JPH09262232A - Bone measurement method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To evaluate accurately the construction deviation and/or relative value in the cavernous bone evaluation by calculating the tensor to indicate the cavernous bone construction applying the image element adjacent data obtained form detailed image corresponding to the bone on the centered cross section view image showing the construction of the centered cross section of the subject bone. SOLUTION: Electron accelerated by a rotating anode 2 of the micro X-ray CT is radiated to the focused area of an X-ray tube 1 so as to generate the X-ray. After having transmitted a subject bone 3, the X-ray is restricted by a slit 4 to have the data with the thickness of a slice, then reaches a sensor 5. A centered part cross section view image of the subject bone is input to the image processing device as the original and binary coded there. From the binary coded image are obtained the barycentric coordinates and the principal axes of inertia to pass through it. After the analysis subject area is set, the bone part in the binary coded image is divided in upper and lower by a line passing its center of gravity of whole the image. Then the cortex bone in the remaining area is deleted, and the reset is set as the cavernous bone so as to execute the analysis on.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a bone measuring method and a bone measuring device. More specifically, according to the present invention, an image obtained by using a means for imaging the structure of a representative cross section of a bone to be inspected is input to an image processing apparatus and a tensor representing a cancellous bone structure is calculated by a predetermined method. The object of the present invention is to provide a method of accurately and rapidly evaluating the relative value of the strength and / or the strength of the bone to be inspected, particularly the strength of the structure anisotropy.

[0002]

2. Description of the Related Art Bone structure evaluation is an important issue in bone strength evaluation. The bone structure is roughly classified into an external shape and an internal structure. In the conventional research, the evaluation index for the internal structure is only a regional index such as cross-sectional area and a mesh index such as bifurcation points, and no specific measurement method is found for the structural anisotropy of the bone cross-section of interest. . Therefore, since the bone structure evaluation index is insufficient, there is a drawback that even the same measurement result has a low correlation with the actual mechanical strength.

On the other hand, in order to observe the internal structure of the bone, the bone is destructively cut out thinly and observed with a microscope, or a low-resolution CT (Computed Tomography) with a resolution of about 200 μm is used.
aphy) device was the only way. Therefore, in the case of the whole company, other tests cannot be performed on the same bone to be inspected, the specimen is cracked when it is cut out (artifacts occur), or in the latter case, even a minute trabecular structure cannot be observed and anisotropy is evaluated. There were various problems such as not being able to do it.

[0004]

The present invention is non-destructive,
In addition, microscopically observing the internal structure of the bone to be examined, we found a measurement method that evaluates the structural anisotropy of the cancellous bone among the indicators that reflect the bone strength well. The purpose of the present invention is to make it possible to evaluate the anisotropy and / or the relative value of strength more accurately and quickly.

[0005]

Means for Solving the Problems As a result of intensive studies to achieve such an object, the present inventors have carried out image processing on an image of a bone to be inspected, and in particular the structural anisotropy and Further, they have found that an index relating to a relative value of strength can be measured, and have reached the present invention. As a means for inputting an image of the cross section of interest of the bone to be inspected, it is non-destructive to use a fine focus X-ray cross-section photography apparatus having a focal dimension and resolution sufficient for measuring the fine structure of cancellous bone. In addition, it is preferable in that the internal structure of the bone to be inspected can be microscopically observed.

That is, according to the present invention, the tensor representing the cancellous bone structure is calculated by a predetermined method from the thinned image corresponding to the bone portion of the cross-sectional image of interest representing the cross-section of the bone to be inspected. It is intended to provide a bone measuring method characterized in that relative values of anisotropy degree and strength can be quantified. The method for obtaining the thinned image in the present invention may be either a method of directly exchanging the image of the bone to be inspected (whether analog or digital) or a method of converting the image once and then converting. Further, the thinning may be performed by any method as long as it corresponds to the bone portion, but a binarized image can be constructed once during the process.

Further, according to the present invention, an image of a cross section of interest of a bone to be inspected, which is obtained by a fine focus X-ray tomography apparatus, is used as an input image, and the input image is binarized to extract a bone portion only. By calculating the tensor representing the cancellous bone structure by the method described above, or by further weighting the trabecular width to the tensor, it is possible to quantify the degree of anisotropy of the strength of the test bone and / or the relative value of the strength. The present invention provides a bone measuring method characterized by the above.

