JP4405716B2 - Image processing apparatus and X-ray CT system - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a tissue fat evaluation method, an image processing apparatus, and an X-ray CT system using an X-ray CT image.
[0002]
[Prior art]
In recent years, with the progress of obesity research, it is becoming known that the amount of intramuscular fat as well as subcutaneous fat is related to the obesity constitution. At this time, it is important to quantitatively know the amount of intramuscular fat. On the other hand, X-ray CT apparatuses have rapidly spread in recent years, and in vivo information is acquired non-invasively. If the intramuscular fat of a living body can be estimated using this X-ray CT image, the relationship between obesity and intramuscular fat can be grasped more clearly and beneficially.
[0003]
In the estimation of intramuscular fat using this X-ray CT image, the muscle tissue portion of the X-ray CT image is extracted, and the CT value of this extracted portion is made into a histogram. In this histogram, the histogram value near the CT value having the CT value of subcutaneous fat is regarded as the intramuscular fat component.
[0004]
[Problems to be solved by the invention]
However, according to the above prior art, the CT value of fat has individual differences and has a large variation, and the fat component in muscle cannot be accurately evaluated.
[0005]
In particular, the CT value histogram of the muscle tissue portion has peaks in the vicinity of the CT values of muscle and fat, and these peaks overlap to form one CT value histogram. When the muscle and fat components are separated and detected from this histogram, they are discriminated based on these peak values. However, the CT values of muscle and fat have individual differences and large variations, and it is often difficult to separate the two peak values described above, causing a large error.
[0006]
Therefore, it is important how to realize a tissue fat evaluation method, an image processing apparatus, and an X-ray CT system that can evaluate fat components in a tissue without being affected by variations in CT values due to individual differences. .
[0007]
This invention was made in order to solve the above-described problems due to the prior art, and can evaluate a fat component in a tissue without being affected by variations in CT values due to individual differences. An object is to provide an image processing apparatus and an X-ray CT system.
[0008]
[Means for Solving the Problems]
In order to solve the above-described problems and achieve the object, the tissue fat evaluation method according to the first aspect of the invention acquires tomographic image information of a subject using an X-ray CT apparatus, and images of the tomographic image information A tissue image is extracted from the image, the tissue image is smoothed, the tissue image is sharpened, a difference between pixel values of the smoothed smoothed image and the sharpened sharpened image is obtained, and generated by the difference The variance of pixel values of the difference image thus obtained is obtained.
[0009]
According to the invention according to the first aspect, the tomographic image information of the subject is acquired by the X-ray CT apparatus, the tissue image is extracted from the image of the tomographic image information, the tissue image is smoothed, the tissue image is sharpened, The difference between the pixel values of the smoothed smoothed image and the sharpened sharpened image is obtained, and the variance of the pixel values of the difference image generated by this difference is obtained. It can be used as an index of the amount of intramuscular fat that does not depend on the CT value of a certain muscle and the fat piece in the muscle, and the accuracy of quantitative evaluation of the intramuscular fat can be improved.
[0010]
An image processing apparatus according to a second aspect of the invention is an image processing apparatus that acquires tomographic image information of a subject from an X-ray CT apparatus and performs image processing on the tomographic image information, wherein the tomographic image information Extracting means for extracting a tissue image from the image of the image, smoothing means for smoothing the tissue image, sharpening means for sharpening the tissue image, the smoothed smoothed image and the sharpened image The image processing apparatus includes: a difference acquisition unit that obtains a difference between pixel values of a sharpened image; and a variance calculation unit that obtains a variance of pixel values of the difference image generated based on the difference.
[0011]
According to the second aspect of the invention, the extraction means extracts the tissue image from the image of the tomographic image information, the smoothing means smooths the tissue image, the sharpening means sharpens the tissue image, and the difference The difference between the pixel values of the smoothed smoothed image and the sharpened sharpened image is obtained by the obtaining means, and the variance of the pixel values of the difference image generated by the difference is obtained by the variance calculating means. Therefore, this variance value can be used as an index of intramuscular fat amount independent of the CT value of muscles and intramuscular fat pieces with individual differences, and further improve the accuracy of quantitative evaluation of intramuscular fat. be able to.
