JP2002171444A - Radiation picture information estimate method and device, and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a radiation picture information estimate method and device that eliminates scattered radiation from an image signal representing radiation picture information so as to obtain an image signal able to reproduce an image without a scattered radiation noise. SOLUTION: A storage means 20 stores a relation between a difference (logarithmic dose difference) of logarithmic values of the doses by each pixel of sheets IP1, IP2, IP3 and an average attenuation coefficient of each sheet as a table T. Since relations of S1-S2=μ(IP2) and S2-S3=μ (IP2) hold in image signal S1-S3 obtained from the sheets IP1-IP3, a blurred image signal Sus1' without scattered radiation is obtained from blurred image signals Sus2, Sus3 of the image signals S2, S3 by referencing the relations. A difference between the blurred image signal Sus1 of the image signal S1 and the blurred image signal Sus1' is obtained as a scattered radiation signal SN and the scattered radiation signal SN is subtracted from the image signal S1 to obtain a scattered radiation eliminated image signal S1'.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a radiation image for estimating a true image signal representing true radiation image information free from scattered radiation, based on a plurality of radiation image information carrying a radiation image of the same subject. The present invention relates to a method and apparatus for estimating information, an energy subtraction method and apparatus using estimated image signals, and a computer-readable recording medium on which a program for causing a computer to execute the method for estimating radiation image information and the energy subtraction method is recorded. .

[0002]

2. Description of the Related Art The applicant has once photographed and recorded a radiation image of a subject such as a human body on a sheet-shaped stimulable phosphor, and scanned the stimulable phosphor sheet with excitation light such as a laser beam to emit light. Light is generated, and the obtained stimulating light is read photoelectrically to obtain a digital image signal representing radiation image information of the subject. Based on the image signal, a radiation image of the subject is recorded on a recording material such as a photographic photosensitive material. , A radiation image recording / reproducing system for outputting a visible image to a CRT or the like has been proposed.

On the other hand, energy subtraction processing of a radiation image has been conventionally known (Japanese Patent Laid-Open No. 7-2).
No. 87330). This energy subtraction processing means irradiating the same subject with radiation having different energy distributions, or changing the energy distribution state of the radiation transmitted through the subject to two radiation detecting means (for example, the stimulable phosphor sheet). ) To obtain an image signal representing two pieces of radiation image information. Thereafter, each pixel of the radiation image represented by the two image signals is made to correspond to each other, and an appropriate weighting coefficient is multiplied between the image signals. In this method, subtraction is performed to obtain a difference signal representing an image of a specific structure. By using the difference signal obtained in this way, a radiation image from which only a specific structure is extracted can be reproduced.

In the radiation image recording / reproducing system using the stimulable phosphor sheet, the radiation image information recorded on the sheet is directly read in the form of an electric image signal. Can be easily performed. In order to perform energy subtraction processing using this stimulable phosphor sheet, for example, image recording (photographing) is performed on two stimulable phosphor sheets so that image information of a portion corresponding to a specific structure is different. More specifically, specifically, a two-shot method in which imaging is performed twice using two types of radiation having different energy distributions, and two stimulable phosphor sheets (for example, (Which may be in contact with or apart from each other) is simultaneously exposed, so that the two sheets are irradiated with radiation having different energy distributions from each other.

In the energy subtraction method using such a stimulable phosphor sheet, for example, when a person is set as a subject, two sheets are irradiated with radiation (high-pressure radiation, low-pressure radiation) having different energy distributions from each other. By obtaining an image signal and performing appropriate weighting on each image signal to obtain a difference signal, it is possible to obtain a radiation image in which only a soft part and a bone part of a human body are extracted. Specifically, the energy subtraction processing is performed as follows. The sheet irradiated with the low-pressure radiation is denoted by IP1, and the sheet irradiated with the high-pressure radiation is denoted by I.
Assuming that L2 and H are logarithmic values (logarithmic radiation doses) of the radiation dose applied to P2 and the sheets IP1 and IP2, L and H are represented by the following equations (1) and (2).

[0006]

(Equation 1) Note that the logarithmic radiation doses L and H are the sheets IP1 and I
The image signal obtained from P2 can be used.

Here, the substance has a radiation attenuation coefficient depending on radiation energy. On the other hand, when the radiation applied to the subject is not monochromatic but distributed in a certain energy range, the energy distribution of the detected radiation (e.g., applied to the stimulable phosphor sheet) is changed according to the substance contained in the subject ( In the case of a human body, a phenomenon called beam hardening that varies depending on the thickness of the bones and soft parts occurs. Therefore, the weighted average of the radiation attenuation coefficient of the substance with the energy distribution of the detected radiation is defined as the average attenuation coefficient. Therefore, the average attenuation coefficient differs depending on the thickness of the substance.

[0008]

(Equation 2) Thus, a difference signal representing a soft part image in which only the soft part not including the thickness of the bone part is extracted can be obtained.

[0009]

(Equation 3) Thus, a difference signal representing a bone part image in which only the bone part not including the thickness of the soft part is extracted can be obtained.

In the above equations (3) and (4),
The average attenuation coefficient multiplied by L and H becomes the weighting coefficient.

[0011]

As described above, when a radiographic image is obtained by photographing a subject, radiation transmitted through the subject is scattered by the subject to generate scattered rays, and the scattered rays are accumulated fluorescent light. The body sheet is irradiated and accumulated and recorded on the sheet. Here, since the scattered radiation has low energy, most of the scattered radiation is accumulated and recorded on the stimulable phosphor sheet provided at the position closest to the radiation source, especially when imaging is performed by the one-shot method. Is done. When scattered radiation is accumulated and recorded on the sheet,
The radiation image information obtained from the stimulable phosphor sheet will contain scattered radiation information.

Therefore, when the scattered radiation information is included in the radiation image information, the image quality of the radiation image obtained by reproducing the image signal representing the radiation image information is deteriorated by the influence of the scattered radiation. It will be. Also, when the energy subtraction process is performed, the information of the scattered radiation affects the subtraction process, and as a result, a specific structure may not be completely removed.

The present invention has been made in view of the above circumstances, and provides a radiation image information estimating method and apparatus capable of obtaining an image signal in which the influence is removed from an image signal representing radiation image information including scattered radiation. It is intended to do so.

It is another object of the present invention to provide an energy subtraction method and apparatus capable of obtaining an energy subtraction image from which a specific structure has been removed without being affected by scattered radiation.

A further object of the present invention is to provide a computer-readable recording medium on which a program for causing a computer to execute the above-described radiation image information estimation method and energy subtraction method is recorded.

[0016]

Means for Solving the Problems The difference between the logarithmic value of the radiation dose or the logarithmic value of the ratio of the radiation dose when the radiation image information is obtained, and the two radiographic images from which the difference or the logarithmic value of the logarithmic value are obtained There is a certain relationship between the information and the average attenuation coefficient, as will be described later.By obtaining this relationship in advance, if the difference in the logarithmic value of the radiation dose or the logarithmic value of the ratio of the radiation dose is known, An average attenuation coefficient can be obtained for two pieces of radiation image information for which a difference between logarithmic values or a logarithmic value of a ratio has been obtained. On the other hand, the radiation image information obtained by photographing the subject has a larger value as the radiation dose irradiated increases, so that the radiation image information and the radiation dose, and furthermore, the image signal representing the radiation image information and the radiation dose Can be associated with each other. Therefore, it can be said that there is also a certain relationship between the logarithmic value difference of the image signal representing the radiation image information or the logarithmic value of the ratio of the image signal representing the radiation image information, and the average attenuation coefficient for each radiation image information. . When three or more pieces of radiation image information are obtained, the average attenuation coefficient for one piece of radiation image information is the number of combinations of logarithmic dose differences or ratios of logarithmic doses using the radiation dose corresponding to the radiation image information. Can be determined according to
The present invention has been made focusing on this point.

That is, the method for estimating radiation image information according to the present invention is a method for estimating at least three or more pieces of radiation image information obtained by radiation having different energy distributions transmitted through the same subject and different in at least a part of image information. Of the plurality of image signals represented, the plurality of other image signals other than the one image signal representing one radiation image information containing more scattered radiation noise compared to the other radiation image information, Calculate the logarithmic difference or ratio of the logarithmic value between the image signals, the logarithmic difference of the radiation dose at each pixel between the respective radiation image information, or the logarithmic value of the ratio of the radiation dose, and the respective radiation image Based on the relationship with the average attenuation coefficient for the information, and the logarithmic difference or ratio of the logarithmic values between the other plurality of image signals, at least one of the other plurality of image signals. Obtaining an average attenuation coefficient for radiation image information represented by one image signal, based on the relationship, the average attenuation coefficient, and one selected image signal selected from the plurality of other image signals, It is characterized by estimating a true image signal representing true radiation image information corresponding to one piece of radiation image information.

In order to obtain an image signal representing radiation image information, radiation having different energy distributions transmitted through a subject is irradiated to radiation detecting means such as a stimulable phosphor sheet or a semiconductor sensor. What is necessary is just to detect by the image signal showing the radiation image information according to the irradiated radiation dose. When the radiation detecting means is a semiconductor sensor, a signal output from the semiconductor sensor may be used as an image signal. When the radiation detecting means is a stimulable phosphor sheet, the signal may be used as in the above-described radiation recording / reproducing system. Then, the stimulable phosphor sheet is irradiated with excitation light to generate stimulated emission light, and an image signal may be obtained by photoelectrically reading the stimulated emission light.

