JPS6311013B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an apparatus for reconstructing an X-ray tomographic image, and more particularly to an apparatus capable of reconstructing an X-ray tomographic image by performing desired data processing.

In general, in the industrial and medical world, X-ray inspections are widely used to determine the internal structure of products and the human body.
Linear computerized tomography equipment (abbreviated as "CT") is being actively developed.

By the way, in this type of conventional X-ray computerized tomography apparatus, first, as shown by the solid line in FIG. The X-ray projection distribution is measured by the X-ray detector c, and then, as shown by the chain line in Fig. 1, the X-ray generator a and the X-ray detector c are rotated by a predetermined angle α (for example, 1°). Then, from this position, the X-ray beam flux is directed toward the X-ray inspection object b again to measure the X-ray projection distribution when moved by the angle α, and the same operation is repeated for a total of 60 to 360 times. After going around,
The data obtained from these large numbers of X-ray projection distributions are processed using the Fourier transform method and the convolution method, and then the tomographic image of the X-ray object b is reconstructed based on the processing results. It is.

However, such a conventional X-ray computer tomography apparatus requires a large number of X-ray projection distributions in order to reconstruct a tomographic image of the subject b, which causes the following problems.

(1) Since it takes a long time to collect data (several seconds to several minutes), for moving object b,
The tomographic image cannot be reconstructed.

(2) Since the amount of X-ray exposure increases, if the subject b is a living body such as a human body, there is a risk of adverse effects on the living body.

(3) Since a large amount of data is required for processing, data processing becomes extremely complicated, and the equipment for processing this data is also large and expensive.

The present invention aims to solve these problems, and allows highly accurate reconstructed images by simply using a single X-ray projection distribution as data for reconstructing a tomographic image of a subject. It is an object of the present invention to provide a single X-ray projection type tomographic image reconstruction device that can obtain a single X-ray projection type tomographic image reconstruction device.

For this reason, the single X-ray projection type tomographic image reconstruction apparatus of the present invention uses a virtual tomographic plane S that includes the tomographic plane of the X-ray subject B and has m pixels vertically and n pixels horizontally. and , from one end to the other end of a single X-ray projection distribution D obtained by irradiating the plurality of unit X-ray beams from the X-ray generator A toward the X-ray inspection object B. M [>N (=m×n)] X
an X-ray density measuring device E that measures linear densities d ₁ to d _M ;
and a data selection device DS that divides the M X-ray concentrations d ₁ to d _M obtained by the X-ray concentration measurement device E into a plurality of measurement value groups each consisting of a plurality of X-ray concentrations;
The virtual tomographic plane S is divided into a plurality of pixel groups consisting of a plurality of pixels based on the plurality of X-ray densities in each of the above measurement value groups, and the X-ray absorption coefficient μ of each pixel constituting each pixel group is calculated. Therefore, the X-ray density obtained by a unit X-ray beam passing only through basic pixel 1
Let d ₁ be the X-ray absorption coefficient μ ₁ of the same basic pixel 1, and then calculate the X-ray absorption of the pixel by subtracting the X-ray concentration d _k ^ _-1 (k^ is a natural number of 2 or more) from the X-ray concentration d _k ^. A means for determining the coefficient μ _k ^ and a relational expression determined from the X-ray concentration vector 〓, coefficient matrix 〓 and residual vector consisting of non-negative elements 〓〓≦〓+〓, 〓〓≧−〓+〓 A projective transformation device F is provided, which includes a calculation device having means for calculating 〓〓≧−〓+〓, and a storage device G capable of storing the calculated X-ray absorption coefficient. The tomographic image of the X-ray subject B is obtained by applying the X-ray absorption coefficients μ ₁ to μ _no to the positions of each of the N pixels 1 to mn constituting the virtual tomographic plane S. It is characterized by being provided with a reconstruction device H that performs reconstruction, and a display device J that displays a tomographic image of the X-ray subject B based on the output from the reconstruction device H.

however Here 〓=(μ _i , μ _i+1 , ..., μ _i+u-1 ) ^T 〓=(d _i , d _i+1 , ..., d _i+v-1 ) ^T 〓=(r _i , r _i+1 , ..., r _i+v-1 ) ^T 〓 is a band matrix of v×u.

