JP2017093879A - X-ray measurement system and X-ray detection data processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an X-ray measurement system for measuring bones and soft tissue, which X-ray measurement system is configured such that bone pixels, air pixels, etc. are not embedded in a soft tissue pixel group (soft tissue area) to be analyzed.SOLUTION: An expansion process is applied to a bone density image 54 to generate a pseudo bone pixel group (margin area) 84 outside an original bone pixel group 82. Under conditions for excluding bone pixel groups (original bone pixel group 82 and pseudo bone pixel group 84) as pixel groups to be excluded, a soft tissue pixel group 90 is determined, based on an L detection value array and an H detection value array. The soft tissue pixel group 90 is composed of two soft tissue pixel groups 86, 88 which are respectively present in both sides of a femur 63. A fat rate, a fat-free rate, etc. are calculated from the soft tissue pixel group 90.SELECTED DRAWING: Figure 3

Description

  The present invention relates to an X-ray measurement system, and more particularly to a system capable of measuring both bone and soft tissue.

  Bone and soft tissue are measured using X-rays for health care or disease diagnosis. Soft tissue is roughly classified into fat and lean body (other than fat). In the case of femur measurement, muscle is dominant as the lean body. For example, for the health management of elderly people, it is desirable to periodically measure bone density and fat percentage (or fat mass). This is because fracture risk increases when bone density is low. Also, fat percentage is an index related to adult diseases.

  When measuring bone and soft tissue using X-rays, if both can be measured simultaneously with a single X-ray irradiation (that is, with a single X-ray beam scan), the burden on the subject can be reduced, In particular, the exposure dose can be reduced. Bone density measuring devices, X-ray imaging devices, CT devices, and the like are known as X-ray measuring devices that measure both or one of bone and soft tissue. The bone density measuring device is a device for measuring a bone density distribution in a subject according to a dual energy X-ray absorption (DEXA) method. It is also possible to calculate the fat percentage and the like based on the DEXA method.

  Patent Document 1 discloses a bone density measuring device capable of measuring fat percentage together with bone density. The apparatus has a function of displaying a composite image composed of a fat percentage image and a bone density image. Patent Document 2 discloses an apparatus that can manually change a pixel type on a bone density image. As an unpublished prior application, there is Japanese Patent Application No. 2015-109141 filed earlier by the present applicant.

JP 2004-147863 A Japanese National Patent Publication No. 10-509075

  When a soft tissue analysis is performed, if a bone pixel enters the analysis target pixel group (or within the analysis target area), a large error occurs in the analysis result of the soft tissue, and the reliability of the analysis result is greatly reduced. This is because the X-ray attenuation observed in the bone pixel is generally much larger than the X-ray attenuation observed in the soft tissue pixel. The shape of the bone is not simple, and it includes a portion such as a thorn (spin), and such a portion may erroneously enter the soft tissue area to be analyzed. It is desired to prevent such intrusion as much as possible and to perform soft tissue analysis. This problem may also occur when an air pixel (a pixel existing in an air region outside the subject) enters the analysis target pixel group (or in the analysis target area). The body side portion of the subject is thin, and due to clothing, the boundary between the soft tissue area and the air area may become unclear.

  In general, it is difficult to strictly discriminate a bone region, a soft tissue region, and an air region from only a pixel value on a simple transmission image (X-ray image) formed based on two-dimensional X-ray detection data. If pixels belonging to other areas enter the analysis target area, the reliability of measurement decreases as described above.

  An object of the present invention is to reduce the possibility of a bone pixel entering a soft tissue pixel group (or soft tissue area) to be analyzed. Alternatively, an object of the present invention is to provide a bone pixel group determined for bone analysis in a system capable of performing computation on bone and computation on soft tissue based on two-dimensional data obtained by scanning with an X-ray beam. It is also used in the determination of the soft tissue pixel group to improve the determination accuracy of the soft tissue pixel group.

  An X-ray measurement system according to the present invention irradiates a subject including bones and soft tissues with X-rays, detects X-rays transmitted through the subject, and two generated by detection of the X-rays. An arithmetic device that processes the dimensional data, the arithmetic device comprising: a raw bone pixel group determining unit that determines a raw bone pixel group based on the two-dimensional data; and the raw bone pixel group based on the raw bone pixel group. An excluded pixel group defining means for defining an excluded pixel group including a bone pixel group; a soft tissue pixel group determining means for determining a soft tissue pixel group based on the two-dimensional data according to a condition for excluding the excluded pixel group; Analyzing means for analyzing the soft tissue based on a soft tissue pixel group.