The present invention estimates a true image signal representing true radiation image information of one radiation image information by using a plurality of image signals representing three or more radiation image information. Note that “true radiation image information” refers to radiation image information obtained in a situation where no scattered radiation is present. Here, by performing imaging using three or more radiation detection units, image signals representing three or more radiation image information corresponding to each radiation detection unit can be obtained.

When a stimulable phosphor sheet is used as the radiation detecting means, excitation light is scanned only on both sides or one side of the stimulable phosphor sheet, and stimulated emission light emitted by the excitation light scanning is accumulated. A double-sided condensing reading method for photoelectrically reading from both sides of a luminescent phosphor sheet can be applied (see, for example, JP-A-55-87970). When such a double-sided condensing reading method is applied, two image signals are obtained from one stimulable phosphor sheet. In the present invention, by applying the double-sided condensing reading method, two image signals obtained from one stimulable phosphor sheet are also treated as image signals representing independent radiation image information. As described above, when the double-sided condensing reading method is applied, more image signals are obtained than the number of radiation detecting units.

Here, for example, when the radiation image information is obtained by the one-shot method, since the scattered radiation has relatively low energy, most of the scattered radiation is absorbed by the radiation detection means located closest to the subject. For this reason, the radiation image information obtained from the radiation detecting means located closest to the subject contains more scattered radiation noise.
Therefore, “one piece of radiation image information that contains more scattered radiation noise than other pieces of radiation image information” refers to radiation image information obtained from radiation detection means located closest to the subject.

The radiation dose refers to the radiation dose transmitted through the subject and irradiated to the radiation detecting means when the subject is photographed. Although the radiation dose can be obtained by directly detecting the radiation irradiated to the radiation detecting means, it is very difficult to detect the radiation dose for each pixel of the radiation image information. On the other hand, since the signal value of the image signal obtained by the radiation detection means increases as the amount of irradiated radiation increases, the signal value of the image signal and the radiation amount can be associated with each other. Therefore, it is possible to regard the image signal obtained by the radiation detecting means (no image processing has been performed) as a radiation dose, and to use this as a relation between the radiation dose and the average attenuation coefficient.

Since the average attenuation coefficient is determined for each specific structure included in the subject, a number of average attenuation coefficients corresponding to the type of the specific structure included in the subject can be obtained for one piece of radiation image information. For example, when a specific structure is a soft part and a bone part of a human body, two average attenuation coefficients of the soft part and the bone part are obtained for one piece of radiographic image information.

The difference between the logarithmic value of the radiation dose or the logarithmic value of the ratio of the radiation dose and the average attenuation coefficient for the radiation image information from which the logarithmic difference or logarithmic value of the ratio is obtained is described later. There is a certain relationship as you do. When three or more pieces of radiation image information are obtained, the average attenuation coefficient for one piece of radiation image information is the number of combinations of logarithmic dose differences or ratios of logarithmic doses using the radiation dose corresponding to the radiation image information. Can be determined according to

For example, when the number of radiation image information is 3 (J1, J2, J3 respectively), the radiation image information J2
The average attenuation coefficient μ2 for the radiation image information J1,
It can be obtained from the difference of the logarithmic dose of the radiation dose or the logarithmic value of the ratio for J2 and the radiation image information J2, J3. Specifically, when the radiation doses of the radiation image information J1, J2, and J3 are I1, I2, and I3, respectively, ln
Since I1-lnI2 = μ2 and lnI2-lnI3 = μ2 (in the case of a difference in log dose), lnI1-lnI is satisfied.
2 and lnI2-lnI3, the average attenuation coefficient μ2 can be determined.

The average attenuation coefficient is an average attenuation coefficient obtained from a logarithmic value difference or a ratio logarithmic value between a plurality of other image signals, and one selected image signal selected from the plurality of other image signals. And a common image with the average attenuation coefficient obtained from the logarithmic value difference or ratio logarithm value between the image signal and the image signal.

The "difference in logarithmic value of the radiation dose at each pixel between the pieces of radiation image information" and "logarithmic value of the ratio of the radiation doses" are lnI1-lnI when the radiation doses are I1 and I2.
Since there is a relation of 2 = ln (I1 / S2), the values are the same.

As the relation between the logarithmic value difference or the logarithmic value of the ratio and the average attenuation coefficient, a table showing these relations or a function formula showing these relations can be used.
When the relationship is a table, the table is referred to, and when the relationship is a functional expression, an operation based on the functional expression is performed to obtain an average attenuation coefficient for the radiation image information.

The relationship between the logarithmic value difference or the logarithmic value of the ratio and the average attenuation coefficient differs depending on the imaging conditions such as the voltage of the radiation source used during imaging, the type of radiation source, and the sensitivity of the radiation detection means. Become. Therefore, a plurality of tables or functions corresponding to various photographing conditions are prepared in advance, a table or a function is selected based on the photographing conditions, and the selected table or function is set as the relationship described above, and the average attenuation coefficient is set. It is preferable to obtain.

Here, assuming that an image signal representing one radiation image information containing a large amount of the scattered radiation noise is S1, other image signals are S2 and S3, and a certain average attenuation coefficient is μ, the radiation dose is Since it is associated with the image signal, lnS1-lnS2 = μ, lnS2-InS3 = μ
Is established. Therefore, lnS1-lnS2 =
The relationship of μ and lnS2−InS3 = μ is determined in advance. Here, among the image signals S1, S2, and S3 obtained by photographing the subject, the image signal S1 contains more scattered radiation noise than the other image signals S2 and S3. On the other hand, the other image signals S2 and S3 are hardly affected by scattered radiation noise. Therefore, by referring to the above relationship, the average attenuation coefficient μ can be obtained based on the image signals S2 and S3. On the other hand, referring to the relationship of lnS1-lnS2 = μ in the above relationship, since the image signal S2 without the influence of the scattered radiation noise and the average attenuation coefficient μ are obtained, the image signal S1 without the influence of the scattered radiation noise is obtained. Can be requested. The image signal S1 obtained in this manner becomes a true image signal.

In the radiation image information estimating method according to the present invention, the scattered radiation signal representing the scattered radiation information included in the one radiation image information is determined based on the true image signal and the one image signal. Preferably, it is calculated.

In this case, by performing noise suppression processing on the plurality of other image signals, a plurality of other noise suppression signals are obtained, and the difference or ratio of the logarithmic value between the plurality of other noise suppression signals is obtained. The logarithmic value of is determined as the logarithmic value of the difference or ratio of the logarithmic values between the plurality of other image signals, the average attenuation coefficient is obtained and the true image signal is estimated, and the one image signal is calculated. It is preferable to obtain one noise suppression signal by performing a noise suppression process, and calculate a difference value between the true image signal and the one noise suppression signal as the scattered radiation signal.

In this case, it is preferable that the scattered radiation signal is subtracted from the one image signal to obtain a scattered radiation removed image signal representing radiation image information from which scattered radiation noise has been removed.

As the noise suppressing process, a filtering process using a low-pass filter can be applied.
In addition, a band-limited image signal representing a frequency response characteristic of each frequency band of the image signal is created, a noise component is separated from each band-limited image signal to obtain a noise band-limited image signal, and this is sequentially performed from the lowest frequency band. A process of obtaining a noise signal by reconfiguration, multiplying the noise signal by a predetermined enhancement coefficient, and subtracting the noise signal from the image signal may be applied (Japanese Patent Application No. 11-3).
No. 63766). Further, a method of performing a wavelet transform on an image signal to obtain an image signal in which noise is suppressed may be applied (Japanese Patent Laid-Open No. 2000-224421).
This method performs a noise extraction process using a morphological operation on a predetermined frequency band image signal of a predetermined frequency band obtained in a process of decomposing into a frequency band image signal representing an image for each of a plurality of frequency bands by a wavelet transform. A noise removal process is performed on a predetermined frequency band image signal based on a result of the noise extraction process to obtain a processed frequency band image signal, and a multi-resolution conversion process is performed on the processed frequency band image signal to obtain a predetermined frequency band image signal. A noise extraction process, a noise removal process, and a multi-resolution conversion process in which a frequency band image signal in a lower frequency band by one step than the frequency band is obtained and the frequency band image signal in the lower frequency band by one step is used as a predetermined frequency band image signal are desired. Repeated processing up to the frequency band to be processed, the processed frequency band image for each frequency band The resulting is a process for obtaining an image signal from which the noise is suppressed by applying an inverse multi-resolution transform processing to further processed frequency band image signal No..

The true image signal obtained by the radiation image information estimating method according to the present invention may be output as a scattered radiation removed image signal.

In the energy subtraction method according to the present invention, of the scattered radiation-removed image signal and the plurality of other image signals obtained by the radiation image information estimating method according to the present invention, Two image signals representing the other image signals are subtracted by multiplying each of the image signals by a predetermined weighting coefficient between signals of corresponding pixels, thereby obtaining an image of the specific structure of the subject. It is characterized in that a difference signal representing the difference is obtained.

In the energy subtraction method according to the present invention, two image signals representing a scattered radiation-removed image signal and other image signals out of a plurality of image signals representing three or more pieces of radiation image information are provided. Is used to perform energy subtraction processing. Here, by performing imaging using three or more radiation detection units, image signals representing three or more radiation image information corresponding to each radiation detection unit can be obtained. In this case, as the “representative image signal”, any two image signals selected from three or more image signals may be used.

