JPS59123067A - Picture display processor - Google Patents

Abstract

PURPOSE:To increase both accuracy and speed of diagnosis by adding the medical picture obtained from the device of different type after filtering the picture with an optionally set space frequency characteristic. CONSTITUTION:The medical picture obtained from different device 1 and 2 filtered with the optionally set space frequency characteristics. These pictures are added and synthesized to be displayed as a composite picture. An optional change is possible by varying the set value with a composite picture having the flat space frequency for each picture through a composite picture having the space frequency characteristics of different bands for each picture. Thus it is possible to extract effectively the information on original pictures.

Description

[Detailed Description of the Invention] [Technical Field of the Invention] The present invention relates to a medical image diagnostic apparatus, and relates to a medical image diagnostic apparatus for performing composite image display diagnosis by superimposing and displaying a plurality of medical images having mutually complementary characteristics. The present invention relates to an image display processing device.

[Technical background of the invention]

Medical image diagnostic equipment includes X-ray diagnostic equipment and ultrasound diagnostic equipment, and in recent years, digital radiography, which converts X-ray fluoroscopic images into digital data and performs image processing, and NMR-CT equipment, which utilizes nuclear magnetic resonance, have emerged. . When the same part of a subject is photographed using each of these diagnostic devices, for example, X-ray C
With T, the distribution of X-ray absorption coefficients of the tomographic plane of the subject can be obtained, with NMR-CT, the distribution of hydrogen nuclei by nuclear magnetic resonance can be obtained, and with nuclear medicine diagnostic equipment, the distribution of the administered radioisotope can be obtained. It will be done. In this way, each diagnostic device provides various types of biological information such as form and function, and usually a complex image diagnosis using various diagnostic devices complementary to the same region is performed to reach a definitive diagnosis of a disease. is necessary. Conventionally, for such composite image diagnosis, a doctor would view the films outputted from each diagnostic device for the same area side by side on a Scherkasten, or multiple images would be displayed in parallel on a single screen on a composite image diagnosis device. In some cases, multi-frame display was performed, and the outlines of other images were displayed superimposed on one image.

However, in the case of multi-frame display, it is the same as arranging multiple images on a shark screen,
It is difficult to intuitively understand the positional relationship between the two images, and even if the method of superimposing the contour lines of one image on the other image makes it easy to understand the mutual positional relationship, it takes a lot of time to extract the contours. Moreover, the contour lines omit minute information contained in the original image, making it impossible to perform a quick and accurate composite diagnosis.

[Purpose of the invention]

The present invention was developed in view of the above circumstances, and it is possible to accurately capture the positional relationship between the two images by overlapping and displaying images from different diagnostic devices having complementary roles for the same part of the subject. It is an object of the present invention to provide an image display processing device for composite image diagnosis that is capable of performing composite image diagnosis quickly and with good operability without damaging the biological information possessed by each of the two images. It is said that

[Summary of the invention]

That is, in order to achieve the above object, the present invention provides an image memory that stores each medical image obtained by photographing the same part of a subject using a plurality of different image diagnostic apparatuses having complementary roles. , a means for setting a spatial frequency characteristic of filtering for each image, a means for filtering each image with a spatial frequency characteristic according to the set value, and a means for adding each of the filtered images. and a means for displaying the added images, filtering each medical image obtained by different types of devices using arbitrarily set spatial frequency characteristics, and adding the filtered images. the composite image is changed from a composite image in which each image has a flat spatial frequency characteristic to an image in which each image has a spatial frequency characteristic in a different band; By varying the values, it is possible to display the image in any desired state.This also makes it possible to effectively extract the information contained in the original image, allowing for highly accurate diagnosis, and filtering the composite image. To shorten arithmetic processing time and quickly perform diagnosis by requiring only processing and addition.

[Embodiments of the invention]

An embodiment of the present invention will be described below with reference to the drawings.

First, a detailed explanation of the present invention will be given.

Two images', i, j 21. When overlapping j, they are combined by addition, and as shown in the first equation, ``s j ' i r J 2 t t j
It is possible to obtain (1), but if the image quality characteristics of the two images are similar, the two images overlap, resulting in an extremely difficult to see image in which it is difficult to distinguish between them.

Therefore, in order to be able to read the two images separately and to reduce the loss of individual image quality, it is effective to separate them on the spatial frequency axis and add them together.

