JP2000069369A - Energy subtraction image forming device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enhance energy discrimination performance in the energy subtraction of a one-shot method. SOLUTION: X-rays are detected by using a sensor 31 which generates a signal corresponding to the energy of radiation ray particles. Comparators 42a, 42b separate an output voltage pulse S0 of the sensor 31 into a low-energy level and a high-energy level. Counters 43a, 43b count output pulses PL, PH of the comparators 42a, 42b to count the numbers of radiant ray particles of the low-energy level and the high-energy level made incident onto the sensor 31 in a unit time respectively. A signal processing means 44 generates low- energy image data DL and high-energy image data DH, based on the radiant ray particle quantities CL, CH of respective counted energy levels. A subtraction processing means 3 conducts subtraction processing based on both image data DL, DH.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an energy subtraction image forming apparatus for radiographic images, and more particularly to an energy subtraction image forming apparatus for obtaining an image having excellent observation performance by one irradiation of radiation.

[0002]

2. Description of the Related Art A stimulable fluorescent recording medium (e.g., a stimulable phosphor sheet or the like, hereinafter simply referred to as a "recording medium") on which radiation image information is accumulated and recorded is irradiated with excitation light such as laser light. There has been known a radiation image information reading apparatus that detects stimulating luminescence light that emits stimulating light in accordance with the radiation image information stored and recorded on the recording medium and reads the radiation image information (Japanese Patent Application Laid-Open No. Sho 62-18536). No.).

Further, in such a radiation image information reading apparatus, an apparatus for reading a plurality of recorded radiation image information to obtain a plurality of image data and performing a subtraction process based on the plurality of image data has also been proposed. Have been.

Here, the subtraction processing of the radiation image information is a subtraction processing of a plurality of pieces of radiation image information taken under different conditions in correspondence with each pixel of both images to obtain a specific structure in the radiation image information. A process for obtaining an image corresponding to a difference signal for forming an image of an object. Specifically, a plurality of digital images corresponding to each piece of radiation image information are read by reading these pieces of radiation image information at a predetermined sampling interval. A signal is obtained, and a subtraction process is performed for each corresponding sampling point of the plurality of digital image signals to obtain radiation image information in which only a specific subject portion in the radiation image information is enhanced or extracted.

There are basically the following two methods for this subtraction processing. That is, the image signal of the radiation image information to which no contrast agent is injected is subtracted from the image signal of the radiation image information in which a specific part of the subject (for example, a blood vessel when a human body is the subject) is enhanced by the injection of the contrast agent. A so-called temporal subtraction process for extracting a specific portion (eg, a blood vessel) of the subject by performing (subtracting) and utilizing that the specific portion of the subject has different radiation absorptivity for radiation having different energies. The same subject is irradiated with radiation having different energies to obtain a plurality of pieces of radiation image information from the respective radiations having different energies. The plurality of pieces of radiation image information are appropriately weighted to determine the difference. By calculating (see the following equations (1) and (2)), There are so-called energy subtraction processing of extracting (an image obtained by the energy subtraction processing (or image information) is referred to as "energy subtraction image (or image information)".). The applicant has
An energy subtraction process for a radiation image using a stimulable phosphor sheet has been proposed (Japanese Patent Laid-Open No. 60-2)
No. 25541, JP-A-3-285475).

S = Ka · H−Kb · L + Kc (1) where S is energy subtraction image information obtained by subtraction processing, Ka and Kb are weighting coefficients, and Kc is a bias component (hereinafter, Ka, Kb, Kc
H and L are radiation image information to be subtracted, H is radiation image information of a component having a relatively high energy level (so-called high pressure side), and L is a relatively low energy. It means radiation image information of a level component (so-called low pressure side).

In some cases, the bias component Kc is represented by the following equation (2), with the bias component Kc being zero.

S = Ka · H−Kb · L (2)

[0009]

SUMMARY OF THE INVENTION Here, JP-A-60-225
The energy subtraction process proposed in Japanese Patent Application No. 541 and Japanese Patent Application Laid-Open No. 3-285475 is a so-called two-shot (two-shot) method. Radiation image information on a recording medium, and then irradiates the same subject with high-energy radiation to record high-voltage radiation image information on the recording medium. The radiation image information is read separately from the recording medium on which the image information is recorded and from the recording medium on which the high-voltage side radiation image information is recorded, and is subjected to the arithmetic processing as in the above equation (1) or (2). It is possible to perform an energy subtraction process excellent in energy separation. However, since the energy subtraction processing by the two-shot method causes a time difference in recording both image information,
There is a problem that the motion of the subject between the two pieces of image information causes blur or artifacts in the energy subtraction image, and a high-precision energy subtraction process cannot be performed.

