement vector set by the movement vector setting unit 27 is read from the movement vector setting unit 27, or vector data 37 to be analyzed is read from the analysis output database 25. The movement in the respective directions (X-axis direction and Z-axis direction) with respect to the nine movement vectors existing in the 3 × 3 pixel inspection region of the movement vector indicated by the reference symbol M described with reference to FIG. For each quantity, the standard deviations σ _x and σ _z are analyzed, and the intermediate value analyzing unit 47 similarly analyzes the intermediate values ΔX _m and ΔZ _m . The analyzed standard deviation and intermediate value are both stored in the analysis output database 25 as the vector statistical data 38 or output from the output unit 23 by the standard deviation analysis unit 46 and the intermediate value analysis unit 47, respectively.
Then, the data of the result is transmitted to the error vector determination unit 48. If the error vector determination unit 48 does not satisfy either of the expressions expressed by the expressions (4) and (5), the movement is performed. The vector is determined as an error vector.
Expression (4) is an expression for the X-axis direction, and Expression (5) is an expression for the Z-axis direction. Further, κ included in the equations (4) and (5) is a predetermined constant, and is this constant also stored in the analysis input database 24 as analysis condition data 34 via the input unit 21 in advance? The error vector determination unit 48 reads out the input from the input unit 21 and inputs to the equations (4) and (5) for analysis.
In the present embodiment, the same κ is used in formulas (4) and (5), but those defined as independent constants may be used as κ _x and κ _z , respectively. Further, in the present embodiment, the case of executing with 3 × 3 pixels has been described, but it goes without saying that the number of pixels may be appropriately increased or decreased.

Data relating to the movement vector determined as the error vector using the standard deviation analysis unit 46, the intermediate value analysis unit 47 and the error vector determination unit 48 are deleted from the vector data 37 by the error vector determination unit 48.
It is one of the specific examples of the first statistical processing in the present invention to judge and eliminate error vectors by the equations shown in equations (4) and (5).
A representation of this situation is shown in FIGS. FIG. 7A shows a state in which the error vector indicated by the symbol K and the movement vector 45 are mixed, and FIG. 7B shows that the error vector is not determined after the error vector is deleted. Vector data 37 consisting of a set of only movement vectors 45 is shown.

Next, a specific analysis of error vector elimination by the movement vector error correlation analysis unit 49 will be described.
Whereas the standard deviation analysis unit 46 and the intermediate value analysis unit 47 analyze the standard deviation and the intermediate value for each movement vector unit, the movement vector error correlation analysis unit 49, for example, shows a correlation value (mutual value in a neighboring examination region). Correlation coefficient sets (this set of correlation values is referred to as a correlation map in the present specification) are further correlated to eliminate random speckle noise that causes an error vector.
A correlation map 1 [R ₁ (i, j)] is created in order to calculate the movement amount of a certain examination region (tissue 1), and the maximum correlation value within the map indicates the movement position of the tissue 1 as described above. The movement vector is determined. This is similarly performed in the examination region (tissue 2) close to the tissue 1, and a correlation map 2 [R ₂ (i, j)] is created to determine the movement position of the tissue 2 to form a movement vector. Is.
When the examination region is set at a very close position so that the organization 1 and the organization 2 overlap, both tissues, that is, the speckle information in both the examination regions overlaps. Correlation map 1 [R ₁ (i, j)] calculated by organization 1 and organization 2 and correlation map 2 [R ₂ (i, j)] also have similar patterns.
On the other hand, it is known that random speckles due to multiple scattering are always mixed in the speckle information of the tissues 1 and 2 as noise. This random speckle noise appears as randomness of the correlation value in the correlation map 1 and the correlation map 2. When this randomness is strong, that is, when the refractive index distribution of the object to be measured changes randomly, the similarity between the correlation map 1 and the correlation map 2 is hindered, and noise is generated in the setting of the movement vector. This is the main cause of reducing the SN ratio of the vector.
Therefore, the correlation map 1 and the correlation map 2 are multiplied or calculated to obtain a correlation. This is because the randomness of the correlation values of the correlation map 1 and the correlation map 2, that is, random speckle noise settles to zero by the randomness by multiplying or correlating, whereas the correlation map 1 and the correlation map The similarity between the two is based on the principle that it can be enhanced.
Specifically, when S (i, j) represented by Equation (6) is analyzed and the maximum value of S (i, j) is obtained, the maximum value is obtained from correlation map 1 and correlation map 2. In order to represent a similar position, this indicates a tissue movement amount to be calculated, that is, a movement vector excluding an error vector.
The statistical process for obtaining the maximum value of S (i, j) is also a specific example of the first statistical process of the present invention.

When the moving vector error correlation analysis unit 49 is used, the random speckle noise itself, which is the main cause of the error vector, converges to zero by multiplication or correlation, so that the error vector itself disappears after all. As a result, only the movement vector 45 that does not include the error vector remains as the vector data 37.
The movement vector determined as the error vector is deleted from the vector data 37 read from the analysis output database 25 by the error vector analysis unit 28.
The vector data 37 from which the error vector is eliminated by using the error vector analysis unit 28 is stored as new vector data 37 in the analysis output database 25 by the error vector analysis unit 28 or directly as it is, the direct movement vector interpolation / correction analysis unit 29. Or output to the output unit 23.

