JPS6185929A - Method and apparatus for improving quality of energy subtraction image - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of the Invention) This invention relates to subtraction processing of radiographic images, specifically, in digital subtraction processing of radiographic images using a stimulable phosphor sheet, to improve the image quality of subtracted images. The present invention relates to a method and an apparatus for doing so.

(Technical Background of the Invention and Prior Art) Digital subtraction of radiographic images has been known for some time. This digital subtraction of radiographic images involves photoelectrically reading out two radiographic images that have been subtracted under different conditions, converting them into digital image signals (q), and then making these digital image signals correspond to each pixel of both images. Subtraction processing to form an image of a specific structure in a radiographic image 1
The difference signal for y is 1! Using the difference signal obtained in this way, it is possible to reproduce a radiation image in which only specific structures are extracted.

There are basically two methods for this subtraction process: That is, from image 1 (!f3) of a radiographic image in which a specific structure is emphasized by contrast agent injection, image φ(F; subtract the arch (
(knob trough i~) to extract a specific structure by so-called temporal sub-1 ~ 1~1 processing, and irradiate the same subject with radiation with different energy distributions. Making use of the fact that a certain structure has a characteristic radiation energy absorption time 11 to cause an image in which the specific structure is different to exist between two radiation images,
After that, appropriate weighting (J) is applied between the image signals of these two radiation images, and subtraction is performed to extract the image of a specific structure (so-called 1-energy retraction). It is processing.

Since this restriction processing is an extremely useful method for diagnosis, especially in the image processing of medical X-ray photographs, it has received much attention in recent years, and its research and development has been carried out by making full use of electronic engineering technology. It is being actively progressed.

More recently, for example, as shown in Publication No. 141-1633/10, stimulable phosphor sheets with an extremely wide and steep 041'i exposure have been used, and these phosphor sheets have the above-mentioned By irradiating the same object with radiation transmitted under different conditions, we accumulate and record J-ray images with different image information of specific structures on the phosphor sheet, and these The image is read out by scanning using the 5() destination and converted into a digital signal'f! Alternatively, it has also been proposed to perform the digital subtraction using these digital signals. -H) The fluorescent phosphor sheet refers to the phosphor sheet (X-rays, α-rays, β-rays, γ-rays, electron beams,
When irradiated with ultraviolet rays, etc., a part of the radiation energy is accumulated in the body at 91) degrees Celsius, and when excited light, such as visible light, is irradiated, the phosphor emits photostimulated light according to the accumulated radiation energy field. It has an extremely wide amplitude (
(exposure range) and extremely high resolving power. Therefore, this i′1? If the digital subtraction described above is performed using the radiofrequency information stored in the nine-body sheet, diagnostic ability will be improved by always providing sufficient image information even if the S gland changes. It is possible to obtain a high radiation 1'JJ line image. .

However, 71 copper wire images of the subject, including bone and soft tissue, are stored and recorded on two accumulation f'tn1 optical sheets as shown in the previous figure, and energy subtraction is performed to create a soft tissue erased image. If a subtraction image in which only I50 is extracted is (q), the extracted bone image signal is a difference signal generated by an arithmetic process called subtraction, so S, -'N is low, and therefore the previous extracted image is There is a problem with poor image quality.

(Objective of the Invention) Therefore, an object of the present invention is to provide a method for improving the image quality of an energy subtraction image that can sufficiently improve the image quality of the pre-extracted image as described above, and an apparatus for implementing the method. It is something.

(Structure of the Invention) A method for improving the image quality of an energy sub-99212 image of the present invention uses two or more storable S7 photosensors to improve radiographic images of a subject including bones and soft tissues. In energy subtraction, in which a difference signal is used to form an image in which the rays are erased, the difference signal is used for pixels in a portion of the back of the radiographic image where soft tissue of a predetermined thickness or more is present, and the difference signal is used in the other portions ( In other words, the original image signal read from the sheet is used for the pixels of the area where soft tissue of less than a predetermined thickness exists and the area where the subject is not present. It is characterized by being adapted to play back.

”-2 original image (No. 5 is a digital image read out from one stimulable phosphor sheet & (No. 5 itself, or a digital image read out from each of multiple stimulable phosphor sheets) A signal obtained by adding and Q-averaging the signal may be used.

