JP2009207048A - Radiograph signal detection system and method thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To shorten a time for reading a radiograph in mammography for obtaining the radiograph while the breasts of a subject are held and pressed between pressing plates. <P>SOLUTION: When reading the charge signal of a region-of-interest (high-definition read area HRA) of a radiograph recorded by a radiation transformer 10, a signal detector is operated in a high-definition read mode and when reading a charge signal in a region-of-no-interest (high-speed read area HSA), the signal detector is operated in a high-speed read mode, thereby greatly shortening the time for image reading in comparison with a case where the signal detector is operated in the high-definition read mode for all the image regions (region-of-interest + region-of-no-interest). <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a radiation converter that records radiation images by receiving radiation emitted from a radiation source and passed through a subject, and an image signal corresponding to the radiation image based on a charge signal output from the radiation converter. The present invention relates to a radiation image signal detection system and a method thereof.

  Conventionally, a photoelectric conversion element such as a CCD (charge coupled device) sensor or a photomultiplier, or a radiation converter that accumulates charges generated by irradiation of radiation and outputs a charge signal corresponding to the accumulated charges Devices that convert light or radiation into charge signals and output them are used in various fields.

  As an electronic component for detecting the charge signal output from the photoelectric conversion element or the radiation converter as described above, an integration amplifier that can be integrated into an IC and has relatively low noise is generally used. The integration amplifier starts integration of the charge signal by switching to the integration mode, outputs an electric signal (integration voltage) corresponding to the integrated charge amount, and integrates the charge by switching to the reset mode. It has a function of discharging the signal (integrated voltage) and returning it to the initial state.

  Here, the integration amplifier is switched to the integration mode by switching a reset switch made of a semiconductor element connected in parallel to the capacitor of the integration amplifier from an on state to an off state. As a result, kTC noise of the reset switch is generated, and this noise is included in the electrical signal of the signal component.

  Therefore, in order to avoid the influence of the kTC noise, a correlated double sampling process is performed. Correlated double sampling processing refers to an electrical signal {reference level held when a predetermined reference track period (also referred to as a reference level sampling period or a base level sampling period) elapses after the integration amplifier switches to the integration mode. (Noise)} and an electric signal (reference level + signal level) held when a signal track period (also referred to as a signal level sampling period) including an integration period in the reference track period has elapsed, This is a process for eliminating the influence of the kTC noise by using the difference as a signal component (signal level).

  Further, in the signal detector using the integration amplifier as described above, a low-pass filter is provided after the integration amplifier in order to reduce noise due to thermal noise generated in the signal line connected to the input terminal of the integration amplifier. The electric signal output from the integrating amplifier passes through this low-pass filter and is output.

  Here, for example, when the line resistance of the signal line connected to the integrating amplifier is large, for example, several hundreds [kΩ], noise due to thermal noise generated in the signal line is large in the line resistance of the signal line. Since it increases in response to this, the charge signal flowing in the signal line becomes very large. Therefore, in order to sufficiently reduce the noise as described above, it is necessary to increase the time constant of the low-pass filter in accordance with the magnitude of the noise.

  In Patent Document 1, in a signal detector that performs correlated double sampling processing, integration is performed by setting the reference track period after resetting the integration amplifier to be 10 times (10τ) or more of the time constant τ of the low-pass filter after the integration amplifier. Noise generated when the amplifier is reset is reduced to obtain a high-quality image.

  In Patent Document 2, in order to shorten 10τ which is a noise reduction period generated when the integration amplifier is reset, the period of the low-pass filter time constant τ is extended to the first half of the reference track period immediately after the integration amplifier reset, and the kTC noise waveform is generated. By converging in a short time, the period of the low-pass filter time constant 10τ in the latter half of the reference track period is shortened, and the reference track period is devised to be shorter than the period of Patent Document 1.

JP 2005-269215 A JP 2006-229581 A

  By the way, it is possible to shorten the reading time of the image from the radiation converter by making it possible to obtain an image at a high speed, but mammography for obtaining a radiation image in a state where the breast of the subject is sandwiched and compressed. Then, even if this reading time is short, there exists a subject of giving a pain to a subject.

  Therefore, it is beneficial (good news) for the subject to shorten the time of pressing.

  The present invention has been made in consideration of the above-mentioned Patent Documents 1 and 2 and the above-described problems, and a radiation image signal detection system and method for further shortening the reading time of an image from a radiation converter. The purpose is to provide.