Next, the movement vector interpolation / correction analysis unit 29 of the analysis unit 22 shown in FIG. 2 will be described. In FIG. 3, the movement vector interpolation / correction analysis step shown in step S5 is performed.
In this movement vector interpolation / correction analysis unit 29, the error vector is excluded from the new vector data 37, and the movement vector 45 is missing in some places. Therefore, processing for compensating for this is performed.
This is because the distortion distribution to be obtained is a spatial differential coefficient of the movement vector 45, and therefore measurement errors increase if there is a gap.
In the movement vector interpolation / correction analysis unit 29, first, new vector data 37 is read from the analysis output database 25, or new vector data 37 transmitted from the error vector analysis unit 28 is read and included in the vector data 37. For the movement vector 45, a least square method, which is one of statistical methods, is locally applied to perform interpolation and correction processing for adding continuity to the vector distribution.

Specifically, description will be made with reference to FIGS. 8 (a) and 8 (b).
As shown in FIG. 8A, a second-order polynomial approximation is performed on the vicinity of the periphery 24 indicated by the reference symbol U around the movement vector indicated by the reference symbol S, and the movement of the vicinity of the peripheral 8 indicated by the reference symbol T is performed. Vector interpolation and correction processing are performed. Interpolation processing is an error vector in which a motion vector is missing in a conceptual cell (region), and a motion vector generated by statistical processing such as quadratic polynomial approximation is added to the vector data 37. This means that the movement vector is added, and the correction process is generated by statistical processing such as quadratic polynomial approximation when a movement vector that is not an error vector exists in a conceptual cell (region). This is a process of replacing the vector data 37 with the replaced movement vector.
The physical quantity to be subjected to quadratic polynomial approximation is an amount related to the magnitude and direction of the vector, which is included in the vector data 37, and is obtained from the coordinates of the start point and end point of the vector.
The interpolation range is overlapped by 50%, and the average value of the vector amount is used as the corrected movement vector in the overlapping portion.

FIG. 8B is a conceptual diagram showing new vector data 37 after the movement vector interpolation / correction by the movement vector interpolation / correction analysis unit 29 is completed.
In FIG. 8B, the movement vector 50a is a movement vector that was originally deleted as an error vector and the missing movement vector was interpolated by quadratic polynomial approximation, and the movement vector 50b is the same as that in FIG. As shown in FIG. 8 (b), the original motion vector originally existed as a motion vector, but its direction was slightly different from that of other motion vectors. The movement vector is as follows.
In the present embodiment, as a specific example of the second statistical processing of the present invention, interpolation and correction processing by quadratic polynomial approximation using the least square method is shown. In any case, any method may be used as long as it is a statistical method, such as polynomial approximation higher than the second order and spline interpolation, as long as interpolation and correction processing of the movement vector is possible.

  Data relating to the movement vector obtained by the interpolation processing and correction processing is stored as new vector data 37 in the analysis output database 25 by the movement vector interpolation / correction analysis unit 29 or output to the output unit 23. Also good. Further, the data related to the approximate expression may be stored as the vector statistical data 38 of the analysis output database 25 by the movement vector interpolation / correction analysis unit 29.

Next, the distortion analysis unit 30 shown in FIG. 2 will be described. In FIG. 3, steps S6 and S7 are shown. First, the flow for performing strain distribution analysis and elastic modulus distribution analysis shown in steps S8 and S9 in a situation where the resolution is not enhanced will be described.
The distortion analysis unit 30 defines the distortion as the spatial change rate of the corrected movement vector shown in Expression (7) and obtains the distortion distribution. In Expression (7), ε ^x _{i, j} represents the strain in the X-axis direction, and the numerator represents the difference between the X-axis components of the movement vectors of adjacent cells. Similarly, ε ^z _{i, j} represents the strain in the Z-axis direction, and its numerator represents the difference between the Z-axis components of the movement vectors of adjacent cells. ΔX ^max _{i, j} , ΔZ ^max _{i, j, and the} like indicate the movement amount ΔX on the X axis and the movement amount ΔZ on the Z axis that give the maximum correlation value shown in Expression (3).
By obtaining the distortion for the movement vector of each cell, it becomes possible to analyze the distortion distribution as a whole.