Also, i, u, v are natural numbers, and u, v are 0<u
<v<mn is satisfied.

Embodiments of the present invention will be described below with reference to the drawings. FIGS. 2 to 4 show a single X-ray projection type tomographic image reconstruction apparatus as the first embodiment, and FIG. FIG. 3 is a schematic diagram showing the distribution measuring means, FIG. 3 is a system configuration diagram thereof, and FIG. 4 is a schematic diagram for explaining its operation.

As shown in FIG. 3 (a diagram corresponding to the claims), an X-ray inspection object B is positioned between an X-ray generator A and an X-ray detector C. This X-ray inspection object B can be irradiated with X-rays from a predetermined direction, and as a result, the X-rays that have passed through the X-ray inspection object B form a single X-ray projection distribution D (Fig. ).

As shown in FIG. 4, this X-ray generating device A has a virtual tomographic plane S (this plane S is the tomographic plane of the X-ray subject B), which has m pixels vertically and n pixels horizontally. ), a plurality of unit X-ray beams are irradiated toward the X-ray inspection object B from a predetermined direction θ so that the unit X-ray beams pass through at least one lower left corner of each pixel 1 to mn. It is designed to generate X-rays with radiation quality (penetrating power) and dose that are compatible with the X-ray inspection object B, which is the object of inspection, and the wavelength of the generated X-rays depends on the applied voltage. In addition, the quality of X-rays is determined by their wavelength.

The applied voltage varies depending on the application, ranging from 50,000 volts to 120,000 volts when used for medical diagnosis, and from 100,000 volts to 300,000 volts when used for non-destructive testing. Ru.

X-ray inspection object B refers to an object that is irradiated with X-rays and whose transmitted dose distribution (X-ray projection distribution) is measured to reproduce an image in a desired tomographic plane. For example, in the case of medical diagnosis, it is a human body (generally a living body), and in the case of non-destructive testing, it is a so-called industrial product.

Furthermore, as the X-ray detector C, an X-ray film, a scintillation detector, a semiconductor detector, a xenon gas detector, etc. are used.

By the way, the single X-ray projection distribution data D obtained by the X-ray detector C is measured by an X-ray projection distribution measuring means (X
The above X-ray projection distribution D is measured by the line density measuring device) E.
Equal intervals w from one end to the other
Each value _dk at _{mn +} m- ₁ (= _{M )} positions k separated by ₊₁ ~d _3n ,
..., d _M-2n+1 d _M , respectively. is now being measured (see Figure 4).

When the X-ray detector C is an X-ray film, the X-ray projection distribution measuring means E includes multiple sets of X-ray density distributions (so-called X-ray photographs) obtained as shading of the degree of blackening on the film. A microdensitometer is used that can measure the value of .

In this way, from one end of the X-ray projection distribution D to the other end, there are multiple positions k spaced at equal intervals w.
In addition to the above-mentioned example, as a means for measuring each value d _k in In the case of a scintillation detector, one scintillation detector may be combined with a mechanism that can move it from one end of the X-ray projection distribution D to the other, or a number of scintillation detectors may be The arrangement may be performed over the entire width of the line projection distribution D.

In addition, if the X-ray detector C is a semiconductor detector that also receives X-rays transmitted through the X-ray inspection object B and outputs a signal corresponding to the X-ray concentration, the above-mentioned scintillation detection can be used as the above-mentioned means. As in the case of detectors, one semiconductor detector and a mechanism for moving it may be combined, or a large number of semiconductor detectors may be arranged.

Furthermore, when the X-ray detector C is a xenon gas detector, the above means includes one xenon gas detector and a mechanism for moving it, as in the case of the scintillation detector or semiconductor detector described above. Combinations and the arrangement of multiple xenon gas detectors may be used.

The signals obtained by the microdensitometer, scintillation detector, and semiconductor detector in this way are analog signals, so in order to input them to a digital computer, this signal is converted to an analog-to-digital converter E' ( (hereinafter referred to as "A/D converter"), converts it into a digital signal,
Thereafter, it is stored on a disk or the like.