When a stimulable phosphor sheet is used as the radiation detecting means, the above-described double-sided condensing reading method can be applied. When such a double-sided condensing reading method is applied, an average signal value of two image signals (including a scattered radiation removal image signal) obtained from one stimulable phosphor sheet is obtained, and this average signal value is referred to as “ One of the two representative image signals "
One may be used.

The radiation image information estimating apparatus according to the present invention comprises:
Compared with other radiation image information among a plurality of image signals representing three or more radiation image information, at least a part of which is obtained by radiation having different energy distributions transmitted through the same subject and different from each other. Then, for a plurality of other image signals other than one image signal representing one radiation image information containing a large amount of scattered radiation noise, a logarithmic difference or a logarithmic value of a ratio between the plurality of other image signals is calculated. Logarithmic calculation means, the difference between the logarithmic value of the radiation dose in each pixel between the respective radiation image information, or the logarithmic value of the ratio of the radiation dose, and the relationship between the average attenuation coefficient for each radiation image information, and Radiation image information represented by at least one image signal of the other plurality of image signals based on a logarithmic value difference or a logarithmic ratio between the other plurality of image signals And the average attenuation coefficient obtaining means for obtaining an average attenuation coefficient of about the relationship, the average attenuation coefficient,
And estimating means for estimating a true image signal representing true radiation image information corresponding to the one radiation image information based on one selected image signal selected from the plurality of other image signals. It is characterized by the following.

In the radiation image information estimating apparatus according to the present invention, the scattered radiation signal representing the scattered radiation information included in the one radiation image information is determined based on the true image signal and the one image signal. It is preferable to further include a scattered radiation signal calculation means for calculating.

In this case, a first noise suppressing means for obtaining a plurality of other noise suppression signals by performing a noise suppression process on the plurality of other image signals; A second noise suppression unit that obtains one noise suppression signal by performing a suppression process, wherein the logarithmic calculation unit and the average attenuation coefficient acquisition unit include a logarithmic value between the plurality of other noise suppression signals. The logarithmic value of the difference or ratio of the plurality of image signals is determined as the logarithmic value of the difference or ratio of the logarithmic value between the plurality of other image signals, and the means for obtaining the average attenuation coefficient and estimating the true image signal, It is preferable that the scattered ray signal calculating means is means for calculating a difference value between the true image signal and the one noise suppression signal as the scattered ray signal.

Also, in this case, the scattered radiation removal image signal calculating means for subtracting the scattered radiation signal from the one image signal to obtain a scattered radiation removal image signal representing radiation image information from which scattered radiation noise has been removed is provided. It is preferable to further provide.

Further, in the radiation image information estimating apparatus according to the present invention, the estimating means outputs the true image signal as a scattered radiation-removed image signal representing radiation image information from which scattered radiation noise has been removed. There may be.

Further, in the radiation image information estimating apparatus according to the present invention, the storage means for storing a table or a function representing a relationship between the previously calculated difference of the logarithmic value or the logarithmic value of the ratio and the average attenuation coefficient is provided. The average attenuation coefficient acquisition unit may be further provided as a unit that sets a table or a function stored in the storage unit as the relation.

In this case, it is assumed that the storage means stores a plurality of tables or functions representing the relationship set according to the imaging conditions for obtaining the radiation image information,
The average attenuation coefficient acquisition unit may be a unit that receives selection of the table or function based on imaging conditions when the radiation image was obtained, and sets the selected table or function as the relationship.

The energy subtraction apparatus according to the present invention includes the scattered radiation-removed image signal and the plurality of other image signals of the scattered radiation-removed image signal and the plurality of other image signals obtained by the radiation image information estimating apparatus according to the present invention. Two image signals representing the other image signals are subtracted by multiplying each of the image signals by a predetermined weighting coefficient between signals of corresponding pixels, thereby obtaining an image of the specific structure of the subject. A subtraction means for obtaining a difference signal representing the difference is provided.

The radiation image information estimating method and the energy subtraction method according to the present invention may be provided by being recorded on a computer-readable recording medium as a program to be executed by a computer.

[0048]

According to the present invention, the difference between the logarithmic value of the radiation dose or the logarithmic value of the ratio of the radiation dose at each pixel between the radiation image information and the logarithmic value of the difference or the ratio of the logarithmic value are obtained.
In view of the fact that there is a certain relationship between the average attenuation coefficient and the two radiation image information, this relationship is determined in advance. Then, a logarithm of a logarithmic value or a logarithmic value of a ratio between a plurality of other image signals other than one image signal representing one radiation image information containing a large amount of scattered radiation noise is obtained, and each radiation is referred to by referring to the above relationship. An average attenuation coefficient for image information is obtained. Here, the plurality of other radiation image information hardly includes scattered radiation noise. Further, the above relationship does not include the influence of scattered radiation noise. Therefore, the average attenuation coefficient for each piece of radiation image information can be acquired based on the logarithmic value difference or ratio logarithmic value between the plurality of other image signals and the above relationship. When the subject includes a plurality of structures, the average attenuation coefficient may be for any of the structures.

Here, when three or more pieces of radiation image information are obtained, the average attenuation coefficient for one piece of radiation image information is:
It can be obtained according to the number of combinations of logarithmic dose differences or ratio logarithmic values using the radiation dose corresponding to the radiation image information. Therefore, if the acquired average attenuation coefficient and one selected image signal selected from a plurality of other image signals that have acquired the average attenuation coefficient are known, the difference between the logarithmic value of one selected image signal and one image signal is obtained. Alternatively, an image signal corresponding to one piece of radiation image information containing a large amount of scattered radiation noise can be obtained by referring to the relationship of the logarithmic value of the ratio. Since the above relationship does not include the influence of scattered radiation noise, the obtained image signal is a true image signal that does not include scattered radiation noise. Therefore, based on the one selected image signal and the logarithmic difference or ratio of the one selected image signal and the one image signal, a true image signal free from the influence of scattered noise is estimated. be able to.

Although the true image signal is free from the influence of scattered radiation noise, the one image signal representing the one radiation image information contains the scattered radiation noise. Therefore, based on the true image signal and one image signal, it is possible to obtain a scattered radiation signal representing scattered radiation information included in one radiation image information.

Further, since an image signal obtained by photographing a subject contains high-frequency noise such as quantum noise of radiation, noise suppression is performed on a plurality of other image signals for obtaining an average attenuation coefficient. By performing the processing, a plurality of other noise suppression signals are obtained, and the average attenuation coefficient and the true image signal are estimated as a plurality of other image signals for obtaining the average attenuation coefficient, thereby suppressing high-frequency noise. A true image signal can be obtained.
Furthermore, by obtaining a difference value between one noise suppression signal obtained by performing noise suppression processing on one image signal and a true image signal, a scattered radiation signal including only information on low-frequency scattered radiation is obtained. Obtainable. Therefore, by subtracting this scattered radiation signal from one image signal, a scattered radiation-removed image signal from which scattered radiation has been appropriately removed can be obtained.

Further, by setting a table or function representing the relationship between the logarithmic value of the radiation dose or the logarithmic value of the ratio and a predetermined weighting factor as the above relationship, the table or function can be referred to to obtain the average attenuation coefficient. Can be easily obtained. Therefore, calculation of a true image signal can be performed efficiently.

Further, by preparing a plurality of tables or functions corresponding to the photographing conditions and selecting these tables or functions according to the photographing conditions and setting the above relationship, an appropriate average value corresponding to the photographing conditions is obtained. The attenuation coefficient can be obtained.

Further, by performing the energy subtraction processing using the scattered radiation-removed image signal and two image signals representing a plurality of image signals, the scattered radiation noise does not affect the subtraction processing, and the specific structure is reduced. It is possible to obtain an energy subtraction image in which objects are appropriately erased.

[0055]

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

FIG. 1 shows three stimulable phosphor sheets IP1, IP1.
Radiation 2 transmitted through the same subject 1 to IP2 and IP3
FIG. 1 is a diagram showing an imaging device for performing so-called one-shot energy subtraction in which irradiation is performed while changing energy. As shown in FIG. 1, the first stimulable phosphor sheet IP1 is disposed closer to the radiation source 3, and the second stimulable phosphor sheet IP2 is disposed at a certain distance from the first stimulable phosphor sheet IP1.
Are disposed, and the third stimulable phosphor sheet IP3 is disposed at a slight distance from the second stimulable phosphor sheet IP2. Further, between each of the sheets IP1, IP2 and IP3, a radiation energy conversion filter 5 made of a copper plate is used.
The radiation source 3 is driven by disposing A and 5B, respectively. As a result, the first stimulable phosphor sheet IP1 is irradiated with the low-pressure radiation 2 including the so-called soft wire, and the second stimulable phosphor sheet IP2 is irradiated with the medium-pressure radiation 2 from which the soft wire is removed. A radiation image of the subject 1 is accumulated and recorded on the stimulable phosphor sheet IP3 of No. 3 by high-pressure radiation 2 from which soft lines have been removed. At this time, the positional relationship of the subject 1 is the same between the stimulable phosphor sheets IP1, IP2, and IP3. As a result, on the three stimulable phosphor sheets IP1, IP2, and IP3, radiation images of at least part of the image information of the subject 1 different from each other are accumulated and recorded.