  A radiation image signal detection system according to the present invention records a radiation converter that records radiation that has passed through a subject as a radiation image, and outputs an image signal corresponding to the radiation image based on a charge signal output from the radiation converter. A signal detector, a region boundary classifier that classifies a boundary between a region of interest and a non-region of interest of the radiographic image recorded in the radiation converter, and the signal detector is partitioned by the region boundary classifier A controller that operates in a high-definition reading mode when detecting a charge signal of the region of interest, and a controller that operates in a high-speed reading mode when detecting a charge signal of the non-region of interest. .

  In the radiological image signal detection method according to the present invention, when a signal detector outputs an image signal corresponding to the radiological image based on a charge signal output from a radiation converter that records radiation passing through a subject as a radiographic image. And dividing the boundary between the region of interest and the non-region of interest of the radiographic image recorded in the radiation converter, and detecting the charge signal of the segmented region of interest in the high-definition reading mode. When operating and detecting a charge signal in the non-interesting region, the high-speed reading mode is used.

  According to the present invention, the signal detector is operated in the high-definition reading mode when reading the charge signal of the region of interest of the radiographic image recorded in the radiation converter, and the signal is read when reading the charge signal of the non-region of interest. By operating the detector in the high-speed reading mode, it is possible to significantly shorten the image reading time compared to operating the entire image region (region of interest + non-region of interest) in the high-definition reading mode.

  A radiation image signal detection system that implements the radiation image signal detection method according to the present invention will be described below.

  FIG. 1 shows a schematic configuration diagram of a radiation image signal detection system 50 according to this embodiment.

  As shown in FIG. 1, the radiation image signal detection system 50 records a radiation image by receiving a radiation source (not shown) and radiation irradiated from the radiation source and passed through the subject, and according to the radiation image. A radiation converter 10 (also referred to as a radiation image recording apparatus) that outputs a charge signal i, a reading light source unit 20 that scans the radiation converter 10 with linear reading light, and scanning of the reading light by the reading light source unit 20. And a signal detector 30 that outputs a digital image signal corresponding to the radiation image based on the charge signal i output from the radiation converter 10.

  The signal detector 30 integrates the charge signal i output from the radiation converter 10, and the first and second holding circuits 31 that hold (track hold) the electric signal integrated by the integration amplifier 33. , 32, and the difference (signal level) between the first electric signal (reference level) and the second electric signal (reference level + signal level) held in the first and second holding circuits 31, 32, respectively. A differential amplifier 34 and an A / D converter 35 that converts the analog signal output from the differential amplifier 34 into a digital signal are provided, and a correlated double sampling process is performed based on the charge signal i output from the radiation converter 10. Is to do.

  The integrating amplifier 33 includes a capacitor 33a for accumulating the charge signal i output from the radiation converter 10 and a reset switch S3 for discharging the charge signal i accumulated in the capacitor 33a.

  The first holding circuit 31 includes a resistance element R1, a switch S1 connected in parallel to the resistance element R1, a switch S4 and a capacitor C1 connected in series to the resistance element R1, and is output from the integrating amplifier 33. The electric signal is held in the capacitor C1.

  The second holding circuit 32 includes a resistance element R2, a switch S2 connected in parallel to the resistance element R2, a switch S5 and a capacitor C2 connected in series to the resistance element R2, and is output from the integrating amplifier 33. The electric signal is held in the capacitor C2.

  The first and second holding circuits 31 and 32 hold the electric signal output from the integrating amplifier 33 at different timings, and perform low-pass filter processing on the electric signal output from the integrating amplifier 33. It is also something to apply.

  In the signal detector 30 of the present embodiment, the reading speed is changed between the region of interest and the non-region of interest of the radiation image detected by the radiation converter 10. In the region of interest, a relatively long reading time is set at a high-definition reading speed (high-definition reading mode), and in the non-interesting region, a relatively short reading speed is set at a high-speed reading speed (high-speed reading mode).

  Specifically, in the first holding circuit 31 that holds the reference level, when the electric signal is accumulated in the capacitor C1, the switch S1 and the switch 4 are turned on in both the high-definition reading mode and the high-speed reading mode. Therefore, a first-order first low-pass filter is formed by the on-resistance of the resistor element R1, the switch S1, the on-resistance of the switch S4, and the capacitor C1 (time constant τ1 as will be described later). The processed filtered electric signal is held in the capacitor C1.