  According to the above configuration, the original bone pixel group is determined based on the two-dimensional data obtained by X-ray irradiation and detection, and the excluded pixel group is defined based on the determination. Under the condition for excluding the excluded pixel group, the soft tissue pixel group is specified based on the two-dimensional data. Desirably, based on the low energy X-ray detection value array (L detection value array) and the high energy X-ray detection value array (H detection value array) as two-dimensional data, two-dimensional bone before tissue discrimination is obtained by the DEXA method. The density distribution is calculated, and based on this, the raw bone pixel group constituting the bone area is determined with high accuracy. Such a current bone pixel group is used as a special exclusion condition in the determination of the soft tissue pixel group. Therefore, the possibility of pixels corresponding to bones entering the soft tissue pixel group can be greatly reduced. That is, the accuracy of specifying the soft tissue pixel group to be analyzed can be increased. Then, the soft tissue is analyzed based on the soft tissue pixel group. It is possible to perform both bone analysis and soft tissue analysis using single scan data. Bone analysis and soft tissue analysis may be performed using separate scan data. Even in such a case, if the previous raw bone pixel group determination result is inherited by the next soft tissue pixel determination, the accuracy of the soft tissue pixel determination can be improved.

  Preferably, the excluded pixel group defining unit generates a pseudo bone pixel group connected to the original bone pixel group outside the original bone pixel group, and the excluded pixel group is formed by the original bone pixel group and the pseudo bone pixel group. Is configured. According to this configuration, the possibility that the bone pixel is erroneously determined to be a soft tissue pixel can be greatly reduced. For example, even if the target bone has a small part such as a thorn (spin) and it is not recognized as a part of the bone, it is handled outside the target bone in the same manner as the bone area in the soft tissue pixel determination. Providing a margin area can prevent such thorns from being discriminated into soft tissue.

  Preferably, the excluded pixel group defining unit generates the pseudo bone pixel group by applying an expansion process to the original bone pixel group. According to this configuration, the pseudo bone pixel group can be easily added to the outside of the original bone pixel group using an already established technique such as an expansion filter or an expansion operator. In general, the expansion degree can be easily controlled by the expansion process.

  Preferably, the raw bone pixel group determining means is means for forming a bone density image based on the two-dimensional data, and the raw bone pixel group is a pixel group constituting a bone area in the bone density image. The excluded pixel group specifying unit applies the expansion process to the raw bone pixel group.

  Preferably, the arithmetic device includes a soft tissue image forming unit that forms a soft tissue image representing the soft tissue based on the two-dimensional data, and a paint processing unit that performs a paint process on the soft tissue pixel group in the soft tissue image. In the soft tissue image after the paint processing, a band-like shape constituted by the pseudo bone pixel group between a bone area constituted by the original bone pixel group and a soft tissue area constituted by the soft tissue pixel group A margin area is generated. By observing the margin area (a kind of gap), it can be easily confirmed that the expansion process has been properly executed and that a part of the bone has not entered the soft tissue area. The soft tissue image is a black and white image in which the X-ray detection value (or attenuation amount) is expressed in luminance. Desirably, when the soft tissue area is painted, the bone area and the margin area are not painted, and they are expressed in black and white as they are.

  Preferably, the arithmetic device includes an air pixel group specifying unit that specifies an air pixel group based on the two-dimensional data, and in the soft tissue image after the paint processing, the bone area, the margin area, and the soft tissue area. In addition, an air area constituted by the air pixel group is generated. According to this configuration, it can be easily confirmed that the soft tissue area (paint area) does not enter the air area. It is desirable to identify the soft tissue area inside the actual boundary between the subject and the air area. In that case, a margin area occurs between the actual boundary and the identified soft tissue area. As a result, air, clothing, and a portion with a thin body thickness are excluded from the soft tissue area, so that the soft tissue pixel determination accuracy can be improved.

  Preferably, the arithmetic unit includes a correction unit that accepts addition of a soft tissue pixel from a user on the soft tissue image after the paint processing, and the correction unit prohibits addition of a soft tissue pixel to the bone area and the margin area. To do. According to this configuration, erroneous correction can be prevented. Further, since the bone pixel group is automatically excluded from the correction target, the correction work becomes easy.

  Preferably, the measurement apparatus sequentially irradiates the subject with low energy X-rays and high energy X-rays, and sequentially detects low energy X-rays and high energy X-rays transmitted through the subject, thereby The first two-dimensional data corresponding to the low energy X-rays and the second two-dimensional data corresponding to the high energy X-rays are generated, and the soft tissue pixel group determining means is configured to output the first two-dimensional data. And the soft tissue pixel group is determined based on the two-dimensional data of the two. As in the case of bone pixel determination, it is desirable to determine soft tissue pixels with reference to two types of data obtained by X-ray beam scanning. According to this, the determination accuracy can be improved without being affected by noise or the like.

  Preferably, the soft tissue pixel group determination means includes a first condition in which a target pixel does not belong to the excluded pixel group, a second condition regarding pixel data corresponding to the target pixel in the first two-dimensional data, and the first When the third condition regarding the pixel data corresponding to the pixel of interest in the two-dimensional data of 2 is satisfied, the pixel of interest is determined to be a soft tissue pixel. According to this configuration, soft tissue pixels can be accurately determined by combining two types of data under conditions for excluding excluded pixels.