FIG. 2 is a schematic diagram showing the configuration of a radiation image reading apparatus to which the radiation image information estimating apparatus and the energy subtraction apparatus according to the first embodiment of the present invention are applied. As described above, of the sheets IP1, IP2, and IP3 on which the radiation images are accumulated and recorded, first, the stimulable phosphor sheet IP1 is moved in the direction of the arrow Y by the endless belt 9 while the laser light (excitation) is emitted from the laser light source 10. The light 11 is deflected by the scanning mirror 12, and the main scanning is performed on the sheet IP1 in the X direction. The excitation light scan causes the stimulable phosphor sheet IP1 to emit stimulating light 13 in an amount corresponding to the radiation image information stored and recorded. The stimulated emission light 13 enters the inside of the light guide 14 from one end surface of the light guide 14 formed by molding a transparent acrylic plate, and travels through the light guide 14 while repeating total reflection therein, thereby forming a photomultiplier. (Photomultiplier tube) 15 receives the light. The photomultiplier 15 outputs an analog output signal Q1 corresponding to the light emission amount of the stimulating light 13, ie, indicating the image information.

This output signal Q1 is logarithmically converted by a logarithmic converter 16 and then input to an A / D converter 17,
It is converted into a digital image signal S1. Next, in exactly the same manner, the other two stimulable phosphor sheets IP2 and IP3
Is read out, and output signals Q2 and Q
3 are obtained, the output signals Q2 and Q3 are logarithmically converted by the logarithmic converter 16, and further converted into digital image signals S2 and S3 by the A / D converter 17.

The image signals S 1, S 2, and S 3 are input to the scattered radiation removing means 18, where the stimulable phosphor sheet closest to the subject 1 is referred by referring to the table T stored in the storage means 20. Scattered radiation noise is removed from the image signal S1 obtained from IP1. Here, the radiation 2 transmitted through the subject 1 includes scattered radiation. Since the scattered radiation has low energy, most of the scattered radiation is absorbed by the sheet IP1, and the radiation 2 transmitted through the sheet IP1 contains almost no scattered radiation, and as a result, the image signals S2 and S3
Scattered radiation noise is hardly included in the image represented by. Therefore, the scattered radiation removing unit 18 refers to the table T stored in the storage unit 20 and
The scattered radiation noise is removed only from the image signal S1, and the scattered radiation removed image signal S1 'from which the scattered radiation is removed is obtained.

The scattered radiation-removed image signal S1 'and the image signals S2 and S3 are input to a subtraction means 21, where an energy subtraction process is performed, and a difference signal S _S , representing a radiation image of a soft part and a bone part in the subject. S _B
Is obtained. The weighting coefficients for the image signals S1 ', S2, and S3 when performing the energy subtraction processing are set by the setting unit 19 with reference to the table T stored in the storage unit 20.

FIG. 3 is a diagram schematically showing the configuration of the setting means 19 and the processing performed by the setting means 19.
As shown in FIG. 3, the setting unit 19 includes an average attenuation coefficient calculating unit 19A that calculates an average attenuation coefficient as described later,
An averaging means 19B for averaging the average attenuation coefficient calculated by the average attenuation coefficient calculating means 19A to obtain a weighting coefficient for the energy subtraction processing performed in the subtraction means 21.

In the subtraction means 21, the image signal S1 'and the image signal S2, the image signal S1' and the image signal S3, or the image signal S2 and the image signal S3 are combined. An energy subtraction process is applied to the energy subtraction process. For example,
When a combination of the scattered radiation-removed image signal S1 'and the image signal S2 is used, the energy subtraction processing is performed as shown in the following equations (5) and (6), and the soft part image in which only the soft part is extracted only the difference signal S _S and the bone portion representative of the difference signals S _B representing the bone image extracted is obtained.

[0063]

(Equation 4) In the above equations (5) and (6), the average attenuation coefficient multiplied by the scattered radiation-removed image signals S1 'and S2 is the weighting coefficient.

Next, the setting of the table T stored in the storage means 20 will be described.

The substance has a radiation attenuation coefficient depending on the radiation energy. On the other hand, when the radiation applied to the subject is not monochromatic but distributed in a certain energy range, the energy distribution of the radiation applied to the stimulable phosphor sheets IP1, IP2, and IP3 is determined by the material (bone) contained in the subject. Beam hardening, which varies depending on the thickness of the part, the soft part). Therefore, in the present embodiment, a value obtained by weighting and averaging the radiation attenuation coefficient of the substance with the energy distribution of the radiation applied to the stimulable phosphor sheet is defined as the average attenuation coefficient. Therefore, the average attenuation coefficient differs depending on the thickness of the substance.

By the way, for a plurality of substances (for example, bones and soft parts of a human body) having a characteristic that the average attenuation coefficient decreases smoothly with respect to the radiation energy, the radiation energy of the radiation transmitted through each substance and the average attenuation coefficient are calculated. Can be approximated by a simple equation. For example, there is a relationship shown in FIG. 4A between the radiation energy and the average attenuation coefficient for each of the bone and the soft part of the human body.
Here, the average attenuation coefficient of the bone is multiplied by a constant p,
Further, by adding a constant q (q> 0), FIG.
As shown in (b) and the following equation (7), the average attenuation coefficient of the soft part can be approximately expressed using the average attenuation coefficient of the bone part. In the present embodiment, the subscript “S” of the average attenuation coefficient represents a soft part, and “B” represents a bone part.

[0067]

(Equation 5) Conversely, by multiplying the average attenuation coefficient of the soft part by a constant p ′ and further adding a constant q ′ (q> 0), the average attenuation coefficient of the bone part is calculated by the following equation (8). Can be approximated using the average attenuation coefficient of

[0068]

(Equation 6) Here, for ease of explanation, sheets IP1, IP2
Assume that you have used only In such an assumption,
The energy emission distribution of the radiation source 3 is S (E), sheet I
The sensitivity of P1 and IP2 to radiation energy is D
Assuming (E), the energy distributions I1 (E) and I2 (E) of the radiation applied to the sheets IP1 and IP2 can be expressed by the following equations (9) and (10).

[0069]

(Equation 7) Note that t _Cu in Expression (10) is the thickness of the filter 5A in FIG.

In equations (9) and (10), A (E) =
S (E) D (E), B (E) = S (E) exp [-μ
_IP (E) t _IP −μ _Cu (E) t _Cu ] D (E), the radiation doses I1 and I2 applied to the stimulable phosphor sheets IP1 and IP2 are expressed by the following equations (11) and (12). ).

[0071]

(Equation 8) The integration is performed over the entire energy range of the radiation.

As described above, the average attenuation coefficient is defined as the weighted average of the radiation attenuation coefficient of a substance by the energy distribution of the detected radiation. Therefore, the average attenuation coefficient of the soft part and the bone part in the sheet IP1 is represented by the following equations (13) and (14).

[0073]

(Equation 9) On the other hand, the average attenuation coefficient of the soft part and the bone part in the sheet IP2 is represented by the following equations (15) and (16).

[0074]

(Equation 10) When the relationship of the above equation (7) is established, the above equation (11),
From (12), the logarithmic value ln (I
The difference between 1) and ln (I2) (hereinafter referred to as log dose) is represented by the following equation (17).

[0075]

[Equation 11] As shown in equation (17), logarithmic doses ln (I1) and ln (I2) expressed by two variables t _s and t _B are obtained by calculating the difference ln (I1) −ln (I2). , P
It can be seen that it is represented by one variable, t _S + t _B. Here, when t = pt _S + t _B for simplicity, equation (17) can be rewritten into the following equation (18).

[0076]

(Equation 12) On the other hand, considering the average attenuation coefficient of the sheet IP1,
By rearranging the equations (13) and (14) by applying the relation of the equation (7), the equations (13) and (14) can be rewritten into the following equations (19) and (20).

[0077]

(Equation 13) Considering the average attenuation coefficient of the sheet IP2,
By applying the relationship of Expression (7) to Expressions (15) and (16), Expressions (15) and (16) are converted into Expression (2) below.
1) and (22) can be rewritten.

[0078]

[Equation 14] Since the difference between the radiation doses IP1 and IP2 of the sheets IP1 and IP2 decreases as the thickness of the object increases, it can be estimated that the above equation (18) is a monotonically decreasing function related to t. Also, as shown in FIG. 4, when the average attenuation coefficient monotonically decreases with respect to the radiation energy, the average attenuation coefficient decreases when the thickness of the object increases and the radiation after passing through the object is biased toward the high pressure side. , (22) can also be inferred to be a monotonically decreasing function for t. here,
The value of a function that monotonically decreases via a certain variable corresponds to one-to-one. Therefore, the above equations (18) and (19) to
(22) corresponds to one-to-one since both decrease monotonically via t. Therefore, as shown in equation (23) to (26) below between the average attenuation coefficients and the logarithmic radiation amount difference, a function _{F S, F B, F S} ', F B' relationship established interposed the Will be done.

[0079]

(Equation 15) Therefore, if the above functions F _S , F _B , F _S ′, and F _B ′ are experimentally obtained in advance, the average attenuation coefficient can be obtained based on the logarithmic dose difference. FIG. 5 shows the sheet IP when the thicknesses of the soft part and the bone part were variously changed experimentally.
1, logarithmic dose difference of radiation irradiated to IP2 and sheet I
It is a figure which shows the plotting result with the average attenuation coefficient in P1. As shown in FIG. 5, it can be seen that the log dose difference and the average attenuation coefficient are located on a curve represented by a certain function.