Next, the data output (digital signal) from this A/D converter is sequentially input to the projective conversion device F via the data selection device DS.

The data selection device DS selects M X-ray concentrations d ₁ to d _M
n sets of measurement value groups d ₁ ~ d _2n , d _n+1 ~ d _3n , ...,
It is divided into d _M-2n+1 to d _M. That is, n sets of 2m measurement value groups are selected by the data selection device DS.

Further, the projective conversion device F converts the measured values in each of the measured value groups based on the digital data output (one-dimensional data) for each of the measured value groups inputted from the A/D converter via the data selection device DS. Sequentially calculate each X-ray absorption coefficient (two-dimensional data) of each pixel in n sets of pixel groups consisting of m pixels, which is less than 2m (however, m for the n-th pixel group). In order to output
The X-ray density d ₁ obtained by a unit X-ray beam passing only through basic pixel 1 is assumed to be the X-ray absorption coefficient μ ₁ of the same basic pixel 1. Hereafter, the X-ray density d _k ^ is calculated from the X-ray density d _k ^ _-1 (k ^ is a natural number greater than or equal to 2).
A means for determining the linear absorption coefficient μ _k ^ and a relational expression determined from the X-ray concentration vector 〓, the coefficient matrix 〓, and the residual vector consisting of non-negative elements 〓〓≦〓+, with the X-ray absorption vector 〓 as an unknown quantity. 〓、〓〓≧〓＋〓
It consists of a storage device G for storing linear absorption coefficients (a three-dimensional internal structure storage device serves as this storage device as described later), and a specific example thereof is a digital computer containing a desired program.

Next, a method for determining the X-ray absorption coefficient of each pixel will be explained.

First, based on each measurement value group consisting of 2m values d _k (one-dimensional data) on a single X-ray projection distribution D, a virtual tomographic plane including the tomographic plane of the X-ray subject B is A method for determining each X-ray absorption coefficient μ _t (two-dimensional data) of mn pixels as constituent elements of S (this plane S consists of n groups of pixels) will be described.

First, a virtual tomographic plane S as a tomographic image reconstruction plane of the X-ray subject B is defined as a column partial plane of the tomographic image reconstruction plane consisting of m pixels divided into small sections, as shown in FIG. It is assumed that n groups of pixels are gathered together, and the center of the plane S is taken as the origin of the xy coordinates.

Note that it is assumed that each pixel has the same X-ray absorption coefficient, and for convenience of explanation, m and n are assumed to be even numbers, and the size of one pixel is assumed to be a square of Δ×Δ.

In addition, the X-ray beam passing through this virtual plane S shall be irradiated in parallel from the direction θ that satisfies θ = tan ^-1 m, and the beam diameter of each unit X-ray beam shall be sufficiently smaller than that of each pixel. do.

First, each pixel 1 to m in the first pixel group _Q1
The purpose is to find the X-ray absorption coefficients μ _{1 to} μ _n of
Adopt d ₁ ~ d _2n . In this way, the following formula can be obtained.

〓 _Q1 〓＝〓 _Q1 ……(1) However, Note that 〓 = (μ ₁ , μ ₂ , μ ₃ , ..., μ _2n ) ^T 〓 _Q1 = (d ₁ , d ₂ , d ₃ , ..., d _2n ) ^T 〓 _Q1 is a 2m × 2m square matrix. be.

Further, α is the length of the X-ray beam passing through the pixel, α=Δ√1+ ² , and the symbol T represents transposition.

Then, solving equation (1) yields the following.

However, 〓=(p ₁ , p ₂ , ..., p _2n ) ^T = 〓 _Q1 /α Here, in calculating μ ₁ , μ _k ^, μ _k ^', Using the beam length α is
This is to obtain an accurate X-ray absorption coefficient for each pixel. Therefore, the two-dimensional data μ _k can be obtained from the one-dimensional data d _k without including α in the coefficient matrix _Q1 . The solution in this case is as follows.

However, since the actual X-ray inspection object B generally includes measurement errors, it is difficult to expect that the two-dimensional data μ ₁ to μ _2n calculated by equation (1) will give good results.