The above method scans excitation light onto a stimulable phosphor sheet on which a J-ray image has been accumulated, and thereby photoelectrically reads out the stimulated luminescent light emitted from the stimulable phosphor sheet. image information of at least a part of the subject is converted into a digital image signal by irradiation with mutually different energy radiation 1>1 rays that have passed through Shuichi's subject 77 through bones and soft tissue; The image 1g! is obtained from each of the two or more stimulable phosphor sheets on which different emission fJJ ray images are stored and recorded. 111 performs denumbing between corresponding pixels of each digital image signal by the H extraction means;
a rib 1 to traction mouth means for forming a difference circle thereby forming an image in which the bone is erased; 'l Form an image that has been removed & fffl
Based on S''i, identify the portion of the radioradius image in which the soft part II'1 weave having a thickness greater than a predetermined thickness is present, and determine the position (
, 1, and the difference fffi 8 that forms the image in which the soft tissue has been deleted, the original image signal read from the sea 1 to 8 is inputted, and The position signal is inputted from fQ, and by referring to the 16th signal n, a pixel signal of a portion of the tIi ray image where soft tissue of a predetermined thickness or more exists is formed, in which the soft tissue is deleted. and a signal 8 selection means which outputs the difference signal of the pixel signals of the other parts, while outputting the original image signal as the pixel signal of the other parts.
J: is executed on the modified@device.

As mentioned above, in the method of the present invention, the difference signal obtained by subtraction and the original image signal are used to
Since two radiographic images are formed, at least one of the two signals is adjusted to the density range or contrast range of the other so that the area where soft tissue of a predetermined thickness or more is present and the other area are combined without any discomfort. Or density correction (image C
When displaying on RT, it is preferable to perform brightness correction, contrast correction, or density and contrast correction.

(Embodiments) Hereinafter, the present invention will be described in detail based on embodiments shown in the drawings.

Figure 1 shows two accumulative phosphor sheets 1-A, (3) x\A2, which have passed through the same subject 1, which covers soft tissues such as lung fields and blood vessels, and bones, respectively. Change and shine!
The X-ray transmission image of the subject 1 is accumulated and recorded on the first stimulable phosphor sheet △ which shows the state of J-1, and then the areal image of the subject 1 is recorded within a short time. - If you replace A and I3, Ii′1 [4, xl! By changing the tube voltage of the illumination source 3, the energy r- of the transmitted X-rays can be adjusted to
The line images are accumulated and recorded.At this time, it is assumed that the positional relationship of the subject 1 is the same in the human body sheets Δ and B.

In this way, at least 8 pieces of thumbnail information differ from each other.
Two radiation images are stored and recorded on two storage phosphor sheets AXB. Next, these two phosphorescent sheets △
, from B, shown in Figure 2": Una image reader "Rini J:
Both X-ray images are read, and a digital image signal representing the image is sent to IF7. First, while moving the stimulable phosphor sheet 1-A in the direction of arrow Y for sub-scanning,
The rake light 11 from the light source 10 is main-scanned in the X direction by the scanning mirror 12, and the accumulated X-ray energy from the phosphor sheet A is converted into stimulated luminescence light 13 according to the accumulated and recorded X-ray image. The stimulated luminescence light 13 enters the inside of the light condensing plate 14 from one end surface of the light condensing plate 14 made by molding a transparent acrylic plate, and reaches the photomultiple 15 while repeating total reflection inside. , the luminescence appearance of the stimulated luminescence light 13 is outputted as an image@signal S. This outputted image signal qS is converted into a logarithmic transformation 151 including an amplifier and an A/D converter.
Digital image signal Si3 of logarithm value (l ogS) by G
It is converted into l OQ SA . This digital image signal 10QsA is stored in a storage medium 17 such as a magnetic disk. Next, in exactly the same way, accumulate another +
1: The recorded image on the phosphor sheet B is read out, and its digital image signal @ l OQ SB is similarly stored in the storage medium 17 .

Next, the digital image signals 10Qs and 10 (JSB) which have been multiplied by 1!
This figure shows the flow of the sub-translation process performed by applying the image quality improvement method for energy sub-1-rough silicon images by fljJ. First, from the image file 17△ and the image file 17F3 in the storage medium 17, the digital image 1=+bow" gS A, 1og
S, is read out and input to the sub 1-rough circuit lb'5 circuit 18. The liptraxis operator Q circuit 18 subtracts the two digital images 9188' OQ S A and l0 QsB by l- for each corresponding pixel by applying an appropriate <7 mdz1'', and produces a digital difference signal SS U b. =
21・l0GsA-b・log3. +c (since a and b are reduction coefficients, C is a bias component that is approximately constant and is 1-diagonal for every 3 degrees). This difference signal Ss u b
Ha-tan1ililf? After being memorized by Ile 19A,
After) HO (CR through 191 selection circuit 20 in August)
T? ! The difference signal S
The subtraction image is reproduced and recorded by b. FIG. 4 shows an image scanning and recording device as an example of a Rib I/Rough Seacon image reproduction and recording system. The photosensitive film 1130 is moved in the sub-scanning direction of arrow Y, and the laser arm 31 is main-scanned over this photosensitive film 30 in the X direction, and the rate arm 31 is moved to the A10 modulator 3.
2, a visible image is formed on the photosensitive film 30 by modulating it with the image signal from the No. 3 supplier 33. As this image signal for modulation, the difference signal 5
If Sub/B is used, an image of only a desired specific structure can be reproduced and recorded on the photosensitive film 30 by digital subtraction processing.

