JPH04138393A - Apparatus for correcting body movement - Google Patents

Abstract

PURPOSE:To accurately correct image data against irregular body movements by comparing camera images to compute displacement quantity of a body movement of each picture element and correcting picture element data of a diagnostic image. CONSTITUTION:A gamma camera 1 detects gamma-rays generating from RI dosed out to an inspected object M and takes a picture of the RI distribution image. A TV camera 2 is arranged to the side part not invading an image pickup space of the camera 1. A modification coefficient of the coordinate system of the camera 2 against that of the camera 1 is computed to have both the coordinate systems of cameras agreed and a preparation to house in a coefficient memory 8 is done. So, photographings by cameras 1 and 2 is made. A CPU 5 computes displacement quantity of a body movement of the inspected object M per each picture element from each image of the body surface of the inspected object M taken picture of by the camera 2. This displacement quantity catchs wholly and finely the body movement of the diagnosis region of the inspected object M and even an irregular body moving displacement can be accurately detected. This displacement quantity is remembered in a displacement memory 6 and the picture element data of the diagnosed image are corrected based on this.

Description

DETAILED DESCRIPTION OF THE INVENTION A. Field of Industrial Application This invention relates to radioisotope (R1) administered to a subject.
The present invention relates to a nuclear medicine diagnostic device such as a gamma camera device that detects and images the distribution of a substance in the body, and particularly relates to a technique for correcting image blur caused by body movements such as breathing of a subject.

B. Prior art gamma camera device converts gamma rays generated from RI administered to a subject into visible light using a scintillator, detects the position information of the light emitting point and the number of times the light is emitted over the required time, and calculates the R1 distribution inside the body. This is a nuclear medicine diagnostic device that creates gray scale images. Generally, the detection time is longer than that of other diagnostic devices such as an X-ray imaging device, and body movements such as breathing of the subject during that time appear as blur in the image.

Therefore, so-called body movement correction is performed in which the body movement displacement of the subject is detected and the light emitting point position information is corrected using the amount of displacement to prevent blurring of the image. Conventional body motion correction is performed as follows.

(1) A respiration sensor is installed near the subject's nose to detect temperature changes or pressure changes during exhalation and inhalation to obtain a respiration signal synchronized with respiration.

Since body movement due to breathing occurs repeatedly in response to this breathing signal, the amount of body movement displacement due to breathing is determined in advance, the amount of body movement displacement is read out in synchronization with the breathing signal, and the above light emitting point position information is corrected. I do.

(2) A point-shaped R1 (correction R1) is adhered to a part of the body surface of the subject as a correction radiation source, and the movement direction and movement amount of this correction R1 are calculated. The calculated movement of correction R1 is taken as the body movement of the subject, and the above-mentioned light emitting point position information is corrected using the calculated data.

Problems to be Solved by the C0 Invention However, the conventional body motion correction has the following problems.

As described in (1), body movement correction using the respiration signal from the respiration sensor and the amount of body movement displacement determined in advance is effective against regularly repeating displacements such as respiratory body movements. Although good correction results can be obtained, the problem is that the correction does not follow the subject's body movement and displacement, and if the body movement is irregular, it cannot be corrected. There is.

In the body motion correction using the point-like correction RI described in (2), since the movement at a specific point on the surface of the subject is represented by the movement of the entire body, the entire body of the subject is corrected. It was not possible to grasp the movement of the robot within one minute, and therefore the correction accuracy was not sufficient.

In this case, the correction accuracy can be improved to some extent by attaching multiple correction RIs to the subject's body surface and detecting the amount of displacement at each point. The image of the correction RI overlaps with the gray scale image obtained from the test R1 administered in the test, resulting in a decrease in image interpretation ability.

The present invention was made in view of the above circumstances, and aims to provide a body movement correction device that can perform accurate correction even for irregular body movements. There is.