Here, ΔL _x and ΔL _z represent spatial resolution. Considering a cell on the XZ plane divided into square meshes as shown in FIG. 5, if the length of the cell is ΔL, this becomes the spatial resolution, and if applied to equation (7), ΔL = ΔL _x = ΔL _z .
In addition, the spatial resolution based on the coherence length l _c shown in the equation (1) represents a theoretical limit resolution that cannot theoretically be increased any more, and ΔL in the equation (7) is delimited when considering the movement vector. For example, it can be said that the spatial resolution is obtained by the size of the square mesh. Accordingly, the mesh by a reduced gradually, but the spatial resolution is increased, spatial resolution in view of the small mesh below the coherence length l _c are those not be a more sophisticated.
Data relating to the strain analysis result obtained by the strain analysis unit 30 is stored as strain distribution data 39 shown in the analysis output database 25 by the strain analysis unit 30. Further, when the elastic modulus distribution is continuously analyzed by the elastic modulus analyzing unit 31, it may be transmitted to the elastic modulus analyzing unit 31 and output to the output unit 23.
In the present embodiment, the distortion is defined and calculated as the spatial change rate of the corrected movement vector shown in Equation (7) to obtain the distortion distribution. However, the definition as shown in Equation (7) is used. In addition, the interpolation processing executed by the movement vector interpolation / correction analysis unit 29 in the previous stage (step S5) and the differential coefficient of the interpolation function (for example, higher-order equation or spline function) obtained during the correction processing are also used. Definitions are possible and may be used. That is, these interpolation functions are stored in advance in the strain analysis unit 30 or stored as analysis condition data 34 in the analysis input database 24 and read out by the strain analysis unit 30 or movement vector interpolation / correction analysis. The distortion analysis unit 30 may read the interpolation function obtained by the unit 29 from the movement vector interpolation / correction analysis unit 29 as it is.

Next, the elastic modulus analysis unit 31 shown in FIG. 2 will be described. In FIG. 3, the process is shown as step S9.
The elastic modulus analyzing unit 31 analyzes the elastic modulus distribution data 40 using the strain distribution data 39 obtained by the strain analyzing unit 30 because the elastic modulus is obtained by dividing the stress by the strain. .
Data relating to stress is stored in advance in the analysis input database 24 as surface pressure data 35. This surface pressure data 35 is obtained by measuring the stress on the surface of the measurement object 3 and is obtained based on the premise that the stress value is uniform and has no distribution. However, if the stress value is obtained quantitatively as some kind of stress distribution, of course, the data may be used. That is, the stress distribution data measured by another system and the analysis data of the stress distribution analyzed by another system may be stored as the surface pressure data 35 in the analysis input database 24, respectively, or during the analysis. It may be input via the input unit 21 and read by the elastic modulus analysis unit 31.
The elastic modulus analysis unit 31 first reads the strain distribution data 39 from the analysis output database 25 or directly reads the strain distribution data 39 from the strain analysis unit 30 and further reads the surface pressure data 35 from the analysis input database 24.
Then, for each cell (region) as conceptualized in FIG. 5, the modulus of elasticity in each cell is calculated by executing the calculation of dividing the surface pressure data 35 by the strain distribution data 39. These are collectively elastic modulus distribution data 40 and are stored in the analysis output database 25 by the elastic modulus analyzer 31. Further, it may be output to the output unit 23.

Further, a description will be added of the step of performing the iteration by comparing the resolution of the analysis result of step S6 shown in FIG. 3 and the resolution of the analysis result of step S7 with the theoretical limit resolution.
One important feature of the present invention is this iteration function. In FIG. 3, the initial correlation analysis resolution is set as shown in step S1, the resolution is enhanced by the steps S6 and S7, and the analysis according to the steps S3 to S5 is further executed. By gradually increasing the resolution of the analysis, it is possible to perform strain distribution analysis or elastic modulus distribution analysis with high accuracy with few errors.
First, the initial correlation analysis resolution is set by the cross correlation analysis unit 26 as described above. The predetermined initial correlation analysis resolution is given as the length of the cell in the inspection area shown in Equation (7).
When the analysis is performed using the initial correlation analysis resolution and the processing is executed up to step S5, the cross-correlation analysis unit 26 next increases (decreases) the initial correlation analysis resolution. As described above, the cross-correlation analysis unit 26 reads the iteration information included in the analysis condition data 34 stored in the analysis input database 24 before analysis, or iterates the output unit 23 during the analysis. It is preferable to display a display for inquiring to the user of this system whether or not to execute, and to read the iteration information input from the input unit 21.
As a specific method for increasing the analysis resolution, a predetermined fixed resolution may be subtracted and set as a new resolution, or may be set as, for example, 90% or 50% of the initial correlation analysis resolution. However, in the present embodiment, the inspection region is divided into ¼, that is, ΔL in Expression (7) is halved and the process proceeds to the next stage where the iteration is started.
In the next stage, since the resolution is increased, one side (ΔL) of the cell in the inspection area and the exploration area is small. However, the data used for the analysis here is the error vector analysis unit 28 or the movement in the previous stage. This is vector data 37 newly obtained by the vector interpolation / correction analysis unit 29.