Therefore, using mathematical programming, the non-negative correction value r ₁ ,
Introduce r ₂ , r ₃ , ..., r _2n . Furthermore, since the absorption coefficient of an X-ray beam transmitted through an object is generally non-negative and does not exceed a certain positive upper limit U, equation (1) can be expressed as follows by adding these constraints.

Here, p ₁ = d ₁ /α, p ₂ = d ₂ /α, ..., p _2n
= d _2n / α Furthermore, non-negative slack variables s ₁ , s ₂ , s ₃ , ...,
When s _3(2n) is introduced, the simultaneous linear inequalities in equation (3) become the following simultaneous linear equations.

Using mathematical programming to find a solution that minimizes the objective function F= _2n 〓 ^k=1 r _k ……(5) under the constraint equation (4) above, we get
After a finite number of calculations, the optimal 2
Dimensional data μ ₁ to μ _2n are calculated.

In the above example, the two-dimensional data μ _t was calculated based on the objective function that minimizes the sum of the absolute values of the correction values in the constraint condition expression, but on the other hand, the following constraint condition expression (6) is also calculated. There is also a method of minimizing the objective function in equation (7).

F=r (7) By solving the above equation, the tomographic two-dimensional data μ ₁ to μ _2n are calculated under the condition that the maximum correction value of the absolute value in the constraint condition equation is the minimum.

Furthermore, there is a method of minimizing the objective function F= _2n 〓 ^k=1 r _k 2 ...(8) under the constraint expression of Equation (4), or a method of minimizing the objective function F= 2n 〓 k=1 r k ² ...(8) under the constraint expression of Equation (6). It is also possible to implement the method by minimizing the objective function F=r ² ...(9).

Two-dimensional data μ ₁ ~ μ _2n obtained in this way
Among them, only the first m pieces, ie, μ ₁ to _{μ n} , are transferred to the three-dimensional internal structure storage device (memory) G.

Next, each pixel m+1~ in the second pixel group _Q2
To find the X-ray absorption coefficient μ _n+1 ~ μ _2n for 2 m, v ₂
(=v ₁ = 2m) If we adopt the projected densities d _n+1 to d _3n and the X-ray absorption coefficient we just found, we can calculate 2m Absorption coefficients μ _n+1 to μ _3n can be determined.

Furthermore, in the same way as when calculating the X-ray absorption coefficient of each pixel in the first pixel group Q ₁ described above, the optimum X-ray absorption coefficient (cross-sectional two-dimensional data) μ _n+ is calculated using appropriate mathematical programming. ₁ to μ _3n are determined, and then only the first m of these two-dimensional data μ _n+1 to μ _3n , that is, only μ _n+1 to μ _2n are transferred to the memory G.

After that, repeat almost the same operation to create the subsequent third
The X-ray absorption coefficient of each pixel in the ~n pixel groups _Q3 to _Qo is determined, and only the first m two-dimensional data of each are sequentially transferred to the memory G.

In addition, when calculating the X-ray absorption coefficient μ _no-n+1 ~ μ _no of each pixel in the last pixel group Q _o , since the number of X-ray density data is larger than the number of X-ray absorption coefficients to be calculated, Optimal solutions for the X-ray absorption coefficients μ _no-n+1 to μ _no are determined by mathematical programming.

That is, according to the above mathematical programming method, as is clear from what has been stated above, the following relational expression holds true, but in this relational expression, residual vector (correction value vector) 〓 Element of _Qo r _{no-n+ 1} ~ _r Sum of squares _M 〓 ^i=mn-m+1 r _i
² [See formula (8)] or absolute value sum _M 〓 ^i=mn-m+1 r _i [See formula (5)]
An optimal solution for the X-ray absorption coefficient μ _no-n+1 to _{μ no} that minimizes is found.

Here, the relational expression is obtained from the above equation (3): (〓 _Qo /α)〓 _Qo ≦〓 _Qo } (9)′ (〓 _Qo /α)〓 _Qo ≧−〓 _Qo +〓 _Qo , that is, 〓 _Qo 〓 _Qo ≦〓 _Qo +〓 _Qo } (9)″ 〓 _Qo 〓 _Qo ≧−〓 _Qo +〓 _Qo .