The average attenuation coefficient is calculated for each sheet IP1, I2
The bone and soft parts are obtained for each of P2 and IP3. For simplicity, the average attenuation coefficient of each sheet IP1, IP2 and IP3 is represented by μ (IP1), μ (IP2) and μ (IP3).
(Including the average attenuation coefficient of the bone part and the soft part, respectively), μ (IP1) and μ (IP2) are obtained from the logarithmic dose difference of the sheets IP1 and IP2 as described above. From the logarithmic dose difference, μ (IP1),
For μ (IP3), μ (IP2) and μ (IP3) are obtained from the logarithmic dose difference between the sheets IP2 and IP3. Therefore, the average attenuation coefficient μ (IP1) for sheet IP1
Are the sheets IP1 and IP2 and the sheets IP1 and IP3
The average attenuation coefficient μ (IP2) for the sheet IP2 is obtained from the logarithmic dose difference between the sheets IP1 and IP2 and the sheets IP2 and IP3, and the average attenuation coefficient μ (IP3) for the sheet IP3 is obtained from the sheets IP1 and IP1. IP
3 and the logarithmic dose difference between sheets IP2 and IP3.

As described above, in this embodiment, the logarithmic dose difference l for the three sheets IP1, IP2 and IP3
n (I1) -ln (I2), ln (I1) -ln (I
3) and ln (I2)-(I3), the logarithmic dose difference and the average attenuation coefficient μ (IP1), μ (IP2), μ for the sheets IP1, IP2 and IP3 corresponding to the logarithmic dose difference.
(IP3) is obtained experimentally in advance as a table T, and is stored in the storage means 20. By referring to this table T, the average attenuation coefficient μ (IP
1), μ (IP2) and μ (IP3) can be obtained. Specifically, the average attenuation coefficients μ (IP1) and μ (IP2) are calculated from ln (I1) −ln (I2) by ln (I
1) Average attenuation coefficient μ (IP1), μ from −ln (I3)
For (IP3), average attenuation coefficients μ (IP2) and μ (IP3) are obtained from ln (I2) − (I3).

The radiation dose at each pixel position of the sheets IP1, IP2, IP3 cannot be directly detected. On the other hand, the signal values of the image signals S1, S2, S3 obtained from the sheets IP1, IP2, IP3 increase as the radiation dose increases.
2, S3 and the radiation doses I1, I2, I3 can be associated with one another. Therefore, in this embodiment, the logarithmically converted and A / D converted image signals S1, S2, S3
Are used as radiation doses I1, I2 and I3. here,
Since the image signals S1, S2, and S3 are logarithmically converted,
Difference signals S1-S2, S1- of image signals S1, S2, S3.
S3, S2-S3 are log dose differences ln (I1), respectively.
-Ln (I2), ln (I1) -ln (I3), ln
(I2) -ln (I3). Therefore, the table T includes the difference signals S1-S2, S1-S3,
It is assumed that the relationship between S2-S3 and the average attenuation coefficient is stored in the storage unit 20.

Next, the processing performed by the scattered radiation removing means 18 and the setting of the weighting coefficients by the setting means 19 will be described in detail.

FIG. 6 is a diagram for explaining the processing performed in the scattered radiation removing means 18. First, in the filtering means 31A, 31B, 31C, the image signal S
1, S2, and S3 are subjected to a filtering process using a low-pass filter LP, and image signals S1, S2, and S3 are processed.
3 is a blurred image signal Sus1, Sus from which high-frequency components have been removed.
2, Sus3 are obtained. Thereby, the image signals S1, S
High frequency noise such as quantum noise is removed from S2 and S3. Here, it is assumed that the value of the image signal S1 is unknown. As described above, the storage unit 20 stores S
The relationship of 1−S2 = μ (IP2) and S2−S3 = μ (IP2) is shown in Table T (here, tables T1 and T2, respectively).
2) is stored. Note that these tables T
1, T2 is similarly established in the blurred image signals Sus1, Sus2, Sus3 of the image signals S1, S2, S3. That is, Sus1-Sus2 = μ (IP2), S
The relationship us2-Sus3 = μ (IP2) holds. Also,
This relationship does not include the influence of scattered radiation noise included in the image signal S1 or the blurred image signal Sus1.
Therefore, the blur image signals Sus2 and Sus3 are known values,
By using the true blur image signal Sus1 'as an unknown value and referring to the relationship between the tables T1 and T2, a true blur image signal Sus1' from which high-frequency noise has been removed and which is free from the influence of scattered noise can be obtained. it can.

For this purpose, first, the difference signal Sus2-Sus3 between the blurred image signals Sus2 and Sus3 is obtained in the subtractor 32, and the average attenuation coefficient μ (IP2) is obtained in the arithmetic unit 41 with reference to the table T2. The average attenuation coefficient μ
(IP2) may be a soft part or a bone part. When the average attenuation coefficient μ (IP2) is obtained, the arithmetic unit 42 refers to the table T1 to calculate the difference signal Sus1′−
Sus2 is obtained. Then, the difference signal S
By adding the value of the blurred image signal Sus2 to us1'-Sus2, a true blurred image signal S free from the influence of scattered radiation noise is obtained.
us1 'can be obtained. The above processing can be represented by the following equation (27).

Sus1 ′ = Sus2 + T1 ⁻¹ (T2 (Sus2−Sus3)) (27) where T1 ⁻¹ () is μ (IP
2) The process of obtaining Sus1'-Sus2 from (2), and T2 () refers to the table T2 to calculate μ (IP2) from Sus2-Sus3.
Is a process for obtaining

In this manner, the true blurred image signal Sus
When 1 ′ is obtained, the difference value between the blurred image signal Sus1 and the true blurred image signal Sus1 ′ is further obtained in the subtractor 34, whereby the scattered ray signal SN representing the scattered ray noise included in the image signal S1 is obtained. Desired. Then, by subtracting the scattered radiation signal SN from the image signal S1 in the subtractor 35, the true image signal S from which the scattered radiation has been removed is obtained.
1 ′, that is, the scattered radiation removed image signal S1 ′ can be obtained.

The high-frequency noise of the image represented by the image signals S1, S2, and S3 is obtained by the method described in Japanese Patent Application No. 11-363766 or Japanese Patent Application Laid-Open No. 2000-224421 instead of the filtering process using the low-pass filter. May be obtained as an image signal representing an image in which is reduced, and used as the blurred image signal Sus1, Sus2, Sus3.

Here, the filtering means 31A sets the second
Filtering means 31B, 31
C is the first noise suppression means, the subtractor 32 is the logarithmic calculation means, the calculator 41 is the average attenuation coefficient acquisition means, and the calculator 4
2 and the adder 33 correspond to the estimating means, the subtractor 34 corresponds to the scattered radiation signal calculating means, and the subtractor 35 corresponds to the scattered radiation removed image signal calculating means. Further, the image signal S1 is one image signal, the image signals S2 and S3 are a plurality of other image signals, the image signal S2 is a selected image signal, and the blurred image signals Sus1 and S2.
us2 and Sus3 correspond to the noise suppression signal, respectively.

In the setting means 19, first, in the mean attenuation coefficient calculating means 19A, the scattered radiation removed image signal S1 'is obtained.
And difference signals S1'-S2, S from the image signals S2, S3.
1′-S3 and S2−S3 are calculated, and the average attenuation coefficient μ (I
P1), μ (IP2), and μ (IP3) are obtained. Here, an average attenuation coefficient μ (IP
1), μ (IP2) is obtained from the difference signal S1-S3, and the average attenuation coefficient μ (IP1), μ (IP3) is obtained from the difference signal S2-S3.
, The average attenuation coefficients μ (IP2) and μ (IP3) are obtained. Here, the average attenuation coefficient obtained from the difference signal S1'-S2 is represented by μ ₁ (IP1), μ ₁ (IP2), and the difference signal S1.
The average attenuation coefficient obtained from −S3 is μ ₂ (IP1),
μ ₂ (IP3) and the average attenuation coefficient obtained from the difference signal S2−S3 are μ ₃ (IP2) and μ ₃ (IP3). In the averaging means 19B, the average attenuation coefficient μ ₁ (IP1),
μ ₂ (IP1) average value μ _T (IP1), average attenuation coefficient μ
₁ (IP2), _{μ 3} the mean value of (IP2) μ _T (IP2) and the average attenuation coefficient μ ₂ (IP3), μ ₂ average of (IP3) μ _T (IP3) is determined, this is the subtraction means 21 It is set as a weighting factor.

In the subtraction means 21, the scattered radiation-removed image signal S1 ', the image signal S2, and the image signal S1
And an image signal S3, or an image signal of the image signal S2 and the image signal S3. When the image signal S1 'and the image signal S2 are used, according to the above equations (5) and (6), when the image signal S1' and the image signal S3 are used,
According to the following equations (28) and (29), when the image signal S2 and the image signal S3 are used, the following equations (30) and (3)
The energy subtraction process is performed according to 1).

[0092]

(Equation 16)

[Equation 17] The combination of the image signals to be subjected to the energy subtraction processing may be selected in advance, but in the present embodiment, since the scattered radiation information has been removed from the image signal S1, the scattered radiation removed image signal S1 'is used. Is preferred.

Next, the operation of the present embodiment will be described. FIG. 7 is a flowchart showing the operation of the present embodiment. The output signals Q1, Q2, Q3 read from the sheets IP1, IP2, IP3 are logarithmically converted by the logarithmic converter 16 (step S1), and A / D converted by the A / D converter 17 to obtain a digital image signal S1. , S2, S3
Is obtained (step S2). Image signals S1, S2, S
3 is input to the scattered radiation removing means 18, where the scattered radiation removed image signal S1 'from which the scattered radiation noise of the image signal S1 has been removed is obtained with reference to the table T as described above.