[0008]

Therefore, the present invention makes it possible to observe the cross-sectional image of the bone to be inspected in a nondestructive and microscopic manner by taking the above-mentioned means. The degree of anisotropy of strength and / or the relative value of strength can be quantified.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT An embodiment of the present invention will be described below with reference to FIGS.

First, before measuring an index for evaluating bone strength, a cross section having a characteristic representative of the bone to be inspected is determined,
A tomographic image of the cross section is acquired.

A cross-sectional photograph requires a resolution that allows detailed measurement of the internal structure of bone. Therefore, as a method of obtaining a cross-sectional photographic image, there is a fine focus X-ray cross-sectional photography (hereinafter referred to as μX-ray CT) in which X-rays are generated and irradiated from a fine focus having a focus size of 20 μm or less, preferably 10 μm or less. In particular,
It is sufficient that a spatial resolution of about 10 μm, which is sufficient for observing the trabecular bone, can be obtained.

It should be noted that obtaining a cross-sectional image of the bone to be inspected using μX-ray CT is also preferable in that there are no artifacts that occur during thin sectioning.

[0013] Examples of subjects applicable to the device of the present invention include the bone growth state of animals, confirmation of the degree of aging, the range of types of bone lesions such as osteoporosis and osteomalacia, and the degree of progression thereof. Examples include animal test bones and the like that are required when performing various bone measurements such as confirmation of effects during treatment. Specific examples of the bone to be inspected include a femur, a tibia, a fibula, a lumbar vertebra, a tail vertebra, a sacrum, and an iliac bone. As the other test object, any object can be used as long as it can photograph and input the obtained sectional image to the image processing apparatus. Furthermore, the measurement area applicable to the device of the present invention is μX of the bone to be inspected.
Part or the whole of the cross-sectional image obtained by the line CT can be mentioned.

The bone to be inspected may or may not have soft tissue or soft tissue-like material attached thereto. In this case, the tube voltage is changed to generate and irradiate X-rays twice or more, and μX-ray CT imaging is performed to obtain X-rays of each wavelength.
Method of canceling soft tissue by utilizing difference in X-ray attenuation coefficient between soft tissue and bone with respect to X-ray (DECT: DualEn
Energy Computed Tomography) has been reported and this method may be applied.

A μX-ray CT and a sectional image acquisition method using the μX-ray CT will be described below with reference to FIG.

As the μX-ray CT used in the present invention, an X-ray tube 1 having a focal size of about 8 μm was used. The electrons accelerated by the rotating anode 2 irradiate an area of the focal dimension of the X-ray tube to generate X-rays. After passing through the bone 3 to be inspected, the X-rays are narrowed down by the slit 4 to the information corresponding to the slice thickness and reach the sensor 5. For example, by gradually rotating the bone to be inspected, the irradiation X-ray direction with respect to the bone to be inspected is changed, and X-ray irradiation and sensor detection are repeated each time to reconstruct a μX-ray CT image. The μX-ray CT image obtained in the present invention has a vertical length of 512.
The number of pixels is 512 pixels horizontally, and the size of each pixel is approximately 15 μm in length × 15 μm in width, and the CT value of each pixel is 2 ¹⁶
Expressed in gradation. Further, when the CT value of each pixel is input to the image processing apparatus, it is converted by the equation 1.

[0017]

[Equation 1]

[However, TL: CT value after conversion, C
T _max : maximum CT value, CT _max : minimum CT value] Image processing is performed based on the image obtained as described above. The image processing method according to the present invention will be described in detail with reference to FIGS. A flowchart of image processing for measuring a tensor representing a cancellous bone structure is shown in FIG. 2, and an explanatory view thereof is shown in FIG. A cross-sectional image of the region of interest of the bone to be inspected, which is imaged by μX-ray CT, is input to the image processing apparatus, becomes an original image, and is then binarized.

As a method of binarizing the original image, for example, a discriminant analysis method can be used. Discriminant analysis method, when the pixels in the image are classified by a certain threshold,
It is a method of binarizing with a threshold value such that variations in altitude of pixels of each class are small and variations between classes are large.