With reference to the new vector data 37, the cross-correlation is again obtained using the cross-correlation analysis unit 26 with respect to the movement vector in the cell that has become finer due to the increased resolution, and the movement vector setting unit 27 is used. The movement vector is set in a fine cell, the error vector analysis unit 28 eliminates the error vector, and the movement vector interpolation / correction analysis unit 29 executes the movement vector interpolation / correction.
That is, the cross correlation analysis unit 26, the movement vector setting unit 27, the error vector analysis unit 28, and the movement vector interpolation / correction analysis unit 29 each have an iteration function. Here, having an iteration function means that each analysis unit or setting unit has a function that enables repeated analysis with improved resolution. In this embodiment, it is the cross-correlation analysis unit 26 that instructs the selection for analysis.
However, the selection of exhibiting the iteration function is not particularly limited to the cross-correlation analysis unit 26, and other analysis units such as the strain analysis unit 30 and the elastic modulus analysis unit 31 may be used.
In the next stage, the data obtained by these analysis units and setting units are stored in the analysis output database 25 as cross-correlation coefficient data 36, vector data 37, and vector statistical data 38, respectively, as in the previous stage. Or is output to the output unit 23. Further, the data of the next stage may be overwritten on the data of the previous stage, or may be newly stored as other data of the next stage.
Using the vector data 37 obtained at the next stage, the strain analysis unit 30 and the elastic modulus analysis unit 31 further analyze the strain distribution in the same manner as described above to analyze the elastic modulus distribution.
Next, the resolution is further increased than the resolution in the next stage. In the present embodiment, as will be described later as an example, the inspection region is further reduced to ¼ and ΔL is further halved.
However, the spatial resolution obtained in this way eventually becomes smaller than the coherence wavelength lc shown in Expression (1). In this case, the analysis ends because the resolution exceeds the theoretical limit. This is expressed in step S7 of FIG. The theoretical limit resolution shown in step S7 is the coherence wavelength lc shown in equation (1). Since the spatial resolution is not improved even if the cells are further divided, the process for increasing the resolution is terminated. In this embodiment, since the resolution is improved as ΔL or the coherence wavelength l _c becomes smaller, an inequality sign is shown as in step S7 in FIG. 3, but other indices are used as the resolution. When expressed, if the resolution increases as the index increases, the inequality sign in step S7 is reversed.
Further, in FIG. 3, it is a process in which step S8 and step S9 are processed through step S6 and step S7, but it is not always necessary to execute an iteration from step S7 to step S2, and from step S5. It goes without saying that it is possible to proceed to steps S8 and 9, and even if steps S6 and S7 are included, the iteration from step S7 to step S2 may be performed after executing steps S8 and 9 each time.
Further, as step S10-1 and step S10-2, an analysis result output process and an analysis result storage process are shown, respectively, but this is also given a process code for convenience rather than a process performed after step S9. As described above, the result analyzed in each step is appropriately output to the output unit 23 or stored in the analysis output database 25 as described above.

As described above, the elastic modulus distribution measurement system 6b and the elastic modulus distribution measurement method have been described as embodiments. In this elastic modulus distribution measurement system 6b, the surface pressure data 35 stored in the elastic modulus analysis unit 31 and the analysis input database 24 is described. The strain distribution measurement system 6a is conceptualized as a component that does not include the above-described components, and the other components commonly include the components described so far and exhibit the same functions and effects. Of course, in the strain distribution measurement system 6a, since there is no elastic modulus analysis unit 31, the elastic modulus distribution data 40 is not analyzed, and is not stored in the analysis output database 25 but may be output to the output unit 23. Absent.
Further, the embodiments of the elastic modulus distribution measuring method and the strain distribution measuring method have been described with reference to FIG. 3, but in the strain distribution measuring method, as in the relationship between the strain distribution measuring system 6 a and the elastic modulus distribution measuring system 6 b. In FIG. 3, step S9 in FIG. 3 does not exist and there is no surface pressure data input process described in the input unit 21, but other processes exist in common with the elastic modulus distribution measuring method.

The strain distribution and the elastic modulus distribution obtained as described above can detect the presence of unstable plaque in the blood vessel with high accuracy in the treatment for acute coronary syndrome such as myocardial infarction. In particular, by obtaining the elastic modulus distribution, it is possible to clarify the existence of unstable plaque because the elastic modulus is different between blood vessels and unstable plaque, and it is also possible to visualize biological tissue structure . Of course, it can be applied to biological tissues other than blood vessels and can be used in various ways.
Further, by measuring data relating to optical interference using a light source such as near-infrared light capable of higher resolution than ultrasonic waves, the effect of high resolution relating to processing of measurement data in the present embodiment In combination with this, it is possible to obtain a more accurate strain distribution or elastic modulus distribution.
With such strain distribution measurement system and elastic modulus distribution measurement system, or strain distribution measurement method and elastic modulus distribution measurement method, in the medical field, not only early detection of circulatory disease, but also diagnosis of disease, quantification of treatment effect It can be used for various purposes in medical practice such as general grasping and healing confirmation, and in the engineering field, it is possible to conduct more accurate diagnosis and inspection in nondestructive inspection etc. as mentioned above Become.
In the present embodiment, the case where optical interference data is used has been described. However, as described at the beginning, an ultrasonic echo signal is measured using an ultrasonic source instead of a light source. It is established as a strain distribution measurement system and an elastic modulus distribution system. In addition, when measuring an ultrasonic echo, it is necessary to use an ultrasonic echo system instead of the OCT optical system 2 shown in FIG. Since this is sufficient, the description will be omitted in the present application.
Hereinafter, since experiments were conducted as specific examples according to the present embodiment, the results will be described.