More specifically, the blurred image signals Sus1, Sus2, Sus3 of the image signals S1, S2, S3 are calculated (step 3), and the difference signal Sus2 of the blurred image signals Sus2, Sus3 is calculated.
−Sus3 is calculated (step 4), and the difference signal Sus2-S
Average attenuation coefficient μ based on us3 and table T2
(IP2) is obtained (step 5). Then, based on the average attenuation coefficient μ (IP2), the table T1 and the blurred image signal Sus2, a true blurred image signal Sus1 ′ free from the influence of scattered radiation noise is calculated (step 6). Next, a true blur image signal Sus1 'is obtained from the blur image signal Sus1.
Is subtracted to calculate a scattered radiation signal SN (Step 7), and the scattered radiation signal SN is subtracted from the image signal S1 to calculate a scattered radiation-removed image signal S1 'from which scattered radiation noise has been removed (Step 8). .

The scattered radiation-removed image signal S 1 ′ and the image signals S 2, S 3 are calculated by the average attenuation coefficient calculating means 1 of the setting means 19.
9A, where the log dose difference l is
n (I1) -ln (I2), ln (I1) -ln (I
3), the difference signal S corresponding to ln (I2) -ln (I3)
1'-S2, S1-S3, S2-S3 are calculated for each corresponding pixel, and the difference signals S1'-S2, S1-S3
Referring to the table T based on S2-S3, the difference signal S
1'-S2, S1-S3, S2-S3, sheet IP
The average attenuation coefficient of the soft part and the bone part for each of IP1, IP2, and IP3 is calculated (step S9).

Then, in the averaging means 19B, the average value of the average attenuation coefficient obtained for each of the difference signals S1'-S2, S1-S3, S2-S3 is calculated as the sheet IP1, IP2, IP2.
3 and set as a weighting coefficient (step S10). The weighting coefficient is input to the subtraction means 21, and the image signals (here, the energy subtraction processing is performed on the scattered radiation-removed image signals S 1 ′ and S 2) are weighted by the weighting coefficient, and the above equations (5) and (6) Are performed to obtain difference signals S _S and S _B representing the radiation image information of the soft part and the bone part in the subject (step S11), and the processing ends. The obtained S _S and S _B are used as reproducing means (not shown) (printer, CR,
T) etc., and provided for diagnosis.

As described above, in this embodiment, in view of the fact that there is a certain relationship between the logarithmic value difference of the radiation dose in the sheets IP1, IP2, and IP3, that is, the logarithmic dose difference and the average attenuation coefficient, Dose difference ln (I1)-
ln (I2), ln (I1) -ln (I3) and ln
(I2)-(I3) (difference signals S1-S2, S1-S3,
The relationship between S2-S3) and the average attenuation coefficient is obtained as a table T, and the image signal S containing a large amount of scattered radiation noise is obtained.
1 is obtained by referring to the table T to obtain a scattered radiation removed image signal S1 'from which scattered radiation noise has been removed. Therefore, the scattered radiation-removed image signal S
By performing energy subtraction processing using 1 'and the image signals S2 and S3, scattered radiation noise does not affect the subtraction processing, and an energy subtraction image in which a specific structure is well erased can be obtained. it can.

Since the image signals S1, S2, and S3 obtained by photographing the subject 1 include high-frequency noise such as quantum noise of radiation, the image signals S2 and S3 are filtered by a low-pass filter. By performing the filtering process, blurred image signals Sus2 and Sus3 in which high-frequency noise is suppressed can be obtained. Further, by referring to the table T2 from these blur image signals Sus2 and Sus3, a true blur image signal Sus1 'from which scattered radiation noise has been removed is obtained without being affected by high frequency noise. Then, a blurred image signal Sus1 obtained by performing a filtering process using a low-pass filter on the image signal S1 and a true blurred image signal Sus1 ′.
By obtaining the difference value from the scattered radiation signal, a scattered radiation signal SN including only information on low-frequency scattered radiation can be obtained. Therefore, by subtracting the scattered radiation signal SN from the image signal S1, it is possible to obtain a scattered radiation eliminated image signal S1 'from which scattered radiation has been appropriately removed.

In the first embodiment, since the table T representing the relationship between the logarithmic dose difference and the average attenuation coefficient is obtained in advance, the average attenuation coefficient can be easily obtained. The calculation of the scattered radiation-removed image signal S1 'and the calculation of the difference signals S _S and S _B can be performed efficiently.

In the first embodiment, 3
Photographing is performed using the stimulable phosphor sheets IP1, IP2, and IP3, and energy subtraction is performed using any two of the image signals S1, S2, and S3 obtained from each of the sheets IP1, IP2, and IP3. Although the processing is being performed, as shown in FIG.
Using IP3, an image is taken by interposing a radiation energy conversion filter 5 between each of the sheets IP1 and IP3. For the sheet IP1, for example, two-sided reading described in JP-A-55-87970 is performed and two sheets are read. The image signals S1 and S2 may be obtained and averaged to obtain an average signal Sm, and the energy subtraction processing may be performed using the addition signal Sm and the image signal S3 obtained from the sheet IP3. Hereinafter, this will be described as a second embodiment.

FIG. 9 is a diagram showing the configuration of a radiation image reading apparatus for performing double-sided reading to which the energy subtraction apparatus according to the second embodiment of the present invention is applied. As shown in FIG. 9, this radiation image reading apparatus is provided with endless belts 9a and 9b which carry the stimulable phosphor sheets IP1 and IP3 and convey the stimulable phosphor sheets IP1 and IP3 in the direction of arrow Y. Above the endless belts 9a and 9b. Is provided with a laser light source 10 for emitting a laser beam 11 and a scanning mirror 12. further,
Above the position where the laser light 11 is scanned (scanning position), a light guide 14a for focusing the stimulated emission light emitted by the scanning of the laser light 11 from above is arranged close to the scanning position. Below the position, a light guide 14b for concentrating stimulated emission light from below is disposed perpendicular to the sheet IP1. Each of the light guides 14a and 14b is a photomultiplier 1 that photoelectrically detects stimulated emission light.
5a and 15b are connected. The photomultipliers 15a and 15b are connected to logarithmic amplifiers 16a and 16b.
It is connected to converters 17a and 17b.

In this radiation image reading apparatus, a radiation image is read as follows. A stimulable phosphor sheet IP1 on which a radiation image of a subject is accumulated and recorded
Is moved by the endless belts 9a and 9b in the direction of arrow Y, the laser beam 11 from the laser light source 10 is deflected by the scanning mirror 12, and the sheet IP1 is scanned in the X direction. From the portion of the sheet IP1 irradiated with the laser beam 11, the stimulable luminescent light 13a, 13b (here, stimulable luminescent light) having an amount of light corresponding to the radiation image information stored and recorded in the stimulable phosphor sheet IP1. 13a, 13
b indicates the divergence from above (front) and below (back) of the sheet IP1). The stimulated emission light 13a enters the inside of the light guide 14a from one end face of the light guide 14a, travels therethrough while repeating total reflection, and is received by the photomultiplier 15a. The photomultiplier 15a outputs an analog output signal Q1 corresponding to the light emission amount of the stimulating light 13a, ie, indicating the image information.

On the other hand, the stimulated emission light 13b is
The light enters the light guide 14b from one end surface of the light guide b, travels therethrough while repeating total reflection, and is received by the photomultiplier 15b. From this photomultiplier 15b, the amount of light of the photostimulated light 13b was
That is, the output signal Q2 indicating the image information is output.

These output signals Q1 and Q2 are logarithmically converted by logarithmic converters 16a and 16b, and then input to A / D converters 17a and 17b to obtain digital image signals S1 and Q2.
1, S2.

On the other hand, a radiation image is read from the stimulable phosphor sheet IP3 as in the case of the sheet IP1, but only the stimulated emission light 13a emitted from the surface of the sheet IP3 by scanning with the laser beam 11 is guided by the light guide. The output signal Q3 is detected by the photomultiplier 14a and the photomultiplier 15a, logarithmically converted by the logarithmic converter 16a, and A / D converted by the A / D converter 17a to obtain a digital image signal S3.

The image signals S1, S2, and S3 are input to the scattered radiation removing means 18, where the scattered radiation noise is obtained from the image signal S1 obtained from the surface of the stimulable phosphor sheet IP1 located closest to the subject 1. Is removed. Here, the radiation 2 transmitted through the subject 1 includes scattered radiation. Since the scattered radiation has low energy, the sheet IP
1, most of the radiation is absorbed, and the radiation 2 transmitted through the sheet IP1 contains almost no scattered radiation. Therefore, the image signal S2 obtained from the back surface of the sheet IP1
In addition, the image represented by the image signal S3 obtained from the sheet IP3 hardly includes noise due to scattered radiation. Therefore, the scattered radiation removing means 18 removes scattered radiation noise only from the image signal S1, and obtains a scattered radiation removed image signal S1 'from which scattered radiation has been removed.

