JP2812875B2 - Bone measurement method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention provides a method for measuring a bone to be examined by using an image obtained by radiography on the bone to be examined. More specifically, there is provided a method of obtaining a bone mass pattern based on the transmitted X-ray amount of the subject bone itself by removing the influence of the soft tissue in the image as rationally as possible, and performing bone measurement using the pattern. .

[0002]

2. Description of the Related Art Various types of bone measurement such as confirmation of the state of growth and aging of human bones, determination of types of bone lesions such as osteoporosis and osteomalacia, progression of the symptoms, and confirmation of effects during treatment. May be performed.

[0003] As a method of such bone measurement, X-rays are applied to a bone to be examined.
MD method for measuring bone density by using a microdensitometer to measure the density of an image on an X-ray photographic film obtained by irradiating the bone (“Bone Metabolism” No. 13)
Vol. 187-195 (1980), "Bone Metabolism", No. 1
4, pp. 91-104 (1981), etc.), there are photon-absorpsiometry, etc., in which bone is measured by irradiating the test bone with gamma rays and measuring the amount of transmitted gamma rays with a detector. .

[0004] The MD method is easy to adopt in that it uses an X-ray film which can be easily obtained using an X-ray image photographing apparatus which is widely used as an apparatus for diagnosing fractures and the like, and is gradually becoming widespread. ing.

[0005] In the conventional bone measurement method, there are many cases in which measurement is mainly performed on typical cortical bone in which there is no other bone close to the test bone and therefore there is little influence on the bone, and the influence of cartilage tissue is small. Was.

[0006]

However, as shown in FIG. 1, there are other bones (ulna) around the bone to be examined (radius) and are likely to be affected by other bones during bone measurement. In addition, as shown in Fig. 2, in the case of bone with a large amount of cancellous bone, the bone mass pattern related to the amount of transmitted radiation becomes complicated, and it is easily affected by other bones and soft tissues. However, in the case of these complicated bone mass patterns relating to the transmitted radiation dose, accurate bone measurement could not be performed. Accordingly, it is an object of the present invention to provide an improved bone measurement method that enables accurate bone measurement even when a bone mass pattern related to such a transmitted radiation dose is complicated.

[0007]

Means for Solving the Problems The inventors of the present invention have conducted intensive studies to achieve the object, and as a result, have found that each value of the bone mass pattern based on the transmitted radiation dose obtained from the influence on the transmitted radiation dose of the subject bone. Is standardized as necessary, and furthermore, using each value of the bone mass pattern, using the product of the second-order difference value and the first-order difference value and / or using the second-order difference value, the region of the subject bone and the soft tissue A plurality of candidate points for a boundary point with only the area are obtained, a point satisfying a predetermined condition is selected from the candidate points, and a primary regression line is determined within a predetermined range according to a pattern until a predetermined condition is satisfied. By repeating the processing according to the above, it was found that accurate and accurate measurement could be performed, and the present invention was reached.

That is, the present invention provides an image inputting step for inputting an image based on the amount of transmitted radiation obtained by irradiating a bone to be inspected, and a method for inputting the image along a measurement line in a portion to be inspected. A step of obtaining a pattern relating to the amount of transmitted radiation of the test part, a step of obtaining two boundary points with soft tissue at both ends of the test bone in the pattern, and approximation by a line connecting the two points Subtracting a pattern portion related to the transmitted radiation amount corresponding to the soft tissue from the pattern to obtain a correction pattern related to the transmitted radiation amount of the test bone itself; Measuring the two points at the boundary between the pattern and the soft tissue at both ends of the bone to be examined in the pattern. A primary neighborhood point is determined, an inner primary regression line is determined in a predetermined range (2) by skipping a predetermined range (1) from the primary neighborhood point toward the center of the subject bone, and further calculates the primary neighborhood. The outer linear regression line is obtained by skipping a predetermined range (3) in the direction away from the center of the bone to be examined from the point and within the predetermined range (4).
A bone measurement method is provided in which the intersection of the next regression line is a secondary neighboring point, and is obtained by repeating such processing until a new neighboring point satisfies a predetermined condition.