More specifically, the within-class variance δW ² is used as an index representing the variation in altitude of pixels in each class.

ΔW ² = ω ₁ σ ₁ ² + ω ₂ σ ₂ ² [where ω _{1 is the} number of pixels of class 1, σ _{1 is} the variance of the luminance of pixels of class 1, ω _{2 is the} number of pixels of class 2, σ ₂ : Dispersion of Luminance of Pixels of Class 2] In addition, inter-class dispersion δB ² is used as an index representing the variation between classes. _{^{δB 2 = ω 1 ω 2 (}} M 1 -M 2) 2 [ however, _{M 1:} Average of the brightness of the pixel class 1,
M ₂ : Average of Luminance of Class 2 Pixels] A threshold value that maximizes the dispersion ratio F ₀ that is the ratio of δB ² and δW ² is obtained.

F ₀ = δB ² / δW ² A method of binarizing with the threshold value is called a discriminant analysis method.

FIG. 3A shows a binarized image obtained by this discriminant analysis method. The barycentric coordinates of the entire image are obtained from the binarized image, and the principal axis of inertia passing through the barycentric coordinates (one-dot chain line in FIG. 3A)
Ask for. The binarized image is rotated so that the principal axis of inertia is parallel to the Y axis.

Next, the analysis target area is set. The analysis target area is an area in which the physical meaning is well reflected, and can be selected, for example, as the most loaded area. Less than,
The lumbar vertebra will be described as an example. In the image which is rotated and positioned as described above, a closed curve (hereinafter, sometimes referred to as a boundary pixel curve) formed of pixels that form a boundary between the spinal cavity and the bone portion is left and right about the inertial principal axis. Divided into 2
The points farthest from the principal axis of inertia in each of the divided boundary pixel curves are defined as P _R and P _L. Of the two points P _R and P _L ,
The length of a perpendicular line from one point closer to the principal axis of inertia to the principal axis of inertia is the spinal cavity radius r.

Next, passing through the center of gravity of the entire binarized image,
A straight line perpendicular to the principal axis of inertia divides the bone part of the bone part binarized image into upper and lower parts. Which of these two halves is selected is determined according to the physical meaning to be analyzed. In this example, since the bone strength is analyzed, the region that contributes little to the bone strength is deleted. That is, the fillet diameter of each of the divided bone parts is calculated, and the bone part having the larger Y-direction fillet diameter is deleted. Further, with respect to the remaining region, a region whose distance from the principal axis of inertia is the spinal cavity radius r × α or more is deleted. Although α may be any value as long as an appropriate analysis target region can be set, in this example, α = 0.9 in order to extract a region meaningful in bone measurement.

Here, when considering a rectangle in which each side is parallel to the X-axis or the Y-axis among the smallest rectangles including bones in the target binarized image, the short side of the rectangle and /
Alternatively, the length of the long side is referred to as a fillet diameter, and the coordinates of two vertices that are diagonal in the rectangle are referred to as fillet coordinates.

Next, the cortical bone portion in the remaining region is manually erased, and the last remaining region is set as the cancellous bone portion as the analysis target. (FIG. 3B).

In calculating the tensor representing the cancellous bone structure of the bone to be inspected, the cancellous bone average width MWT is calculated as a representative value of the cancellous bone trabecular width based on the image of the cancellous bone portion to calculate the cancellous bone. The length of the tensor indicating the orientation of the structure is determined, and then the orientation of the cancellous bone structure is examined based on the image obtained by thinning the cancellous bone image.

The thinning of a binarized image means that the connectivity of the target figures is not changed, that is, the figures are cut or holes are not formed, and finally the line width is reduced. This refers to the operation of extracting the center line having a width of 1.

First, the average width MWT of the trabecular bone of the cancellous bone is determined with reference to the explanatory view of FIG. 3B.

Area Bb of the cancellous bone image Tb. BV is calculated by the equation 2.

[0032]

## EQU2 ## V = Σ _i Σ _j φ [f (i, j)] [where V: area, bone part: φ [f (i, j)] = 1,
Others: φ [f (i, j)] = 0] where i and j are the X axis (horizontal direction of the image) and the Y axis (also vertical direction) of the two-dimensional image space consisting of 512 × 512 pixels. Means the coordinate value of. Also, the range of i and j is

[0033]

(Equation 3)

Is as follows.