[0012]
The image processing apparatus according to the third aspect of the invention is characterized by comprising threshold means for extracting an image area having a pixel value exceeding the threshold or a pixel value not exceeding the threshold.
[0013]
According to the third aspect of the invention, the threshold value means is used to extract an image region having a pixel value exceeding the threshold value or a pixel value not exceeding the threshold value. And the background can be extracted and removed from the region of interest.
[0014]
An image processing apparatus according to a fourth aspect of the invention is characterized in that the extraction means includes a region of interest setting means for setting a region of interest.
According to the fourth aspect of the invention, since the extraction unit sets the region of interest by the region-of-interest setting unit, only the muscle tissue portion on the image can be extracted as the region of interest.
[0015]
The image processing apparatus according to the fifth aspect of the invention is characterized in that the smoothing means includes a smoothing filter having a size of a mask region for smoothing fat pieces in the tissue image.
[0016]
According to the fifth aspect of the invention, since the smoothing means has the size of the mask region for smoothing the fat piece in the tissue image by the smoothing filter, the smoothing means can smooth the fat piece. it can.
[0017]
The image processing apparatus according to the sixth aspect of the invention is characterized in that the sharpening means is provided with a sharpening filter having a size of a mask region for sharpening the fat fragment image in the muscle.
[0018]
According to the sixth aspect of the invention, the sharpening means has the size of the mask region for sharpening the fat piece image in the muscle by the sharpening filter, and therefore the fat piece that does not depend on the CT value. Only image information can be acquired.
[0019]
An image processing apparatus according to a seventh aspect of the invention is characterized in that the smoothing filter and the sharpening filter are provided with changing means for changing the size of a mask region.
[0020]
According to the seventh aspect of the invention, since the smoothing filter and the sharpening filter change the size of the mask region by the changing means, the mask region corresponding to the size of the fat piece having individual differences. Can be.
[0021]
The X-ray CT system according to the invention of the eighth aspect irradiates a subject with an X-ray beam, collects projection data of the subject, and performs tomography of the subject by image reconstruction of the projection data. An X-ray CT apparatus for acquiring image information, a transfer means for transferring the tomographic image information of the X-ray CT apparatus, and an image processing apparatus for performing image processing on the tomographic image information transferred by the transfer means. An X-ray CT system comprising: the image processing apparatus: an extraction unit that extracts a tissue image from an image of the tomographic image information; a smoothing unit that smoothes the tissue image; and a sharpening of the tissue image Sharpening means, difference obtaining means for obtaining a difference between pixel values of the smoothed smoothed image and the sharpened sharpened image, and variance for obtaining a variance of pixel values of the difference image generated by the difference Calculation Characterized in that it comprises a means.
[0022]
According to the eighth aspect of the invention, the image processing apparatus extracts the tissue image from the image of the tomographic image information by the extracting means, smoothes the tissue image by the smoothing means, and smoothes the tissue image by the sharpening means. The difference acquisition unit obtains the pixel value difference between the smoothed smoothed image and the sharpened sharpened image, and the variance calculation unit obtains the pixel value of the difference image generated by the difference. Since the variance is to be obtained, this variance value can be used as an index of the amount of intramuscular fat, independent of the CT value of muscles and muscle fat pieces with individual differences. The accuracy of evaluation can be improved.
[0023]
DETAILED DESCRIPTION OF THE INVENTION
Exemplary embodiments of a tissue fat evaluation method, an image processing apparatus, and an X-ray CT system according to the present invention will be described below with reference to the accompanying drawings. Note that the present invention is not limited thereby.
[0024]
First, the overall configuration of the image processing apparatus and the X-ray CT system according to the present embodiment will be described. FIG. 1 shows a block diagram of an X-ray CT apparatus. As shown in FIG. 1, the apparatus includes a scanning gantry 2 and an operation console 6.
[0025]
The scanning gantry 2 has an X-ray tube 20. An X-ray (not shown) emitted from the X-ray tube 20 is shaped by a collimator 22 into, for example, a fan-shaped X-ray beam, that is, a fan beam X-ray, and a detector array (array). ) 24 is irradiated.