On the other hand, JP-A-59-83147 and JP-A-5-14
The method using the laminate proposed in No. 2713,
This is a method in which radiation image information having different energy levels can be captured and energy subtraction processing can be performed by one radiation irradiation (one shot, one shot), and the movement of the subject does not matter. However, in the energy subtraction processing by the one-shot method, the radiation absorption region of the laminated body is not completely separated, and the radiation energy generally has a certain width, so that the radiation image information on the low voltage side and the high voltage side Cannot be completely separated from the radiation image information, that is, the energy separation property is not sufficient.

As described above, it is difficult to perform the energy subtraction processing by the conventionally proposed method, which does not cause blurring or artifacts and has both excellent energy separation properties. Met.

The present invention has been made in view of the above circumstances, and provides an energy subtraction image forming apparatus that forms and generates an energy subtraction image having excellent energy separation while employing a one-shot method that does not cause blurring or artifacts. It is intended to provide.

[0013]

An energy subtraction image forming apparatus according to the present invention comprises a sensor for generating a signal corresponding to the energy of radiation particles, and a plurality of radiation particles having different energy levels incident on the sensor in a unit time. Image signal generating means for counting the number of pixels for each pixel, and generating radiation image signals of different energy components based on the counted amount of radiation particles for each energy level,
Energy subtraction processing means for performing an energy subtraction process between the generated radiation image signals of different energy components.

The image signal generating means of the image forming apparatus comprises:
The counting is performed as a plurality of energy levels as a noise level, an energy level higher and lower than the noise level, and an energy level higher than and higher than the noise level, and the radiation image signal having the lower energy component and the lower A radiographic image signal having a high energy component is generated, and the energy subtraction processing means performs an energy subtraction process between the radiographic image signal having a relatively low energy component and the radiographic image signal having a relatively high energy component. It is desirable that

In this case, the image signal generating means counts energy near the energy corresponding to the K absorption edge of the main component of the subject as a boundary value between a relatively low energy level and a relatively high energy level. Is desirable.

The image signal generating means of the image forming apparatus converts the plurality of energy levels into a noise level, an energy level higher and lower than the noise level,
The counting is performed with the energy level being higher than the noise level and a relatively high energy level, and the energy level higher than the noise level and a medium level, and a radiation image signal having a relatively low energy component and a radiation image signal having a relatively high energy component are generated. Preferably, the energy subtraction processing means performs energy subtraction processing between a radiation image signal having a relatively low energy component and a radiation image signal having a relatively high energy component.

In this case, the image signal generating means sets the region including the energy near portion corresponding to the K absorption edge of the main component of the subject to a medium energy level, and a relatively low energy level and a relatively high energy level before and after the region. It is desirable that the counting be performed as an energy level.

In the above, the "main component of the subject" is the main portion of the subject to be separated by the energy subtraction process. For example, when a contrast agent such as iodine I or barium Ba is injected into the subject,
This is the part of the main subject where the contrast agent is deposited.

[0019]

According to the energy subtraction image forming apparatus of the present invention, X-rays are detected by using a sensor that generates a signal corresponding to the energy of the radiation particles, and different X-rays that enter the sensor within a unit time are detected. The number of radiation particles at a plurality of energy levels is counted for each pixel, and a radiation image signal of a different energy component is generated based on the counted amount of radiation particles at each energy level. Good image signals at each energy level can be obtained, whereby the subtraction image after the energy subtraction processing can be obtained as an image having no blur or artifact and having excellent energy separation.

Further, since the counting of the number of radiation particles corresponds to the so-called one-shot method, no blur occurs due to the movement of the subject between the two images.

Further, if an energy near portion corresponding to the K-absorption edge of the main component of the subject is subjected to energy discrimination as a boundary between high and low energy levels or a medium energy level, an image signal of each energy component is generated. By injecting a contrast agent into a subject and extracting a portion where the contrast agent is deposited by subtraction processing, an image having excellent energy discrimination performance can be obtained.