And 〓 _Qo = (μ _no-n+1 , ..., μ _no ) ^T 〓 _Qo = (r _no-n+1 , ..., r _M ) ^T 〓 _Qo = (p _no-n+1 , ... …, p _M ) ^T =〓 _Qo ／α 〓 _Qo = (d _no-n+1 , …, d _M ) ^T Also, _〓Qo is a (M-mn+m)×m band matrix.

By the way, it is clear from looking at the state of the unit X-ray beam that passes through the n _- th pixel group _Qo in FIG.

By the way, as mentioned above, the three-dimensional internal structure memory measurement device G, which also serves as a storage device, receives a set of n signals from the projective transformation device F and assembles each pixel group Q ₁ to _{Q o} according to their mutual order. , that is, the first to
This refers to a memory that can store signals from the projective transformation device F in order to assemble the n-th pixel group from the left in their ordinal order, and further refers to a memory that calculates three-dimensional internal structure data of the X-ray inspection object B.

By the way, from the projective transformation device F, m pieces are sequentially n
Two-dimensional data μ _t (t=1, 2, 3,
..., mn) is for a certain cross section of the X-ray inspected object B, but by changing the measurement location with the X-ray projection distribution measuring means E, other X-ray projection distributions can be obtained.
D′, and from this, two-dimensional data μ _t ′ for other cross sections can be easily obtained in almost the same way as in the previous case, so two-dimensional data μ _t , μ _t ′ for several different cross sections can be obtained. , μ _t ″, ..., it becomes possible to memorize the three-dimensional internal structure of the X-ray inspection object B, but in order to construct a complete three-dimensional internal structure, Since interpolation between each section data is required, the memory G is also used as a memory device having a calculation function for this purpose.

An arbitrary tomographic image constructing device (reconstruction device) H is connected to this memory G, and this arbitrary tomographic image constructing device H stores the three-dimensional internal structure of the This is an apparatus that selectively extracts two-dimensional data regarding a specified arbitrary cross section of the X-ray subject B from the data to construct a tomographic image.

Note that the arbitrary cross section here refers to a cross section of the X-ray inspected object B in a horizontal, vertical, or diagonal direction.

In this way, the two-dimensional data of an arbitrary tomographic image obtained by this arbitrary tomographic image constructing device H is consistently mathematically calculated based on the X-ray projection distribution obtained by the X-ray projection distribution measuring means E. Since it has been calculated without contradiction, it is sent to an arbitrary tomographic image display device J via an appropriate digital-to-analog converter F' (hereinafter referred to as "D/A converter") to obtain a virtual tomographic plane S. By applying the X-ray absorption coefficients μ ₁ to μ _no to the positions of each of the N pixels 1 to mn, it is possible to display a tomographic image of the X-ray subject B. Since this includes elements that degrade image quality such as noise and blur, there is no guarantee that a suitable image will be obtained.

Therefore, in order to correct the data from this optional tomographic image construction device H, this data is sent to the optional tomographic image quality improvement device I.

This arbitrary tomographic image quality improvement device I improves the image quality by removing noise, smoothing, and emphasizing the sharpness of arbitrary tomographic image data sent from the arbitrary tomographic image construction device H. , a digital filter is used to remove noise,
A smoothing circuit is used for smoothing, and a differentiation circuit is used for emphasizing sharpness.

The signal whose image quality has been improved in this way is D/A
It is sent to the arbitrary tomographic image display device J via the converter F'.

This arbitrary tomographic image display device J receives a signal sent from the arbitrary tomographic image quality improvement device I and displays an arbitrary tomographic image of the X-ray inspected object B on a color or monochrome cathode ray tube (braun tube) monitor. A device that displays images, and as mentioned above, cathode ray tubes are generally used.

With the above configuration, in order to reconstruct a tomographic image of the X-ray subject B, first the X-ray generator A irradiates the X-ray subject B with X-rays from a predetermined direction, and then the X-ray detector C Single X-ray projection distribution D obtained with
, using the X-ray projection distribution measuring means E, calculate each value d _k at M positions k at equal mutual intervals w from one end of the X-ray projection distribution D to the other end of the X-ray projection distribution D. One-dimensional data d _k is obtained by measuring until .