  On the other hand, in the second holding circuit 32 that holds both the reference level and the signal level, when the electric signal is stored in the capacitor C2, the switch S5 is turned on in both the high-definition reading mode and the high-speed reading mode, and the switch S2 Is turned off in the high-definition reading mode and turned on in the high-speed reading mode. Therefore, in the high-definition reading mode, the first-order second low-pass filter is configured by the resistance element R2, the on-resistance of the switch S5, and the capacitor C2 (time constant τ2, τ2> τ1 as described later), and in the high-speed reading mode. The resistance element R2, the on-resistance of the switch S2, the on-resistance of the switch S5, and the capacitor C2 constitute a first-order second low-pass filter (time constants τ2 ′, τ2′≈τ1 as described later).

(A) Further detailed description of the configuration when operating in the high-definition reading mode in the region of interest of the radiographic image In the region of interest in the radiographic image, the operation is performed in the high-definition reading mode in order to obtain a high-definition image signal with little noise. Is preferred.

  When operating in this high-definition reading mode, as shown in Patent Document 2, the time constant τ1 of the first low-pass filter circuit and the time constant τ2 of the second low-pass filter circuit are τ1 <τ2. It is set to a value that satisfies

When the resistance value of the resistance element R1 is r1, the resistance value of the on-resistance of the switch S1 is RON1, the resistance value of the on-resistance of the switch S4 is RON4, and the capacitance value of the capacitor C1 is c1, the time constant τ1 of the first low-pass filter is Calculated by equation (1).
τ1 = (RON1 × r1 / (RON1 + r1) + RON4) × c1 (1)

Therefore, if RON1 << r1, τ1 = (RON1 + RON4) × c1
It becomes.

When the resistance value of the resistance element R2 is r2, the resistance value of the on-resistance of the switch S5 is RON5, and the capacitance value of the capacitor C2 is c2, the time constant τ2 of the second low-pass filter is calculated by the following equation (2). .
τ2 = (r2 + RON5) × c2 (2)

  Therefore, if RON5 << r2, then τ2 = r2 × c2.

  Since RON1 and RON5 are several Ω to several hundred Ω, when r1 and r2 are set to values of several hundred kΩ to several tens of MΩ, τ1 <τ2 can be set as described above. it can. For example, when the line resistance of the linear electrode 15a that is a signal line is about several hundreds kΩ, it is desirable that the time constant τ2 of the second low-pass filter is several hundreds μs. Therefore, c1 and c2 are set to values of several tens of pF to several thousand pF with respect to the resistance values of r1 and r2. For example, r1 = r2 = 10 MΩ and c1 = c2 = 10 pF. By setting as described above, a sufficient length can be secured as the time constant τ2, and the time constant τ1 can be shortened. In order to achieve both high speed and low noise in the high-definition reading mode, it is desirable to set τ1 = 1 to 10 μs and τ2 = 10 to 100 μs (τ2 = 10τ1). The above is the description of (A).

(B) More detailed description of the configuration in the case of reading in the high-speed reading mode in the non-interest area of the radiographic image In the non-interest area of the radiographic image, priority is given to shortening of the reading time, and even in the presence of some noise, high speed is achieved in a short time. In order to obtain an image signal, it is operated in a high-speed reading mode (high-speed reading speed).

  In the case of operating in the high-speed reading mode, the second low-pass filter is set to a value in which the signal level (precisely, the reference level + signal level) held in the capacitor C2 of the second holding circuit 32 converges in the shortest response time. The time constant τ2 ′ is set.

In this embodiment, the time constant τ2 ′ of the non-interesting region (reading region at a high reading speed) of the second low-pass filter is set by turning on the switch S2. The time constant τ2 ′ is calculated by the following equation (3).
τ2 ′ = (RON2 × r2 / (RON2 + r2) + RON5) × c2 (3)

Therefore, if RON2 << r2, τ2 ′ = (RON2 + RON5) × c2
It becomes.

  For example, as described above, when r1 = r2 = 10 MΩ and c1 = c2 = 10 pF are set, τ2′≈τ1. Hereinafter, in order to facilitate understanding, the time constant τ2 ′ in the high-speed reading mode is described as τ2 ′ = τ1. The above is the description of (B).

  The signal detector 30 further includes buffer amplifiers 36 and 37 that output the filtered electrical signals output from the first and second holding circuits 31 and 32 to the differential amplifier 34, and a reset switch S3 of the integration amplifier 33. And a controller 38 for controlling the operation timing of the switches S1, S2, S4, S5 and the A / D converter 35 of the first and second holding circuits 31, 32. The controller 38 includes a timing generator, CPU, ROM, RAM, and other interfaces. The CPU operates as various functional units by executing programs stored in the ROM, and controls the timing generator. .