03 Means for Solving the Problems The present invention has the following configuration to achieve the above object.

That is, the body motion correction device of the present invention provides an I.
A displacement calculation method that compares each image output from the imaging means that images the body surface of the subject and calculates the amount of body movement displacement of the subject for each pixel from the difference in pixel data of each image. A displacement amount memory for storing the calculated displacement amount for each pixel, and a correction means for correcting pixel data of a diagnostic image based on the displacement amount for each pixel in the displacement amount memory. It is said that

E1 Effect According to the present invention, the amount of body movement displacement of the subject is calculated for each pixel from each image of the body surface of the subject taken by the imaging means, so that the amount of body movement displacement can be used for diagnosis of the subject. It captures the entire body movement in detail. Therefore, even irregular body movement displacements can be detected with high accuracy, and as a result, the accuracy of body movement correction can be improved.

F. Example Hereinafter, one embodiment of the present invention will be described based on the drawings.

FIG. 1 is a simplified perspective view showing an embodiment of the body motion correction device of the present invention.

In the figure, numeral 1 indicates T generated from R1 administered to subject M.
A gamma camera detects lines, and in this example, it is installed in a space above the subject M so as to capture an R1 distribution image in the chest. Region R indicated by notching is the imaging region.

The TV camera 2 as the imaging means of this invention aims at the region R diagonally above and from the side, and in this way the imaging space of the gamma camera 1 (the space from the region R to the detection surface of the gamma camera 1) Install it in a location where it will not be infringed upon. As long as this condition is met, it does not matter where the camera 2 is installed.

FIG. 2 is a partially cutaway front view showing the internal structure of the gamma camera l.

As shown in this figure, inside a casing 12 made of a radiation shielding material such as lead, there is a scintillator 14 that emits light according to the intensity of radiation transmitted through a collimator 13, and a pulsed electrical signal that controls the light emission of the scintillator 14. A photomultiplier tube 15 converts the light emission into a photomultiplier tube 15, a glass plate 16 serves as a light guide that guides the light emitted from the scintillator 14 to the photomultiplier tube 15, and a pulse signal from the photomultiplier tube 15 is used to obtain dimensional light emitting point position information and the amount of light emitted energy at that position. It is determined whether the amount of emitted light energy at the light emitting point position calculated by the output position calculation circuit I7 and the position calculation circuit 17 is within a predetermined setting range, and if it is within the set range, An energy amount determination circuit 18 and the like is provided which outputs the light emitting point position information X and Y as valid information. Here, Y is position information in the body axis direction, and X is position information in the direction perpendicular to the body axis (see FIG. 1). This energy amount determination circuit 18 determines that the energy is scattered and enters the scintillator 14.
=I are excluded, and radiation that is perpendicularly incident on the collimator 13 is selected.

Light emitting point position information X, Y and TV from gamma camera 1
The output video signal from the camera 2 is configured to be applied to the image processing system shown in the block diagram of FIG.

Each component of the image processing system will be explained below.

3 is a D/D converter that converts the analog video signal from the TV camera 2 into digital image data, 4 is an interface, and 5 is a device that measures the displacement of each image output from the TV camera 2 every 1/30 seconds. The CPU, which calculates the time (calculation processing is performed within 1 frame scanning period of 1/30 seconds), extracts the characteristic pattern of the image of the frame 2, and calculates the amount of movement from the difference in the characteristic pattern from the subsequent image. It is designed to be calculated.

6 stores the calculated displacement amount two-dimensionally corresponding to each pixel of the image, and stores the light emitting point position information X from the gamma camera l,
A displacement memory whose reading address is Y; 7 is a RAM for storing the feature pattern; 8 is a memory for correcting the error between the coordinate values of the image taken by the TV camera 2 and the coordinate values of the image taken by the gamma camera 1; A coefficient value memory is stored in advance as a numerical value, and 9 is an image memory that stores images taken by the TV camera 2 for each frame.

20 is the light emitting point position information x output from the gamma camera l
, Y, a correction circuit corrects the light emitting point position information X, Y by applying the displacement data output from the displacement amount memory 6 to the light emitting point position information X, Y (X', Y '
) is stored two-dimensionally to store the corrected R1 distribution image, and 22 is a monitor that displays the number of light emitting points corresponding to the light emitting point position information X and Y in the display memory 21 as a gray scale image. It is a display.

Next, the operation of the body motion correction device described above will be explained.