Next, these one-dimensional data d _k are suitably analog-digital converted and then sent to a projective conversion device F.
Then, using the method described above, find the X-ray absorption coefficient of each pixel for each 2m pixel group, and then transfer only the first m of these X-ray absorption coefficients to the memory G. By this, the virtual tomographic plane S
Each X-ray absorption coefficient μ ₁ ~ _{μ no} of mn pixels in
(two-dimensional data) are obtained, and these X-ray absorption coefficients are stored in the memory G in a predetermined order.

Thereafter, these two-dimensional data μ ₁ to _{μ no} are sent to an arbitrary tomographic image display device J via a memory G, an arbitrary tomographic image construction device H, an arbitrary tomographic image quality improvement device I, and a D/A converter F′. Then, it is reconstructed and displayed as a tomographic image of the X-ray subject B.

FIG. 5 is a schematic diagram for explaining the operation of a single X-ray projection type tomographic image reconstruction apparatus as a second embodiment of the present invention.

In the case of this second embodiment, an X-ray generator A,
When the X-ray inspection object B is close to the object B, that is, when the X-ray beam is a Fan beam, in this case, the tomographic two-dimensional data should be composed of pixels with modified polar coordinates as shown in Figure 5. For example, equation (1) can be obtained and the above theory can be applied as is.

In FIG. 5, the X-ray source is the origin O, the inner radius of the polar coordinate pixel is R _I , the outer radius R _O , and the pixels are composed of m×n. The pixels are numbered as shown in FIG. 5, and a straight line passing through OO' is used as the reference for angle. The radii R ₁ , R ₂ and angles φ ₁ , φ ₂ , φ ₃ , φ ₄ are expressed as follows.

Now, let L ₁ be the straight line passing through the points (R ₁ , φ ₁ ) and (R ₂ , φ ₃ ), and let L 1 be the straight line passing through the points (R ₁ , φ ₂ ) and (R ₂ , φ ₄ ).
If L ₂ , pixel t has radii R ₁ , R ₂ and two straight lines L ₁ ,
This is the area surrounded by L ₂ .

However, θ' is the apex angle far inside (radius R _I ) of the polar coordinate pixel.

In this way, even when the X-ray beam is a Fan beam, two-dimensional data can be obtained for each row partial plane of the tomographic image from 2m pieces of one-dimensional data in the same way as in the case of a parallel beam as in the first embodiment. As a result, a tomographic image of the X-ray object B can be reconstructed from a single X-ray projection distribution D.

Note that, instead of each pixel group consisting of m pixels forming a row partial plane of the tomographic image of the X-ray subject B as in each of the above-mentioned embodiments,
The column partial plane of the above-mentioned tomographic image may be constructed by appropriately combining pixel groups consisting of these pixels.

In addition, in order for each of the above pixel groups to constitute a column partial plane of the above tomographic image, this pixel group is divided into m to m·(n-1).
This pixel group is composed of n ~ (m-1) pixels in order to construct the row partial plane of the above-mentioned tomographic image.
It may be composed of n pixels.

Furthermore, each pixel group can be made up of fewer pixels than m and n.

In other words, one pixel group can be made up of any number of pixels as long as it is less than mn, but the number of pixels that make up one pixel group depends on the processing for data processing. It is desirable to determine the optimal number based on the capacity of the device and the data processing time.

Note that instead of using 2m measured values each as one-dimensional data, it is also possible to use an appropriate number of measured values as long as the number is greater than m.

As described in detail above, the single X-ray projection type tomographic image reconstruction apparatus of the present invention provides the following effects and advantages.

(1) Since the tomographic image of the X-ray subject B can be reconstructed from a single X-ray projection distribution D, the data collection time is extremely short (30 milliseconds or less), and this makes it possible to The tomographic image of the X-ray object B (for example, the heart) can also be clearly reconstructed.