  Desirably, the second condition and the third condition are such that the soft tissue pixel group is determined via a band-shaped margin area inside the outer edge of the subject. In other words, the first condition and the second condition have a margin or more severe conditions so that air, clothing, and a thin body part do not enter the soft tissue area to be analyzed, that is, from the viewpoint of misjudgment. It stipulates. Preferably, the analysis means analyzes at least one of fat percentage, fat mass, lean mass rate, and lean mass.

  According to the method of the present invention, an original bone pixel group is obtained based on two-dimensional data obtained by irradiating a subject including bone and soft tissue with X-rays and detecting the X-ray transmitted through the subject. Determining, applying a dilation process to the raw bone pixel group to generate a pseudo bone pixel group connected to the outside of the raw bone pixel group, the raw bone pixel group and the pseudo bone pixel And determining a soft tissue pixel group to be analyzed based on the two-dimensional data according to a condition for excluding the excluded pixel group consisting of a group.

  This method is executed in an information processing apparatus such as a dedicated arithmetic device or a general-purpose computer. Each process is usually realized as a software function. The program for executing the above method is distributed via a recording medium or transmitted via a communication line.

  Preferably, a bone area composed of the original bone pixel group, a soft tissue area composed of the soft tissue pixel group, and a band-shaped margin area composed of the pseudo bone pixel group between the bone area and the soft tissue area. Displaying the image.

  According to the present invention, it is possible to reduce the possibility that a bone pixel enters a soft tissue pixel group (or a soft tissue area) to be analyzed. Alternatively, according to the present invention, in a system that can analyze bone and soft tissue based on two-dimensional data obtained by scanning with an X-ray beam, the bone density calculation result is also used in the determination of the soft tissue pixel group. The determination accuracy can be improved.

1 is a block diagram showing a preferred embodiment of an X-ray measurement system according to the present invention. It is a block diagram which shows the some function which the arithmetic unit shown in FIG. 1 has. It is a figure for demonstrating the determination method of a soft tissue pixel group. It is a figure which shows the determination conditions of pixel classification. It is a flowchart which shows the operation example of the system shown in FIG.

  DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, preferred embodiments of the invention will be described with reference to the drawings.

  The X-ray measurement system according to the present invention is installed in a medical institution such as a hospital, for example, and includes a measurement device 10 and a calculation device 12 as shown in FIG. The measurement device 10 is a device that performs irradiation and detection of X-rays, and the arithmetic device 12 is a device having a control function and a calculation function. The arithmetic device 12 is constituted by a computer, for example.

  The measuring apparatus 10 has a bed 14 as an inspection table. A subject 16 is placed on the bed 14. The position and posture of the subject 16 are adjusted so that the measurement target site (test site) is included in the X-ray irradiation area at a predetermined position and posture. In the present embodiment, the measurement target site is the soft tissue existing at the proximal end of the femur and its surroundings. In other words, the hip joint and its vicinity. Actually, X-ray measurement is performed on a two-dimensional range centered on the femoral head. Measurements may be made on other sites containing bone and soft tissue.

  The bed 14 has a top plate as a mounting table, and a lower mechanism 18 is provided below the top plate. The lower mechanism 18 has an X-ray generator 20. The X-ray generator 20 generates an X-ray beam having a fan beam shape, for example. The X-ray generator 20 is mechanically scanned so that the X-ray beam moves in the left-right direction of the subject (the penetration direction in FIG. 1). An upper mechanism 22 is provided above the top plate, specifically above the subject 16. The upper mechanism 22 has an X-ray detector 24. The X-ray detector 24 includes a plurality of detection elements arranged in the X-ray beam spreading direction. The X-ray detector 24 moves with the X-ray generator 20. By moving the X-ray generator 20 and the X-ray detector 24, that is, by scanning the X-ray beam, a two-dimensional measurement region is formed. It is also possible to reverse the arrangement of the X-ray generator and the X-ray detector.

  In the present embodiment, the X-ray energy is switched during scanning of the X-ray beam, and specifically, low energy and high energy are alternately switched. Thereby, a detection value pair including a low energy X-ray detection value (L detection value) and a high energy X-ray detection value (H detection value) is obtained in pixel units (that is, in bone density calculation units). One-time scanning of the X-ray beam, that is, measurement on a two-dimensional region, two-dimensional low-energy X-ray detection value array (L detection value array, first two-dimensional data) and two-dimensional high-energy X-ray detection And a value array (H detection value array, second two-dimensional data). That is, the first two-dimensional data and the second two-dimensional data are obtained corresponding to the two-dimensional pixel array.

  In the bone density calculation described later, the L attenuation amount is specified in units of pixels from the L detection value and the L air detection value (detection value obtained when low energy X-rays are irradiated to the air region), and The H attenuation amount is specified from the H detection value and the H air detection value (detection value obtained when low energy X-rays are irradiated to the air region), and the bone density is calculated from the L attenuation amount and the H attenuation amount. Such operations are well known.

  The computing device 12 includes a computing unit 26, an input unit 28, and a display unit 30. The calculation unit 26 has a bone density image forming function, a soft tissue image forming function, a soft tissue analysis function, and the like, as will be described later with reference to FIG. The calculation unit 26 is constituted by a CPU and a program. As will be described later, a bone analysis region of interest is set on the bone density image by using the input unit 28 or by the user's coordinate designation or automatically. In addition, a region of interest for soft tissue analysis is set on the soft tissue image.