The experiment was performed with the configuration shown in FIGS. Although this experiment is carried out as a case where optical interference data is acquired, the same applies to the case where ultrasonic echo data is acquired by adopting an ultrasonic echo system instead of the OCT optical system.
As the superluminescent diode 7 of the light source, a superluminescent diode having a center wavelength λ _c = 831 nm and a full width at half maximum Δλ = 55 nm was used. The theoretical value of the coherence wavelength l _{c in} equation (1) is 5.54 μm, but considering the spectral shape, the spatial resolution in the optical axis direction (Z direction) is considered to be about 10 μm due to the influence of side lobes.
An aspherical molded glass lens (NA = 0.4, f = 6.25 mm) was used as the objective lens 10 and focused on the object 3 to be measured.
Since the beam spot diameter of 6.61 μm is the spatial resolution in the direction perpendicular to the optical axis (X direction), scanning in the X direction was performed every 4 μm. In the system shown in FIG. 1, by scanning the reference mirror 9 at a constant speed (Vm / s), the Doppler shift frequency component (f _D = 2ν / λ _c ) of the interference signal is amplified using the amplifier 4, Optical interference data was obtained.
The depth of field is 330 μm, and a coherence gate is applied in this region to obtain optical interference data. A linear driver 12 is installed to compress the sample that is the object 3 to be measured. The uniform sample is compressed by 50 μm and the non-uniform sample is compressed by 20 μm in the direction perpendicular to the optical axis, and optical coherence tomographic images before and after compression (optical interference data). )
Since the sample is as small as 1 cm on a side, the linear driver 12 can be regarded as compressing the entire sample end surface uniformly.
The range of the optical coherence tomographic image was X × Z = 1200 × 1200 μm, and 4 μm was imaged as one pixel (X × Z = 300 × 300 pixels). The test sample was a uniform sample prepared by dissolving 1 g of agar powder (elastic modulus: 3.92 to 4.90 kPa) in 60 ml of hot water and mixing fine polystyrene particles (particle diameter: 1 μm).
A compression experiment was conducted using a non-uniform sample in which a fishing line (elastic modulus: 1.47 to 2.45 GPa) having an outer diameter of 275 μm was mixed.

FIG. 9A shows an optical coherence tomographic image of a uniform sample. The white spots in the image are speckle patterns due to the mixed polystyrene particles. FIG. 10 shows the movement vectors obtained by this method for each stage. Note that the sizes of the inspection area and the exploration area in the first stage were set to M _x = M _z = 320 μm and N _x = N _z = 400 μm.
As is clear from FIGS. 10A to 10C, each time the stage is moved up from the first stage, the resolution of the movement vector is increased by the recursive correlation processing by iteration. The final stage having the highest resolution is determined by the spatial resolution of the optical coherence tomographic image and the information amount of the speckle pattern in the examination region. In the experiment in this example, the speckle pattern particle image is about 10 μm, and the spatial resolution of the optical coherence tomographic image is considered to be about 10 μm. Therefore, the fourth stage having a spatial resolution of ΔL = 20 μm was set as the final stage. This is shown in FIG. From the movement vector distribution of each stage, it can be confirmed that the stage moves uniformly in the compression direction (X direction). FIG. 12 shows the movement amount distribution in the X direction in the fourth stage. The average movement amount in the X direction was 18.26 μm, and the average movement amount in the Z direction on the same stage was 0.22 μm. Since the average movement amount in the Z direction is less than the resolution of the optical coherence tomographic image, it can be confirmed that the movement is only in the compression direction. As shown in FIG. 12, the moving amount change rate calculated from the approximate straight line concerning the moving amount in the X direction is −0.68%, and it is detected with a high resolution of 20 μm that the moving amount is decreasing in the compression direction. Yes. Here, when it is assumed that the end face of the uniform sample is uniformly compressed, the change rate in the compression direction can be calculated as −0.43% by using the measured value of the sample width. An error of 0.25% occurs in both the change rates, but when converted to an error distance in the measurement region of 1200 μm, it becomes 3.0 μm, and the resolution in the vertical direction of the optical axis is 6.61 μm or less.
Furthermore, the variation errors regarding the movement amounts in the X direction and the Z direction are ± 1.10 μm and ± 1.00 μm, and it can be seen that they are within the error range when the resolution in each direction is taken into consideration. As a result, it was considered that the result was consistent with the result based on the assumption of uniform compression for the uniform sample, and the validity of the method according to the present invention was verified. For comparison with the conventional method, a direct cross-correlation method with sub-pixel processing was applied at the same spatial resolution.