(2) Also, since the amount of X-ray exposure is extremely small (several tenths to hundreds of times compared to conventional methods),
Even when the object B to be inspected is a living body, there is no adverse effect.

(3) For each pixel group consisting of u pixels, the X-ray absorption coefficient of each of the u pixels can be determined based on a measurement value group consisting of a larger number of measured values than u pixels. , the amount of data required for one process is significantly reduced, which greatly simplifies data processing, and in turn, it is possible to sufficiently reduce the size and cost of data processing equipment.

(4) It is possible to reconstruct an X-ray tomographic image based on a larger number of measured values than the number u of pixels constituting a pixel group as a partial plane of the tomographic plane S that includes the X-ray subject B. Therefore, the reproduction accuracy can be greatly improved.

[Brief explanation of the drawing]

FIG. 1 is a schematic diagram for explaining a conventional X-ray tomographic image reconstruction means, and FIGS. 2 to 4 are a single X-ray projection type tomographic image reconstruction apparatus as a first embodiment of the present invention. Fig. 2 is a schematic diagram showing the means for measuring the X-ray projection distribution, Fig. 3 is a system configuration diagram, Fig. 4 is a schematic diagram for explaining its operation, and Fig. 5 is a schematic diagram showing the means for measuring the X-ray projection distribution. FIG. 2 is a schematic diagram for explaining the operation of a single X-ray projection type tomographic image reconstruction apparatus as a second embodiment of the present invention. A...X-ray generator, B...X-ray inspection object, C...X
ray detector, D...X-ray projection distribution, DS...data selection device, E...X-ray projection distribution measuring means (X-ray density measurement device), E'...A/D converter, F...projection conversion device,
F'...D/A converter, G...Three-dimensional internal structure storage device (memory) that also serves as a storage device, H...Arbitrary tomographic image construction device (reconstruction device), I...Arbitrary tomographic image quality improvement device, J...Arbitrary Tomographic image display device, _Q1 to _Qo ...pixel group, S...virtual tomographic plane.

Claims (1)