  In this embodiment, prior to setting these regions of interest, an initial region of interest is set on the bone density image automatically (or by user coordinate designation). In the present embodiment, the initial region of interest is a region including the region of interest for bone analysis and the region of interest for soft tissue. As will be described later, in the pixel correction process on the bone density image, bone pixels can be added (or filled) and bone pixels can be deleted (or removed) by using a pointing device in the initial region of interest. . The result of automatic bone pixel determination and the result of manual correction thereto are used as an exclusive condition in the determination of the soft tissue pixel group. That is, the bone determination result is inherited by the soft tissue calculation. The display unit 30 displays various images and various measurement results.

  In FIG. 2, a plurality of functions included in the calculation unit 26 illustrated in FIG. 1 are represented by a plurality of blocks. Individual blocks are implemented as software functions. Individual blocks may be configured by dedicated hardware (such as a processor). The configuration shown in FIG. 2 is an example including the connection relationship.

  The bone density image forming unit 32 is a module that forms a bone density image as a two-dimensional bone density distribution according to the DEXA method based on the L detection value array and the H detection value array obtained by X-ray beam scanning. At that time, L air detection values (or L air detection value trains) obtained by detecting low energy X-rays transmitted only through the air layer and high energy X-rays detected only through the air layer are obtained. The detected H air detection value sequence (or H air detection value sequence) is also used. The bone density image forming unit 32 functions as a bone pixel determination unit as described below.

  In the bone density calculation, a histogram is created based on a plurality of bone densities (before bone discrimination) calculated for a plurality of pixels. The histogram typically has a mountain corresponding to bone and a mountain corresponding to soft tissue. A threshold is specified between the two peaks. In the bone density pixel array, each pixel having a bone density equal to or higher than a threshold value is determined to be a bone pixel (raw bone pixel). In general, pixels to be imaged are bone pixels, and pixels corresponding to soft tissue and air are not imaged. As described above, the bone density image forming unit 32 determines the bone pixel group (original bone pixel group) in the process of forming the bone density image. In the illustrated configuration example, the determination result is registered in the pixel type table 40. The entity of the pixel type table 40 is a memory. Bone density image data is sent from the bone density image forming unit 32 to the display processing unit 52.

  The transmission image forming unit 34 is a module that forms a subject transmission image as an X-ray image. In the present embodiment, the transmission image forming unit 34 forms a subject transmission image based on the L detection value array (or L attenuation amount array). In this case, the L detection value (or L attenuation amount) is associated with the luminance. The specimen transmission image is an image representing both soft tissue and bone, but if viewed in contrast to the bone density image, that is, from the viewpoint of an image representing soft tissue that does not appear in the bone density image, the transmission image is It can be referred to as a soft tissue image. It is also possible to form a soft tissue image from the H detection value array (or H attenuation array) or considering it simultaneously. Transmission image data is sent from the transmission image forming unit 34 to the display processing unit 52.

  The expansion processing unit 36 performs expansion processing on the bone density image. Specifically, in a soft tissue analysis ROI (region of interest), which will be described later, a process of adding a band-like pseudo bone pixel group (first margin area) to the outside of the original bone pixel group (bone area) Execute. This is for excluding bone pixel groups (original bone pixel group and pseudo bone pixel group) in the soft tissue pixel determination process. The bone pixel group is understood as an excluded pixel group having an exclusion action or exclusiveness in view of its function. That is, the expansion processing unit 36 functions as an excluded pixel group defining unit. By executing the expansion filter process on the bone density image a predetermined number of times, the bone area can be expanded outwardly in stages. Even when a missing portion is generated inside the bone area, if the size is not more than a certain value, the missing portion is eliminated by the expansion process. That is, the portion is filled with one or a plurality of pseudo bone pixels. By excluding the expanded bone area from the soft tissue area, that is, by using the expanded bone area as an exclusive condition, it is effective that the bone pixel is erroneously determined to be a soft tissue pixel. It can be avoided or the possibility can be greatly reduced. Incidentally, in the present embodiment, the expansion process is executed after the bone pixel group is corrected by the correction unit 42 described later.

  It should be noted that bone pixels (pseudo bone pixels) generated after the fact or additionally as a result of the expansion process may be registered on the pixel type table 40. As long as it is possible to prevent a bone pixel from being mistaken as a soft tissue pixel, a process other than the expansion process may be employed. The expansion amount in the expansion process is, for example, 5 pixels. The condition is variable.

  The soft tissue pixel determination unit 38 is a module that determines whether or not it is a soft tissue pixel (and whether or not it is an air pixel) for each target pixel in the ROI for soft tissue analysis. Specifically, as will be described later with reference to FIG. 4, each pixel of interest is not a bone pixel after expansion processing (first condition), and the L detection value of the pixel of interest (low-energy X-ray detection) Value) is not more than A% of the L air detection value (low energy X-ray air detection value) (second condition), and the H detection value (high energy X-ray detection value) of the target pixel is the H air detection value. When all three conditions (third condition) of B% or less (high energy X-ray air detection value) are satisfied, it is determined that the target pixel is a soft tissue pixel. The determination result is registered on the pixel type table 40.