  In detail, as shown in FIG. 2, the radiation converter 10 is charged by receiving a first electrode layer 11 that transmits radiation carrying a radiation image, and radiation irradiated through the first electrode layer 11. The photoconductive layer 12 for recording, which acts as an insulator for charges generated in the photoconductive layer 12 for recording, and charge transport which acts as a conductor for transport charges of the opposite polarity to the charges The layer 13, the photoconductive layer 14 for reading that generates charges when irradiated with the reading light, and the second electrode layer 15 in which linear electrodes 15 a that extend linearly that transmit the reading light are arranged in parallel are formed They are laminated in order. A power storage unit 16 is formed at the interface between the recording photoconductive layer 12 and the charge transport layer 13 in which charges generated according to the radiation dose are stored.

  In FIG. 1, only the signal detector 30 connected to one linear electrode 15a of the radiation converter 10 is shown, and the signal detectors 30 connected to the other linear electrodes 15a are not shown. It is.

  Further, the A / D converter 35 may be provided for each linear electrode 15a, or a multiplexer is provided to switch the analog signal output from the differential amplifier 34 for each linear electrode to provide one A / D. You may make it input into the converter 35. FIG.

  Further, the radiation converter 10 and the reading light source unit 20 include the length direction (main scanning direction X) of the reading light source of the reading light source unit 20 and the length direction (sub-scanning direction Y) of the linear electrode 15a of the radiation converter 10. And are installed so as to be substantially orthogonal. The reading light source unit 20 scans the reading light by moving the linear reading light source in the length direction (sub-scanning direction Y) of the linear electrode 15a. Is omitted in FIG.

  Next, the operation of the radiation image signal detection system 50 will be described.

  As shown in FIG. 3A, first, in a state where a voltage is applied so that the first electrode layer 11 of the radiation converter 10 is negatively charged and the second electrode layer 15 is positively charged, the radiation source is applied to the subject. Radiation L1 is irradiated toward 40.

  The radiation L1 emitted from the radiation source is applied to the entire subject 40, and the radiation that has passed through the transmission part 40a that transmits the radiation in the subject 40 is applied from the first electrode layer 11 side of the radiation converter 10. Note that the radiation irradiated to the blocking unit 40b that does not transmit radiation in the subject 40 is not irradiated to the radiation converter 10.

  The radiation L <b> 1 irradiated to the radiation converter 10 passes through the first electrode layer 11 and is irradiated to the recording photoconductive layer 12. Then, a charge pair is generated by irradiation of radiation in the recording photoconductive layer 12, and the positive charge is combined with the negative charge charged in the first electrode layer 11 and disappears, and the negative charge is a latent image charge. As a result, the radiation image is recorded in the power storage unit 16 formed at the interface between the recording photoconductive layer 12 and the charge transport layer 13.

  Then, as shown in FIG. 3B, in the state where the first electrode layer 11 is grounded, the reading light L2 is irradiated from the second electrode layer 15 side, and the reading light L2 passes through the linear electrode 15a. Then, the reading photoconductive layer 14 is irradiated. The positive charge generated in the reading photoconductive layer 14 by irradiation with the reading light L2 is combined with the latent image charge in the power storage unit 16, and the negative charge is charged on the linear electrode 15a of the second electrode layer 15. Combined with the charge.

  On the other hand, the reset switch S3 of the integrating amplifier 33 in the signal detector 30 is turned on before the radiation converter 10 is irradiated with the reading light. Thereafter, the reset switch S3 is turned off, and reading light irradiation is started after the baseline sampling is completed. Then, the negative charge generated in the reading photoconductive layer 14 of the radiation converter 10 as described above is combined with the positive charge charged on the linear electrode 15a of the second electrode layer 15, whereby the coupling is achieved. A charge signal i having a magnitude corresponding to the amount of charge thus obtained is accumulated in the capacitor 33a of the integrating amplifier 33 and integrated.

(C) Operation when Reading Radiation Image in Region of Interest in High-Definition Reading Mode Here, the operation when reading in the region of interest from the start of integration in the high-definition reading mode will be described using the time chart shown in FIG. To do. FIG. 4 shows a schematic diagram of ON / OFF control timings of the switches S3, S1, S2, S4, and S5, voltage waveforms at V1, V2, and V3 in FIG.