This body movement correction device is a device that corrects the light emitting point position information X, Y of the gamma camera l from the displacement amount of each image taken by the TV camera 2, and prevents body movement blur of the subject M. The TV camera 2 is installed diagonally above and to the side of the subject M so as not to invade the imaging space of the gamma camera 1. Therefore, the coordinate systems of the image captured by the TV camera 2 and the image captured by the gamma camera 1 do not match at all. Therefore, before starting the actual photographing, a correction M (correction coefficient) for matching these coordinate systems is determined in advance as a preprocessing process as follows.

The positions of the gamma camera 1 and the TV camera 2 are fixed while the subject M is not placed on the head 10 yet.

A phantom f (not shown in FIG. 4) is mounted on the head 10 corresponding to the imaging space of the gamma camera 1. This phantom r is a plate in which a plurality of dotted lines #(RI) are arranged in a grid pattern (R1-R117).

This phantom f is photographed by a gamma camera 1 and a TV camera 2 whose positions are fixed. Since the gamma camera 1 photographs the phantom f from the true position, the photographed image is shown in Figure 5 (
Since the TV camera 2 photographs the phantom f diagonally from above, the photographed image becomes as shown in FIG. 2(b).

Then, coordinate data of the RI of each point of the phantom in both images is obtained.

Here, the coordinate data of the image taken by the gamma camera 1 is (XgiLYgij: i, j=] ~ n), and T
The coordinate data of the image taken by the V camera 2 is (XLij,
If YLij : i, j -1-n), then the correction coefficient of the coordinate system of 1゛■ camera 2 for gamma camera 1 is αXij = (Xgij -Xtij 1,
Calculated as αYij-1Ygij-Y[iJ 1.

Calculated (αxij, αYij: i,
j-1 to n) are stored in the coefficient value memory 8. This completes the preprocessing.

Next, the phantom f is removed from the head 10, and the subject M to which R1 has been administered is placed in its place. Gamma camera 1
At this time, imaging of the subject M is started, and at the same time, imaging by the TV camera 2 is started. The CPU 5 executes T as described below.
Once the amount of displacement of the captured image of the V camera 2 is calculated, the value obtained by adding the correction coefficient to the amount of displacement is stored in the amount of displacement memory 6 as true amount of displacement data.

The calculation of the amount of displacement by the CPU 5 will be described below with reference to the flowchart in FIG.

In step S1, the J-th frame image Fl (see FIG. 7) captured by the TV camera 2 in synchronization with the start of imaging by the gamma camera 1 is stored in the image memory 9. The frame number of the image memory 9 in which the image Fl is stored is assumed to be N011.

In Step 7 S2, a characteristic pattern P1 is found in the image R1.
Set and extract.

As the characteristic pattern P1, it is desirable to set a part where the density of the image changes suddenly and a part where the movement of the subject M is noticeable, and as in this example, an R1 distribution image in the chest is imaged. If so, for example, as shown in FIG. 7, a region including both shoulders of the subject M is set as the feature pattern PI.

Then, the set characteristic pattern P1 is cut out and its image data is stored in the RAM 7. In step “S3,
The second frame image F2 captured by the TV camera 2 is captured and stored in the N002 frame of the image memory 9.

In step S4, a pattern P2 having the same area as the feature pattern P1 is cut out from the image F2 (see FIG. 8), and the image data in the feature pattern Pl and the cut out pattern P are cut out from the image F2.
A comparison is made with the image data in 2. To cut out the pattern P2, first set the upper left corner of the image F2 to the pixel Pl+ and cut out, and then change the cutting start position pixel by pixel in the χ and Y directions to extract the feature pattern P from the entire pixel area.
An operation is performed to cut out pattern P2 in the same area as i.

As a result of comparing and matching the characteristic pattern P1 and each pattern P2, a pattern P having the same image data as the characteristic pattern P1 is found.
2 (equivalent pattern P) is extracted (step S5).

Here, as shown in FIG. 9, the coordinate value (XP.

An equivalent pattern P with yr) as the left corner coordinate point is extracted, and a positional shift (displacement) has occurred between it and the feature pattern P1 with (XP I , YP l ) as the upper left corner coordinate point. shall be.