[0026]
The detector array 24 includes a plurality of X-ray detection elements arranged in an array in the fan beam X-ray spreading direction. The detector array 24 is a multi-channel X-ray detector in which a plurality of X-ray detection elements are arranged in an array.
[0027]
The plurality of detector arrays 24 as a whole form an X-ray incident surface curved in a cylindrical concave shape. The detector array 24 is composed of, for example, a combination of a scintillator and a photodiode. However, the present invention is not limited to this. For example, a semiconductor X-ray detection element using cadmium tellurium (CdTe) or an ionization chamber type X-ray detection element using Xe gas (gas) may be used. The X-ray tube 20, the collimator 22, and the detector array 24 constitute an X-ray irradiation / detection device.
[0028]
A data collection unit 26 is connected to the detector array 24. The data collection unit 26 collects detection data of individual X-ray detection elements of the detector array 24. X-ray irradiation from the X-ray tube 20 is controlled by an X-ray controller 28. Note that the connection relationship between the X-ray tube 20 and the X-ray controller 28 and the connection relationship between the collimator 22 and the collimator controller 30 are not shown. The collimator 22 is controlled by a collimator controller 30.
[0029]
The above components from the X-ray tube 20 to the collimator controller 30 are mounted on the rotating unit 34 of the scanning gantry 2. Here, the subject 1 is placed in a recumbent state on a cradle 31 in a bore 29 located at the center of the rotating unit 34. The rotator 34 rotates while being controlled by the rotation controller 36, bombards the X-rays from the X-ray tube 20, and detects the transmitted X-rays of the subject 1 in the detector array 24. The connection relationship between the rotating unit 34 and the rotation controller 36 is not shown.
[0030]
The operation console 6 has an image reconstruction unit 60. The image reconstruction unit 60 is configured by, for example, a computer. A control interface (interface) 62 is connected to the image reconstruction unit 60. The scanning gantry 2 is connected to the control interface 62. The image reconstruction unit 60 controls the scanning gantry 2 through the control interface 62.
[0031]
The data acquisition unit 26, the X-ray controller 28, the collimator controller 30 and the rotation controller 36 in the scanning gantry 2 are controlled through the control interface 62. Note that illustration of individual connections between these units and the control interface 62 is omitted.
[0032]
A data collection buffer 64 is also connected to the image reconstruction unit 60. A data collection unit 26 of the scanning gantry 2 is connected to the data collection buffer 64. Data collected by the data collection unit 26 is input to the image reconstruction unit 60 through the data collection buffer 64.
[0033]
The image reconstruction unit 60 performs image reconstruction using a transmission X-ray signal collected through the data collection buffer 64, that is, projection data. A storage unit 66 is also connected to the image reconstruction unit 60. The storage unit 66 stores projection data collected in the data collection buffer 64, reconstructed tomographic image information, a program (program) for realizing the functions of the apparatus, and the like.
[0034]
The image reconstruction unit 60 is also connected with a display device 68 and an operation device 70. The display device 68 displays tomographic image information and other information output from the image reconstruction unit 60. The operation device 70 is operated by an operator, and inputs various instructions and information to the image reconstruction unit 60. An operator operates the apparatus interactively using the display device 68 and the operation device 70.
[0035]
The image processing apparatus 40 is connected to the storage unit 66 of the operation console 6 through a storage unit 44 by communication means (not shown). The image processing device 40 is disposed on a console independent of the operation console 6, transfers the reconstructed tomographic image information and the like stored in the storage unit 66 to the storage unit 44, and the image processing unit 41. Perform image processing with. Here, the image processing unit 41 is configured by, for example, a computer.
[0036]
In addition, a display device 42 and an operation device 43 are connected to the image processing unit 41. The display device 42 displays tomographic image information and other information output from the image processing unit 41. The operation device 43 is operated by an operator and inputs various instructions and information to the image processing unit 41. The operator uses the display device 42 and the operation device 43 to interactively operate the device.