  First, the integration of the charge signal i is started in the integration amplifier 33 from the time t1 shown in FIG. 4 when the reset switch S3 of the integration amplifier 33 is turned off.

  At the integration start time t1 in the integrating amplifier 33, the switches S1 and S4 of the first holding circuit 31 and the switches S2 and S5 of the second holding circuit 32 are in the on state as shown in FIG. The filtered electric signal integrated by the integrating amplifier 33 and subjected to the low-pass filter processing by the first low-pass filter is accumulated in the capacitor C1 of the first holding circuit 31.

  Further, in the second holding circuit 32, a second low-pass filter circuit is configured by the resistor element R2, the on-resistance of the switch S2, the on-resistance of the switch S5, and the capacitor C2, and is integrated by the integrating amplifier 33 to be integrated with the second The filtered electric signal subjected to the low-pass filter processing by the low-pass filter circuit is accumulated in the capacitor C2 of the second holding circuit 32.

  Between the time points t1 and t2, the time constant τ2 ′ of the second low-pass filter and the time constant τ1 of the first low-pass filter are set to the same magnitude (τ2 ′ = τ1). With this setting, the reference level related to the kTC noise waveform is a period (time points t1 to t1) corresponding to the time constant τ1, as shown in the rising region 100 represented by the dotted line V1 representing V1 and V2 in FIG. It converges in a short time at t2). Therefore, the switches S1 and S2 are turned off at time t2.

  Thereafter, the switch S4 is turned off at time t3, and the reference track period (period from time t1 to time t3) is ended.

  Therefore, during the reference track period (t1 to t3), V1 = V2 and V3 (V2-V1) = 0. In this way, the time constant τ2 ′ of the second low-pass filter is changed to the time constant of the first low-pass filter until V1 and V2 are converged in the first half period (t1-t2) of the reference track period (t1-t3). By adopting a short time constant equal to τ1 (τ2 ′ = τ1), the reference track period (t1 to t3) can be shortened.

  After integration by the integrating amplifier 33 is started at time t1, at time t3 when the reference track period (t1 to t3) has elapsed, the switch S4 of the first holding circuit 31 is turned off and stored in the capacitor C1. The first filtered electrical signal is held in capacitor C1.

  Then, as described above, at the time t2 going back to the time t3 immediately after the first filtered electric signal is held by the capacitor C1, the switch S1 of the first holding circuit 31 and the switch S2 of the second holding circuit 32 Is turned off. In other words, the electric signal output from the integrating amplifier 33 after the time point t2 is subjected to the low-pass filter processing by the second low-pass filter circuit, and the second filtered electric signal is supplied to the second holding circuit 32. Accumulated in capacitor C2.

  Then, after the integration period (time points t3 to t4) has elapsed during the signal track period (time points t1 to t4), the switch S3 is turned on at time point t4 to reset the integration amplifier 33. At a time point immediately before the reset time point t4 (the symbol is t4), the switch S5 of the second holding circuit 32 is turned off, and the second filtered electric signal accumulated in the capacitor C2 is held. (Hold) At the time point t4, the switch S3 of the integrating amplifier 33 is turned on, the resetting of the integrating amplifier 33 is started, and the switches S1 and S2 are turned on again.

  In practice, in the integration period, the time obtained by multiplying the time constant τ2 by 10 is added to the output time (detector output time) of the charge signal i (pulse signal and input signal of the integration amplifier 33) from the radiation converter 10. (Integration period = detector output time + 10 × τ2 = detector output time + 10 × 10τ1).

  The first filtered electric signal V1 held in the capacitor 32c of the first holding circuit 31 and the second filtered electric signal V2 held in the second holding circuit 32 as described above are Are output to the differential amplifier 34 through buffer amplifiers 36 and 37, respectively. The difference amplifier 34 calculates the difference (V2−V1) between the two filtered electrical signals V1 and V2 and outputs the difference to the A / D converter 35. The A / D converter 35 digitally converts the difference signal, which is an input analog image signal, during a period between time points t4 and t5 shown in FIG. 4 and outputs a digital image signal.

  At the time point t5 when the A / D conversion is completed as described above, the switch S4 of the first holding circuit 31 and the switch S5 of the second holding circuit 32 are turned on again. The switch S3 is turned off, and integration in the integrating amplifier 33 is started again.

  For each irradiation of one line in the main scanning direction X of the reading light source unit 20, the processing from the start of integration as described above to the output of the digital image signal is performed for each signal detector 30 connected to each linear electrode 15a. Thus, detection of a digital image signal for one line is performed. The reading light source unit 20 detects the digital image signals for one line in synchronization with the scanning of the linear reading light in the direction of the arrow Y in FIG. 1, and finally the radiation converter 10. The digital image signal for the entire surface is detected.