[0022]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an embodiment of an energy subtraction image forming apparatus according to the present invention will be described in detail with reference to the drawings. FIG. 1 is a schematic configuration diagram of an image forming apparatus according to an embodiment of the present invention.

As shown in FIG. 1, this image forming apparatus 1
Comprises a radiation image reading unit 2 and an energy subtraction processing unit 3. A CRT display device 4 is connected to the subtraction processing means 3, and a subtraction image is displayed on the CRT display device 4.

The radiation image reading unit 2 scans the X-ray L1
X-ray irradiating means 20 for irradiating X-ray particles (radiation particles) transmitted through the subject 5, and a sensor unit 30 formed in a linear shape with a sensor 31 for detecting X-ray particles (radiation particles). The image signal generating means 40 counts the number of energy level X-ray particles for each pixel, and generates radiation image signals of different energy components based on the counted amount of radiation particles for each energy level. X
The energy of the X-rays L1 emitted from the beam irradiation means 20 may have a certain width.

FIG. 2 is a diagram showing a scanning method when reading a radiation image. As shown in FIG. 2, the radiation image reading unit 2 has a configuration of a scanning type reading device.
That is, the sensor unit 30 is
(FIG. 1), it moves in the sub-scanning direction Y substantially perpendicular to the sensor unit 30. In synchronization with the movement of the sensor unit 30, the drive unit 22 causes the detection crystal 31a of the sensor 31 to move.
X-ray L1 with constant intensity only in the fan beam area
The position of the slit 24 provided at the tip of the arm 23 of the X-ray irradiation means 20 and the angle of the X-ray tube 21 are moved so as to irradiate the X-ray.

Many sensors constituting the sensor unit 30
Reference numeral 31 corresponds to a pixel, and the main scanning is performed by sequentially reading out a large number of sensors 31 arranged in a line.

In the image forming apparatus according to the present invention, it is needless to say that a two-dimensional area sensor may be used instead of the linear sensor as described above.

The slit 24 has an effect of reducing scattered radiation similarly to the grit used in the conventional film screen method.
It is provided so that an image with little fog can be obtained.

FIG. 3 is a block diagram showing details of the circuit configuration of the image forming apparatus 1. FIG. 3 shows one of the sensors 31 arranged in a line and constituting the sensor unit 30.
One is shown. The sensor 31 is a detection crystal 31 that generates a charge amount according to the energy of the incident X-ray particles.
a and an amplifier 31 for converting the voltage into a voltage corresponding to the generated electric charge
b. As the detection crystal 31a, it is desirable to use a semiconductor detector using a CdTe (cadmium telluride) crystal. The output signal S0 of the sensor 31 is input to the image signal generating means 40. Although not shown in FIG. 2, the output of each sensor 32 is input to a changeover switch, which is sequentially switched by the switch according to main scanning and is input to the image signal generating means 40.

The image signal generating means 40 includes a plurality of reference voltages V _REF1 and V _REF1 , each of which can register a plurality of different energy levels.
Reference voltage output means 41 for outputting V _REF2 , two comparators 42a and 42b for comparing the output signal S0 of the sensor 31 with the reference voltages V _REF1 and V _REF2 as thresholds, and the sensor 31 within a unit time.
Counters 43a, 43b for counting the number of X-ray particles having energy equal to or higher than the respective energy levels corresponding to the reference voltages V _REF1 , V _REF2 incident on
The signal processing means 44 generates radiation image signals having different energy components based on the count values a and 43b, that is, the amount of X-ray particles. Here, the “different plurality of energy levels” in this example are a noise level, an energy level higher and relatively lower than the noise level, and an energy level higher and higher than the noise level. The "energy component radiation image signal" is a relatively low energy component (low energy component) radiation image signal (low energy image data) DL
And relatively high energy components (high energy components)
Is a radiation image signal (high-energy image data) DH.

The subtraction processing means 4 performs an energy subtraction process between the low energy image data DL and the high energy image data DH. The reference voltage output means 41 is connected to a keyboard 45, and each of the reference voltages V _REF1 and V _REF2 can be changed by data input from the keyboard 45.

The operation of the image forming apparatus 1 having the above configuration will be described below.

While irradiating the subject 5 with the X-rays L1 from the X-ray tube 21, the sensor unit 30, the X-ray tube 21 and the slit 24 are moved in the sub-scanning direction Y.