13 and 14 show the movement vector distribution and the X-direction movement amount distribution according to the conventional method. The size of the search area (N _x = N _z ) is set to a condition that minimizes the error. The attenuation of the movement amount in the X direction shows the same result as the method according to the present invention. However, since the variation error of the movement amount in the X direction and the Z direction is ± 7.81 μm and ± 4.66 μm, respectively, which is larger than the resolution in each direction, it is impossible to calculate a high-resolution movement vector distribution using the conventional method. Have difficulty. As can be confirmed from FIG. 13, this is considered to be an influence of an error vector. However, since the method of the present invention suppresses the generation of error vectors and limits the moving direction from a large inspection region where the occurrence is small and performs recursive correlation processing, it is possible to calculate a moving vector distribution with high spatial resolution. Is possible. FIG. 15 and FIG. 16 show the distortion distribution in the X direction calculated from the method of the present invention and the movement vector distribution of the conventional method. Although uniform distortion can be confirmed with this method, it can be visually observed that local distortion has occurred in the conventional method. Based on the amount of distortion calculated on the assumption that the uniform sample end face was uniformly compressed, the mean square error of this measurement result was 2.12%, and the result by the conventional method was 45.88%. When both are compared, a 95.4% reduction in distortion amount error is realized. With respect to the amount of strain in the Z direction, an error reduction of 91.6% was possible. Since the distortion amount is a spatial differential amount, an error is increased in the conventional method. However, in the method of the present invention, spatial interpolation and correction processing is performed by the least square method after removing an erroneous vector, and thus the continuity of the spatial differential amount is increased. It has improved.

FIG. 9B is an image of a non-uniform sample. An elliptical black portion centered at (X, Z) = (500 μm, 600 μm) represents a fishing line having a different elastic modulus, and is imaged together with a speckle pattern of polystyrene particles. The movement vector distribution and the distortion distribution in the fourth stage (ΔL = 20 μm) to which the method of the present invention is applied are shown in FIGS. The hatched portion in the figure represents the fishing line portion. The movement vector distribution does not move uniformly in the compression direction (X direction), and a large movement vector distorted around the fishing line can be confirmed. It can be understood that large deformation occurred due to compression in the vicinity of the fishing line portion having a high elastic modulus with respect to the uniform Agar powder sample. Comparing FIG. 15 and FIG. 18, with respect to the strain distribution of the non-uniform sample, local deformation is observed in the vicinity of (X, Z) = (300 μm, 300 μm) and (X, Z) = (300 μm, 800 μm) where the deformation is severe. It can be confirmed that large distortion occurs. Under the assumption of uniform compression, it can be predicted that distortion will occur in the periphery of the fishing line in the compression direction, but from FIG. 9 (b), separation can be observed at the boundary between the Agar powder and the fishing line. Since it is different from the ideal conditions for compression, it is difficult to make a comparative study. However, it was verified that characteristic tissue deformations appearing in heterogeneous samples with different elastic moduli can be observed.

As in the case of the uniform sample described above, a conventional method was applied for comparison with the same spatial resolution. The results are shown in FIG. 19 and FIG. As is clear from both figures, local distortion appears in the form of spots due to the frequent occurrence of error vectors. In the direct cross-correlation method with high spatial resolution, the occurrence of erroneous vectors increases due to the frequent occurrence of correlation peaks accompanying a decrease in the amount of information in the examination region. In addition, since a complicated movement phenomenon accompanied with rotation occurs in a non-uniform sample, it is considered that the movement vector cannot be identified by the conventional method. However, according to the method of the present invention, a complicated movement phenomenon can be traced with high spatial resolution by a combination of spatial interpolation processing by a statistical method such as a recursive speckle cross correlation method and a least square method.
From these results, it becomes clear that the mechanical properties can be detected with high spatial resolution in unstable plaques interspersed with tissues with different elastic moduli such as fibrous capsules, lipid cores, and calcareous materials. Has been demonstrated.

As described above, the inventions described in the first to twelfth aspects of the present invention can be used as an apparatus and system for health diagnosis for early detection of circulatory diseases. It can be used in various medical settings such as diagnostic devices and systems, quantitative grasping of therapeutic effects, and devices and systems for confirming healing.
It can be used not only in humans but also in veterinary clinics that handle animals.
Furthermore, it can be applied not only to animal and plant biological tissues but also to non-destructive inspection devices.