[0037]
FIG. 2 is a functional block diagram of the image processing apparatus 40. The image processing apparatus 40 includes the tomographic image information 200 from the operation console 6, and further includes a region-of-interest setting unit 210 and a threshold unit 220 serving as extraction units, a smoothing unit 230 serving as an image processing unit, a sharpening unit 240, and It has functions such as a difference acquisition unit 260, a histogram process 250 that forms an image evaluation unit, and a variance calculation unit 270.
[0038]
The tomographic image information 200 is tomographic image information of an axial section of the subject 1 transferred from the storage unit 66 to the storage unit 44.
The region-of-interest setting unit 210 sets an image region for executing image processing when executing an image processing unit to be described later. In this setting, a region of interest (also referred to as a ROI; region of interest) is set on the tomographic image of the display device 42, and various calculations are performed on the pixels in this region. The region of interest is set in position and shape by input from the operation device 43.
[0039]
The threshold value means 220 receives a threshold value from the operation device 43 of the image processing device 40, discriminates a pixel value in the tomographic image that exceeds or does not exceed this threshold value, extracts or removes an area where this pixel value exists, and displays it. Display on the device 42. The threshold means 220 is effectively used to select regions having greatly different pixel values in the tomographic image.
[0040]
The smoothing means 230 smoothes the image of the designated image area, sets a neighboring area called a mask around a pixel in the tomographic image, and sets all pixel values in the mask. This is a smoothing filter that is added or weighted to obtain a new pixel value. The size of the mask is experimentally determined to be a size for smoothing fat pieces generated in a muscle image, which will be described later, and has a size of approximately 3 × 3 pixel matrix. The smoothing unit 230 also includes a changing unit that allows the operator to change the size of the pixel matrix from the operation device 43 of the image processing apparatus 40 in consideration of individual differences of the subject 1.
[0041]
The sharpening means 240 sharpens the image in the designated image area, sets a neighboring area called a mask around a pixel in the tomographic image, and sets all pixel values in the mask. It is a sharpening filter that is weighted and added to obtain a new pixel value. As the sharpening filter, for example, a Laplacian filter is used. The size of the mask is experimentally determined to be a size that sharpens fat pieces generated in a muscle image, which will be described later, and has a size of approximately 3 × 3 pixel matrix. The sharpening unit 240 also includes a changing unit that allows the operator to change the size of the pixel matrix from the operation device 43 of the image processing apparatus 40 in consideration of individual differences of the subject 1.
[0042]
The histogram processing 250 generates a histogram of pixel values included in the region of interest of the tomographic image. In the histogram processing 250, the frequency for each pixel value included in the region of interest is calculated and displayed on the display device 42 as a graph.
[0043]
The difference acquisition unit 260 takes a difference between pixel values of two different tomographic images and generates one tomographic image using the difference as a new pixel value. The execution of the difference is performed not only on the tomographic image but also on the region of interest in the tomographic image.
[0044]
The variance calculating unit 270 detects a peak from the histogram of the region of interest obtained by the histogram processing 250 and calculates the variance of this peak. This dispersion expresses the spread of the peak or the variation of the pixel value from the peak value. The standard deviation sigma, ² the variance sigma, P an average value of pixel values, the pixel value variables a i a parameter P _i, when the total frequency is N,
σ ² = ΣSQR (P _i −P) / (N−1)
σ = root (ΣSQR (P _i −P) / (N−1))
Is calculated by Note that the sum Σ is performed for all parameters i.
[0045]
Next, tissue fat evaluation by the image processing unit 41 will be specifically described with reference to the flowchart of FIG. First, the operator images the subject 1, for example, the thigh (step S301), and obtains projection data of the subject 1. The image reconstruction unit 60 generates tomographic image information of the subject from the projection data and stores it in the storage unit 66. Then, the tomographic image information in the storage unit 66 is transferred to the storage unit 44 of the image processing unit 41 by a transfer unit (not shown).