  By applying the first condition, it is possible to reliably prevent a bone pixel and a pixel having the possibility to be erroneously determined as a soft tissue pixel, or to reduce the possibility of the erroneous determination considerably. By applying the second condition and the third condition, it is possible to reliably prevent an erroneous determination that the pixel corresponding to the air pixel or clothing is a soft tissue pixel, or the possibility of the erroneous determination can be significantly reduced. The A% is 50%, for example, and the B% is 70%, for example. When such second condition and third condition are employed, a strip-shaped margin area (second margin area) is generated inside the actual outer edge of the subject (soft tissue). A soft tissue pixel group is determined inside the margin area. The various conditions described above are merely examples. The contents of these conditions can be changed depending on the situation.

  In the X-ray measurement system, the intensity of X-rays is generally lowered from the viewpoint of reducing the exposure of the subject. Moreover, the transmission power of low energy X-rays is small. Therefore, it is generally difficult to clearly discriminate soft tissues only by the L detection value. That is, there is a possibility that soft tissue cannot be clearly distinguished from air or clothing. On the other hand, although the transmission power of high energy X-rays is large, a filter for monochromatizing the spectrum is inserted on the X-ray beam path upon irradiation with high energy X-rays due to the actual apparatus configuration. Further, when acquiring the air detection value, a filter for further reducing the X-ray intensity is inserted on the X-ray beam path. Under such circumstances, the high-energy X-ray detection value (particularly the air detection value) becomes small, and statistical fluctuations are likely to occur in the detection value. Therefore, it is generally difficult to accurately discriminate soft tissues using only the H detection value. Therefore, in the present embodiment, both the L detection value and the H detection value given to the target pixel are referred to when determining the pixel type in the target pixel unit. The second and third conditions are taken into consideration. By referring to the two types of detection values, there is an advantage that erroneous determination due to noise having different properties can be prevented.

  By providing a certain margin between the soft tissue pixel group and the boundary, it is possible to prevent the air pixel, the clothing pixel, and the like from being erroneously determined as the soft tissue pixel, or to greatly reduce the possibility thereof. Its function is the same as the function of the expansion process. By setting multiple margin areas, the number of pixels that make up the soft tissue pixel group is slightly reduced, but the fat percentage and lean percentage are not affected by the reduction, but rather have the advantage of improved computation accuracy. can get. Regarding the calculation of fat mass and lean mass, as long as the reproducibility of measurement is ensured, there is no particular problem even if the number of pixels constituting the soft tissue pixel group decreases. According to the present embodiment, the accuracy of soft tissue determination can be increased by using the previously obtained determination result of the bone pixel and referring to the two types of detection values. In addition, soft tissue analysis can be reliably performed using a pixel group belonging to the soft tissue.

  The soft tissue pixel determination unit 38 has a function of determining air pixels in addition to a function of determining soft tissue pixels. In the present embodiment, when the target pixel is not a bone pixel (original bone pixel, pseudo bone pixel) or a soft tissue pixel, it is determined that the target pixel is an air pixel. The determination result is also registered on the pixel type table. The pixel type table can be configured as a storage cell array having the same two-dimensional array as the pixel array. A flag indicating the pixel type is stored in each memory cell. However, without configuring such a pixel type table, the determined fat tissue pixel group may be directly referenced or used to calculate the fat percentage, the lean body percentage, and the like.

  The fat percentage calculation unit 48 is a module that calculates the fat percentage from the L detection value, the H detection value, the L air detection value, and the H air detection value for each soft tissue pixel. The fat mass may be calculated for the soft tissue pixel group. The fat-free rate calculating unit 50 calculates the fat-free rate from the L detection value, the H detection value, the L air detection value, and the H air detection value, or from a predetermined coefficient (ratio α) generated in the fat rate calculation process. It is a module that calculates The lean mass for the soft tissue pixel group may be calculated. A method for calculating the fat percentage and the lean percentage is known, and is disclosed in Patent Document 1. Values other than those described above may be calculated as values indicating the composition of soft tissue.

  The display processing unit 52 is provided with data indicating the fat percentage and the lean percentage for each soft tissue pixel. The display processing unit 52 forms a fat percentage image showing a two-dimensional fat percentage distribution based on the fat percentage array calculated for the soft tissue pixel array having a two-dimensional array, and also calculates the fat-free rate calculated in the same manner. Based on the array, a lean body ratio image showing a two-dimensional lean body distribution is formed.