The elastic modulus distribution measurement system according to the present embodiment of the present invention is included, and the biological tissue that is the object to be measured is compressed, and the optical interference data obtained before and after that is acquired to measure the elastic modulus distribution in the biological tissue. It is a conceptual diagram of the tissue elasticity image measuring system for doing. 1 is a system configuration diagram of an elastic modulus distribution measurement system (strain distribution measurement system) according to the present embodiment. It is a flowchart which shows the process of the elastic modulus distribution measuring method (strain distribution measuring method) which concerns on embodiment of this invention. FIG. 4 is a conceptual diagram for explaining cross-correlation analysis performed in a cross-correlation analysis unit 26. It is a conceptual diagram of a movement vector. It is a system block diagram of an error vector analysis part. (A) And (b) is a conceptual diagram for demonstrating exclusion of an error vector performed by an error vector analysis part, or performed by step S4. (A) And (b) is a conceptual diagram for demonstrating the interpolation / correction process of the vector distribution performed by a movement vector interpolation and correction | amendment analysis part, or is performed as step S5. It is an optical coherence tomographic image obtained by the experiment described in the column of Example, (a) is an image obtained in a uniform sample, (b) is an image obtained in a non-uniform sample. It is the image of the movement vector obtained by the experiment described in the column of Example, (a) has a spatial resolution of ΔL = 160 μm, (b) has a spatial resolution of ΔL = 80 μm, (c) Has a spatial resolution of ΔL = 40 μm. It is the image of the movement vector obtained by the experiment described in the column of the embodiment, and is in the fourth stage set as the final stage, and has a spatial resolution of ΔL = 20 μm. It is a graph which shows the movement amount distribution of the X direction calculated from the movement vector distribution of a 4th stage. It is an image which shows the movement vector distribution obtained by the conventional method. It is a graph which shows the X direction movement amount distribution calculated from the movement vector distribution obtained by the conventional method. It is a distortion distribution in the X direction calculated from the movement vector distribution obtained by the method of the present invention. It is a distortion distribution in the X direction calculated from the movement vector distribution obtained by the conventional method. It is a movement vector distribution in the fourth stage (ΔL = 20 μm) obtained by applying the technique of the present invention. It is a strain distribution in the fourth stage (ΔL = 20 μm) obtained by applying the method of the present invention. It is the movement vector distribution obtained in the same spatial resolution (ΔL = 20 μm) using a non-uniform sample by the conventional method. This is a strain distribution obtained at the same spatial resolution (ΔL = 20 μm) using a non-uniform sample by a conventional method.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Tissue elastic image measurement system 2 ... OCT optical system 3 ... Object to be measured 4 ... Amplifier 5 ... A / D converter 6a ... Strain distribution measurement system 6b ... Elastic modulus distribution measurement system 7 ... Super luminescent diode 8 ... Beam splitter 9 ... Reference mirror 10 ... Objective lens 11 ... Photo detector 12 ... Linear drive 21 ... Input unit 22 ... Analysis unit 23 ... Output unit 24 ... Analysis input database 25 ... Analysis output database 26 ... Cross correlation analysis unit 27 ... Movement vector setting unit 28 ... Error vector analysis unit 29 ... Movement vector interpolation / correction analysis unit 30 ... Strain analysis unit 31 ... Elastic modulus analysis unit 32 ... Optical interference data 33 ... System data 34 ... Analysis condition data 35 ... Surface pressure data 36 ... cross-correlation coefficient data 37 ... vector data 38 ... vector statistical data 39 ... distortion distribution data 40 ... Sex ratio distribution data 45 ... moving vector 46 ... standard deviation analyzer 47 ... intermediate value analysis unit 48 ... malpractice vector determining unit 49 ... moving vector error correlation analysis section 50a, 50b ... moving vector

Claims (12)

