JPH08266529A - Quantitative analysis of osteosalt - Google Patents

Abstract

PURPOSE: To make it possible to correct correction curves without laboriousness in spite of a fluctuation in the energy distribution of radiation by regarding the deviation rate between the signal value of the reference of a radiation image and the true osteodensity of the standard osteosalt reference to be approximately the same as the deviation rate of another correction curve and correcting the other correction curve with the deviation rate. CONSTITUTION: A data updating processing section 34 for forming calibration functions executes processing to correct the deviation in the calibration function in accordance with the data based on the result of the measurement of an osteodensity measurement processing section 33. The measurement of the osteodensity is executed in an osteodensity measurement processing section 33 in accordance with the data for forming the calibration function for the purpose of correcting the calibration function determined in the data updating processing section 34 for forming calibration functions. The deviation rate between the signal value of the standard osteosalt reference of the radiation image obtd. by irradiating the standard osteosalt reference of a prescribed thickness with radioactive rays and the signal value of the standard osteosalt reference of the prescribed thickness is calculated and the deviation of the plural correction curves by lapse of time is corrected in accordance with the deviation rate.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for quantifying bone mineral in a human body, and more particularly to a method for quantitatively analyzing bone mineral using an energy subtraction technique.

[0002]

2. Description of the Related Art Bone mineral quantification, that is, quantitative measurement of the amount of calcium in bone, is necessary for preventing fractures. That is, knowing a minute change in calcium in bone enables early detection of osteoporosis and has an effect of preventing fracture.

Conventionally, various methods for quantifying bone mineral as listed below have been proposed and implemented.

I) MD method (Microdensitometry): This is an X-ray photography of the middle phalange with an aluminum step wedge (stepwise pattern), and the density is measured by a densitometer,
The amount of X-ray absorption is converted corresponding to the aluminum step wedge, and the value is corrected according to the bone width to quantify bone mineral. The device configuration is simple, but there is a problem in quantitative accuracy. In addition, there is a drawback in that the vertebra, which most commonly represents osteoporosis, cannot be measured.

Ii) SPA method (Single Photon Absorptiome
try): This is 15 cm after low-energy gamma rays penetrate the bone
It is detected by a scintillation detector that is far away, and the weight per unit length of the bone is calculated from the analog calculation by the change of the count number of γ-rays, which is more accurate than the MD method. However, it also has the drawback that it cannot measure the vertebrae, and it requires special management due to the use of radioisotopes and has the drawback that the radiation source must be replaced because it has a half-life period.

Iii) DPA method (Dual Photon Absorptiome)
try): This uses ¹⁵³ Gl, which is a nuclide having two energy peaks of 44 KeV and 100 KeV, as a radiation source.
The amount of bone mineral is measured by the difference in the amount of bone penetration of the energy rays of the species.
Also, there is an advantage that it is possible to measure the bone mineral content and the fat content of the whole body with high accuracy, but this also has a difficulty associated with the use of the radioisotope. In addition, since irradiation is performed by a scanning method, there is a problem that it takes 10 to several minutes for the lumbar spine and 30 to 40 minutes for the whole body.

Iv) QDR method (Quantitative Digited Radi
(ography): (aka DPX method) This is almost the same as the DPA method,
Two types of energy are obtained by combining a pulsed X-ray with a filter instead of a radioisotope, which has good reproducibility and shortens inspection time (about 1/3 of DPA).
) Is effective. Although this method is the most expected method from the viewpoints of both simplicity and performance, even though the examination time has been shortened, it takes about 6 minutes to image the lumbar spine, and further reduction is desired.

V) QCT method (Quantitative Computer Tom
(ography): This is a method for quantitatively determining the bone mineral content of the third lumbar vertebra mainly by CT number using X-ray CT, and it is possible to quantify by cross-section, but the problem is that the device becomes large-scale. There is.

Vi) DQCT method (Dual energy Quantitativ
e Computer Tomography): This is to quantify bone mineral by performing energy subtraction using two kinds of energy in the QCT method, and it is possible to quantify bone mineral without the influence of fat in bone tissue. However, this also has a problem that the device becomes large-scale.

As mentioned above, the conventional methods for quantifying bone mineral have a problem that a simple method has a low accuracy, and a high accuracy method requires a large-scale apparatus and a long examination time.