  In this embodiment, the ROI setting unit 44 is a function for setting an initial ROI (common ROI for bone and soft tissue) on a bone density image based on coordinate information automatically determined or specified by a user, user specification Function for setting ROI for bone density analysis on bone density image based on (or automatically determined) coordinate information, and for soft tissue analysis on soft tissue image based on user specified (or automatically determined) coordinate information A function of setting an ROI. In the present embodiment, a bone density analysis ROI and a soft tissue analysis ROI are set in the initial ROI. In this embodiment, the initial ROI is automatically set to include the femoral head and its periphery by automatic analysis of the bone density image. A single ROI or two ROIs may be set.

  The correction unit 42 is a module that adds or deletes bone pixels (original bone pixels) in accordance with a user designation within the initial ROI set on the bone density image. Prior to that, a paint process for painting a bone density pixel group (original bone density pixel group) is applied to the bone density image, and the bone density area is expressed in a predetermined hue. When a pixel that is not a bone pixel is recognized as a bone pixel, the coloring of the pixel is deleted, and the bone pixel certification for the pixel is canceled. When the pixel is recognized as another pixel despite being a bone pixel, the pixel is colored, and the pixel is newly recognized as a bone pixel. In the present embodiment, the pixel type is manually corrected, but the correction may be automated or semi-automated.

  The average bone density calculation unit 46 is a module that calculates the average bone density based on a plurality of bone densities in the bone density analysis ROI set on the bone density image. When the above pixel type correction is executed, the average bone density is calculated based on the corrected bone pixel group belonging to the bone density analysis ROI. The calculation result is sent to the display processing unit 52.

  When the pixel type correction is executed, the expansion processing unit 36 performs an expansion process on the corrected bone pixel group (original bone pixel group) in the soft tissue analysis ROI, and the bone pixel after the expansion process. Create a group. That is, a pseudo bone pixel group that forms a belt-like margin area is added outside the original bone pixel group. The bone pixel group after such expansion processing is used in the soft tissue pixel determination as the excluded pixel group.

  On the soft tissue image, the determined soft tissue pixel group is subjected to paint processing, that is, the soft tissue area is expressed with a predetermined hue. While observing such a soft tissue image, it is possible to manually add or delete soft tissue pixels. Even in this case, the correction unit 42 functions. The correction unit 42 receives the correction of the pixel type only in the pixel group that is not in the bone density pixel group in the ROI for soft tissue analysis. That is, it is prohibited to correct the bone density pixel to the soft tissue pixel. In that case, it may be further prohibited to modify the air pixel to a soft tissue pixel. By applying such a prohibition condition, it is possible to effectively prevent bone, air, clothing, etc. from entering the soft tissue pixel area to be analyzed. In the present embodiment, the pixel type is changed by changing the contents of the pixel type table 40. Based on the pixel type table 40, the display processing unit 52 performs a paint process using the first color of the bone area and a paint process using the second color of the soft tissue area.

  The image generated by the display processing unit 52 is displayed on the display. On the screen of the display, a bone density image, a soft tissue image, a fat percentage image, a lean percentage image, an average bone density, an average fat percentage, an average lean percentage, a fat mass, a lean mass, and the like are displayed.

  FIG. 3 shows the contents of image processing executed in the X-ray measurement system shown in FIG. An initial ROI 62 is automatically set on the bone density image 54. The initial ROI 62 is defined by two horizontal lines a and b and two vertical lines c and d. The horizontal line a circumscribes the apex of the femur (femur image) 63 on the head (bone head image) 64, and the vertical line c circumscribes the end of the head 64 on the body center axis side. For example, a bone head is recognized by image analysis, and two lines are set based on the bone head. However, the lines a and c may be set at positions away from the bone head 64. In the illustrated example, the horizontal line b is on the lower side of the bone density image 54, and the vertical line d is on the right side of the bone density image 54. For this reason, the initial ROI 62 has a relatively large area. The horizontal line b and the vertical line d may be set inside the outer edge of the bone density image 54. However, the initial ROI 62 is set so as to include at least the ROI 80 for soft tissue analysis.

  In the initial ROI 62, the bone pixel group (current bone pixel group) is subjected to paint processing, and necessary correction processing is applied to the bone pixel group. Thereafter, the ROI 72 for bone density analysis is set on the bone density image. In this case, in this example, the upper side and the left side in the initial ROI 62 are directly set as the upper side and the left side of the ROI 72 for bone density analysis. On the other hand, the lower side of the ROI 72 for bone density analysis is manually set to the height of the lower limit (lower end) of the small trochanter (small trochanter image) 68, and the right side thereof is the large trochanter (large trochanter image). It is manually set on the outside of the greater trochanter 70 so as to ensure that 70 enters the ROI 72 for bone density analysis. The average bone density is calculated based on the bone density distribution in the ROI 72 for bone density analysis.

  On the other hand, the ROI 80 for soft tissue analysis is set on the soft tissue image 56. In the present embodiment, the upper side and the left side of the ROI 80 for soft tissue analysis are the same as the upper side a and the left side c in the initial ROI 62. The lower side of the ROI 80 for soft tissue analysis is set at a position that is lowered downward from the lower limit or lower end height of the lesser trochanter 68 by a predetermined distance (for example, 5 cm). The right side of the ROI 80 for soft tissue analysis is set in an air region (air pixel group) 78 so that the boundary 76 is surely included in the ROI 80. Reference numeral 74 denotes a soft tissue area (soft tissue pixel group).