A spatial filter is a means for changing spatial frequency characteristics. That is, using this spatial filter, for example, one image is subjected to a high-pass filter to enhance edges, and the other image is subjected to a low-pass filter for smoothing, and then the two images are added to form a single image. Create an image. Then, since the smoothed image of one image overlaps the edge-enhanced image of the other image, they can be distinguished from each other on the same image, and the image quality of both images is less likely to be degraded. Figure 1 shows the situation on the spatial frequency axis.

That is, if (,) is the spatial frequency distribution of the X-ray CT image and (b) is the spatial frequency distribution of the NMR-CT image, then (
, ) is filtered by a high-frequency emphasizing filter with characteristics like (c), and (b) is filtered by a low-frequency emphasizing filter having characteristics like (d), then 0 becomes (e)
A filtered image of a spatial frequency distribution like this! D, (
In b), a filtered image of the spatial frequency distribution as shown in (f) is obtained. Adding this (e) # (f) gives (h
) is a superimposed image of the overall characteristics as shown by the dotted line, and
The line CT image is superimposed as an edge-enhanced image with fine changes, and the NMR-CT image is a smoothed image with low-frequency components, so that they are differentiated from each other. Moreover, such operations can be performed freely by changing the spatial filter characteristics, including the reverse combination of the above, so it is possible to create any state according to the diagnostic purpose, without damaging the information contained in the original image. It is possible to effectively extract images and change them into images of various spatial frequency bands to observe the examination site.

By the way, as a simple spatial filter, for each pixel, 98 filters around the drawing position are summed up and filtered with a support size of 3×.
There are three linear stationary non-recursive two-dimensional digital filters.

If the input image is htj, the output image is gi, j, and the filter coefficients are bc ab bc, then gitj""fi-1tj-11,j-1i+1. j-
1+bf t-4, j+bf l, j+ bf ,,
, j+c'i-1, J+1 i, j+1
i+1. , J+1 °−(2)+bf, and its spatial frequency characteristic is H(u,v)=2ccol(u+v)+2bcosv+
2acos(u-v)+2bcosu+a ・
-(a). Now, filter coefficients a, b, and c are all 1
Then H(u,v) (a7u+v)+casv+cas
(u-v)4cosu+'t-(4)2 Then, when the orthogonal coordinates uyv are converted to polar coordinates r and O, and the characteristics are illustrated, the result is as shown in FIG. Considering that the original image and the filtered image of the above fourth equation are linearly combined with weights 1- and k, the spatial frequency characteristic is 1- k + k H (u, o)
=-(5). Here, with k as a parameter, the fifth
If we draw the characteristics of the equation, we can obtain multi-spatial frequency characteristics by changing the weight k from a low-pass filter with a weight of k = 10 to a high-pass filter with a weight of =-1. can be changed continuously.

That is, by changing k, two images can be extracted with different spatial frequency bands, and by superimposing them, the edges of one image are emphasized, while the other image is focused on the low frequency band. It is possible to superimpose images extracted from two images with arbitrarily different spatial frequency bands, such as displaying them as images with similar tones or displaying them as images with similar tones. It is possible to change the state arbitrarily and make it observable.

Next, an embodiment of the present invention will be described with reference to FIG.

FIG. 4 is a system block diagram showing one embodiment of the present invention. Here - as an example, NMR-CT image and X-ray CT image
The explanation is given using a composite image diagnosis with images as an example. In the figure, 1 is an NMR-CT device, 2 is an X-ray CT device,
Each tomographic image is obtained for the same part of each subject. Image data of tomographic images obtained by each of these devices is stored in corresponding image memories 3 and 4, respectively. A voltage of ±V is applied between the two poles, and 5 is used to specify the spatial frequency characteristics for each image during spatial filtering to divide and extract this voltage. The output of this variable resistor 5 is used to specify the parameters, and is given to a converter 6 where it is converted into a digital quantity and then given to the arithmetic unit 1.
This arithmetic unit 7 calculates and outputs k and -k, which are inputted as meters for defining the jfC characteristics of spatial filter devices 8 and 9 provided corresponding to the OCT devices 1 and 2, respectively. Here, the reason why the signs of the parameters given to both filters are reversed is to give mutually opposite characteristics so that when one has a low-pass characteristic, the other has a high-pass characteristic.

(The relationship between the spatial frequency characteristics of the filter and the 74 rammeter is as shown in Figure 2.) The images stored in the image memory 3 and 4 are simultaneously read out at the same pixel position and sent to the corresponding spatial filter device. 8,9
After being filtered by the spatial filter devices 8 and 9, the images are added together by an adder 10, and 1
Two images are stored in the display image memory 11 and displayed on the display device 12.