[0046]
Thereafter, the operator extracts a region of interest consisting only of a tissue image, for example, a muscle tissue, from the tomographic image of the tomographic image information 200 in the storage unit 44 (step S302). In this extraction, the bone region and the background region are selected from the tomographic image by the threshold means 220 and removed, and then a region of interest consisting only of muscle tissue is set. Further, when other tissues such as subcutaneous fat are present in the tomographic image, they are removed by edge detection by differential images or manual setting by eye measurement by the region-of-interest setting means 210, and the like, and only the muscle tissue is included. Set the region of interest. FIG. 4A schematically shows a tomographic image of the thigh and an image of only the muscle tissue that is the region of interest extracted and generated from the tomographic image. In the tomographic image of the thigh in FIG. 4A, a fat part, a muscular tissue, and a bone part are depicted. Here, fat pieces are distributed in the muscle tissue, and the amount of the fat pieces in the muscle tissue is an important parameter for evaluating the degree of obesity.
[0047]
Here, the operator performs three independent processes in parallel on the extracted tissue image that is the region of interest. First, the operator creates a histogram of the tissue image by histogram processing 250 (step S303). An example of this histogram is shown in FIG. The horizontal axis is the CT value or pixel value, and the vertical axis is the frequency of the pixel. Muscle has a CT value of approximately 30-80 HU (Hounsfield Unit), and fat has a CT value of approximately -150 to -50 HU. For this reason, in the histogram of FIG. 5, a peak appears at the location of the CT values of muscle and fat. From this histogram, the amount of intramuscular fat related to human obesity can be estimated. However, there are individual differences in the CT values of muscle and fat, and there are cases where the muscle and fat peaks in FIG. 5 are difficult to separate depending on the situation.
[0048]
Further, the operator smoothes the tissue image by the smoothing means 230 (step S304). FIG. 4B schematically shows a smoothed image of the smoothed tissue image. Fat pieces found in the muscular tissue disappear by smoothing, and a uniform image is obtained.
[0049]
The operator also sharpens the tissue image by the sharpening means 240 (step S305). FIG. 4C schematically shows a sharpened tissue image. Fat pieces seen in the muscular tissue are rendered as high pixel value regions by sharpening.
[0050]
Thereafter, the operator takes the difference between the smoothed image and the sharpened image that are processed images (step S306). FIG. 4D is a diagram schematically showing a difference image of the tissue image of the muscular tissue subjected to the difference. In FIG. 4D, a sharpened fat piece and an afterimage of muscle tissue that has not been removed due to the difference are depicted. Moreover, since the difference of a process image is taken, the pixel value of a difference image turns into a value which does not depend on CT value of a muscle tissue and the fat piece in a muscle.
[0051]
Thereafter, the operator uses the variance calculation unit 270 to calculate a variance value from the difference image obtained in step S306 (step S308). This dispersion value is a numerical value that quantitatively expresses the distribution around the peak reflecting the pixel value of the fat piece and the number of fat pieces that have been clarified.
[0052]
After that, the operator evaluates intramuscular fat based on the histogram of the tissue image of the muscular tissue obtained in step S303 of FIG. 5 and the variance value of the difference image obtained in step S308 (step S309). In this evaluation, a quantitative evaluation of intramuscular fat is performed. For example, considering the variance value obtained in step S308, the histogram of FIG. 5 is separated into a muscle region and a fat region, and the muscle mass and the intramuscular fat mass are calculated from each area. In particular, when the CT values of muscle tissue and intramuscular fat are close to each other, it is difficult to separate the histogram of FIG. 5 into a muscle region and a fat region. In this case, the variance value obtained in step S308 is used as an index of the intramuscular fat mass that does not depend on the CT value.
[0053]
In addition, the operator can obtain a dispersion value of a sample that can measure intramuscular fat mass, such as beef, in step 308 and experimentally create a correspondence table of intramuscular fat mass and dispersion value. Here, since the variance value does not depend on the CT value, this correspondence table can be applied to the human body, and a specific intramuscular fat mass can be estimated from the variance value of the subject 1.
[0054]
As described above, in the present embodiment, a tissue image of a muscular tissue that is a region of interest is extracted from a tomographic image, the tissue image is smoothed, and the tissue image is also sharpened. Since the difference of the processed image is taken and the variance value of the peak is calculated from the histogram of this difference image, this variance value does not depend on the CT value of the muscle tissue and muscle fat pieces with individual differences, It can be used as an index of intramuscular fat mass.