Here, the image based on the difference signal 5sub is based on the soft part [i5 and bone 6 Assuming that P7ia and b are appropriately selected, the soft tissue 5 is eliminated as shown in FIG. 5B, and f)6 Only the following will be extracted. The above difference (: No. 3 S LJ b is -ta/V image file 19A is stored.

Next, the above-mentioned subtraction calculation circuit 18 performs the same calculation as above 5sub-a・1oasp-b′・1OQsB+c′(
a' and b' are weighting coefficients, and a' is an if bias component so as to have a substantially constant density of 1 or more. Here, by appropriately selecting the coefficients a', l)', and a', when the difference signal 5sub'' is input to the reproduction/recording head 21 as described above and the X-ray image is reproduced, the As shown in FIG. 5(C), the image is one in which the bones 6 have been deleted and only the soft tissue 5 has been extracted.However, this difference (
M4Ssub' is rebirth record 3a! It is not input to the device 21 but is stored in the image file 19B.

As mentioned above, 5sub-a-10GsA-b・IOQSg
If the calculation 10 is performed and the X-ray image is reproduced using the difference signal 3sub, the soft tissue 5 will be erased and only the mother 6 will be extracted, resulting in an image with excellent diagnostic ability.
The image quality of the mother 6 portion of this soft tissue erased image is the original image, that is, the digital image signal 10QSA or the digital image signal logs.

f) The image quality is inferior to that of the bone 6 portion in the image formed from s+. That is, the image signal @ha (that is, the difference signal 5sub) forming the soft tissue erased image was obtained by performing the above-mentioned calculation (subtraction) on the digital image signal 10g5A and the digital image signal tog3B. Of course, the digital image signal is 1oos.
S/N than A or Digital Drawing Sui No. 3 10QsB
This is because

Therefore, as shown in FIG.
0Qs, 4 and the digital image signal 10c)sB are input to the υ1st) averaging circuit 22, and both signals 10QSA, 1og
33 was appropriately weighted.''2 is the additive average Δ,
The added average signal is then subjected to appropriate density J3 and/or contrast correction in the contrast correction circuit 23, and then stored in the -Iζ image file 24. The signal does not represent the original image and is input to the signal selection circuit 20 as the original image 1Δ1°QSo.

The signal selection circuit 20 also receives a position signal output from an identification circuit 25 that receives the difference 1g3sub' (carrying an image in which only the soft tissue 5 is extracted) stored in the image file 19B. 0Sp is input. The identification circuit 25 identifies the portion in the X-ray image where the soft portion 111R5 of a predetermined value or more is present. That is, compare the difference @5sub' with a predetermined threshold,
If the image signal 5sub' of a certain pixel exceeds the threshold value, it is determined that the thickness of the soft part group Jf15 at that pixel is equal to or greater than a predetermined value. In this way, the soft part 1'1 weave 5 of a predetermined thickness or more is formed. ! ? The position where the position is located is identified and the position is indicated by the signal selection circuit 2 as described above.
It is input to 0.

The signal selection circuit 20 receives the above-mentioned position signal Sp and plans the X-ray image, and selects the difference signal 5sul+ from the image file 24 for the pixel at the position specified by the position 11 Sp. , and for pixels in other areas (areas where soft tissue is present below the predetermined corner and where the subject is not present, if there are any areas), +'! from the image funfill 19Δ. Original image signal 1O
Qso) is selected and output.

In this way, the difference signal 3sub is appropriately selected according to the pixel.
Alternatively, using the original image signal 10g5O, in the X-ray image reproduced by the reproduction/recording device 21, there is a soft tissue 5 (for example, the mediastinum) with a thickness greater than a predetermined thickness (the part marked E L is the same as described above). Only the bone 6 is extracted because it is formed by the passing difference signal 5sub.