The X-ray particles transmitted through the subject 5 enter each sensor 31 of the sensor unit 30. When one X-ray particle enters the detection crystal 31a, the detection crystal 31a generates a charge amount according to the energy of the X-ray particle. The generated electric charge is converted into one voltage pulse signal S0 having a peak corresponding to the energy by the amplifier 31b, and the pulse signal S0 is input to each of the comparators 42a and 42b of the image signal generating means 40.

The reference voltage output means 41 of the image signal generation means 40
Thus, as shown in FIG. 4, a reference voltage V _REF1 corresponding to the boundary ENL between the noise level EN and the low energy level EL and a reference voltage V _REF2 corresponding to the boundary ELH between the low energy level EL and the high energy level EH are output. And the reference voltage V
_REF1 is input to the comparator 42a, and the reference voltage V _REF2 is input to the comparator 42b.

The voltage pulse signals S0 input to the comparators 42a and 42b are compared with each other using the reference voltages as thresholds.
Pulses PL and PH are output from each comparator only when the signal S0 is larger than each reference voltage. Therefore, if the signal S0 has a voltage level as shown in FIG.
a and 42b output pulse signals PL and PH, respectively. If the voltage level is as shown by b, the comparator 42b outputs the pulse signal
No comparator outputs a pulse signal as long as it outputs PH and has a voltage level as shown by c. As a result, the X-ray particles incident on the sensor 31 within the unit time are changed to each level EN.
Energy discrimination is performed using L and ELH as thresholds. A multi-channel pulse-height analyzer (MPHA) is known as a measuring instrument for detecting the peak value of a pulse signal at different levels as described above. Means 4
1. Each of the comparators 42a and 42b and the keyboard 45 can be used in place of this multi-channel analyzer.

The output pulse signals PL and PH of each comparator are input to counters 43a and 43b, respectively. Each counter 43a, 43
b counts the pulse signals PL and PH, respectively, for a certain period of time. Thus, the number of X-ray particles at each energy level incident on the sensor 31 within a unit time is counted. Therefore, according to the image forming apparatus 1 of the present invention, the X-ray particles having an energy of a certain level or more incident on a certain time (unit time) are counted one by one, and the value is used as the value of each pixel. Naturally, as the X-ray absorption of the subject 5 increases,
This count value becomes smaller, and conversely, the smaller the X-ray absorption rate of the subject 5, the larger the value is output from each counter. This is performed for each output signal S0 of each sensor 31 corresponding to a pixel in the main scanning direction, and is performed while moving the sensor unit 30 and the like in the sub-scanning direction. The signals CL and CH are obtained.

The details of the method of obtaining X-ray images by capturing X-ray particles by the sensor 31 and counting the number of particles are described in “Medical Electronics and Biotechnology”, No. 32.
Volume: P311 and “INNERVISION (10 ・ 7)” 1995 P47-P
48 for details.

The image signals CL and CH are input to the signal processing means 44, where the signal processing is performed according to the following equation, whereby the energy components higher than the low energy level EL and lower than the high energy level EH are obtained. , And high-energy image data DH, which is an image signal consisting only of energy components higher than the high energy level EH, is obtained.

DL = CL-CH DH = CH In the conventional subtraction processing, first, relative positioning of each X-ray image carried by each of two input image signals is generally performed on the image signals. (For example, see Japanese Patent Application Laid-Open No. 58-163338), the image reading unit 2 according to the present invention employs both image data DL,
Since DH is obtained at the same time (corresponding to one shot), and the corresponding pixel positions do not shift, the alignment process is not required.

The subtraction processing means 3 performs subtraction processing based on the input low energy image data DL and high energy image data DH. That is, the two image data DL and DH are signals containing information on both the soft part and the bone part, although the density of the soft part and the bone part are different from each other. By calculating D1 = A.DL-B.DH + C (where A, B and C are coefficients),
The bone image data (subtraction image signal) D1 representing the bone image from which the shadow of the soft tissue of the subject 5 is eliminated and only the shadow of the bone is extracted is obtained. By changing the coefficients A, B, and C in the above equation, it is also possible to obtain soft part image data D2 representing a soft part image in which the shadow of the bone tissue is eliminated and only the shadow of the soft part is extracted.

The soft part image data thus obtained
The D1 or bone image data D2 is input to the CRT display device 4, and a soft image based on the soft image data D1 or a bone image based on the bone image data D2 is displayed on the CRT as a visible image.