It has an input unit, a data storage unit, an analysis unit, and an output unit, and is based on ultrasonic echo data or optical interference data before and after compression of the measurement target (hereinafter collectively referred to simply as measurement data). A strain distribution measurement system for analyzing strain distribution,
The input unit is a means capable of inputting measurement data before and after the compression to the data storage unit,
The analysis unit includes a cross-correlation analysis unit that calculates a cross-correlation coefficient related to a positional displacement of the measurement data before and after compression in a desired evaluation region of the measurement data, and the cross-phase relationship analyzed in the cross-correlation analysis unit Select a starting point and an ending point of the position displacement where the number is maximum, a moving vector setting unit for setting a moving vector by compression having these starting point and ending point, and a plurality of moving vectors in the vicinity of the moving vector An error vector analysis unit that executes a first statistical process to remove an error vector from the movement vectors, and a second statistical process for a plurality of movement vectors in the vicinity of the movement vector. A movement vector interpolation / correction analysis unit for making the movement vector distribution continuous, and a distortion distribution analysis unit for calculating a spatial change rate of the continuous movement vector, Has,
The data storage unit stores output data of at least one analysis executed in the analysis unit;
The strain distribution measuring system, wherein the output unit is means capable of reading out and outputting output data of at least one analysis executed by the analysis unit from the analysis unit or the data storage unit.   The first statistical processing includes a predetermined coefficient of the standard deviation from the intermediate value with reference to a statistical analysis value of standard deviation and intermediate value obtained from a plurality of movement vectors in the vicinity of the movement vector. 2. The distortion distribution measuring system according to claim 1, wherein the double deviation width is a process of determining an error vector when the difference width exceeds or equals a predetermined width.   The first statistical processing includes a set of a plurality of cross-correlation coefficients in the vicinity of the movement vector and a set of a plurality of cross-correlation coefficients in the vicinity of another movement vector adjacent to the movement vector. The strain distribution measuring system according to claim 1, which is a process for obtaining a correlation between the two.   The cross-correlation analysis unit, the movement vector setting unit, the error vector analysis unit, and the movement vector interpolation / correction analysis unit have an iteration function, and the evaluation that the movement vector continuous by the movement vector interpolation / correction analysis unit exists In the evaluation region divided by dividing the region, the cross-correlation analysis unit again calculates a cross-correlation coefficient regarding the positional displacement of the measurement data before and after compression, and the movement vector setting unit is analyzed by the cross-correlation analysis unit Select a starting point and an ending point of the position displacement that maximizes the cross-correlation coefficient, set a movement vector by compression having these starting point and ending point, and the error vector analysis unit determines and removes the error vector, 4. The movement vector interpolation / correction analysis unit executes continuation of a distribution of movement vectors. Strain distribution measurement system according to claim. An elastic modulus distribution measurement system that has an input unit, a data storage unit, an analysis unit, and an output unit, and analyzes an elastic modulus distribution from measurement data before and after compression of a measurement target,
The input unit is a means capable of inputting measurement data before and after the compression and surface pressure data of the measurement target into the data storage unit,
The analysis unit includes a cross-correlation analysis unit that calculates a cross-correlation coefficient related to a positional displacement of the measurement data before and after compression in a desired evaluation region of the measurement data, and the cross-phase relationship analyzed in the cross-correlation analysis unit Select a starting point and an ending point of the position displacement where the number is maximum, a moving vector setting unit for setting a moving vector by compression having these starting point and ending point, and a plurality of moving vectors in the vicinity of the moving vector An error vector analysis unit that executes a first statistical process to remove an error vector from the movement vectors, and a second statistical process for a plurality of movement vectors in the vicinity of the movement vector. A movement vector interpolation / correction analysis unit for making the movement vector distribution continuous, and a distortion distribution analysis unit for calculating a spatial change rate of the continuous movement vector, And an elastic modulus distribution analysis unit for calculating an elastic modulus by using the spatial rate of change computed in the strain distribution analysis unit and the surface pressure data,
The data storage unit stores output data of at least one analysis executed in the analysis unit;
The elastic modulus distribution measurement system, wherein the output unit is a means capable of reading out and outputting output data of at least one analysis executed by the analysis unit from the analysis unit or the data storage unit.   The first statistical processing includes a predetermined coefficient of the standard deviation from the intermediate value with reference to a statistical analysis value of standard deviation and intermediate value obtained from a plurality of movement vectors in the vicinity of the movement vector. 6. The elastic modulus distribution measuring system according to claim 5, which is a process of determining an error vector when the double deviation width exceeds or is equal to a predetermined width.   The first statistical processing includes a set of a plurality of cross-correlation coefficients in the vicinity of the movement vector and a set of a plurality of cross-correlation coefficients in the vicinity of another movement vector adjacent to the movement vector. The elastic modulus distribution measurement system according to claim 5, wherein the process is a correlation process.   The cross-correlation analysis unit, the movement vector setting unit, the error vector analysis unit, and the movement vector interpolation / correction analysis unit have an iteration function, and the evaluation that the movement vector continuous by the movement vector interpolation / correction analysis unit exists In the evaluation region divided by dividing the region, the cross-correlation analysis unit again calculates a cross-correlation coefficient regarding the positional displacement of the measurement data before and after compression, and the movement vector setting unit is analyzed by the cross-correlation analysis unit. Select a starting point and an ending point of the position displacement that maximizes the cross-correlation coefficient, set a movement vector by compression having these starting point and ending point, and the error vector analysis unit determines and removes the error vector, The movement vector interpolation / correction analysis unit executes continuation of the distribution of movement vectors. Modulus distribution measurement system according to claim. In the strain distribution measurement method in which the computer analyzes each strain distribution from the measurement data before and after compression of the measurement target while executing each process,
An input unit stores the measurement data before and after compression in a data storage unit so that the measurement data can be read, and a cross-correlation analysis unit reads the measurement data from the data storage unit and stores the measurement data before and after compression in a desired evaluation region. A step of calculating a cross-correlation coefficient related to the position displacement, and the movement vector setting unit selects a start point and an end point of the maximum position displacement using the analyzed cross-correlation coefficient as a key, and includes the start point and the end point. A moving vector set by compression, and an error vector analyzing unit executes a first statistical process on a plurality of moving vectors in the vicinity of the moving vector, and sets an error vector out of the moving vectors. And the movement vector interpolation / correction analysis unit executes a second statistical process on a plurality of movement vectors in the vicinity of the movement vector. A step of making the distribution of the movement vector continuous, a step of calculating a spatial change rate of the movement vector by the strain distribution analysis unit, and a step of outputting the distribution of the spatial change rate of the movement vector by the output unit A strain distribution measuring method characterized by comprising:   The first statistical processing includes a predetermined coefficient of the standard deviation from the intermediate value with reference to a statistical analysis value of standard deviation and intermediate value obtained from a plurality of movement vectors in the vicinity of the movement vector. 10. The strain distribution measuring method according to claim 9, wherein the double deviation width is a process of determining an error vector when a predetermined width exceeds or equals a predetermined width.   The first statistical processing includes a set of a plurality of cross-correlation coefficients in the vicinity of the movement vector and a set of a plurality of cross-correlation coefficients in the vicinity of another movement vector adjacent to the movement vector. The strain distribution measuring method according to claim 9, which is a process of obtaining a correlation between the two. An elastic modulus distribution measuring method obtained by adding a new process to the strain distribution measuring method according to any one of claims 9 to 11.
The step of storing the measurement data before and after compression in the data storage unit so that the input unit can be read includes a step of storing the surface pressure data of the measurement target, and the strain distribution analysis unit is a continuous motion vector An elastic modulus distribution measuring method, comprising: a step of calculating an elastic modulus using the surface pressure data and the spatial change rate after the step of calculating a spatial change rate.