Therefore, the present applicant has proposed a bone mineral quantitative analysis method using energy subtraction using a stimulable phosphor sheet (see JP-A-4-11473). The method using this energy subtraction means that each of two or more stimulable phosphor sheets is irradiated with radiation having different energies transmitted through a subject including soft tissue and bone tissue, and a radiation image of the subject. Is stored and recorded, these sheets are scanned with excitation light to photoelectrically read the radiation image and converted into a digital image signal, and the digital image signal is subtracted between corresponding pixels of each image to obtain a radiation image. In energy subtraction to obtain a difference signal forming an image of only the bone tissue, the bone mineral reference simulating a human bone in which the bone mineral content is changed stepwise is obtained at the same time when the radiation image of the subject is obtained. , By comparing the density of the shadow of the bone tissue and the density of the bone mineral reference on the image of the bone tissue alone (bone image) A method of quantifying the amount of salt.

Furthermore, this method corrects so-called shading such as unevenness of the stimulable phosphor sheet, irradiation unevenness of radiation, and reading unevenness from the stimulable phosphor sheet irradiated with radiation having different energies that does not pass through the subject. Is obtained and subtraction is performed between this image signal and the radiation image to correct shading and perform a more accurate bone mineral quantitative analysis.

However, in the energy subtraction as described above, since an image is photographed using X-rays having a broad energy distribution, the energy distribution of the X-rays transmitted through the subject is so-called beam which is biased toward the high energy side as a whole. The phenomenon of hardening will occur. That is, due to this beam hardening, for example, on an image from which soft tissue has been erased, even in the same bone structure, the bone in the thick part of the subject has a lower density of the bone than the bone in the thin part of the subject. Happens. This causes a problem that the bone density and the amount of bone mineral measured in the above-described bone mineral quantification are affected by the thickness of the soft tissue overlapping the bone, and the measurement accuracy is deteriorated.

Therefore, the applicant of the present invention has proposed an analysis method for reducing the effect of beam hardening so as to measure bone mineral with high measurement accuracy without being affected by the thickness of the subject (Japanese Patent Laid-Open No. 6-90941). issue).

In this method, a sheet for approximating the thickness of soft tissue and a reference bone mineral reference are overlaid, the thickness of the sheet is changed, and radiography for performing energy subtraction processing is performed each time, and the reference is taken. After performing the subtraction process for extracting the bone mineral reference, a plurality of calibration curves for each thickness of the sheet are created, and the signal value of the sheet is read from the low-voltage image together with the calibration curves. Then, the signal value of the soft tissue is obtained from the low-pressure image obtained by radiography for the energy subtraction processing of the subject having the soft tissue and the bone tissue, and the calibration curve is obtained by correlating this value with the signal value of the sheet. The amount of bone mineral is selected from the signal value of the region of interest after the subtraction processing for selecting and extracting the bone tissue and its calibration curve.

However, in the method described in Japanese Patent Laid-Open No. 6-90941, the calibration curve is obtained only from the soft tissue around the bone in the bone image. When the energy distribution fluctuates, the value of the calibration curve changes corresponding to the fluctuation, and as a result, it becomes impossible to obtain an accurate bone mineral content.

Therefore, a certain function is obtained based on the bone mineral reference, the obtained function is used as a basic calibration curve, and the calibration curve is corrected by the function obtained from the soft tissue around the bone in the bone image. A method using two types of functions has been proposed so that a final calibration curve can be obtained.

[0018]

However, in an apparatus for capturing a radiation image, the fluctuation of the energy distribution of radiation becomes large, or even if the fluctuation is small, the fluctuation is accumulated over time, resulting in fluctuation of the energy distribution. Can be large. In the method of obtaining the calibration curve by the above-mentioned two kinds of functions, there is a certain degree of control width for allowing the energy distribution of the radiation, and therefore some fluctuation of the energy distribution is acceptable. However, if the variation of the radiation energy distribution exceeds this control range, the variation cannot be corrected completely, and it is necessary to re-calculate the function for correcting the basic calibration curve.

On the other hand, if the variation of the energy distribution of the radiation exceeds the control range, if it is not corrected, the value of the calibration curve will change, and it will not be possible to obtain an accurate bone mineral content.

In view of the above circumstances, it is an object of the present invention to provide a bone mineral quantitative analysis method capable of correcting a calibration curve without trouble even if the energy distribution of radiation during imaging changes. .