  According to the radiation image signal detection system 50 described above, when reading in the high-definition reading mode in the region of interest of the radiation image, the time constant τ1 of the low-pass filter processing in the first half of the reference track period (baseline sampling), and the signal track period Since the time constant τ2 of the low-pass filter processing is set to a value that satisfies τ1 <τ2, the kTC noise waveform is converged in a short time in this small time constant τ1 period by shortening τ1, thereby speeding up the signal detection. In addition, the noise in the signal component can be sufficiently reduced by setting the time constant of τ2 to an appropriate length. That is, the S / N of the signal after the correlated double sampling can be improved while shortening the reading time. The above is the description of (C).

  As shown in FIG. 5, when the subject 40M is a breast or the like, the subject 40M is divided into a region where the subject 40M exists and a region where the subject 40M does not exist on the sub-scanning direction Y of the radiation converter 10.

  In this case, a region where the subject 40M exists is a region of interest, and a region where the subject 40M does not exist is a non-region of interest. Here, radiation image region = region of interest + non-region of interest. In the non-interest region, it is not necessary to consider much noise as an image signal, and it is necessary to optimize (minimize) the noise in the region of interest.

  Therefore, in this embodiment, the region of interest in which the subject 40M exists in the sub-scanning direction Y is set as the high-definition reading region HRA that reads in the above-described high-definition reading mode, and the non-interesting region in which the subject 40M does not exist is described later. The high-speed reading area HSA for reading in the reading mode is set.

  The controller 38 that also functions as a region boundary classifier that separates the boundary between the region of interest and the non-region of interest automatically discriminates the image edge in the regions (edge regions) Ea and Eb where the edges of the image exist. In the area Ea, the high-speed reading mode is switched to the above-described high-definition reading mode, and in the edge Eb, the high-definition reading mode is switched to a high-speed reading mode described later.

  Automatic discrimination (automatic segmentation) of the image edge can be performed based on a change in the difference signal V3 of the charge signal i being read. For example, the difference signal V3 of each pixel of the reading line adjacent in the main scanning direction X in the sub-scanning direction Y is compared, or the average value of the difference signal V3 of the reading line adjacent in the main scanning direction X is compared. be able to.

  It is also possible for a radiographer or the like to visually set the size of the subject 40M and manually set an area slightly larger than the visually checked size in the high-definition reading area HRA using a keyboard or a display (not shown).

  Alternatively, a high-definition area HRA set at the time of the previous image reading of the same subject 40M or an area having a little margin can be set in advance using a keyboard or a display. In this case, instead of setting the entire subject 40M shown in FIG. 5 as the high-definition reading area HRA, a part of the area can be set in advance as a region of interest (high-definition reading area).

(D) Action when Reading Radiation Image in Non-Interest Area in High-Speed Reading Mode Here, the action when reading in the non-interest area from the start of integration in the high-speed reading mode is described with reference to the time chart shown in FIG. explain. FIG. 6 shows a schematic diagram of ON / OFF control timings of the switches S3, S1, S2, S4, and S5 and voltage waveforms at V1, V2, and V3 in FIG.

  The operation of the reference track period from time t11 to t13 is the same as that from time t1 to t3 in the high-definition reading mode described above with reference to FIG.

  That is, at the time t13 when the reference track period (t11 to t13) has elapsed after the integration by the integrating amplifier 33 is started, the switch S4 of the first holding circuit 31 is turned off and the first accumulated in the capacitor C1. One filtered electrical signal is retained.

  From this holding time t13, the switch S1 of the first holding circuit 31 and the switch S2 of the second holding circuit 32 are turned on.

  The electric signal output from the integrating amplifier 33 from this time t13 is subjected to low-pass filter processing by the second low-pass filter circuit, and the second filtered electric signal is applied to the capacitor C2 of the second holding circuit 32. However, since the time constant τ2 of the second low-pass filter is set to a short period of τ2 ′ = τ1, the kTC generated by the charge signal i as shown in the rising region 102 surrounded by the dotted line of V2 The signal level related to the noise waveform converges in a short time in a period corresponding to the time constant τ1 (= τ2 ′).