Further, the present invention provides a bone measuring method in which the step of obtaining the primary neighborhood uses a second-order difference value and / or a product of the second-order difference value and the first-order difference value in the pattern. .

[0010] In addition, a more preferable embodiment of the bone measuring method of the present invention includes the following.

That is, in the step of obtaining the boundary points between the bone and the soft tissue at both ends of the subject bone in the pattern, a primary neighboring point in the pattern is obtained by a predetermined method, and the subject is examined from the primary neighboring point. A predetermined range (1) is skipped in the direction of the center of the bone, an inner linear regression line is obtained in the predetermined range (2), and a predetermined range (3) is set in a direction away from the center of the subject bone from the neighboring point. If the predetermined range (4) does not satisfy the predetermined condition by skipping, instead of the linear regression line, an average value of the transmitted radiation dose pattern is obtained in the range and a straight line of a constant radiation dose passing through the average value is obtained. Is defined as a new neighboring point, and the process is repeated as necessary until the new neighboring point satisfies a predetermined condition.

Further, the step of obtaining the first-order neighboring point includes first obtaining a plurality of candidate points by using a product of the second-order difference value and the first-order difference value and / or a second-order difference value, and then obtaining the plurality of candidate points. There is a bone measurement method having a step of selecting a point satisfying a predetermined condition from the points.

[0013] Further, the predetermined range (2) is a step from the primary neighboring point to a neighboring point of a point where the inclination of the neighboring point has a large change in the direction of the center of the subject bone, and the predetermined range. (4) The method includes the step of setting a range from a point at which the change in inclination in a direction away from the center of the bone to be examined is away from the primary neighboring point to the vicinity of a point at which the change in inclination is large from a point satisfying the predetermined condition (1). There is a bone measurement method.

Further, there is a bone measurement method including a step of using a condition that the product of the first-order difference and the second-order difference is smaller than a predetermined value as the predetermined condition (1).

Further, there is a bone measurement method characterized in that the processing is performed after each value of the transmitted radiation dose pattern is standardized using a representative value of the pattern.

Further, the image input step is an image reading step by detecting the amount of transmitted light obtained by irradiating the X-ray film of the subject's bone photographed with the reference material having a changed thickness with light. And a step of obtaining the pattern includes a conversion step of converting a density pattern into a thickness of a standard substance based on the relationship between the thickness of the standard substance obtained from the X-ray film and the amount of transmitted light.

In the following, referring to the drawings as needed,
The present invention will be described in more detail. X-rays and γ-rays are preferably used as the radiation in the present invention, and X-rays are particularly preferable. In a generally obtained transmitted radiation dose pattern composed of bone and soft tissue, when measuring only the target bone, the target bone (for example, the second metacarpal bone, the distal end of the radius, the calcaneus, etc.) Therefore, the thickness distribution of the soft tissue around the bone to be examined is different. In addition, since there are individual differences (such as weight differences), it is necessary to correct the influence of soft tissue.

Here, the correction method in the present invention will be described. First, as shown in FIG. 3, for example, the amount of transmitted light obtained by irradiating light on an X-ray photographic film obtained by irradiating an X-ray on a subject bone (shown in a sectional view in FIG. 3A). From the obtained pattern relating to the transmitted radiation dose (hereinafter also referred to as “transmitted radiation dose pattern”; see FIG. 3B), two boundary points between the bone and the soft tissue including the thickness of the soft tissue are obtained. The points are divided by a predetermined line connecting the points. The upper part of the line is the transmitted radiation dose pattern of the bone part, and the lower part is the soft tissue radiation pattern.

Here, the soft tissue is corrected using the soft tissue pattern approximated by the line connecting the two points. That is, by subtracting the approximated soft tissue pattern from the transmitted radiation pattern, a corrected pattern ((c) in FIG. 3) relating to the transmitted radiation of the subject bone itself is obtained. A curve or a straight line is used as a line connecting the two points according to the thickness distribution of the soft tissue around the bone to be examined. For example, when measuring the second metacarpal bone and the distal end of the radius, it is preferable to use a straight line because the soft tissue corresponding to the bone in the transmitted radiation dose pattern is substantially uniform.