[0021]

A method for quantitatively analyzing bone mineral according to the present invention is a method for quantitatively measuring bone mineral in a bone tissue based on a signal value of a region of interest in the bone tissue and the calibration curve as described above. In the bone mineral quantitative analysis method for performing analysis, the calibration curve is defined by a first function calculated based on a signal value of the bone mineral reference in the bone image, and one of the radiation images of the plurality of radiation images. The first function is corrected based on the thickness of the soft tissue around the region of interest in the image, and is obtained based on a second function calculated in advance,
Further, a sheet for approximating the thickness of the soft tissue composed of a substance having a radiation absorption coefficient equivalent to that of the soft tissue is superposed on the second function with a reference bone mineral reference having a predetermined bone mineral content. The signal value of the sheet corresponding to the thickness of the soft tissue in one radiation image of the plurality of radiation images obtained by irradiating the radiations with different energies each time while changing the thickness of Performing energy subtraction processing to obtain an image of only the reference bone mineral reference by correcting the signal value according to the thickness of the sheet using the radiation image of
Based on the reference bone mineral reference image obtained by the energy subtraction processing, a plurality of correction curves representing the relationship of the bone mineral content to the signal value of the radiation image are obtained for each signal value of the sheet corresponding to the soft tissue having different thickness. Calculated by, among the reference bone mineral reference different thickness of the sheet, the signal value of the reference bone mineral reference in the radiographic image obtained by irradiating the reference bone mineral reference of a predetermined thickness with radiation, It is characterized in that a deviation amount of a signal value of a reference bone mineral reference having a predetermined thickness from a reference value is calculated, and the deviations of the plurality of correction curves with time are corrected based on the deviation amount. .

Here, the first function and the second function are
A function that represents the relationship between the signal value that is the basis for obtaining the calibration curve and the actual bone mineral content. Specifically, the first function is obtained from the signal value of the bone mineral reference in the bone image and the actual amount of bone mineral of the bone mineral reference, and the change of the energy distribution of radiation by this first function. Is to absorb. The second function is for correcting the actual fluctuation of the amount of bone mineral represented by the first function due to the beam hardening and the scattered radiation due to the fluctuation of the subject thickness.

Further, the reference value of the signal value of the reference bone mineral reference refers to a value representing the actual amount of bone mineral contained in the reference bone mineral reference by a signal value.

Further, the deviations of the plurality of correction curves with time include deviations due to the passage of time, deviations in energy distribution due to fluctuations in the voltage between the radiation sources, deviations due to fluctuations in the thickness of the subject, and the like. .

In the bone mineral quantitative analysis method described above, the deviation amount is calculated in a plurality of radiation images obtained by irradiating the reference bone mineral reference having the predetermined sheet thickness with radiation a plurality of times. It is preferable to perform it based on a representative value such as an average value of a plurality of signal values of the reference bone mineral reference and a predetermined reference value.

[0026]

According to the method for quantitatively analyzing bone mineral according to the present invention, the calibration curve is obtained by the above-mentioned first function and the second function as the correction curve for correcting the first function. In the quantitative analysis method of bone mineral, the deviation of the correction curve over time was determined by photographing radiation on the reference bone mineral reference of a predetermined thickness among the reference bone mineral reference of various thicknesses for which the correction curve was obtained. The amount of deviation between the signal value of this reference in the obtained radiation image and the true bone density of the reference bone mineral reference, which is the reference value of the signal value of this reference, is calculated, and this deviation amount is the deviation amount of the other correction curve. It is considered that they are substantially the same, and the other correction curves are corrected by this shift amount.
In this way, since the deviation amount of the correction curve with time is calculated and the correction curve is corrected based on this deviation amount, an accurate correction curve is always obtained regardless of the change of the energy distribution or the change of the subject thickness with time. Therefore, it is possible to obtain the calibration curve with high accuracy and improve the accuracy of the bone mineral quantitative analysis.

Further, since it is not necessary to obtain the correction curve again, the labor of the operator can be reduced.

Further, by correcting the correction curve based on the representative value of the plurality of signal values in the plurality of radiographic images obtained by irradiating the reference bone mineral reference with the radiation a plurality of times, the deviation amount at a certain photographing time. Even when the values are significantly different, the deviation amount is averaged, so that the correction curve can be corrected with higher accuracy.

[0029]

Embodiments of the present invention will be described below with reference to the drawings.

First, means and steps used in the method of the present invention will be described.