In step S6, the amount of positional deviation (amount of displacement) between the characteristic pattern PI and the equivalent pattern P is calculated using the following equation.

d X= l Xr --XPI l, d y-IYP
In the YPI step S7, the correction coefficient value (αχIJ + αYij
: i, j-1-n) and calculate the above displacement amount (
dx, dy), and the true displacement amount of each pixel between image Fl and image F2 ΔX1j=(dx+αxij), Δ
Yij-(dy+αYij) is calculated.

Then, the calculated displacement data is stored in the displacement memory 6.

This process is performed for each image output from the TV camera 2, such as the third image, fourth image, etc., and the displacement memory 6 is updated each time with the calculated displacement data. do.

The light emitting point position information X, Y outputted from the gamma camera I is given to the displacement amount memory 6 and also given to the correction circuit 20.

The displacement memory 6 inputs the light emitting point position information X and Y as read addresses, and reads each light emitting point position information X.

The displacement amount data for Y is output to the correction circuit 20.

The correction circuit 20 adds displacement data to the given light emitting point position information X, Y, and outputs body movement correction position data expressed by the following equation to the display memory 21. Here, the light emitting point position information χ, Y is X=X, □~X7, Y -Y 1+~Y
7. , and therefore, the body movement corrected position data is X' 1j - Xij + ΔXij, Y
'1j=Yij+ΔYij.

Suppose that the light emitting point position information This becomes a grayscale image showing the R1 distribution.

In contrast, with this device, the number of detected gamma rays is stored in the body movement correction position of the above formula, which is obtained by adding the displacement data to the light emitting point position information X and Y, and the grayscale image becomes an image that has been subjected to body movement correction. .

The body movement correction process as described above is performed every time the displacement amount memory 6 is updated.

In other words, body movement correction is performed every 1730 seconds, and
Since the body motion correction is performed using the displacement data of the characteristic pattern P1, rather than the rhythmic correction using the displacement data at a specific point as in the past, the correction accuracy is significantly improved.

Note that the region setting of the feature pattern P1 is arbitrary and may be set over substantially the entire image area. In this case, body movement displacement in substantially the entire body of the subject M can be detected, and further,
Correction accuracy can be improved.

G1 Effects of the Invention According to this invention, the amount of body movement displacement of the subject is calculated for each pixel from each image of the body surface of the subject captured by the imaging means, so that the body movement at the diagnostic site of the subject is calculated. Displacement can be grasped as a whole and in detail. therefore,
Even irregular body movement displacements can be detected with high accuracy, and the accuracy of body movement correction can be improved.

[Brief explanation of the drawing]

Figures 1 to 9 relate to one embodiment of the present invention;
The figure is a perspective view schematically showing an embodiment of a body motion correction device;
Figure 2 is a partially cutaway front view showing the internal configuration of the gamma camera, Figure 3 is a block diagram showing the schematic configuration of the body motion correction device, and Figure 4 is the difference in coordinate systems between the gamma camera and the TV camera. Fig. 5 is a diagram showing a captured image of the phantom, Fig. 6 is a flowchart showing the processing procedure for calculating the amount of body movement displacement, and Fig. 7 is an example of setting the equivalent pattern P. The illustrated diagrams, FIG. 8, are diagrams for explaining the cutting out of the pattern P2, and FIG. 9 is a diagram showing the positional deviation between the characteristic pattern P1 and the equivalent pattern P. 2...TV camera (imaging means) 5...CPLJ (displacement amount calculation means) 6...Displacement amount memory 20...Correction circuit (correction means) Patent applicant: Shimadzu Corporation

Claims (1)

[Claims] (1) Compare each image output from the imaging means that images the body surface corresponding to the diagnostic site of the subject and this imaging means, and calculate the amount of body movement displacement of the subject for each pixel from the difference in pixel data of each image. a displacement amount calculation means for calculating the displacement amount for each pixel; a displacement amount memory for storing the calculated displacement amount for each pixel; and a correction means for correcting pixel data of the diagnostic image based on the displacement amount for each pixel in the displacement amount memory. A body motion correction device characterized by: