JP6166821B2 - Radiation imaging apparatus and radiation imaging system - Google Patents

Description

  The present invention relates to a radiation imaging apparatus and a radiation imaging system.

  In order to synchronize between the radiation generation apparatus and the radiation imaging apparatus, a configuration in which the radiation imaging apparatus detects the presence or absence of radiation irradiation has been proposed. Patent Document 1 proposes a radiation imaging apparatus that detects the presence / absence of radiation irradiation by utilizing the fact that a current flows through a bias line that supplies a bias potential to a pixel when radiation is applied to the radiation imaging apparatus. Specifically, when the current flowing through the bias line exceeds a predetermined threshold value, it is determined that radiation has been applied. Noise current flows through the bias line due to various factors other than the current caused by radiation. When the noise current is large, it may be erroneously detected that the radiation imaging apparatus has been irradiated even though the radiation is not irradiated. In order to prevent erroneous detection, Patent Document 1 has a sample-and-hold circuit for holding the dark current flowing through the bias line, and the held dark current is subtracted from the current flowing through the bias line at a certain time. Compare the value to the threshold.

Special Table 2002-543684 Publication

  The method of reducing the dark current as proposed in Patent Document 1 can remove only noise of a very slow frequency component (for example, 1 Hz or less) such as a temperature drift of the bias current. Therefore, it cannot cope with erroneous detection due to noise (50 to 60 Hz) mixed from the commercial power supply or noise (several Hz to several kHz) generated when pressure or impact is applied to the housing. In view of this, an object of one aspect of the present invention is to provide a technique for suppressing erroneous detection of determination as to whether or not a radiation imaging apparatus is irradiated with radiation.

In view of the above problems, according to one aspect of the present invention, a plurality of pixels including a conversion element that converts and accumulates radiation into electric charge and a switch element that connects the conversion element to a signal line includes a plurality of rows and a plurality of columns. A pixel array arranged to configure, a bias line for applying a bias potential to the conversion elements of the plurality of pixels, and a plurality of drive lines connected to control terminals of the switch elements of the plurality of pixels; , a driving unit for supplying a driving signal to one or more drive lines your capital, each drive signal, said switching element from oFF voltage to the non-conductive state, the oN voltage of the switching element to a conducting state A switching unit that repeats an initialization operation to switch back to the off-voltage, and a drive signal to one or more drive lines is switched to the on-voltage, and then a drive signal to the next one or more drive lines is on For each driving cycle before switching to pressure, the signals representative of the current flowing through the bias line a acquisition unit for sampling with a period shorter than the drive period, a predetermined delay time after switching the drive signal to the on-voltage Based on the effective value and the noise value , an acquisition unit that acquires a signal sampled after the elapse of time as a valid value, acquires a signal sampled after the delay time has elapsed since the drive signal was switched to the off voltage, and A radiation imaging apparatus comprising: a calculation unit that calculates radiation information; and a determination unit that determines whether or not the pixel array is irradiated with radiation based on the radiation information.

  The above means provides a technique for suppressing erroneous detection of determination as to whether or not radiation imaging apparatus is irradiated with radiation.

The figure explaining the structural example of the radiation imaging device of 1st Embodiment. FIG. 3 is a diagram illustrating a configuration example of a detection circuit in FIG. 1. The flowchart explaining the operation example of the radiation imaging device of FIG. 2 is a timing chart for explaining an operation example of the radiation imaging apparatus of FIG. 1. The figure explaining the detailed operation example of the radiation imaging device of FIG. The figure explaining the detailed operation example of the radiation imaging device of FIG. The figure explaining the effect of the detailed operation example of the radiation imaging device of FIG. The figure explaining another detailed operation example of the radiation imaging device of FIG. The figure explaining the effect of another detailed operation example of the radiation imaging device of FIG. FIG. 6 is a diagram for explaining still another detailed operation example of the radiation imaging apparatus of FIG. 1. The figure explaining the effect of another detailed operation example of the radiation imaging device of FIG. FIG. 6 is a diagram for explaining still another detailed operation example of the radiation imaging apparatus of FIG. 1. FIG. 6 is a diagram for explaining still another detailed operation example of the radiation imaging apparatus of FIG. 1. FIG. 6 is a diagram for explaining still another detailed operation example of the radiation imaging apparatus of FIG. 1. FIG. 6 is a diagram for explaining still another detailed operation example of the radiation imaging apparatus of FIG. 1. FIG. 6 is a diagram for explaining still another detailed operation example of the radiation imaging apparatus of FIG. 1. The timing chart explaining another example of operation of the radiation imaging device of FIG. The figure explaining the structural example of the radiation imaging device of 2nd Embodiment. The figure explaining the structural example of the radiation imaging device of 3rd Embodiment. The flowchart explaining the operation example of the radiation imaging device of 4th Embodiment. The figure explaining the structural example of the integration circuit of the radiation imaging device of 2nd Embodiment. The figure explaining the structure of the radiation imaging system of other embodiment.