- The portion where the force f) 6 overlaps with the soft tissue 5 (for example, lung field, etc.) having a thickness less than a predetermined thickness, the original image signal to
Formed by rso but overlapping soft phase 15
This original image signal is 1 oas because it is thin and the radiation r) 1-ray absorption by the soft phase 15 is small. Bone 6 is clearly shown even when using . The areas other than the areas where the soft tissue 5 having a thickness greater than or equal to a predetermined thickness, including the portion where the bone 6 and the soft tissue 5 having a thickness less than a predetermined thickness are small, are divided into two original image signals 10QsA and 10QsB. The original image signal MIO, which naturally has a higher S/N than the difference signal 3sub obtained by processing
Since the reproduced X-ray image is formed by QSo, the reproduced X-ray image is ii! ii\'7 has improved. In other words, in this reconstructed X-ray image, the portion where image quality deterioration occurs due to the retraction process called ribtraction is limited to the portion where the soft part Ill n 5 with a predetermined thickness or more exists. 7ru.

Note that the above-mentioned sea urchin original image signal 1OQso is subjected to the density and/or contrast correction circuit 23 for density and/or contrast.
This correction is performed in the 1 quintillion image signal QlooSo in areas other than those in which soft tissue 5 with a predetermined thickness or more is present, and in the difference f:'jSSUb where there is a soft tissue group 15 with a predetermined thickness or more. The two parts are arranged so as to be synthesized within one X KA image in a way that gives a natural feel with an image level and/or a contrast of 1 to 11. This correction of IHIf J3 and/or contrast 1 to last may be performed manually based on experience, or may be performed manually using the original image signal logso and the difference signal 5sub.
ii I! ' to CRT, etc. kl: '8 Show L l lf
i la? It may also be performed by manual operation. Furthermore, the pixel difference signal S S I at a specific point of the predetermined mother 6
Jb8vA degree and/or contrast correction circuit 23
, and the original image signal +0QsO may be automatically corrected so that this difference signal 5sub and the original image signal logso of the pixel at the specific point have the same value.

Further, such a degree and/or contrast correction may be applied to the difference signal 3sub instead of being applied to the original image signal @lOQS o, or in some cases, the original image signal 10g5° and Difference 8 S S LJ
It may be applied to both of b. Above concentration +15 J
: and/or contrast correction is not necessarily necessary, but if the actual IJ color is used, a natural-looking reproduced image can be obtained in which the two parts of the different signals are combined without any discomfort.

Further, in the identification circuit 25, a soft tissue 5 having a predetermined thickness is used.
The threshold value indicating the value may be fixed, or the image based on the original image signal logso may be displayed on a CRT or the like, and the operator may be able to change it interactively while viewing the display. . Soft tissue 5
The predetermined thickness may be predetermined and fixed, or may be changed as appropriate depending on each radiation image.

Further, as described above, the identification circuit 25 selects the image signal 5su.
By comparing b with a predetermined threshold value, the area where soft silk N5 of a predetermined thickness or more is present is automatically identified.In addition, the image based on the image signal 3subb is displayed on a CRT, etc., and the operator can may be formed by manually inputting the positional information of the portion where the soft tissue 5 having a predetermined thickness or more is present.

In addition, the above fruit 1i! ! In the aspect, the original image signal qlog
79971 images obtained from two storage bamboo light sheets as S0 & (M '3 + oo S A
An average signal of J2 and logS8 is used, but l
ogso as log S'A d; J: bilog
It cannot be said that either one of the SBs may be fl+.
I don't like it.

I have explained the embodiment in which the above two sheets are stored (^1 to the phosphor sheet 1, △, B (required),
1 [5' rays on each of the phosphor sheets] - energy casts a 'fAQ' shadow on the sheet, and these sheets 1- to 1q et al [sign γ
The difference signal logso is obtained by attenuation processing of the Zicle image signal 8, where a, b, and c are correction coefficients, and d is a bias component that makes the difference signal 1og3o approximately constant density. Therefore, the present invention is also applicable to cases where three or more such stimulable phosphor sheets are used.

In addition, in the above-mentioned embodiment, to the two sheets 1-,
The storage and recording of radiation images on B is performed by separately irradiating each sheet with radiation of different energies, but the method of storing and recording radiation images on multiple sheets is not limited to this. , especially in 1984
83486, stimulable phosphor sheet laminate or stimulable phosphor sheet-filter 8
The radiation irradiation may be performed once using an N-layer body. That is, there is no particular restriction on the 1iil shadowing method as long as a plurality of stimulable phosphor sheets are irradiated with bIl rays having different energies that have passed through the same subject.