FIG. 1 shows a bone thickness reference 2 used in the method of the present invention, and a thickness for approximating the thickness of the soft tissue composed of the bone mineral reference 2 and the soft tissue and a substance having an equivalent radiation absorption coefficient. It is a figure showing the variable reference which overlaps with the sheet which can be changed according to it. The bone mineral reference 2 is a known bone mineral reference of the X-ray absorption amount, which has a pattern in which the X-ray absorption amount is gradually changed. As shown in FIG. 1, the bone mineral reference 2 has a structure in which references 2a, 2b, 2c, 2d, 2e and 2f in which the amount of bone mineral, that is, the content (wt%) of CaCO ₃ are different stepwise are arranged. The amount of CaCO ₃ is known in advance. FIG. 1 (a) shows a bone mineral reference 2 and FIG. 1 (b) shows this bone mineral reference 2.
2 shows a variable reference 8 formed by stacking sheets whose thickness can be changed by stacking the sheets 4a, 4b and 4c.

Before explaining the method for quantitatively analyzing bone mineral of the present invention, an energy subtraction process will be described taking a subject having soft tissue and bone tissue as an example.

FIG. 2A shows two stimulable phosphor sheets A,
FIG. 3B shows a state in which the radiation 14 transmitted through the same subject 12 having soft tissue and bone is irradiated with different energies. That is, the radiation transmission image of the subject 12 is accumulated and recorded on the first stimulable phosphor sheet A, and then the stimulable phosphor sheets A and B are quickly replaced within a short time, and at the same time, the tube voltage of the radiation source 16 is changed. Then, the radiation image of the subject 12 having different transmitted radiation energy is accumulated and recorded in the stimulable phosphor sheet B. At this time, the positional relationship of the subject 12 is the same between the stimulable phosphor sheets A and B.

Further, FIG. 2 (b) shows a state in which two stimulable phosphor sheets are superposed, and a radiation F transmitted through a subject 12 is radiated with a filter F for partially absorbing radiation energy interposed therebetween. Thus, the stimulable phosphor sheets A and B are simultaneously irradiated with radiation having different energy levels (so-called one-shot energy subtraction). The one-shot energy subtraction is described in detail in JP-A-59-83486.

Next, these two stimulable phosphor sheets A,
From B, a radiation image is read by an image reading means as shown in FIG. 3, and a digital image signal representing the image is obtained.
First, while moving the stimulable phosphor sheet A in the direction of arrow Y for sub-scanning, laser light is emitted from the laser light source 20.
21 is main-scanned in the X direction by the scanning mirror 22, and the accumulated radiation energy from the phosphor sheet A is diverged as stimulated emission light 23 according to the accumulated and recorded radiation image.
The stimulated emission light 23 enters the inside of this light guide 24 from one end surface of a light guide 24 made by molding a transparent acrylic plate, and while repeating the total reflection inside, reaches the photomultiplier 25, and stimulated emission light is emitted. The light emission amount of the light 23 is output as the image signal S. The output image signal S is converted into a digital image signal log S _A of logarithm by the logarithm converter 26 comprising an amplifier and A / D converter (log S). This digital image signal log S
_A is stored in the storage medium 27 such as a magnetic disk.
Next, in exactly the same manner, the recorded image of the other stimulable phosphor sheet B is read out, and its digital image signal log
S _B is likewise stored in the storage medium 27.

FIG. 4 is a diagram schematically showing the subtraction processing based on the two digital image signals log S _A and log S _B obtained as described above. Image 28 is an image obtained by photographing the image signal log S _A is the tube voltage of the radiation source 16 upon capturing an image carrying low pressure (e.g., 60 kV). The image 29 is an image obtained by photographing with a tube voltage of the radiation source being high voltage (120 kV). In these two images 28 and 29, the shadows 28a and 29a of the bones of the human foot and the shadows 28b and 29b of the soft tissue are shown.

Here, the soft tissues 28b and 29b differ greatly in thickness and the like between individuals, and it is difficult to quantify the bone mineral content as it is. Therefore, the subtraction process, that is, the two images 28 and 29 corresponding to each other is performed. for each pixel _{log S = a · log S a} -B · log S B + C ( where, a, B, C represents a coefficient) soft tissue shadow 28b by performing arithmetic processing, 29b
A bone part image 30 in which is deleted is obtained.