  Embodiments of the present invention will be described below with reference to the accompanying drawings. Throughout various embodiments, the same elements are denoted by the same reference numerals, and redundant description is omitted. In addition, each embodiment can be appropriately changed and combined.

[First Embodiment]
An overall configuration of a radiation imaging apparatus 100 according to one embodiment of the present invention will be described with reference to FIG. The radiation imaging apparatus 100 is configured to capture an image formed by radiation. The image can be formed by radiation emitted from a radiation source (not shown) and transmitted through the subject. The radiation can be, for example, X-rays, alpha rays, beta rays, or gamma rays.

  The radiation imaging apparatus 100 includes a pixel array 101, a drive circuit (drive unit) 102, a readout circuit 103, a detection circuit 120, a reference bias potential generation circuit 126, a control unit 106, a calculation unit 130, and a determination unit 131. Including. In addition, the radiation imaging apparatus 100 may include a signal processing unit (processor) 105.

  The pixel array 101 has a plurality of pixels PIX that are two-dimensionally arranged to form a plurality of rows and a plurality of columns. In the example shown in FIG. 1, the pixels PIX are arranged to form 3 rows and 3 columns, but in reality, more pixels PIX are arranged to form more rows and columns. . For example, a 17-inch radiation imaging apparatus has about 2800 rows and about 2800 columns of pixels. Each pixel PIX includes a conversion element 201 that converts radiation or light into electric charge and a switch element T that outputs an electric signal corresponding to the electric charge to the signal line Sig, and is configured to detect the radiation.

  The conversion element 201 is an indirect conversion element including, for example, a photoelectric conversion element S that converts light into electric charges and a wavelength converter (scintillator) that converts radiation into light having a wavelength that can be detected by the photoelectric conversion element. sell. Alternatively, the conversion element 201 may be a direct type conversion element that directly converts radiation into electric charges. The photoelectric conversion element S can be, for example, a PIN type photodiode mainly made of amorphous silicon disposed on an insulating substrate such as a glass substrate. When the photoelectric conversion element of the conversion element 201 is a PIN photodiode, the conversion element 201 may have a capacitor Cs.

  The switch element T may be a transistor having a control terminal and two main terminals, for example, a thin film transistor (TFT). One electrode of the conversion element 201 is electrically connected to one electrode of the two main terminals of the switch element T, and the other electrode of the conversion element 201 is electrically connected to a common bias line Vs. A bias potential VVs is supplied to the bias line Vs by the detection circuit 120.

  A control terminal (gate) of the switch element T is connected to a drive line G driven by the drive circuit 102. The drive circuit 102 drives the drive line G of the row to be selected in the pixel array 101 to an active level. When an active level drive signal is supplied to the gate of the switch element T through the drive line G, the switch element T becomes conductive. As a result, signals corresponding to the charges accumulated in the conversion elements 201 of the pixels PIX in the selected row are output in parallel to the plurality of signal lines Sig.

  The signal output to the signal line Sig is read by the reading circuit 103. The read circuit 103 includes a plurality of amplifier circuits 207 and a multiplexer 208. The plurality of amplifier circuits 207 are provided so that one amplifier circuit 207 corresponds to one signal line Sig. The signals of the pixels PIX in the selected row output in parallel to the plurality of signal lines Sig are amplified in parallel by the plurality of amplifier circuits 207.

  Each amplifier circuit 207 includes, for example, an integrating amplifier 203, a variable amplifier 204 that amplifies the signal from the integrating amplifier 203, a sample hold circuit 205 that samples and holds the signal from the variable amplifier 204, and a buffer amplifier 206. sell. The integration amplifier 203 can include, for example, an operational amplifier that amplifies the difference between the signal output to the signal line Sig and the reference potential Vref1 from the reference power supply 107, an integration capacitor, and a reset switch. The integrating amplifier 203 can change the amplification factor by changing the value of the integrating capacitor. The signal output to the signal line Sig is supplied to the inverting input terminal of the operational amplifier, the reference voltage Vref1 is supplied from the reference power supply 107 to the non-inverting input terminal, and the output terminal is connected to the input terminal of the variable amplifier 204. It is connected. The integration capacitor and the reset switch are connected in parallel between the inverting input terminal and the output terminal of the operational amplifier. The sample hold circuit 205 can be configured by, for example, a sampling switch and a sampling capacitor.