Here, in the second embodiment, two image signals S1 and S2 are obtained from the sheet IP1. In addition, the storage unit 20 stores a table T representing the relationship between each logarithmic dose difference and the average attenuation coefficient for the front surface of the sheet IP1, the back surface of the sheet IP1, and the sheet IP3 for each logarithmic dose difference. Therefore, sheet IP1
A logarithmic dose difference is obtained by assuming that the amount of radiation applied to the front surface of the sheet IP1 is I1, the amount of radiation applied to the back surface of the sheet IP3 is I2, and the amount of radiation applied to the sheet IP3 is I3. By associating the back surfaces of the sheet IP1 and the sheet IP1 with the sheet IP2 in the first embodiment, the scattered radiation-removed image signal S1 'can be obtained in the same manner as in the first embodiment.

The scattered radiation-removed image signal S1 'and the image signals S2 and S3 are input to the setting means 19, where the sheet is referred to by referring to the table T stored in the storage means 20, as in the first embodiment. An average attenuation coefficient for IP1 and IP3 is obtained, and further averaged to set a weighting coefficient when the subtraction unit 21 performs the energy subtraction processing.

The average attenuation coefficient is obtained by calculating the difference signal between the scattered radiation-removed image signal S1 'for the front surface of the sheet IP1, the image signal S2 for the back surface of the sheet IP1, and the image signal S3 for the sheet IP3. Are the sheet IP1 and the sheet IP1 in the first embodiment.
By associating the back surface with the sheet IP2 in the first embodiment, it can be obtained in the same manner as in the first embodiment.

Thus, the surface of the sheet IP1 and the sheet I
The average attenuation coefficient for the P back surface and the sheet IP3 is obtained for each difference signal, and the average of these average attenuation coefficients is obtained for the sheet IP1 as in the first embodiment.
Mean attenuation coefficient μ _T (IP1 front surface), μ _T (IP1 back surface) for front surface, sheet IP1 back surface and sheet IP3,
μ _T (IP3) (including soft and bone parts) is set as a weighting factor.

Here, in the second embodiment, the scattered-ray-removed image signal S obtained from the surface of the sheet IP1 and from which the scattered radiation has been removed in the subtraction means 21.
1 'and an image signal S2 obtained from the back surface of the sheet IP1.
Is obtained, and an energy subtraction process is performed using the average image signal Sm and the image signal S3. Therefore, the setting means 19
In, the average attenuation coefficient μ _T (IP1 surface) and μ _T (I
The average value of the P1 back surface) determined as an average attenuation coefficient μ _Tm (IP1), using the average attenuation coefficient mu _Tm (IP1) as a weighting factor for the average image signal Sm. In addition,
To be precise, the average attenuation coefficient μ _Tm (IP1) is the average attenuation coefficient for each of the soft part and the bone part.

(Equation 18) In the subtraction means 21, the average image signal S
With respect to m and the image signal S3, the following equations (32) and (3)
An energy subtraction process is performed as shown in 3).

[0112]

[Equation 19] In this way, the average value image signal Sm of the image signals S1 and S2 obtained from both sides of one sheet IP1 is obtained, and the average image signal Sm and the image signal S obtained from the sheet IP3 are obtained.
In the case where the energy subtraction process is performed on the image signal S1, the scattered radiation noise included in the image signal S1 can be removed in the same manner as in the first embodiment. Therefore, the scattered radiation removed image signal S1 'and the image signals S2, S
By performing the energy subtraction processing using No. 3, scattered radiation noise does not affect the subtraction processing, and an energy subtraction image in which a specific structure is well erased can be obtained.

In the second embodiment, 1
Although the stimulable phosphor sheet IP1 is scanned by the laser light 11 generated from the two laser light sources 10, the present invention is not limited to this. Laser light and scanning are applied to both the front side and the back side of the sheet IP1. A mirror may be provided, and laser light may be scanned on both sides of the stimulable phosphor sheet IP1 to read the stimulated emission light to obtain two image signals S1 and S2.

In the first and second embodiments, the logarithmic dose difference ln between the radiation doses I1, I2, and I3 is used.
The relationship between (I1) -ln (I2) and the like and the average attenuation coefficient is set, but lnI1-lnI2 = ln (I1 / I
2), the radiation doses I1, I2, I3
Logarithmic value l of the ratio I1 / I2, I1 / I3, I2 / I3 between
n (I1 / I2), ln (I1 / I3), ln (I2 /
A relationship between I3) and the average attenuation coefficient may be set. The radiation doses I1, I2, I3 and the image signals S1, S2,
Since the image signals S1, S2, and S3 are logarithmically converted, the ratio signals S1 / S2, S1 /
By obtaining the relationship between S3, S2 / S3 and the average attenuation coefficient, the average attenuation coefficient is obtained based on the ratio signal.

In the first and second embodiments, the relationship between the logarithmic dose difference and the average attenuation coefficient is obtained as the table T. However, a function representing the relationship between the logarithmic dose difference and the average attenuation coefficient is obtained. May be stored in the storage means 20 to calculate the scattered radiation removal image signal S1 'and obtain the average attenuation coefficient from the logarithmic dose difference ln (I1) -ln (I2) using each function.

Here, the relationship between the logarithmic dose difference and the average attenuation coefficient is based on the voltage of the radiation source
Of the sheet IP1, IP2, IP3, and the like, and the photographing conditions such as the sensitivity. Therefore, like the radiation image reading apparatus to which the energy subtraction apparatus according to the third embodiment of the present invention shown in FIG. 10 is applied,
A plurality of tables T1, T2,... According to various photographing conditions
Are prepared in advance and stored in the storage means 20, and the input of the photographing conditions from the input means 22 including a keyboard, a mouse, and the like is accepted, and a table T suitable for the photographing conditions is prepared according to the inputted photographing conditions. Alternatively, calculation of the scattered radiation removed image signal S1 'and acquisition of the average attenuation coefficient may be performed.

In the first and second embodiments, the image signals S1, S2, and S3 are obtained using the stimulable phosphor sheets IP1, IP2, and IP3 as radiation detecting means. Other radiation detecting means such as a semiconductor sensor may be used.

Further, in the first and second embodiments, a radiation image representing a soft part and a bone part of a human body is obtained with the subject as a human body. 2. Description of the Related Art An energy subtraction process is performed on an image signal representing a radiation image taken before and after a durability test of a product to detect a structural change. In the field of food production, an image signal representing a radiation image of a normal food and a food to be shipped may be subjected to energy subtraction processing to perform a foreign substance inspection of the produced food. The present invention can be applied to energy subtraction processing other than medical use.

Further, in the first and second embodiments, the obtained scattered-ray-removed image signal S1 'is used for energy subtraction processing. Signal S3
Are added by the corresponding pixels to obtain S /
The scattered radiation-removed image signal S1 'may be used in a superposition process for obtaining an addition signal representing an image with good N. Alternatively, only the scattered radiation-removed image signal S1 'alone may be reproduced.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating recording of radiation image information on a stimulable phosphor sheet according to a first embodiment.

FIG. 2 is a schematic diagram illustrating a configuration of a radiation image reading apparatus to which the radiation image information estimating apparatus and the energy subtraction apparatus according to the first embodiment of the present invention are applied;

FIG. 3 is a diagram schematically illustrating a configuration of a setting unit and a process performed by the setting unit;

FIG. 4 is a diagram showing a relationship between radiation energy and an average attenuation coefficient.

FIG. 5 is a diagram showing a relationship between a logarithmic dose difference and an average attenuation coefficient.

FIG. 6 is a diagram for explaining processing performed in a scattered radiation removing unit.

FIG. 7 is a flowchart showing the operation of the embodiment.

FIG. 8 is a view for explaining recording of radiation image information on a stimulable phosphor sheet according to the second embodiment.

FIG. 9 is a schematic diagram showing a configuration of a radiation image reading apparatus to which a radiation image information estimating apparatus and an energy subtraction apparatus according to a second embodiment of the present invention are applied;

FIG. 10 is a schematic diagram illustrating a configuration of a radiation image reading apparatus to which a radiation image information estimating apparatus and an energy subtraction apparatus according to a third embodiment of the present invention are applied;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Subject 2 Radiation 3 Radiation source 10 Laser light source 11 Laser light 12 Scanning mirror 13 Stimulated emission light 14 Light guide 15 Photomultiplier 16 Logarithmic converter 17 A / D converter 18 Scattered ray elimination means 19 Setting means 19A Average attenuation coefficient Setting means 19B averaging means 20 storage means 21 subtraction means 22 input means 31A, 31B, 31C filtering means 32, 34, 35 subtractor 33 adder 41, 42 arithmetic unit IP1, IP2, IP3 stimulable phosphor sheet

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) H04N 1/04 H04N 1/04 E 5C072 F term (Reference) 2H013 AC06 4C093 AA16 CA07 EA07 EB05 FD02 FF03 FF34 5B047 AA17 BA02 BB06 CA05 CB04 CB12 5B057 AA07 BA03 CE02 CE04 CE09 DB02 5C024 AX16 AX17 CX03 HX21 HX23 HX28 HX29 HX55 5C072 AA01 EA02 VA01

Claims (24)