Here, the method for quantitative analysis of bone mineral of the present invention will be described.

Regarding this method, the first function based on the bone mineral reference for obtaining the calibration curve, each step for obtaining the second function for correcting this first function, and the bone mineral content are calculated. The step of obtaining and dividing the step will be described.

First, the step of obtaining the first function will be described.

First, as shown in FIG. 4, the signal value of the bone mineral reference 2 imaged in the bone image 30 obtained by the subtraction process is obtained. This signal value is obtained for each of the references 2a to 2f of the bone mineral reference 2, and the signal values (Sba, Sbb, Sbc, Sbd, Sbe, Sb corresponding to the obtained signal values 2a to 2f are associated with each other.
f).

On the other hand, the bone density values (Ba, Bb, Bc, Bd, Be, Bf) of the references 2a to 2f are known in advance. Then, the bone density values Ba to Bf and the signal value S
Based on ba to Sbf, the coefficients a, b, and c of the following formula (1) f ₁ = a · Sb ² + b · Sb + c (1) are determined by the least squares method, and the first function f ₁ As a function of the signal value Sb of the bone image 30.
That is, the signal value Sba corresponding to each of the bone density values Ba to Bf
The ~Sbf into Equation (1), Ba = a · Sba 2 + b · Sba + c Bb = a · Sbb 2 + b · Sbb + c Bc = a · Sbc 2 + b · Sbc + c ... (2) Bd = a · Sbd 2 + b · sbd + c Be = a · Sbe 2 + b · Sbe + c Bf = asked six equations ^{a · Sbf 2 + b · Sbf} + c, coefficients a by the least square method from the equation,
b and c are obtained. The first function thus obtained is as shown in FIG.

The calculation of this function f ₁ is performed for each photographing.
That is, it is obtained each time a subtraction image is obtained.

Then, the second function for correcting the first function
The step of obtaining the function of will be described step by step.

(1) Forming a high-voltage / low-voltage image: First, a variable reference shown in FIG. 1 (b) is formed. Using this variable reference as an object, the high-voltage image and the low-voltage image are accumulated and recorded on the stimulable phosphor sheet by, for example, one-shot energy subtraction imaging shown in FIG. This photographing is performed every time the thickness of the sheet included in the variable reference is changed. Here, three types of sheet thickness of 3 cm, 5 cm, and 7 cm are photographed. That is, the high-voltage image and the low-voltage image are recorded only 3 times the thickness of the sheet. At this time, the bone mineral reference shown in FIG. 1 (a) is also accumulated and recorded on the stimulable phosphor sheet together with the variable reference. The images recorded and recorded are shown in Fig. 3.
The digital image signal of each image can be obtained by being read by the image reading device shown in FIG.

(2) Obtaining a sheet signal value: A digital image signal is read from each low-voltage image to obtain a sheet signal value. When the signal values a, b, and c of this sheet are equal to the soft tissue signal values obtained in the step of obtaining the bone mineral content, the sheet is considered to indicate the soft tissue.

(3) Perform energy subtraction processing: For each sheet thickness, perform energy subtraction processing to obtain an image in which only the bone mineral reference of the variable reference is extracted using the high voltage image and the low voltage image. The obtained image has a one-to-one correspondence with the signal value of the sheet. That is, the bone mineral reference image can be identified by the signal value of the sheet.

(4) Creating a correction curve: Using three bone mineral reference images obtained by the energy subtraction processing, the signal value of each section of the bone mineral reference and the known bone mineral content of that section are associated. At least three correction curves as shown in FIG. 6 are created. In FIG. 6, the correction curves A, B, and C correspond to the sheet signal values a, b, and c, respectively. This correction curve is obtained as follows.

First, as shown in FIG. 4, the signal values (Sbp1) of the respective references 2a to 2f of the variable reference imaged in the bone image 30 obtained by the subtraction process.
, Sbp2, Sbp3, Sbp4, Sbp5, Sbp6) and the signal values Sbp1 of the respective references 2a to 2f of the variable reference.
~Sbp6 a first function f ₁ by the values obtained f described above
_{1 (Sbp1), f 1 (} Sbp2), f 1 (Sbp3), f 1
_{(Sbp4), f 1 (Sbp5} ), obtains the f ₁ (Sbp6). On the other hand, the bone density values (Bp1, B of each reference 2a to 2f)
p2, Bp3, Bp4, Bp5, Bp6) are known in advance. And this bone density values Bp1～Bp6, and the function value f ₁ (Sbp
1) to f ₁ (Sbp6) are used to determine the coefficients (pk, qk, rk) of the following formula (3) f ₂ = pk · Sb ² + qk · Sb + rk (k = 1, n) (3) To do.