  On the bone density image 54A after the correction process, the expansion process is applied to the inside of the ROI 80 for soft tissue analysis. As a result, a band-shaped margin area is generated around the original bone pixel group 82. It is constituted by the pseudo bone pixel group 84. The result may be reflected on the screen. For example, the paint process may extend to the margin area. In the soft tissue pixel determination, both the original bone pixel group 82 and the added pseudo bone pixel group are treated as constituting the bone pixel group without distinction, that is, as the excluded pixel group.

  When a missing portion of a bone pixel occurs in the bone area, the missing portion is usually filled with the bone pixel by an expansion process. The dilation process is an effective process for compensating and correcting such a small missing portion and preventing erroneous mixing of bone pixels into the soft tissue pixel group.

  When the soft tissue pixel determination is executed after the expansion processing, the soft tissue image 56A subjected to the paint processing is displayed. Specifically, in the ROI 80 for soft tissue analysis, a pixel that satisfies the three conditions described above is determined to be a soft tissue pixel, and a pixel that is not a bone pixel or a soft tissue pixel is determined to be an air pixel. In the ROI 80 for soft tissue analysis, paint processing is performed on the soft tissue area 90 after the determination. The paint area is an area other than the bone area (original bone pixel group) 82, the margin area (pseudo bone pixel group) 84, and the air area 78. The soft tissue area 90 is composed of two soft tissue areas 86 and 88 existing on both sides of the femur 63. Within the soft tissue analysis ROI 80, soft tissue pixels can be added or deleted as long as the above prohibition conditions are not met.

  FIG. 4 shows pixel type determination conditions. It is determined that the bone pixel (original bone pixel) (reference numeral 100) before the expansion processing (but after correction) is a bone pixel as it is (reference numeral 102). When all three conditions (reference numerals 104, 106, and 108) are satisfied for the target pixel (reference numeral 110), it is determined that the target pixel is a soft tissue pixel (reference numeral 112). If the pixel of interest is neither a bone pixel nor a soft tissue pixel (reference numeral 114), it is determined that the attention pixel is an air pixel (reference numeral 116). The pixel type may be determined using other determination conditions. However, even in that case, it is desirable to increase the determination accuracy by considering the bone pixel group after the expansion processing in the soft tissue pixel determination.

  FIG. 5 is a flowchart showing an operation example of the system shown in FIG. In S10, the position and posture of the subject are adjusted so that the body axis or the target bone axis is substantially orthogonal to the beam scanning direction and the measurement site is reliably included in the scanning range. But you may make it perform the adjustment using rotation and parallel movement of the image after imaging.

  In S12, the X-ray beam is scanned, whereby the L detection value array and the H detection value array corresponding to the two-dimensional pixel array are acquired and recorded in the memory. In S14, a bone density image is formed and displayed. At that time, the bone pixel group belonging to the temporary ROI (initial ROI) is subjected to paint processing. When the bone pixel group needs to be corrected, the process proceeds from S16 to S18, and the bone pixel is manually added or removed in S18. In S20, a bone density analysis (bone density calculation) ROI is set, and in S22, the bone density is calculated and displayed.

  In S24, it is determined whether or not to perform a soft tissue analysis subsequent to the bone analysis. When the user instructs execution of the soft tissue analysis, each step after S26 is performed. In S26, a soft tissue image is formed and displayed based on the L detection value array, for example. In S28, the ROI for soft tissue analysis is set by the user on the image. In S30, the expansion process is applied to the bone density image, specifically, to the bone density pixel group belonging to the ROI for soft tissue analysis. In S32, based on the bone density pixel group (excluded pixel group), the L detection value array, and the H detection value array after the expansion process, it is determined whether each pixel of interest is a soft tissue pixel. At that time, air pixel determination is also executed. The soft tissue pixel group is subjected to paint processing.

  In S34, it is determined whether or not the soft tissue pixel group is to be corrected. If it is necessary, manual soft tissue pixel addition / correction is executed in S36. In that case, the prohibition condition mentioned above is applied. In S38, composition analysis (for example, fat percentage calculation and lean percentage calculation) is executed on the soft tissue pixel group, and the execution result is displayed.

  As described above, according to the present embodiment, the bone density calculation result based on the DEXA method can be used for soft tissue analysis, and in particular, the determination accuracy of the soft tissue pixel group can be improved. Since the soft tissue pixel group is specified under the condition of excluding the bone pixel group after applying the process of adding the pseudo bone pixel group to the bone density pixel group (the process of expanding the bone), the target bone is Even if it has a complicated shape, it is possible to reduce the possibility that a part thereof will enter the soft tissue pixel group. Further, according to such processing, even if there is an unnatural defect in the bone area, the bone pixel can be automatically filled there. Since the margin area constituted by the pseudo bone pixel group can be visually confirmed on the image, it can be easily confirmed that the generation of the pseudo bone pixel group is appropriate and that there is no omission in determining the bone area. Furthermore, if a margin area is formed adjacent to the air area, it can be effectively reduced that the air pixel or the like is erroneously determined to be a soft tissue pixel.