[Claims] 1. A plurality of image signals of three or more radiation image information, at least a part of which are obtained by radiations having different energy distributions transmitted through the same subject and different from each other, are represented by other radiation signals. For a plurality of other image signals other than one image signal representing one radiation image information containing more scattered radiation noise compared to the radiation image information, the difference or ratio of the logarithmic value between the plurality of other image signals The logarithmic value of the difference between the logarithmic value of the radiation dose in each pixel between the respective radiation image information, or the logarithmic value of the ratio of the radiation dose, and the relationship between the average attenuation coefficient for each of the radiation image information, And a radiation image information represented by at least one image signal of the other plurality of image signals based on a logarithmic value difference or a logarithmic ratio between the other plurality of image signals. The average radiation attenuation coefficient is obtained, and the true radiation corresponding to the one radiation image information is obtained based on the relation, the average attenuation coefficient, and one selected image signal selected from the plurality of other image signals. A radiation image information estimating method, wherein a true image signal representing image information is estimated. 2. A scattered radiation signal representing scattered radiation information included in the one radiation image information is calculated based on the true image signal and the one image signal. Radiation image information estimation method. 3. A plurality of other noise suppression signals are obtained by performing noise suppression processing on the plurality of other image signals, and a difference or ratio of logarithmic values or ratios between the plurality of other noise suppression signals is obtained. The logarithmic value is obtained as the logarithmic value of the difference or ratio of the logarithmic value between the plurality of other image signals, and the average attenuation coefficient is obtained and the true image signal is estimated. 3. A noise suppression signal is obtained by performing a noise suppression process, and a difference value between the true image signal and the one noise suppression signal is calculated as the scattered radiation signal. Radiation image information estimation method. 4. The radiation according to claim 3, wherein the scattered radiation signal is subtracted from the one image signal to obtain a scattered radiation-removed image signal representing radiation image information from which scattered radiation noise has been removed. Image information estimation method. 5. The radiation image information estimating method according to claim 1, wherein the true image signal is output as a scattered radiation-removed image signal representing radiation image information from which scattered radiation noise has been removed. 6. The method according to claim 1, wherein the relationship is set as a table or a function representing a relationship between a previously calculated difference between the logarithmic values or a logarithmic value of the ratio and the average attenuation coefficient. 6. The radiation image information estimating method according to any one of claims 1 to 5. 7. A plurality of tables or functions representing the relationship set according to the imaging conditions when obtaining the radiation image information are prepared, and the plurality of tables or functions are prepared based on the imaging conditions when the plurality of radiation image information is obtained. 7. The radiation image information estimating method according to claim 6, wherein selection of a table or a function is received, and the selected table or function is set as the relation. 8. The scattered radiation-removed image signal and the plurality of other image signals obtained by the radiation image information estimating method according to any one of claims 1 to 7. Two image signals representing the plurality of other image signals are subtracted by multiplying each of the image signals by a predetermined weighting coefficient between signals of corresponding pixels, thereby performing the specific structure of the object. An energy subtraction method characterized by obtaining a difference signal representing an image of the image. 9. Among a plurality of image signals representing three or more pieces of radiation image information, at least a part of which is obtained by radiation having different energy distributions transmitted through the same subject and different from each other, among other image signals, For a plurality of other image signals other than one image signal representing one radiation image information containing more scattered radiation noise compared to the radiation image information, the difference or ratio of the logarithmic value between the other image signals Logarithmic calculation means for calculating the logarithmic value of the difference between the logarithmic value of the radiation dose at each pixel between the respective pieces of radiation image information, or the logarithmic value of the ratio of the radiation dose, and the average attenuation coefficient for each piece of the radiation image information And at least one of the other plurality of image signals based on a logarithmic difference or ratio of logarithmic values between the other plurality of image signals. Average attenuation coefficient acquisition means for acquiring an average attenuation coefficient for the ray image information, based on the relationship, the average attenuation coefficient, and one selected image signal selected from the plurality of other image signals, An estimating unit for estimating a true image signal representing true radiation image information corresponding to the radiation image information. 10. A scattered radiation signal calculating means for calculating a scattered radiation signal representing scattered radiation information included in the one radiation image information based on the true image signal and the one image signal. The radiation image information estimating apparatus according to claim 9, wherein: 11. A first noise suppression unit that obtains a plurality of other noise suppression signals by performing a noise suppression process on the plurality of other image signals, and noise suppression for the one image signal. A second noise suppression unit that obtains one noise suppression signal by performing a process, wherein the logarithmic calculation unit and the average attenuation coefficient acquisition unit include:
Obtaining the logarithmic value of the difference or ratio of the logarithmic value between the plurality of other noise suppression signals or the logarithmic value of the difference or ratio of the logarithmic value between the plurality of other image signals, obtaining the average attenuation coefficient and A scattered radiation signal calculating unit that calculates a difference value between the true image signal and the one noise suppression signal as the scattered radiation signal. The radiation image information estimation device according to claim 10, wherein 12. A scattered-ray-removed image signal calculating means for calculating a scattered-ray-removed image signal representing radiation image information from which scattered-ray noise has been removed by subtracting the scattered-ray signal from the one image signal. The method of claim 11, wherein:
The radiation image information estimating apparatus according to the above. 13. The apparatus according to claim 9, wherein said estimating means is means for outputting the true image signal as a scattered radiation-removed image signal representing radiation image information from which scattered radiation noise has been removed. Radiation image information estimation device. 14. A storage unit storing a table or a function representing a relationship between a difference between the logarithmic values or a logarithmic value of the ratio calculated in advance and the average attenuation coefficient, wherein the average attenuation coefficient acquisition unit includes: 14. The radiation image information estimating method according to claim 9, wherein the relation is a means for setting a table or a function stored in the storage means as the relation. 15. The storage unit stores a plurality of tables or functions representing the relationship set according to imaging conditions for obtaining the radiation image information. The average attenuation coefficient obtaining unit stores the radiation attenuation information. 15. A means for receiving selection of the table or function based on a photographing condition at the time of obtaining an image, and setting the selected table or function as the relation.
The radiation image information estimating apparatus according to the above. 16. The scattered radiation removal image signal and the plurality of other image signals obtained by the radiation image information estimating apparatus according to claim 9.
The scattered radiation-removed image signal and the two image signals representing the plurality of other image signals are subtracted by multiplying each of the image signals by a predetermined weighting coefficient between signals of corresponding pixels. And a subtraction unit for obtaining a difference signal representing an image of the specific structure of the subject. 17. Among a plurality of image signals representing three or more pieces of radiation image information, at least a part of which is obtained by radiations having different energy distributions transmitted through the same subject and different from each other, among other image signals, For a plurality of other image signals other than one image signal representing one radiation image information containing more scattered radiation noise compared to the radiation image information, the difference or ratio of the logarithmic value between the plurality of other image signals And calculating the logarithmic value of the difference between the logarithmic value of the radiation dose at each pixel between the respective radiation image information, or the logarithmic value of the ratio of the radiation dose, and the average attenuation coefficient for each of the radiation image information. Radiation represented by at least one image signal of the other plurality of image signals based on a relationship and a logarithmic difference or ratio of logarithmic values between the other plurality of image signals A step of obtaining an average attenuation coefficient for the image information, and corresponding to the one radiation image information based on the relationship, the average attenuation coefficient, and one selected image signal selected from the plurality of other image signals. A computer-readable recording medium storing a program for causing a computer to execute a radiation image information estimating method having a procedure of estimating a true image signal representing true radiation image information. 18. The method according to claim 18, further comprising a step of calculating a scattered radiation signal representing scattered radiation information included in the one radiation image information, based on the true image signal and the one image signal. The computer-readable recording medium according to claim 17. 19. A procedure for obtaining a plurality of other noise suppression signals by performing a noise suppression process on the plurality of other image signals; and performing a noise suppression process on the one image signal. Acquiring the one noise suppression signal, further comprising: acquiring the average attenuation coefficient; and estimating the true image signal, wherein a difference in logarithmic value between the plurality of other noise suppression signals or The logarithmic value of the ratio is determined as the logarithmic value of the difference or the logarithmic value of the plurality of other image signals, a step of obtaining the average attenuation coefficient and estimating the true image signal, the scattered radiation 19. The computer-readable recording medium according to claim 18, wherein the step of calculating the signal is a step of calculating a difference value between the true image signal and the one noise suppression signal as the scattered radiation signal. body. 20. The method according to claim 20, further comprising a step of subtracting the scattered radiation signal from the one image signal to obtain a scattered radiation-removed image signal representing radiation image information from which scattered radiation noise has been removed. 20. A computer-readable recording medium according to claim 19. 21. The computer-readable computer according to claim 17, further comprising a step of outputting the true image signal as a scattered radiation-removed image signal representing radiation image information from which scattered radiation noise has been removed. recoding media. 22. The step of obtaining the average attenuation coefficient comprises: setting a table or a function representing a relationship between a logarithm of the logarithm value or a logarithmic value of the ratio calculated in advance and the average attenuation coefficient as the relationship. 22. The computer-readable recording medium according to claim 17, wherein said step is a step of acquiring said average attenuation coefficient. 23. When preparing a plurality of tables or functions representing the relations set according to imaging conditions for obtaining the radiation images, the step of obtaining the average attenuation coefficient comprises obtaining the plurality of radiation images. 23. The method according to claim 22, further comprising: receiving a selection of the table or the function based on a photographing condition at the time of setting, setting the selected table or function as the relation, and acquiring the average attenuation coefficient. The computer-readable recording medium according to claim 1. 24. The scattered radiation-removed image signal and the plurality of other image signals obtained by the radiation image information estimating method according to any one of claims 1 to 7. Two image signals representing the plurality of other image signals are subtracted by multiplying each of the image signals by a predetermined weighting coefficient between signals of corresponding pixels, thereby performing the specific structure of the subject. A computer-readable recording medium on which a program for causing a computer to execute an energy subtraction method having a procedure of obtaining a difference signal representing an image of the computer is recorded.