  Then, during the signal track period (t11 to t14), at the time t14 immediately before the integration amplifier 33 is reset (switch S3 is turned on) after the integration period (time t13 to t14) has elapsed, the second holding circuit 32 Switch S5 is turned off, and the second filtered electrical signal stored in capacitor C2 is held (held). Immediately after time t14, the switch S3 of the integrating amplifier 33 is turned on, and the integrating amplifier 33 is reset.

  In practice, in the integration period (time points t13 to t14), a time constant τ1 (τ2 ′) × 10 is added to the output time (detector output time) of the charge signal i (pulse signal) from the radiation converter 10. Time.

  As described above, the first filtered electric signal V1 held in the capacitor 32c of the first holding circuit 31 and the second filtered electric signal V2 held in the second holding circuit 32 are respectively The signal is output to the differential amplifier 34 via the buffer amplifiers 36 and 37. Then, the difference amplifier 34 calculates the difference (V2−V1) between the two filtered electrical signals V1 and V2 and outputs the difference to the A / D converter 35. The A / D converter 35 digitally converts the difference signal, which is an input analog image signal, during a period between time points t14 to t15 illustrated in FIG. 6 and outputs a digital image signal.

  At the time t15 when the A / D conversion is completed as described above, the switch S4 of the first holding circuit 31 and the switch S5 of the second holding circuit 32 are turned on again, and the switch of the integrating amplifier is turned on at the time t15. S3 is turned off, and integration in the integrating amplifier 33 is started again from time t15.

  In this way, in each line in the main scanning direction of the high-speed reading area HSA, the processing from the start of integration to the output of the digital image signal as described above for each irradiation of one line in the main scanning direction X of the reading light source unit 20. However, it is performed for each signal detector 30 connected to each linear electrode 15a to detect a digital image signal for one line. The above is the description of (D).

  As described above, according to the embodiment described above, the radiation converter 10 records radiation emitted from a radiation source (not shown) and passing through the subject 40M as a radiation image. The signal detector 30 outputs a digital image signal corresponding to the radiation image based on the charge signal i output from the radiation converter 10. The controller 38, which also operates as a region boundary classifier, classifies the boundary between the region of interest and the non-region of interest in the radiation image recorded in the radiation converter 10. When the controller 38 detects the charge signal i of the segmented region of interest (the high-definition reading region HRA in which the radiation image of the subject 40M is present in FIG. 5), the controller 38 is in the high-definition reading mode. When the charge signal i of the divided non-interest area (the high-speed reading area HRA where the radiographic image of the subject 40M does not exist in FIG. 5) is detected, the operation is performed in the high-speed reading mode.

  In this way, the image reading time can be significantly shortened as compared to operating the entire image region (region of interest + non-region of interest) in the high-definition reading mode.

  Since the reading time of the image can be greatly shortened, in the mammography for obtaining a radiation image from the radiation converter 10 with the breast of the subject held between the compression plates and compressed, the compression time is shortened. The time which gives pain to a subject can be minimized.

  Specifically, the reading time (image processing cycle) of the image of one main scanning line is the reading time L2 from the reset time t1 to the next reset time t5 in FIG. 4 in the high-definition reading mode. Then, it is the reading time L3 from the reset time t11 to the next reset time t15 in FIG. 6 (L3 << L2).

  In practice, when the charge signal i of the non-interesting region is read in the high-speed reading mode, the time constant τ1 of the first low-pass filter of the first holding circuit 31 is set to a value at which the reference level converges with the shortest response time, The time constant τ2 ′ of the second low-pass filter of the second holding circuit 32 is set to a value at which the signal level converges with the shortest response time (τ2 ′ = τ1).

  Further, when the charge signal of the region of interest is read in the high-definition reading mode, the time constant τ1 of the first low-pass filter of the first holding circuit 31 is set to a value at which the reference level converges with the shortest response time, and the second The time constant τ2 of the second low-pass filter of the holding circuit 32 is set to a value longer than the time constant τ1 of the first low-pass filter (τ2 = 10 × τ1).

  Further, in the above-described embodiment, the description has been made using the so-called optical reading type radiation converter 10 as the output of the charge signal i input to the signal detector 30. However, the present invention is not limited to this. A TFT type radiation converter may be used, or a radiation converter that detects the stimulated emission light emitted from the stimulable phosphor sheet by the photoelectric conversion element and outputs the charge signal i is used. It may be.

  Moreover, in the said embodiment, although the radiation image signal detection system 50 was comprised from the radiation source, the radiation converter 10, the reading light source part 20, and the signal detector 30, the radiation converter 10 is provided without providing a radiation source. The radiation image signal detection system may be configured by the reading light source unit 20 and the signal detector 30.