  The multiplexer 208 sequentially selects and outputs signals read in parallel from the plurality of amplifier circuits 207 respectively corresponding to the plurality of signal lines Sig. The read circuit 103 can include a buffer amplifier 209 that buffers the signal from the multiplexer 208. The buffer amplifier 209 can function as an impedance converter. The read circuit 103 can include an A / D converter 210. The A / D converter 210 can be arranged to convert an analog signal output from the buffer amplifier 209 into a digital signal, for example.

  The signal output from the reading circuit 103 can be provided to the signal processing unit 105. The signal processing unit 105 can be configured to process the signal output from the readout circuit 103 and supply the processed signal to the computer 108. The signal processing unit 105 may be built in the radiation imaging apparatus 100 as illustrated in FIG. 1 or may be provided as an external device of the radiation imaging apparatus 100.

  The control unit 106 controls the overall operation of the radiation imaging apparatus 100 and generates, for example, a control signal for controlling the drive circuit 102 and a control signal for controlling the readout circuit 103. In FIG. 1, lines representing connections from the control unit 106 to each circuit are omitted. In response to the control signal from the control unit 106, the drive circuit 102 turns on the switch element T of the pixel PIX in the row from which the signal is to be read. Control signals for controlling the read circuit 103 can include, for example, a reset signal RC, a sample hold signal SH, and a clock signal CLK. The reset signal RC is a signal for controlling the reset switch of the integrating amplifier 203, the sample hold signal SH is a signal for controlling the sample hold circuit 205, and the clock signal CLK is a signal for controlling the multiplexer 208.

  Hereinafter, the detection circuit 120 will be described with reference to FIG. The detection circuit 120 detects a current flowing through the bias line Vs (hereinafter referred to as a bias current), and provides a bias current signal VSD indicating the current to the calculation unit 130. That is, the detection circuit 120 can function as an acquisition unit that acquires the bias current signal VSD. The detection circuit 120 can include, for example, a current-voltage conversion amplifier 310, a voltage amplification amplifier 320, a filter circuit 330, and an A / D converter 340. The current-voltage conversion amplifier 310 converts the current flowing through the bias line Vs into a voltage. The voltage amplification amplifier 320 amplifies the signal (voltage signal) output from the current-voltage conversion amplifier 310. The voltage amplification amplifier 320 can be constituted by an instrumentation amplifier, for example. The filter circuit 330 is a filter that limits the band of the signal output from the voltage amplification amplifier 320, and may be, for example, a low-pass filter. The A / D converter 340 supplies the bias current signal VSD obtained by converting the signal (analog signal value) output from the filter circuit 330 into a digital signal value to the calculation unit 130.

  The detection circuit 120, specifically, the current-voltage conversion amplifier 310 detects a current flowing through the bias line Vs and supplies a potential corresponding to the reference bias potential Vs_ref supplied from the reference bias potential generation circuit 126 to the bias line Vs. To do. The current-voltage conversion amplifier 310 can be a transimpedance amplifier. The current-voltage conversion amplifier 310 includes, for example, an operational amplifier 311 and a feedback path 312 disposed between the inverting input terminal (second input terminal) and the output terminal of the operational amplifier 311. The non-inverting input terminal (first input terminal) of the operational amplifier 311 is supplied with the reference bias potential Vs_ref. The feedback path includes, for example, a first path in which the inverting input terminal and the output terminal of the operational amplifier 311 are short-circuited by the resistor Rf1, a second path in which the inverting input terminal and the output terminal are short-circuited by the resistor Rf2, and the inversion A third path for short-circuiting the input terminal and the output terminal by the conductive line CL may be included.

  A phase compensation capacitor Cf1 can be connected in parallel to the resistor Rf1. A phase compensation capacitor Cf2 can be connected in parallel to the resistor Rf2. The phase compensation capacitors Cf1 and Cf2 are effective for preventing the current-voltage conversion amplifier 310 from oscillating, for example. A switch SWC can be arranged in series on the path including the resistor Rf2. A switch SWB can be arranged in series on the path constituted by the conductive lines CL.

  The control unit 106 supplies the control signal VSX to the detection circuit 120 to control the feedback impedance by selecting a path to be effective among a plurality of paths including the first path, the second path, and the third path. Here, when the switch SWB is closed, the third path constituted by the conductive line CL becomes valid, and the first path including the resistor Rf1 and the second path including the resistor Rf2 become invalid. When the switch SWB is opened and the switch SWC is closed, the third path becomes invalid and the first path and the second path become valid.