[Claims] 1. Regarding a virtual tomographic plane S that includes the tomographic plane of the X-ray inspected object B and has m pixels vertically and n pixels horizontally, the unit X-ray beam is located at the corner of each pixel 1 to mn. The above X should pass through at least one part of the
A plurality of units X from a predetermined direction toward the object B to be inspected
An X-ray generator A that irradiates a ray beam, and a single X-ray obtained by irradiating the plurality of unit X-ray beams from the X-ray generator A toward the X-ray inspection object B. M [>N (=m×n)] X-ray densities from one end to the other end of the projection distribution D
An X-ray concentration measuring device E that measures d ₁ to d _M and the above-mentioned M X-ray concentrations d ₁ obtained by the same X-ray concentration measuring device E.
~d is equipped with a data selection device DS that divides _M into a plurality of measurement value groups consisting of a plurality of X-ray densities, and selects the virtual tomographic plane S from a plurality of pixels based on the plurality of X-ray densities in each of the measurement value groups. In order to calculate the X-ray absorption coefficient μ of each pixel constituting each pixel group, the X-ray density d ₁ obtained by a unit X-ray beam passing only through basic pixel 1 is Assuming that the X-ray absorption coefficient μ of basic pixel 1 is ₁ , the X-ray density is
A method for obtaining the X-ray absorption coefficient μ _k ^ of a pixel by subtracting the X-ray concentration d _k ^ _-1 (k^ is a natural number of 2 or more) from d _k ^, and an unknown quantity of the X-ray absorption vector 〓 The relational expression determined from the line density vector 〓, the coefficient matrix 〓, and the residual vector consisting of non-negative elements 〓〓≦
〓+〓, 〓〓≧-〓+〓 A projective transformation device F is provided, which includes a calculation device having means for calculating In response to the signal from the projective transformation device F, the X-ray absorption coefficient μ ₁ to
An image composition device H that reconstructs a tomographic image of the X-ray subject B by applying μmn, and a display device that displays the tomographic image of the X-ray subject B based on the output from the reconstruction device H. 1. A single X-ray projection type tomographic image reconstruction device, characterized in that it is provided with: J. however Here 〓=(μ _i , μ _i+1 , ..., μ _i+u-1 ) ^T 〓=(d _i , d _i+1 , ..., d _i+v-1 ) ^T 〓=(r _i , r _i+1 , ..., r _i+v-1 ) ^T 〓 is a band matrix of v×u. Also, i, u, v are natural numbers, and u, v are 0<u
<v<mn is satisfied. 2. The X-ray concentration measuring device E includes a microdensitometer capable of measuring multiple sets of X-ray concentration values on the X-ray concentration distribution obtained on the X-ray film, and an analogue from this microdensitometer. A single X-ray projection type tomographic image reconstruction apparatus according to claim 1, comprising an analog-to-digital converter capable of converting a signal into a digital signal. 3. The X-ray concentration measuring device E includes a scintillation detector with a moving mechanism that receives X-rays passing through the X-ray inspection object B and outputs a signal corresponding to the X-ray concentration, and scintillation detection of the same. 2. A single X-ray projection type tomographic image reconstruction apparatus according to claim 1, comprising an analog-to-digital converter capable of converting an analog signal from a device into a digital signal. 4 The X-ray concentration measuring device E includes a number of scintillation detectors that receive X-rays transmitted through the X-ray inspection object B and output signals corresponding to the X-ray concentration, and these scintillation detectors. 2. A single X-ray projection type tomographic image reconstruction apparatus according to claim 1, comprising an analog-to-digital converter capable of converting an analog signal from a device into a digital signal. 5 The X-ray concentration measuring device E includes a semiconductor detector with a moving mechanism that receives X-rays passing through the X-ray inspection object B and outputs a signal corresponding to the X-ray concentration, and a semiconductor detector from the semiconductor detector. A single X-ray projection type tomographic image reconstruction apparatus according to claim 1, comprising an analog-to-digital converter capable of converting an analog signal into a digital signal. 6 The X-ray concentration measuring device E includes a large number of semiconductor detectors that receive X-rays transmitted through the X-ray inspection object B and outputs signals corresponding to the X-ray concentration, and from these semiconductor detectors. A single X-ray projection type tomographic image reconstruction apparatus according to claim 1, comprising an analog-to-digital converter capable of converting an analog signal into a digital signal. 7 The X-ray concentration measuring device E includes a xenon gas detector with a moving mechanism that receives X-rays passing through the X-ray inspection object B and outputs a signal corresponding to the X-ray concentration, and a xenon gas detector. Claim 1 comprising an analog-to-digital converter capable of converting an analog signal from a device into a digital signal.
The single X-ray projection type tomographic image reconstruction device described in 2. 8 The X-ray concentration measuring device E includes a large number of xenon gas detectors that receive X-rays passing through the X-ray inspection object B and output signals corresponding to the X-ray concentration, and these xenon gas detectors. 2. A single X-ray projection type tomographic image reconstruction apparatus according to claim 1, comprising an analog-to-digital converter capable of converting an analog signal from a device into a digital signal. 9 The projective transformation device F includes a digital computer that calculates and outputs a digital signal corresponding to the X-ray absorption coefficient of each pixel based on the data output for each group of measured values. A single X-ray projection type tomographic image reconstruction device according to any one of items 1 to 8. 10 The storage device G also serves as a memory capable of storing tomographic image information for each tomographic portion of the plurality of tomographic portions in the X-ray subject B and displaying any tomographic image on the display device J. A single X-ray projection type tomographic image reconstruction apparatus according to any one of claims 1 to 9. 11 Claims 1 to 10, wherein a digital filter is interposed between the reconstruction device H and the display device J in order to improve the quality of the image displayed on the display device J. The single X-ray projection type tomographic image reconstruction device according to any one of the above. 12 Claims 1 to 10, wherein a smoothing circuit is interposed between the reconstruction device H and the display device J in order to improve the quality of the image displayed on the display device J. The single X-ray projection type tomographic image reconstruction device according to any one of the items. 13 Claims 1 to 10, wherein a differentiation circuit is interposed between the reconstruction device H and the display device J in order to improve the quality of the image displayed on the display device J. The single X-ray projection type tomographic image reconstruction device according to any one of the above.