DESCRIPTION OF SYMBOLS 10 Measurement apparatus, 12 Arithmetic apparatus, 54 Bone density image, 54A Bone density image after expansion processing, 56 Soft tissue image (subject transmission image), 56A Soft tissue image after paint processing, 90 Soft tissue pixel group (soft tissue as analysis object) area).

Claims (13)

A measuring apparatus that irradiates a subject including bone and soft tissue with X-rays and detects X-rays transmitted through the subject;
An arithmetic device for processing two-dimensional data generated by the detection of the X-ray;
Including
The arithmetic unit is:
Raw bone pixel group determination means for determining a raw bone pixel group based on the two-dimensional data;
An excluded pixel group defining means for defining an excluded pixel group including the original bone pixel group based on the original bone pixel group;
Soft tissue pixel group determination means for determining a soft tissue pixel group based on the two-dimensional data according to a condition for excluding the excluded pixel group;
Analyzing means for analyzing the soft tissue based on the soft tissue pixel group;
An X-ray measurement system comprising: The system of claim 1, wherein
The excluded pixel group defining means generates a pseudo bone pixel group connected to the original bone pixel group outside the original bone pixel group,
The exclusion pixel group is configured by the raw bone pixel group and the pseudo bone pixel group.
X-ray measurement system characterized by this. The system of claim 2, wherein
The excluded pixel group defining means generates the pseudo bone pixel group by applying an expansion process to the original bone pixel group.
X-ray measurement system characterized by this. The system of claim 3, wherein
The raw bone pixel group determination means is a means for forming a bone density image based on the two-dimensional data,
The raw bone pixel group is a pixel group constituting a bone area in the bone density image,
The excluded pixel group defining means applies the expansion processing to the raw bone pixel group.
X-ray measurement system characterized by this. The system of claim 2, wherein
The arithmetic unit is:
Soft tissue image forming means for forming a soft tissue image representing the soft tissue based on the two-dimensional data;
Paint processing means for performing paint processing on the soft tissue pixel group in the soft tissue image;
Including
In the soft tissue image after the paint processing, a band-like margin area constituted by the pseudo bone pixel group is provided between a bone area constituted by the original bone pixel group and a soft tissue area constituted by the soft tissue pixel group. Arise,
X-ray measurement system characterized by this. The system of claim 5, wherein
The arithmetic unit includes an air pixel group determination unit that determines an air pixel group based on the two-dimensional data,
In the soft tissue image after the paint processing, in addition to the bone area, the margin area, and the soft tissue area, an air area formed by the air pixel group occurs.
X-ray measurement system characterized by this. The system of claim 5, wherein
The arithmetic device includes a correction unit that accepts addition of soft tissue pixels from a user on the soft tissue image after the paint processing,
The correction means prohibits addition of soft tissue pixels to the bone area and the margin area;
X-ray measurement system characterized by this. The system of claim 1, wherein
The measurement apparatus sequentially irradiates the subject with low energy X-rays and high energy X-rays, sequentially detects low energy X-rays and high energy X-rays transmitted through the subject,
Thereby, the first two-dimensional data corresponding to the low energy X-ray and the second two-dimensional data corresponding to the high energy X-ray are generated,
The soft tissue pixel group determining means identifies the soft tissue pixel group based on the first two-dimensional data and the second two-dimensional data;
X-ray measurement system characterized by this. The system of claim 8, wherein
The soft tissue pixel group determination means includes a first condition in which a target pixel does not belong to the excluded pixel group, a second condition regarding pixel data corresponding to the target pixel in the first two-dimensional data, and the second two Determining that the target pixel is a soft tissue pixel when a third condition related to pixel data corresponding to the target pixel in the dimension data is satisfied,
X-ray measurement system characterized by this. The system of claim 9, wherein
The second condition and the third condition are such that the soft tissue pixel group is determined via a band-shaped margin area inside the outer edge of the subject.
X-ray measurement system characterized by this. The system of claim 1, wherein
The analysis means analyzes at least one of fat percentage, fat mass, lean mass, and lean mass;
X-ray measurement system characterized by this. Determining an original bone pixel group based on two-dimensional data obtained by irradiating a subject including bone and soft tissue with X-rays and detecting X-rays transmitted through the subject;
Generating a pseudo bone pixel group connected to the outside of the raw bone pixel group by applying an expansion process to the raw bone pixel group; and
Determining a soft tissue pixel group to be analyzed based on the two-dimensional data according to a condition for excluding an excluded pixel group consisting of the original bone pixel group and the pseudo bone pixel group;
An X-ray detection data processing method comprising: The method of claim 11 wherein:
An image including a bone area composed of the bone pixel group, a soft tissue area composed of the soft tissue pixel group, and a band-shaped margin area composed of the pseudo bone pixel group between the bone area and the soft tissue area is displayed. Including steps,
An X-ray detection data processing method characterized by the above.

Images