Here, how to obtain the parameters D and BMD for bone measurement will be described. D is a bone width, which is determined from a distance between two boundary points between the bone and the soft tissue. On the other hand, BMD
(Bone Mineral Density) converts the correction pattern shown in FIG. 3 (c) into the thickness of the reference material based on the relationship between the thickness of the reference material and the amount of transmitted light, and then starts from the area S of the entire region or the center of the bone width. Area Si corresponding to allocated area width X
Is calculated by dividing by each region width, that is, the bone width D or the region width X allocated from the center of the bone width.

Therefore, in order to accurately and accurately measure the bone to be examined, it is necessary to first determine the boundary between the bone and the soft tissue correctly and accurately. This is because, for example, when the boundary point shifts, the bone mass index BMD also changes.

Next, a method of obtaining the boundary between the bone and the soft tissue in the present invention will be described in detail. As shown in FIG. 4, first, if necessary, each value of the transmitted radiation dose pattern is standardized using a representative value of the pattern. Generally, the transmitted radiation pattern differs depending on the site of the subject bone or individual differences. The pattern level also changes depending on the density of the X-ray film or the intensity of the light source irradiated by the X-ray film. It is effective to perform standardization in order to set predetermined conditions at each stage for obtaining bone and soft tissue. For example, for an 8-bit A / D, the representative value is set to the maximum value of the pattern, and this maximum value is set to 2
It is preferable that each value of the pattern is linearly enlarged so as to be 55.

Next, in order to find a boundary point between the test bone and the soft tissue in the present invention, a method of finding a point having a large change in inclination as a primary neighboring point or a neighboring point of the primary neighboring point will be described. . Generally, for mathematically continuous patterns, 2
A point having a large change in the slope due to the differential is easily obtained. However, in the present invention, for a discrete pattern in a digital processing system, using a second-order difference and / or a product of the second-order difference and the first-order difference depending on the inspected bone has a larger change in a target inclination. It has been found that a point or its neighboring points can be obtained correctly.

Here, in the pattern in which the soft tissue of the cortical bone has a small change as shown in FIG. 5 and the bone and the soft tissue can be clearly distinguished, the transmitted radiation pattern corresponding to the boundary point between the bone and the soft tissue or a point in the vicinity thereof. The peak of the second-order difference y ″ can be easily detected as a peak of y ″ having a certain value or more by searching from the left, and a peak of y ″ having a certain value or more having been searched from the right.

However, in the pattern in which the change of the soft tissue is large (FIG. 6), the difference between the peak (A) corresponding to the boundary point between the bone and the soft tissue or its vicinity and the peak (B) due to the change of the soft tissue itself is small. However, it has been impossible to correctly detect the boundary point or a point near the boundary point under a certain condition for many patterns.

The present inventors have conducted intensive research on such a problem, and as a result, have found that it is preferable to use the product of the second-order difference and the first-order difference, and have reached the present invention. Here, the product of the first-order difference and the second-order difference is mathematically expressed by the following equation.

[0027]

(Equation 1)

(K = 2, 4, 6... Determined in accordance with the transmitted radiation dose pattern of the bone to be examined) In other words, only the point where the change in inclination is large can be emphasized, and in the example of FIG. The difference between the peak (A ′) corresponding to the point or its neighboring point and the peak (B ′) due to the change of the soft tissue itself that is the disturbance is five times as compared with the case where only the second order difference is used,
It has become easy to correctly detect the target boundary point or nearby point.

Next, a case where another bone exists around the bone to be examined will be described. First, it is necessary to select a candidate point of the subject bone from a plurality of candidate points obtained by the method using the product of the first-order difference and the second-order difference described above and the second-order difference. For example, when measuring the radius as shown in FIG. 7, candidate point 1 (ulnar part) and candidate point 2 (radial part) are obtained, but since the level of soft tissue is lower than that of bone, the level of the transmitted radiation dose pattern , The candidate point 2 of the subject bone can be selected. Thus, even if there is another bone around the bone to be examined and the measurement line is over the bone to be examined, the bone to be examined can be measured without having to adjust the length of the measurement line.