Next, the signal value Ssk of the region of the soft tissue in the low-voltage image is obtained, and p, q, and r of the following equations (4) to (6) are obtained.
Is a quadratic function of the signal value Ss by the least squares method (4)
~ (6) coefficients (pa, pb, pc, qa, qb, qc,
Ra, rb, rc) are determined.

[0051] p (Ss) = pa · Ss 2 + pb · Ss + pc ... (4) q (Ss) = qa · Ss 2 + qb · Ss + qc ... (5) r (Ss) = ra · Ss 2 + rb · Ss + rc (6) And the second function f ₂ for finally correcting the first function f ₁ by the above (3) to (6) is f ₂ (Ss, Sb) = p (Ss) It is calculated as × Sb ² + q (Ss) × Sb ² + r (Ss) (7).

The second function is obtained in advance before capturing a radiographic image of the subject.

The calibration curve is expressed by the following equation (8) using the first function f ₁ and the second function f ₂ thus obtained.

BMD = f ₁ (Sb) + f ₂ (Ss, Sb) (8) BMD: Bone Mineral Density Ss: Average OL value in the original image of the soft tissue around the bone Sb: Erase of the soft tissue of the bone Average OL Value in Image Next, the step of obtaining the bone mineral content using the first function f ₁ and the second function f ₂ will be described step by step.

(1) Forming high-voltage / low-voltage images: Similar to the above-described steps, one-shot energy subtraction imaging is performed on a subject having soft tissue and bone tissue as shown in FIG. A high-voltage image and a low-voltage image are accumulated and recorded on a stimulable phosphor sheet. The accumulated and recorded images are read by the image reading device shown in FIG. 3 to obtain digital image signals.

(2) Obtaining Soft Tissue Signal Value: Obtain the soft tissue signal value in the peripheral portion of the region of interest from the digital image signal of the low-voltage image. This region of interest indicates the bone part for which the bone mineral content is to be determined, and the peripheral part means the peripheral part of this bone.

(3) Selection of the second function: As described above, the signal value of the soft tissue obtained here is compared with the signal value of the sheet, and the correction curve of the signal value of the sheet having the closest value is obtained. select. However, other selection methods may be used if desired. For example, it is possible to select a value such that the absolute value of the difference between the signal value of the sheet and the signal value of the soft part image is the smallest.

(4) Perform energy subtraction processing: Perform energy subtraction processing to obtain an image in which soft tissue is erased using a high-voltage image and a low-voltage image,
An extracted bone image of the bone tissue is obtained.

(5) The first function f ₁ is calculated.

Based on the signal value of the bone mineral reference in the bone image and the bone density of each reference that is known in advance, the first function f ₁ is calculated by the above equations (1) and (2).
To calculate.

(6) A calibration curve is obtained by correcting the first function.

The obtained first function f ₁ is corrected by the selected second function f ₂ to obtain a calibration curve.

(7) Obtain the signal value of the region of interest: Obtain the signal value of the region for which the amount of bone mineral is to be determined, that is, the region of interest, from the bone image.

(8) Obtaining the amount of bone mineral: In the above-mentioned embodiment, the variable reference is photographed by changing the sheet thickness to 3 cm, 5 cm, and 7 cm, and the second function, which is three kinds of correction curves, is obtained. However, the present invention is not limited to this, and it is also possible to add a variable reference of the sheet thickness of 9 cm and 11 cm to calculate the 5 types of the second functions. .

By thus obtaining the calibration curve by the first function and the second function, the first function absorbs the influence of the fluctuation of the energy distribution of the radiation, and the second function causes the object or soft tissue to be absorbed. It is possible to absorb the influence of the variation of the thickness of the bone, which makes it possible to obtain the calibration curve regardless of the variation of the energy distribution of the radiation and the variation of the thickness of the soft tissue, and therefore the bone mineral quantitative analysis with high accuracy can be performed. It can be performed. Further, since the calibration curve is obtained from the two kinds of functions in this way, there is a certain degree of control width for allowing the energy distribution of the radiation applied to the subject, so that some fluctuation in the energy distribution is allowed. It is possible.