FIG. 5 is a detailed explanation of the spatial filter device part in FIG. 4, and is a spatial filter device that realizes the spatial frequency characteristic of formula 5 (FIG. 3) specified by the parameter.

In FIG. 5, reference numeral 21 denotes an original image storage memory for storing original images obtained by the image diagnostic apparatus), in which original images to be subjected to filtering processing are stored. 22 is a 3×3 filter circuit used for the filtering described above, and the filtering processing result is stored in an image memory 23. 24.25 is an image constant multiplier, which multiplies the filtered image in the image memory 23 and the original image in the original image storage memory 21 by A and 8, and further multiplies the filtered image in the image memory 23 and the original image in the original image storage memory 21 by A times
Addition is performed using an image addition device, and the result is 27
output to and store the processed image storage memory. 24,2
5.degree. 26 is a portion for linearly combining the original image and the filtered image, and the weighting coefficients A and B at this time are given by the arithmetic unit 28 from the parameter k.

\ As a result, an image in the spatial frequency band corresponding to the specified weighting coefficient k is obtained.

FIG. 6 is a block diagram showing details of the 3×3 filter circuit 220 portion in FIG. 5. All filter coefficients are 1
The 3×3 filter that is F14. If the appearance image is CB, then Gi, j: Fi-1, j+1"i, j-
1+F1. j+1+Fl-1*j Lj i
+1. Koju Fi-1,j+1 1. J+1 1+1. j
Calculated by +1. In other words, the X and Y coordinates are 1. In order to calculate pixel 'l t j which is j, pixel F is extracted from image memory 31 in which one input image F is stored.
h, > and its 8 pixels are read out and added together. The signals X, Y, which are output from the 3×3 filter circuit 32, are for giving addresses in the X and Y directions when reading pixels from the image memory in 1n 31, and the signals X, Y, which are output from the 3×3 filter circuit 32 are for giving addresses in the X and Y directions when reading pixels from the image memory in 1n 31. are sequentially input to the filter circuit 32. The pixel G calculated by the 3×3 filter circuit 32 is stored in the image memory 33. The signal that gives the x and y coordinates to be stored for i and j is X. u
t and Y. It is ut.

FIG. 7 explains in detail the internal circuit of the 3×3 filter circuit 32 in FIG. 6, and FIG. 8 shows a timing chart of its operation.

Signals a-t in FIG. 7 correspond to a-t in FIG. That is, the signal a is a basic clock, and this basic clock a is input to a counter 111 having a 2-bit configuration. Therefore, by having the counter 111 count the basic clock a, the counter 111 can output @Q# every clock period.

An output signal is obtained which repeats the 2-bit values of "1#" and ""2". This signal is a counter 11 with a 2-pit configuration.
2. The counter 112 counts the signal KO, 1, every b period.

An output signal C is obtained which repeats the 2-bit value of 2. This signal C is input to the counter 113. counter 11
3, the signal C is counted, and every period of C is counted 0.1.
An output signal d is obtained which repeats a value of 2...25508 bits. This signal d is input to the 8-bit counter 114, and an output signal e is obtained from the counter 114, which repeats 8-bit values of 0°1.2, . . . , 255 every cycle of the signal d. It will be done. Said signal! :d is added by an adder 132 to obtain a signal f. Further, the signals C and e are added by an adder 31 to obtain a signal g. Signal f corresponds to xin in FIG. 6, and signal g corresponds to signal Y1n in FIG. That is, Xln, Y, and n sequentially scan the 493x3 area of the point to be calculated. Next, it moves by one pixel in the X direction and similarly scans a 3×3 area. When one line is completed in this way, it moves one pixel in the y direction and repeats the same scanning. X1n and yin act as address signals in the X and Y directions for reading out the pixel memory, and the pixels read out by these are input as signals F to latches 141 to 149 in parallel.

The timing at which the signal F is held in the latch is determined by the control signals h1 to h9 output from the shift register. The output of the shift register 22 is synchronized with the basic clock. F(0,0) for latches 141, 142 and 149, respectively.

F(1#0), F(2,0), F(0,1).

F(1,1), F(2,1), F(0,2).

F(1,2) and F(2,2) are stored and appear at outputs iJ, 12...19. The signal was 51. The result is held in the latch 135 at a timing determined by the output signal k of the shift register 121, and a signal is obtained. t is the output data G of the 3×3 filter circuit,
Output to image memory for storing filter processing results.