As described above, when obtaining the boundary between the bone and the soft tissue (point where the inclination changes greatly) from the discrete (digital) pattern, the second order difference and / or the first order difference and the second order difference Using the product of the differences, first-order neighboring points (also called candidate points)
It is found that the difference between the candidate point and the boundary point between the soft bone and the soft tissue occurs at the candidate point depending on the number of differences or the transmitted radiation dose pattern.

As a method 1, this candidate point (P _{1 in} FIG. 8)
Was used to correct soft tissue from the transmitted radiation dose pattern consisting of bone and soft tissue. However, there was an error in the measurement of bone mass, and measurement could not be performed with good accuracy. In FIG. 8, when the BMD is calculated in the entire bone width region, the amount of error is shown by this method.

As method 2, an even more accurate candidate point is obtained from the candidate points, and the soft tissue is corrected from the transmitted radiation dose pattern composed of the bone and the soft tissue. FIG. 8 shows how to find the new candidate points. That is, from the candidate point toward the center of the subject bone, a primary regression line is obtained near the maximum point of the inclination,
A primary regression line is determined in a predetermined area in a direction away from the center of the bone, and the intersection (P ₂ ) of the regression line is set as a new candidate point.

As shown in FIG. 8, although the error of the data processing in the bone mass measurement is smaller than that of the method 1, the required accuracy may not be satisfied in the sense that a correct boundary point is obtained. As a result of further research to solve this problem, it was found that the above problem could be solved and a better data processing measurement accuracy could be obtained by setting a linear regression region and repeating the linear regression process.

Next, the contents will be described in detail with reference to FIGS. First, a predetermined range (1) to
(4) will be described. That is, as shown in FIG. 9, the predetermined range (1) is determined by statistical means according to the number of differences and the bone to be examined when obtaining the neighboring points from the transmitted radiation dose pattern of the bone to be examined. For example, in the case of the distal end of the radius, the number of differences is 11 and the sampling is 63.5 μm (about 0.
7 mm long) is preferable. Predetermined range (2)
It is preferable to use a point near the point where the inclination changes greatly on the side with the bone (corresponding to peak 1 and peak 2 in FIG. 7). This is because it is possible to avoid regression in an inappropriate range (near the maximum slope point in the method 2) as shown on the left side of FIG.

As shown in FIG. 10, when the influence of other bones is large, it is desirable to perform the regression in a predetermined range (4) where the change in inclination is small. This is because the cartilage grows around the test bone by the test bone or the test subject, so that the pattern may change gradually. That is, the predetermined range (3) may be set to a point from the neighboring point to a point where the product of the first-order difference and the second-order difference is smaller than the predetermined value. Here, for example, in the case of the distal end of the radius, it is preferable that the number of the first-order difference and the number of the second-order difference be 1, when the transmitted radiation dose pattern is standardized, when both are 11. When another bone is present around the test bone, the predetermined range (4) is determined by the distance between the radius and the ulna as shown in the example of FIG. When the distance between the radius and the ulna is short (for example, 0.5 mm), instead of regression in the predetermined range (4), an average value of the pattern in the range is obtained, and a straight line of a constant radiation dose passing through the average is obtained. It is desirable to ask.

Even if the above processing is performed only once, regression must be performed a plurality of times for a pattern as shown in FIG. 11 where sufficient accuracy cannot be obtained. FIG. 11 shows a new candidate point P2 obtained by the first regression from the candidate point P1 by the above method.
Is required. Further, a new candidate point P3 is obtained from the new candidate point P2. Thus P1 → P2 → P3 → P
(Correct boundary point). The number of times of the regression processing is determined based on the transmitted radiation dose pattern to be measured and the required measurement accuracy.