Further, by injecting a contrast agent such as iodine I or barium Ba into the subject 5, it is possible to obtain an image in which the main part of the subject on which the contrast agent has been deposited by the subtraction processing is extracted.

In this case, as shown in FIG. 5, when the value of the radiation energy is increased, the X-ray absorption coefficient of the contrast agent sharply increases near the energy corresponding to the K absorption edge wavelength of the contrast agent. Except for the vicinity of the K absorption edge, it exhibits a characteristic of gradually decreasing. Therefore, when the reference voltage V _REF2 is set so that the energy corresponding to the vicinity of the K absorption edge becomes a boundary value between the low energy component and the high energy component, the subject portion where the contrast agent is deposited and the subject portion where the contrast agent is not deposited are set. Parts can be separated. For example, in FIG. 5, if energy discrimination is performed using the energy level ELH near the energy corresponding to the K absorption edge wavelength of the contrast agent as its boundary value, the energy level indicated by a corresponding to the living body part absorbing the contrast agent And the energy level indicated by b corresponding to the part of the living body that does not absorb the contrast agent. Therefore, if the energy discrimination is performed such that the boundary between the two energy levels is near the K absorption edge (preferably at the center as much as possible), the contrast agent is deposited by setting the coefficient to an appropriate value in the subtraction processing. It is possible to obtain an image of the subject portion or the image of the subject portion where no contrast agent is deposited.
The energy corresponding to the wavelength at the K absorption edge is about 33 KeV when the contrast agent is iodine, and is about 37 KeV when barium is used.

In the above description, the case of separating into the noise level component, the low level component, and the high level component has been described.
Further, by separating the middle level component which is intermediate between the low level component and the high level component, the energy discrimination performance is further improved. Hereinafter, this aspect will be described.

FIG. 6 is a block diagram showing details of the circuit configuration of the image forming apparatus 7 of this embodiment. This device 7 allows the image signal forming means 46 to extract a medium level component.
Comparator 42c and counter 43c for counting the output pulse PC
Are different from the image signal forming means 40 in that the image signal forming means 40 is further provided.

The reference voltage output means 41 of the image signal generation means 40
7, the reference voltage V _REF1 corresponding to the boundary ENL between the noise level EN and the low energy level EL, and the boundary ELH between the low energy level EL and the middle energy level EC, as shown in FIG.
Reference voltage V _REF3 and medium energy level EC corresponding to
And the reference voltage V _REF3 corresponding to the boundary ECH between the high energy level EH and the reference voltage V _REF1 are output from the comparator 42a.
The reference voltage V _REF2 is input to the comparator 42b, and the reference voltage V _REF3 is input to the comparator 42c.

If the voltage pulse signal S0 inputted to each of the comparators 42a, 42b and 42c has a voltage level as shown in FIG. 7A, all the comparators 42a, 42b and 42c respectively output the pulse signals.
The comparators 42b and 42c output the pulse signals PH and PC if they have the voltage level as shown by b, and the comparators if they have the voltage level as shown by c. 42
If b is a pulse signal PH and has a voltage level as indicated by d, none of the comparators outputs a pulse signal.

Each of the counters 43a, 43b, 43c, to which the output pulse signals PL, PH, PC of each comparator are inputted, counts the pulse signals PL, PH, PC for a predetermined time. As a result, the number of X-ray particles having an energy level equal to or higher than ENL, equal to or higher than level ELC, and equal to or higher than level ECH in the unit time is counted, and the image signal CL composed of components equal to or higher than the respective energy levels incident in the unit time. , CH, CC are obtained.

Each of the image signals CL, CH and CC is processed by signal processing means 44.
Where low-energy image data DL consisting only of low-energy components whose energy level is higher than ENL and lower than ELC, and whose energy level is higher than ELC, by performing signal processing according to the following equation: Medium energy image data DC consisting only of medium energy components lower than ECH and high energy image data DH being image signals consisting only of energy components having an energy level higher than EH are obtained.

DL = CL-CC DC = CC-CH DH = CH If the medium energy image data DC is also extracted in this way, the low energy image data DL and the high energy image data DH can be more clearly distinguished. It is possible to do.

The low energy image data DL and the high energy image data DH thus discriminated are input to the subtraction processing means 3 (FIG. 3), and
A subtraction process is performed in the same manner as described above.