[0055]
In the present embodiment, muscle tissue is extracted and the dispersion value of fat pieces in the muscle is obtained. Similarly, the dispersion value of fat pieces in the liver can be obtained.
[0056]
【The invention's effect】
As described above, according to the present invention, the tomographic image information of the subject is acquired by the X-ray CT apparatus, the tissue image is extracted from the image of the tomographic image information, and the tissue image is smoothed. Since the difference between the pixel values of the smoothed smoothed image and the sharpened sharpened image is obtained and the variance of the pixel values of the difference image generated by this difference is obtained, this variance value Can be used as an index of the amount of fat in the tissue that does not depend on the CT value of the tissue having different individuals and the fat piece in the tissue, and the accuracy of quantitative evaluation of the fat in the tissue can be improved.
[Brief description of the drawings]
FIG. 1 is a block diagram showing the overall configuration of an X-ray CT system.
FIG. 2 is a functional block diagram of the image processing apparatus according to the embodiment.
FIG. 3 is a flowchart illustrating an operation of an image processing unit according to the embodiment.
FIG. 4 is a schematic diagram illustrating image processing of a region of interest according to the embodiment.
FIG. 5 is a diagram illustrating a histogram of a region of interest according to the embodiment.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Subject 2 Scanning gantry 6 Operation console 20 X-ray tube 22 Collimator 24 Detector array 26 Data acquisition part 28 X-ray controller 30 Collimator controller 34 Rotation part 36 Rotation controller 40 Image processing apparatus 40 Image processing part 41 Image processing part 42 Display Device 43 Operation device 44 Storage unit 60 Image reconstruction unit 62 Control interface 64 Data collection buffer 66 Storage unit 68 Display device 70 Operation device 200 Tomographic image information 210 Region of interest setting means 220 Threshold means 230 Smoothing means 240 Sharpening means 250 Histogram Process 260 Difference acquisition means 270 Variance calculation means

Claims (8)

An image processing apparatus that acquires tomographic image information of a subject from an X-ray CT apparatus and performs image processing on the tomographic image information,
Extracting means for extracting a tissue image from the image of the tomographic image information;
Smoothing means for smoothing the tissue image;
Sharpening means for sharpening the tissue image;
Difference obtaining means for taking a difference between pixel values of the smoothed smoothed image and the sharpened sharpened image;
A variance calculating means for obtaining a variance of pixel values of a difference image including an image relating to tissue fat generated by the difference;
An image processing apparatus comprising: The image processing apparatus according to claim 1 , wherein the extraction unit includes a threshold unit that extracts an image region having a pixel value exceeding a threshold value or a pixel value not exceeding the threshold value. It said extracting means, the image processing apparatus according to claim 1 or 2, characterized in that it comprises a region of interest setting means for setting a region of interest. It said smoothing means, the image processing apparatus according to any one of claims 1 to 3, characterized in that it comprises a smoothing filter having a size of the mask region to smooth the fat pieces in the tissue image . It said sharpening means, the image processing apparatus according to any one of claims 1 to 4, characterized in that it comprises a sharpening filter with the size of the mask region to sharpen the fat pieces images of the intramuscular . The image processing apparatus according to claim 4, wherein the smoothing filter includes a mask area size changing unit. The image processing apparatus according to claim 5, wherein the sharpening filter includes a mask area size changing unit. An X-ray CT apparatus that irradiates a subject with an X-ray beam, collects projection data of the subject, and obtains tomographic image information of the subject by image reconstruction of the projection data;
Transfer means for transferring the tomographic image information of the X-ray CT apparatus;
An image processing apparatus that performs image processing on the tomographic image information transferred by the transfer means;
An X-ray CT system comprising:
The image processing apparatus includes an extraction unit that extracts a tissue image from the image of the tomographic image information, a smoothing unit that smoothes the tissue image, a sharpening unit that sharpens the tissue image, and the smoothed image Difference obtaining means for obtaining a difference between pixel values of the smoothed image and the sharpened sharpened image, and variance calculating means for obtaining a variance of pixel values of the difference image including an image relating to tissue fat generated by the difference When,
An X-ray CT system comprising:

Images