When the bone mass index BMD is obtained by using the pattern related to the radius as shown in FIG. 11, the difference from the BMD value obtained by using a boundary point which is considered to be correct manually is different from the BMD value.
34% by the above method (1), 3.2% by one time linear regression, 2
The linear regression was 1.1%.

The image input means for inputting an image based on the amount of transmitted radiation obtained by irradiating the test bone with radiation includes irradiating the film from the upper or lower surface of the X-ray photographic film with a band light source (LED). The transmitted light is read by a line sensor (CCD). Means for obtaining a pattern relating to the amount of transmitted radiation of the subject along the measurement line in the subject with respect to the input image, and a boundary point 2 between the pattern and the soft tissue at both ends of the subject bone in the pattern. Means for obtaining a corrected pattern relating to the transmitted radiation dose of the subject bone itself by subtracting from the pattern a region relating to the transmitted radiation dose corresponding to the soft tissue approximated by a line connecting the two points. The arithmetic means for performing the arithmetic processing for measuring the bone to be inspected using the correction pattern includes a ROM in which an arithmetic program for the arithmetic processing is input and a RAM for arithmetic and temporary storage. Computer means such as a microcomputer.

FIG. 12 is a flowchart illustrating the bone measuring method of the present invention. FIG. 13 shows an example of a preferred embodiment of the bone measuring device for implementing the present invention. That is, the automatic reading unit 11 arranges a line sensor (CCD) at right angles to the film moving direction, irradiates the film with a band light source (LED) from the upper or lower surface of the X-ray film, and focuses the transmitted light on the line sensor. A pulse motor capable of condensing light by a rod lens arranged so as to obtain a signal such as the intensity of transmitted light corresponding to the density of the X-ray film and at the same time, slightly moving in a direction perpendicular to the line sensor and the band light source. And a microfilm running means using the same.

The film feed controller is a control means for enabling detection of transmitted light by squeezing a specific portion of the X-ray film and controlling the film to run intermittently at a predetermined speed. The CCD driver has a function of controlling so that data stored in the CCD can be extracted at a predetermined timing. Alternatively, the LED controller is a light intensity adjusting means of the light source for adjusting the intensity of the light source according to the density level of the X-ray film.

FIG. 14 shows an example of a radius enlarged and displayed on the CRT image display means in the bone measurement data processing unit 12 in FIG. 1 is a display screen, 2 is a radius, and 3, 4, 5, and 6 show the positions of reference points required for bone measurement. Specifically, it is preferable to connect the midpoints of 3, 4 with the midpoints of 5, 6 and take a perpendicular line at a predetermined position from 3 to serve as a reference measurement line in order to ensure position reproducibility. As the point input means, there are a cursor position display, an instruction control means, a light pen type input means, a method of externally inputting with a touch panel, and the like. The data group read by the automatic reading unit 11 in FIG. 13 is stored by image storage means mainly including an image input / output unit and an image memory in the data processing unit 12, and the stored data group relating to the image is transmitted from the CRTC and the CRT. It is displayed as a pattern of the subject bone enlarged by the main image display means.

Further, the calculating means included in the measuring apparatus of the present invention includes a predetermined position to be measured in the image of the bone to be measured stored in the image storing means based on the reference point input by the point input means. Any device may be used as long as it can be determined and a calculation for bone measurement can be performed using the stored data group relating to the image of the subject bone at the predetermined position. As an example, computer means such as a microcomputer including a ROM to which a calculation program for bone measurement is input and a RAM for calculation and temporary storage are given.

Note that RS232C and MODE shown in FIG.
M is for providing a communication function connected to communication means when used in a bone evaluation system via means of a bone measurement device, and PIO is an interface for inputting and outputting digital control inputs to and from a computer system. It functions as.

In the above-described embodiment of the present invention, an apparatus using an X-ray photographic film has been described. However, the present invention is also applicable to an apparatus for irradiating a subject bone with X-rays on an X-ray image sensor to form an image. Easy to apply.