However, when the variation of the energy distribution of radiation exceeds the control width with the passage of time, the variation cannot be corrected, and it is necessary to re-calculate the function for correcting the basic calibration curve. Although the work was extremely troublesome, the present invention detects this variation and corrects this shift amount without re-determining the calibration curve. The correction of this shift amount will be described below.

FIG. 7 is a diagram showing an outline of an apparatus for correcting the deviation of the above-mentioned calibration curve. In FIG. 7, the bone density measurement processing unit 33 uses the subtraction signal log S as described above.
And the bone density is measured based on the above equation (8), and the calibration function creation data update processing unit 34, based on the measurement result data in the bone density measurement processing unit 33, shifts the calibration function. The correction process is performed. Then, the bone density measurement processing unit 33 measures the bone density based on the calibration function creation data for correcting the calibration function obtained by the calibration function creation data update processing unit 34.

Hereinafter, the calibration function creation data update processing unit 34
The details of the processing performed in step 1 will be described.

Specifically, at the same time as the measurement of the reference bone mineral reference having a predetermined thickness (5 cm in this embodiment), which is performed once a day for bone mineral quantification, Sb1, ..., Sb5: each step of the reference phantom Energy sub-image density f ₃ (Sb1), ..., F ₃ (Sb5): (true value)-(measured value) of bone density at each step of the reference phantom is saved and stored in a database (hereinafter referred to as a simple correction database) ).

Then, at the time of bone density measurement, Sb, f of the simple database of the past n times from the shooting date of the measurement data as a starting point
_The following quadratic equation (9) is fitted to the set of ₃ (Sb) and all 6 × n sets by the method of least squares.

F _f (Sb) = A * Sb * Sb + B * Sb + C (9) The bone term concentration obtained by the first and second functions from the correction term f _f (Sb) obtained in this way Add to the bone density curve, ie (f ₁ + f ₂ ) + f _f to obtain the desired bone density.

In the present embodiment, as described above, a simple correction database is created. This database is used to measure the reference bone mineral reference or the photographing date / time of the calibration curve change and the reference bone mineral. The density measurement value Sb of each block of the reference and the simple correction term f ₃ (Sb) are arranged in order. Then, when the reference bone mineral reference is measured, new simple correction data is added to the file.

Specifically, the processing is performed as follows. First, the calculated bone density value BMD1 and the true bone density BMD _R
The absolute value of the difference value | BMD _R −BMD1 |
This absolute value is obtained for each of the 6 blocks of the reference bone mineral reference to obtain 6 values. The maximum value of these is B
_Let it be MD _c1 . When BMD _c1 ≧ 0.02 g / cm ² , the correction is performed as follows.

First, the bone density value is obtained by the same method as the conventional method.
Based on the shooting date / time attached to the image signal in the database storing a plurality of image signals, the correction data that is older than and closest to the shooting date of the data to be corrected is retrieved. This is referred to as reference start data. From the reference start data, the simple correction data for the past n times including itself,
Use as simple correction reference data. In the simple correction reference data,
If all values other than the shooting date and time are negative numbers, simple correction is not performed. And using in the simplified correction reference data, the concentration Sb _i of reference with reference _{phantoms, j} and simple correction term _{f 3, j (Sb i,} j) pairs, all 6 × n sets, Sb _i, and _j f _{3, j} (Sb
_{i, j} ) is expressed by a quadratic expression [f _f (Sb) = A · Sb ·
Sb + B · Sb + C]. As described above, the least squares method is used for the quadratic fitting. A value obtained by adding the obtained f _f (Sb) to the obtained bone density value is taken as a corrected bone density value.

The amount of deviation of the second function f ₂ obtained from the reference bone mineral reference having the other thicknesses of 3 cm and 7 cm is the same as that of the second function obtained from the reference bone mineral reference having the thickness of 5 cm. The same correction is performed by assuming that it is the same as the shift amount.

As described above, the method for quantitatively analyzing bone mineral according to the present invention, the signal value of this reference and the signal value of this reference in the radiation image obtained by photographing the radiation on the reference bone mineral reference having a predetermined thickness. The amount of deviation from the true bone density of the reference bone mineral reference, which is the reference value of, is obtained, and it is considered that this deviation is approximately the same as the deviation of other correction curves, and other correction curves are The correction is made. In this way, since the deviation amount of the correction curve over time is calculated and the correction curve is corrected based on this, an accurate correction curve is always obtained regardless of changes in the energy distribution over time and changes in the subject thickness. Therefore, the calibration curve can be obtained with high accuracy, and the accuracy of quantitative analysis of bone mineral can be improved.