Although the present invention has been explained above using the terms ``bone'' and soft tissue 1, the energy relief 1 recovery using a stimulable phosphor sheet may be used, for example, in which a treatment fitting or the like is embedded. In order to create an image in which the metal fittings have been removed from the mouth 9 image of a human body, to create an image in which the J contrast medium has been removed from a radiographic image of a human body in which the JB contrast medium has been injected, etc. In this specification, the term "bone" refers to "total flexion," which is considered to be an element equivalent to bone in carrying out the present invention. The above-mentioned [bone J and radiation absorption characteristics are considered to be carp, and include all of the energy extracted on the radiographic image by energy ribs 1 to rays.

(Effects of the Invention) As explained in detail above, according to the present invention, it is possible to reliably improve the image quality of a soft tissue urinary specimen whose soft tissue has been erased by energy subtraction, and The ability to diagnose the problem will be improved to a greater extent than ever before.

[Brief explanation of the drawing]

FIG. 1 is an explanatory diagram showing the step of accumulating and recording radiation images in the method of the present invention, FIG. 2 is a schematic diagram illustrating reading of radiation image information from the stimulable phosphor sheet on which the above-mentioned accumulation and recording has been performed, and FIG. 3 is one implementation of the method of the present invention! A schematic diagram illustrating the details of energy subtraction processing performed by applying the FA-like method. Figure 4 is a schematic diagram showing an example of a subtraction image reproduction recording system. Figure 5 is an energy subtraction process according to the present invention. It is an explanatory view explaining traction. '1...Subject 2...X-ray 3...X-ray source 5...Soft part 1 weave 6...f)
10... Laser light source 11... Laser light
12... Scanning mirror 13... Stimulated luminescent light
15...For 1 to 18...Subtraction performance circuit 20...Signal selection circuit 21...Reproduction/recording device 2
5...Identification circuit A, B...Storage phosphor sheet 1og38, log3B...Digital image signal 3su
b, 3sub'...Difference signal of digital image signal logso...Original image signal S L)...Isho Shinkyu Figure 1 Figure 3

Claims (4)

[Claims] (1) Two or more stimulable phosphor sheets are each irradiated with radiation having different energies that have passed through an object including bones and southern tissue, so that at least a portion of the object is irradiated onto these phosphor sheets. Accumulate and record radiation images with different image information, scan these phosphor sheets with excitation light to convert the radiation images into stimulated luminescence light, and photoelectrically read out the amount of luminescence of this stimulated luminescence light. In energy subtraction, the radiographic image is converted into a digital image signal by subtracting the digital image signal between corresponding pixels of each image to obtain a difference signal forming an image in which the soft tissue of the radiographic image is erased. The radiation image is reproduced by using the difference signal for pixels in a portion where the soft tissue of a thickness greater than a predetermined thickness exists, and by using the original image signal read from the sheet for pixels in other portions. A method for improving the image quality of energy subtraction images. (2) The energy according to claim 1, wherein at least one of the difference signal and the original image signal is subjected to density and/or contrast correction in accordance with the density range and/or contrast range of the other. How to improve the image quality of subtraction images. (3) Energy subtraction according to claim 1 or 2, wherein an average signal of digital image signals read from the two sheets is used as the original image signal. How to improve image quality. (4) Excitation light is scanned across a stimulable phosphor sheet on which a radiation image has been stored and recorded, and the stimulable luminescent light emitted from the stimulable phosphor sheet is thereby read out photoelectrically and converted into a digital image signal. and two or more radiographic images in which image information of at least a part of the object differs from each other by irradiation with radiation having different energies that have passed through the same object including bones and soft tissues. From each of the stimulable phosphor sheets, a subtraction is performed between corresponding pixels of each digital image signal obtained by the image reading means, whereby difference signals forming an image in which the bone and the soft tissue are erased. subtraction calculation means for obtaining a difference signal forming an image in which bones have been erased; and identifying a portion of the radiation image where soft tissue of a predetermined thickness or more is present based on the difference signal forming an image in which bones have been erased. an identification means for outputting a position signal from the identification means; an original image signal read from the sheet and a difference signal forming an image in which the soft tissue has been erased are input, and the identification means outputs the position signal; is input, and with reference to this position signal, outputs a difference signal that forms an image in which the soft tissue is removed as a pixel signal of a portion of the radiographic image where soft tissue of a predetermined thickness or more is present, while other and signal selection means for outputting the original image signal as a pixel signal of a portion.