Here, the addresses X and Y in the X direction and Y direction for the image memory where G is to be stored are rounded out.
out X at the timing of control signal h5 by chips 133 and 134
This can be obtained by latching i□' Yin.

In this way, at each pixel position of the image, a value is obtained by adding up 9 pixels including the 8 surrounding pixels around that pixel, and by using the value of V9 of this value as pixel data, It can be obtained as data that has been subjected to 3×3 filtering.

Figure 9 shows the original image in Figure 5 and the 3× circuit created by the circuit in Figure 7.
This figure shows the details of the configuration 4, 5.6 for performing linear combination with the image that has been subjected to 3-filtering. In FIG. 9, the signal a is a basic clock, which is input to the counter 61, and a repeating signal of 8 pits of 0, 1, . . . 255 is obtained. b is input to the counter 62, and a signal C is obtained that counts 8-bit values of 0, 1, . . . 255 every cycle of b. Signals d and c are used as address signals in the X direction and y direction, respectively, for accessing the image memory 1.3.7 in FIG. 5. In this way, the reading of the input image and the writing of the processing results are accomplished by scanning the image line by line from left to right. The read original image data is input as FIG. 9g, and the 3×3 filtered image data is input as j. The registers 64 and 63 respectively hold weighting coefficients A (=k) and B (=1-k) inputted by the arithmetic unit 28 in FIG. 5 (corresponding to 7 in FIG. 4),
Multipliers 65 and 66 in FIG. 9 calculate the products of the filtered image data and the weighting coefficients for each image set by variable resistor 5, and the results are added by adder 67. Otherwise, a line combination operation is performed. The result is output to the address of the processed image storage memory indicated by the address signals d and c in the X and y directions.

〔Effect of the invention〕

As described in detail above, the present invention includes an image memory that stores each medical image obtained by photographing the same part of a subject using a plurality of different image diagnostic apparatuses having complementary roles, and means for setting the spatial frequency characteristic of filtering for the image; means for filtering each image with the spatial frequency characteristic according to the set value; means for adding the filtered images; It has a wide variety of image display means, filters each medical image obtained from different types of devices using arbitrarily set spatial frequency characteristics, and synthesizes the images by adding them. Since the images are displayed as a composite image, the setting values can be changed from a composite image in which each image has a flat spatial frequency characteristic to a composite image in which each image has a spatial frequency characteristic in a different band. This allows the information contained in the original image to be effectively extracted, diagnoses can be made with high accuracy, and complex images can be filtered. It is possible to provide an image display processing apparatus having such characteristics that processing time can be shortened and diagnosis can be performed quickly because it can be obtained by simply adding .

[Brief explanation of the drawing]

FIG. 1 is a diagram for explaining an example of the processing of the present invention, and FIG.
The figure shows the change in spatial frequency characteristics depending on the direction OK.
The figure shows the spatial frequency and spatial frequency characteristics using linear combination coefficients as meters, Figure 4 is a block diagram showing an embodiment of the present invention, and Figure 5 shows the filter device part and its vicinity. A block diagram showing the configuration, FIG. 6 is a diagram for explaining the relationship between the filter device and the image memory, and FIG.
FIG. 8 is a block diagram showing the configuration of the ×3 filter circuit, FIG. 8 is a time chart for explaining its operation, and FIG. 9 is a block diagram of a portion that applies weighting and a portion that performs addition and synthesis. 1...X-ray CT device, 2...M-CT device, 3,
4゜11... Image memory, 5... Variable resistor, 6.
...φ converter, 7... Arithmetic device, 8, 9... Filter device, 10... Adder, 12... Display device. Applicant's agent Patent attorney Takehiko Suzue Figure 3 procedural amendment Di! Iゎ58ya2・? 3.8 Commissioner of the Japan Patent Office Kazuo Wakasugi 1, Display of the case Patent Application No. 1987-233537 2, Name of the invention Image display processing device 3, Person making the amendment Relationship with the case Patent applicant (307) Tokyo Shibaura ~, Ki Co., Ltd. 4, agent

Claims (1)

[Claims] An image memory that stores each medical image obtained by photographing the same part of a subject by a plurality of different image diagnostic apparatuses having complementary roles; means for setting spatial frequency characteristics for filtering; means for filtering each image with the spatial frequency characteristics according to the set value; means for adding the filtered images; and the added image. Image display processing device 0 characterized by comprising means for displaying