Next, the perimeter length Tb. S
Ask for. From these, the average width of the trabecular bone of trabecular bone is calculated by the equation 4.

[0036]

(Equation 4)

Next, the structural anisotropy of cancellous bone will be examined with reference to the explanatory view of FIG.

The product of the thinned image obtained by thinning the target image and the cancellous bone image is obtained to obtain the cancellous bone thinned image.
For each bone pixel in the cancellous bone thinned image, the inter-pixel adjacency information is checked to see if the neighboring eight pixels are bone parts.

Considering the thinned image as shown in FIG. 4, as shown by the numbers around the pixel of interest (i, j) in the bone part,
Considering adjacent pixels of 0 to 7, if the adjacent pixel is a bone part, 1 is added to the count n _{k of} each number, and if it is not a bone part, n _k
Is left as it is. That is, in the case of the pixel of interest in FIG.
Nos. 3, 5 and 7 are bones, so n ₁ , n ₃ , n ₅ and n ₇
Only 1 will be added. Each count n _k (k
= 0,1, ..., 7) is cleared to 0,

[0040]

(Equation 5)

The pixel of interest (i, j) in the bone portion is scanned in the area of.

FIG. 4 shows a thinned image simplified by iend = 6 and jend = 5. In this case, each n _k is as follows.

N ₀ = 1 n ₄ = 0 n ₁ = 4 n ₅ = 4 n ₂ = 1 n ₆ = 1 n ₃ = 5 n ₇ = 5 Therefore, the adjacencies in the X direction and the Y direction are counted in duplicate. In consideration of the fact that

[0044]

[Formula 6] X direction N _X = (n ₀ + n ₁ + n ₃ + n ₄ + n ₅ + n ₇ ) / 2 Y direction N _Y = (n ₁ + n ₂ + n ₃ + n ₅ + n ₆ + n ₇ ) / 2 Therefore, cancellous bone Average trabecular width MWT and X direction / Y
The tensor showing the anisotropy degree of the cancellous bone structure and the relative value of the cancellous bone strength from the number of adjacent directions can be calculated by Equation 7.

[0045]

(Equation 7)

Mathematical expression 7 will be explained with reference to FIG. 5. From equation 6, the tensor showing the structural anisotropy is (N _X , N _Y ), but the length of the tensor is the average width of the trabecular bone of trabecular bone. If converted again as shown, the tensor of Equation 7 is obtained. Therefore, X
The ratio of the component to the Y component indicates the anisotropy of cancellous bone structure. Also,
The length of the tensor indicates the relative value of cancellous bone strength.

FIG. 6 shows the tensor measurement results for the lumbar spine sections of healthy rats and ovariectomized rats. From these results, it is considered that there is a difference in strength between the healthy rat and the ovariectomized rat due to a difference in the cancellous trabecular mean width, but it can be said that both structural directions are almost isotropic. (Here, isotropic means that there is no difference between the X-axis direction and the Y-axis direction.) The bone measuring device of the present invention is characterized by having a configuration for performing such a bone measuring direction. . FIG. 7 schematically shows an example of the aspect of the bone measuring device of the present invention. The measuring device can also be used as computer means of the CT device.

In FIG. 7, the present apparatus is provided with an arithmetic means 12 for performing various arithmetic operations and an image processing means provided in the arithmetic means 12 for processing an image obtained by μX-ray CT imaging and measuring various indexes. Means 6, input means 7 for inputting an instruction to start image processing, text display means 8 for displaying the input instruction, and μX-ray CT for starting image processing
It is provided with an image display means 9 for displaying a process of processing an image obtained by photographing, and an output means 10 for outputting the obtained measurement result.