In the case of such an apparatus embodying the present invention, X
FIG. 15 schematically shows the flow from radiography to bone measurement.
In an apparatus for directly irradiating an X-ray image sensor with an X-ray from an X-ray source 20 to an X-ray image sensor together with a bone 19 to be imaged, an image is used instead of a cassette in which an X-ray film is sandwiched in a conventional X-ray imaging method X-ray photography is performed using the imaging plate 21 and the imaging plate 2 is irradiated with the laser light irradiation means 22 and the light detection sensor 23.
By irradiating the X-ray information stored and recorded in 1 with a laser beam, information proportional to the X-ray intensity can be read as an optical signal. An A / D conversion of the read photoelectric information is performed by the image processing device 25 to obtain an X-ray image 24 of the bone to be inspected. Based on the X-ray image, a bone measurement equivalent to the bone measurement method of the present invention is performed. It can be carried out.

The present invention also includes a method using photon absorption symmetry which detects an image based on a transmitted γ-ray amount obtained by irradiating a test bone with γ-rays and performs bone measurement.

[0047]

According to the present invention, the boundary between bone and soft tissue can be automatically detected correctly when reading an image.
An excellent effect is obtained in which accurate bone measurement can be easily performed on bone with abundant cancellous bone and the like.

[Brief description of the drawings]

FIG. 1 shows an example of a pattern related to a transmitted radiation dose for a radius.

FIG. 2 illustrates an example of a pattern for transmitted radiation dose for cancellous bone-rich bone.

FIG. 3 shows an example of bone measurement using a pattern relating to the amount of transmitted radiation in the present invention.

FIG. 4 is an example of pattern normalization relating to the amount of transmitted radiation in the present invention.

FIG. 5 is an example of a method of obtaining a first-order neighboring point by a second-order difference for a boundary point according to the present invention.

FIG. 6 is an example of a method of obtaining a first-order neighboring point by a product of a first-order difference and a second-order difference according to the present invention.

FIG. 7 shows an example of a method of obtaining a primary neighboring point according to the present invention.

FIG. 8 shows an example of a method for obtaining a boundary point according to the present invention.

FIG. 9 shows an example of a method for obtaining a boundary point in the present invention.

FIG. 10 shows an example of a method for obtaining a boundary point according to the present invention.

FIG. 11 shows an example of a method for obtaining a boundary point according to the present invention.

FIG. 12 is an example of a flowchart of a bone measurement method of the present invention.

FIG. 13 shows an example of a bone measurement device that implements the present invention.

FIG. 14 is an example of image reading in bone measurement according to the present invention.

FIG. 15 shows an example of an apparatus for irradiating a subject bone with X-rays to image it on an X-ray image sensor embodying the present invention.

Claims (2)

(57) [Claims] 1. An image inputting step for inputting an image based on the amount of transmitted radiation obtained by irradiating radiation to a bone to be inspected, and the inspection of the input image along a measurement line in a part to be inspected. Obtaining a pattern relating to the amount of transmitted radiation of the part, and obtaining two boundary points between the pattern and the soft tissue at both ends of the subject bone in the pattern;
A step of obtaining a correction pattern relating to the transmitted radiation dose of the subject bone itself by subtracting from the pattern a pattern portion relating to the transmitted radiation dose corresponding to the soft tissue approximated by the line connecting the points, using the correction pattern A step of calculating the bone to be measured by performing arithmetic processing, and a step of obtaining two boundary points with soft tissue at both ends of the bone to be tested in the pattern, the method comprising: In step (1), a primary neighboring point is obtained, an inner primary regression line is obtained in a predetermined range (2) by skipping a predetermined range (1) from the primary neighboring point toward the center of the subject bone. A predetermined range (3) in a direction away from the center of the bone to be examined from the neighboring point
Is skipped, and an outer linear regression line is obtained in a predetermined range (4), and the intersection of the inner linear regression line and the outer linear regression line is set as a secondary neighboring point, and the new neighboring point satisfies a predetermined condition. A bone measurement method characterized in that the bone measurement method is obtained by repeating such processing until the condition is satisfied. 2. The bone measurement method according to claim 2, wherein the step of obtaining the primary neighborhood uses a second-order difference value and / or a product of the second-order difference value and the first-order difference value in the pattern.