Also in this example, iodine I
Alternatively, by injecting a contrast agent such as barium Ba,
It is possible to obtain an image in which the contrast agent has been extracted.

In this case, if a portion near the energy corresponding to the K-absorption edge of the main component of the subject is subjected to energy discrimination with the medium energy level to generate an image signal of each energy component, the contrast agent is injected into the subject. Then, by extracting the portion where the contrast agent is deposited by the subtraction processing, an image having more excellent energy discrimination performance can be obtained.

For example, in FIG. 8, the boundary of each energy component is near the energy corresponding to the K absorption edge wavelength of the contrast agent, and the energy level ELC of the X-ray is the boundary between the low level component and the medium level component, and the level ECH. If energy discrimination is performed so that the boundary between the medium level component and the high level component is obtained, the energy level indicated by a corresponding to the living body part absorbing the contrast agent and the energy level b corresponding to the living body part not absorbing the contrast agent Can be clearly distinguished from the energy level indicated by the above, whereby the separability between the subject portion where the contrast agent is deposited and the subject portion where the contrast agent is not deposited is superior to that of the image forming apparatus 1. Images can be obtained.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram showing an embodiment of an energy subtraction image forming apparatus according to the present invention.

FIG. 2 is a diagram showing a scanning method at the time of image reading of the image forming apparatus.

FIG. 3 is a block diagram illustrating details of a circuit configuration of the image forming apparatus.

FIG. 4 is a diagram showing a relationship between a voltage pulse signal detected by a sensor and a reference voltage.

FIG. 5 is a diagram showing an example of setting an X-ray absorption coefficient characteristic of a contrast agent and an energy discrimination region.

FIG. 6 is a block diagram illustrating details of a circuit configuration of an image forming apparatus according to another embodiment of the present invention.

FIG. 7 is a diagram showing a relationship between a voltage pulse signal detected by a sensor and a reference voltage.

FIG. 8 is a diagram illustrating an example of setting of an X-ray absorption coefficient characteristic of a contrast agent and an energy discrimination region in the image forming apparatus according to another embodiment.

[Explanation of symbols]

 1, 7 energy subtraction image forming apparatus 2 radiation image reading unit 3 energy subtraction processing means 4 CRT display device 5 subject 20 X-ray irradiation means 30 sensor unit 31 sensor 40, 46 image signal generating means

Claims (5)

[Claims] 1. A sensor for generating a signal corresponding to the energy of radiation particles, and the number of radiation particles having a plurality of different energy levels incident on the sensor in a unit time is counted for each pixel. Image signal generating means for generating radiation image signals of different energy components based on the amount of radiation particles for each energy level; and energy subtraction processing means for performing energy subtraction processing between the generated radiation image signals of different energy components. An energy subtraction image forming apparatus comprising: 2. The image signal generating means counts the plurality of energy levels as a noise level, an energy level higher and lower than the noise level, and an energy level higher and higher than the noise level. And generating a radiation image signal of a relatively low energy component and a radiation image signal of a relatively high energy component, wherein the energy subtraction processing means is configured to generate the radiation image signal of the relatively low energy component and the radiation image signal of the relatively low energy component. 2. The energy subtraction image forming apparatus according to claim 1, wherein the energy subtraction processing is performed with a radiation image signal having a high energy component. 3. The image signal generating means performs the counting using energy near the energy corresponding to the K absorption edge of the main component of the subject as a boundary value between the relatively low energy level and the relatively high energy level. The energy subtraction image forming apparatus according to claim 2, wherein 4. The image signal generating means sets the plurality of energy levels as a noise level, an energy level higher and relatively lower than the noise level, an energy level higher and higher than the noise level, and the noise level. Performing the counting as a higher and medium energy level to generate a radiation image signal having a relatively low energy component and a radiation image signal having a relatively high energy component, wherein the energy subtraction processing means is configured to perform the comparison. The energy subtraction image forming apparatus according to claim 1, wherein the energy subtraction processing is performed between a radiation image signal having a relatively low energy component and the radiation image signal having a relatively high energy component. 5. The image signal generating means sets an area including a portion near energy corresponding to a K absorption edge of a main component of the subject to the medium energy level, and sets the relatively low energy level before and after the medium energy level. 5. The energy subtraction image forming apparatus according to claim 4, wherein the counting is performed as the relatively high energy level.