The bone measuring device according to the present invention is the image storage device 11
Is preferably provided. As such an image storage means, a data group corresponding to positions in a cross section where a digital signal regarding the magnitude of a CT value in an image obtained by μX-ray CT imaging is associated, various structural indexes, and a process of calculating the structural indexes Any memory may be used as long as it can store the data group obtained in step 1. The storage memory size is selected according to the purpose of bone measurement. As a specific example, a high-speed and large-capacity storage device such as a hard disk and a μX-ray CT
A storage device or the like having a large capacity and suitable for carrying, such as a magneto-optical disk, may be used for inputting image data obtained by photographing to the calculation means. In this case, the image data input to the calculation means by the magneto-optical disk is transferred to the hard disk so that it can be processed at high speed.

In the apparatus of the present invention, an image display means 9 such as a CRT (Cathode Ray Tube) for displaying the image of the bone to be inspected stored by the image storage means as shown in FIG. Point input means 13 for inputting a reference point necessary for bone measurement in the image of the bone to be inspected, and calculation for bone measurement relating to the image of the bone to be inspected stored by using the input reference point. The calculation means 12 for performing is provided.

The image display means 9 may be any one as long as it can display a data group consisting of the relationship between the digital signal stored in the image storage means 11 and the position as an image. Examples of CRT and the like are preferable from the viewpoint of resolution and cost. Further, the text display means 8 and the image display means 9 can be combined.

The point input means 13 may be of any type as long as the position can be specified and input as a reference point on the image display means 9. As a specific example, a cursor position such as 13 shown in FIG. There are a display instruction control means, a light pen type input means, a mouse type input means, an external input method using a touch panel, and an automatic input method from a stored image of the bone to be inspected.

Further, as the bone measurement result output means 10 in the bone measurement apparatus of the present invention, any device can be used as long as it can output the measurement result obtained by the calculation, and a concrete example is a hard copy. Examples include dot-type ink printers, thermal printers, laser printers, video printers, and other CRT screens.

[0054]

INDUSTRIAL APPLICABILITY Among the bone strength evaluation indexes, the present invention makes it possible to quantitatively measure the anisotropy degree of the cancellous bone structure and / or the relative value of the cancellous bone strength. In addition, quick bone strength evaluation can be performed. Furthermore, by using μX-ray CT, it is non-destructive and
Microscopically evaluate bone strength.

In the examples, the healthy rat and the ovariectomized rat are compared, and it is considered that the strength of the healthy rat is larger because there is a difference in the average width of the trabecular bone of the cancellous bone, but the structural directions of both are almost the same. The result is isotropic. However, the present invention will be effective in evaluating and diagnosing the lumbar spine section in humans, other animals, or other animal models for drug efficacy evaluation.

[Brief description of drawings]

FIG. 1 is an explanatory diagram of μX-ray CT and a cross-sectional image acquisition method using μX-ray CT.

FIG. 2 is an image processing flowchart for calculating a tensor representing cancellous bone structure.

FIG. 3A shows a binarized image of a cross section of a bone to be examined, and B is a binarized image of a region of interest.

FIG. 4 is an explanatory diagram of a cancellous bone thinned image.

FIG. 5 is an explanatory diagram of a tensor representing cancellous bone structure.

FIG. 6 shows an example of tensor measurement results.

FIG. 7 is a mode example of a bone measuring device of the present invention.

[Explanation of Codes] 1 X-ray tube 2 Rotating anode 3 Bone to be inspected 4 Slit 5 Sensor 6 Image processing means 7 Input means 8 Text display means 9 Image display means 10 Output means 11 Image storage device 12 Calculation means 13 Point input means

Claims (4)

[Claims] 1. A tensor representing a cancellous bone structure is calculated from inter-pixel adjacency information obtained from a thinned image corresponding to a bone portion of a cross-sectional image of interest representing the structure of a cross-section of interest of the bone to be inspected. A bone measuring method characterized in that the degree of bone anisotropy can be quantified. 2. The bone measuring method according to claim 1, wherein the thinned image is obtained from a binarized image obtained by binarizing the cross-sectional image of interest and extracting only a bone part. 3. A relative value of anisotropy and strength can be quantified by weighting a tensor with a cancellous bone trabecular width obtained from the sectional image of interest or the binarized image. The bone measuring method according to claim 1 or 2. 4. The bone measuring method according to claim 1, wherein a sectional image of interest of the bone to be inspected, which is obtained by a micro-focus X-ray tomography apparatus, is used as an input image.