Further, since it is not necessary to obtain the correction curve again, the labor of the operator can be reduced.

In the above embodiment, the correction term f _f (Sb) is added to the calibration curve f ₁ (Sb) + f ₂ (Ss, Sb).
) Is added to correct the calibration curve, but the present invention is not limited to this, and the second function f ₂ may be directly updated based on the calculated shift amount. Is.

[Brief description of drawings]

FIG. 1 is a diagram in which a bone mineral reference used in the method for quantitatively analyzing bone mineral according to the present invention and a sheet made of a substance having an equivalent radiation absorption coefficient for approximating the bone mineral reference and the thickness of soft tissue are superposed. Figure showing

FIG. 2 is a view showing a photographing means for obtaining an image for carrying out a quantitative analysis of bone mineral according to the present invention.

FIG. 3 is a diagram showing a reading device for reading a radiation image from a stimulable phosphor sheet.

FIG. 4 is a diagram schematically showing a subtraction process.

FIG. 5 is a diagram showing a first function based on a bone mineral reference.

FIG. 6 is a diagram showing a second function that corrects the first function.

FIG. 7 is a diagram for explaining a deviation of a calibration curve.

[Explanation of symbols]

 2 Bone mineral reference 4 Sheet 6 Reference reference 8 Variable reference 12 Subject 14 X-ray 16 X-ray tube 20 Laser light source 21 Laser light 22 Scanning mirror 23 Photostimulated emission light 24 Optical guide 25 Photomul 26 Log converter 27 Recording medium 28, 29, 30 images

Claims (2)

[Claims] 1. A plurality of subjects formed by radiation of different energies, each of which is a subject including a soft tissue and a bone tissue and a bone mineral reference in which the amount of bone mineral in the human subject is simulated stepwise. Of the bone tissue in the subject is extracted or enhanced based on the radiation image of the bone image, and the bone portion is generated based on the signal value of the bone mineral reference appearing in the bone image. A bone mineral quantitative analysis method for obtaining a calibration curve of the amount of bone mineral with respect to the signal value of the tissue, and quantitatively analyzing the bone mineral in the bone tissue based on the signal curve of the region of interest in the bone tissue and the calibration curve. In the above-mentioned calibration curve, the first function calculated based on the signal value of the bone mineral reference in the bone image and the soft tissue around the region of interest in one of the radiographic images A method for quantitatively analyzing bone mineral based on a second function calculated in advance, which corrects the first function based on the thickness, wherein the second function has a predetermined bone mineral content. A sheet for approximating the thickness of the soft tissue composed of a substance equivalent to the soft tissue and the radiation absorption coefficient is overlaid on the reference bone mineral reference, and radiation of different energies is changed each time while changing the thickness of the sheet. A signal value of the sheet corresponding to the thickness of the soft tissue in one radiation image among the plurality of radiation images obtained by irradiation is obtained, and a signal corresponding to the thickness of the sheet using the plurality of radiation images The energy subtraction process is performed to obtain an image of only the reference bone mineral reference by correcting the value, and the reference bone mineral reference obtained by the energy subtraction process. Based on the scanning image, it is calculated by obtaining a plurality of correction curves representing the relationship of the amount of bone mineral with respect to the signal value of the radiographic image for each signal value of the sheet corresponding to soft tissue of different thickness, and the thickness of the sheet is Of the different reference bone mineral references, the signal value of the reference bone mineral reference in the radiographic image obtained by irradiating the reference bone mineral reference having a predetermined thickness with radiation and the signal of the reference bone mineral reference having the predetermined thickness. A method for quantitative analysis of bone mineral, comprising: calculating a deviation amount of a value from a reference value, and correcting a deviation of the plurality of correction curves with time based on the deviation amount. 2. A plurality of signal values of the reference bone mineral reference in a plurality of radiographic images obtained by irradiating the reference bone mineral reference having the thickness of the predetermined sheet with radiation a plurality of times to calculate the deviation amount. 2. The method according to claim 1, wherein the determination is performed based on a representative value and a predetermined reference value.
The method for quantitatively analyzing bone mineral as described.