It is a schematic block diagram of the radiographic image signal detection system using one Embodiment of the signal detector of this invention. It is a schematic block diagram of the radiation converter in the radiographic image signal detection system shown in FIG. FIG. 3A is an explanatory diagram of a state in which radiation is irradiated on the radiation converter and a radiation image is recorded, and FIG. 3B is an explanatory diagram of a state in which reading light is irradiated on the radiation converter and a charge signal is read out. It is a time chart for demonstrating the effect | action of the high-definition reading mode in the radiographic image signal detection system shown in FIG. It is explanatory drawing provided when dividing a region of interest and a non-region of interest. It is a time chart for demonstrating the effect | action of the high-speed reading mode in the radiographic image signal detection system shown in FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Radiation converter 30 ... Signal detector 38 ... Controller (area | region boundary classifier) 40M ... Subject

Claims (7)

A radiation converter that records radiation passing through the subject as a radiation image;
A signal detector that outputs an image signal corresponding to the radiation image based on the charge signal output from the radiation converter;
An area boundary classifier for classifying a boundary between a region of interest and a non-region of interest of the radiographic image recorded in the radiation converter;
The signal detector is operated in a high-definition reading mode when detecting the charge signal of the region of interest divided by the region boundary classifier, and is read at high speed when detecting the charge signal of the non-region of interest. A controller operating in mode,
A radiation image signal detection system comprising: The radiation image signal detection system according to claim 1.
The signal detector is
Integrating the charge signal to generate an integrated voltage, and an integrating amplifier having a function of resetting the integrated voltage;
A first and a second holding circuit connected in parallel to the output side of the integrating amplifier, constituting a correlated double sampling circuit, and tracking and holding a reference level and a signal level by first and second low-pass filters;
A differential amplifier that outputs a difference signal between the outputs of the first and second holding circuits as the image signal;
The controller is
When reading the charge signal of the non-interesting region in the high-speed reading mode, the time constant of the first low-pass filter of the first holding circuit is set to a value at which the reference level converges with the shortest response time. A radiation image signal detection system, wherein the time constant of the second low-pass filter of the 2 holding circuit is set to a value at which the signal level converges with the shortest response time. The radiation image signal detection system according to claim 1.
The signal detector is
Integrating the charge signal to generate an integrated voltage, and an integrating amplifier having a function of resetting the integrated voltage;
A first and a second holding circuit connected in parallel to the output side of the integrating amplifier, constituting a correlated double sampling circuit, and tracking and holding a reference level and a signal level by first and second low-pass filters;
A differential amplifier that outputs a difference signal between the outputs of the first and second holding circuits as the image signal;
The controller is
When the charge signal of the region of interest is read in the high-definition reading mode, the time constant of the first low-pass filter of the first holding circuit is set to a value at which the reference level converges with the shortest response time. A radiation image signal detection system, wherein a time constant of the second low-pass filter of the 2 holding circuit is set to a value longer than a time constant of the first low-pass filter. In the radiographic image signal detection system according to any one of claims 1 to 3,
The region classifier is
When the boundary between the region of interest and the region of non-interest of the radiographic image recorded in the radiation converter is divided, the region of interest and the region of non-interest of the region of interest are preliminarily determined based on the size of the subject before reading the radiographic image. A radiation image signal detection system characterized by setting a boundary. In the radiographic image signal detection system according to any one of claims 2 to 4,
The region classifier is
When the boundary between the region of interest and the non-interest region of the radiographic image recorded in the radiation converter is divided, the boundary between the region of interest and the non-region of interest is set based on the difference signal of the charge signal being read. The radiation image signal detection system characterized by the above-mentioned. The radiation image signal detection system according to claim 5, wherein
The region classifier is
The difference signal of each pixel of the adjacent reading line is compared, or the average value of the difference signal of each pixel of the adjacent reading line is compared, and the boundary between the region of interest and the non-interesting region is determined based on the comparison result. Radiation image signal detection system characterized by setting. When the signal detector outputs an image signal corresponding to the radiation image based on the charge signal output from the radiation converter that records the radiation that has passed through the subject as a radiation image,
Dividing the boundary between the region of interest and the region of non-interest in the radiographic image recorded in the radiation transducer;
The signal detector is operated in a high-definition reading mode when a charge signal of the segmented region of interest is detected, and is operated in a high-speed reading mode when a charge signal of the non-region of interest is detected. A radiation image signal detection method.

Images