  A switch SWA and a resistor R may be arranged in series between the inverting input terminal of the operational amplifier 311 and the ground. A capacitor C may be disposed between the inverting input terminal of the operational amplifier 311 and the ground. The current-voltage conversion amplifier 310 includes the feedback path 312, and thereby, a potential corresponding to the reference bias potential Vs_ref given to the non-inverting input terminal (first input terminal) of the operational amplifier 311 is inverted to the inverting input terminal (second input terminal). Function to occur. More specifically, the current-voltage conversion amplifier 310 functions to generate a potential substantially the same as the reference bias potential Vs_ref given to the non-inverting input terminal of the differential amplifier circuit 211 at the inverting input terminal. Here, the impedance of the feedback path 312 of the current-voltage conversion amplifier 310 (hereinafter referred to as feedback impedance) is controlled by the control unit 106.

  A large feedback impedance means that the gain of the current-voltage conversion amplifier 310 is large. On the other hand, if the feedback impedance is large, this limits the magnitude of the bias current I_Vs, and the potential of the bias line Vs may become unstable. Therefore, it is desired that the feedback impedance be controlled in accordance with the operation of the radiation imaging apparatus 100, for example, the detection operation for detecting the irradiation of the radiation to the pixel array 101 and the operation for reading the signal from the pixel PIX. Hereinafter, this will be described more specifically.

  In this embodiment, the calculation unit 130 calculates radiation information based on the output from the detection circuit 120, that is, the bias current signal VSD, and the determination unit 131 starts irradiation of radiation to the pixel array 101 based on the radiation information. Is detected. In response to the detection, the control unit 106 controls the charge accumulation operation by the plurality of pixels PIX. That is, in order to quickly detect the start of radiation irradiation to the pixel array 101, it is necessary to detect the current flowing through the bias line Vs with high sensitivity by the detection circuit 120. Therefore, it is desirable that the feedback impedance be large in the detection operation for detecting radiation irradiation to the pixel array 101.

  On the other hand, when the charge accumulated in the capacitor Cs of the conversion element 201 is transferred to the signal line Sig via the switch element T, if the feedback impedance is large, the bias line Vs from the bias line Vs to the second electrode s2 side of the conversion element 201 Current supply is slow. In particular, when intense radiation is partially incident on the pixel array 101, noise is added to the captured image due to a delay in current supply from the bias line Vs to the second electrode s2 side of the conversion element 201. Is likely to occur. Therefore, when transferring the charge accumulated in the capacitor Cs of the conversion element 201 to the signal line Sig via the switch element T, it is desirable to reduce the feedback impedance.

  Therefore, the control unit 106 controls the feedback impedance so that the feedback impedance during the detection operation for detecting radiation irradiation is larger than the feedback impedance during the signal readout operation from the pixel PIX. An example in which the resistance Rf2 is smaller than the resistance Rf1 is shown below. In this case, the gain increases when the second path including the resistor Rf2 is selected.

  For example, the control unit 106 opens the switch SWB in the detection operation for detecting radiation irradiation, and closes the switch SWB in the signal readout operation from the pixel PIX. In this case, the state of the switch SWC may be either in the detection operation for detecting the irradiation of radiation and in the operation for reading out the signal from the pixel PIX.

  Instead, the control unit 106 opens the switch SWB and closes the switch SWC in the detection operation for detecting radiation irradiation, and closes the switch SWB in the signal reading operation from the pixel PIX (because the switch SWB is closed). The switch SWC may be either.

  The switch SWA and the resistor R are not essential, but when the switch SWA and the resistor R are provided, the switch SWA is closed during a non-operation period of the detection circuit 120 and can be opened in a detection operation for detecting radiation irradiation. For example, the non-operation period of the detection circuit 120 may be a period excluding a detection operation for detecting radiation irradiation, an accumulation operation, and an image output operation described later. Further, in the signal reading operation from the pixel PIX, the switch SWA may be closed or opened. Here, the resistance R may be larger than the resistance Rf1 and the resistance Rf2. For example, the resistor R can be set to 10 KΩ, the resistor Rf1 can be set to 1 KΩ, and the resistor Rf2 can be set to 1050Ω.

  The voltage amplification amplifier 320 can be configured as an amplifier having a variable gain. For example, the gain of the voltage amplification amplifier 320 can be changed depending on whether the switch SWD is closed or opened.

  The operation of the radiation imaging apparatus 100 includes an initialization operation, an accumulation operation, and a readout operation. The initialization operation is an operation for initializing the plurality of pixels PIX of the pixel array 101 in units of rows. The accumulation operation is an operation for accumulating charges generated by radiation irradiation in each pixel PIX of the pixel array 101. The read operation is an operation of reading a signal corresponding to the electric charge accumulated in each pixel PIX of the pixel array 101 by irradiation of radiation to the pixel array 101 from the pixel array 101 and outputting it as an image (image signal).

  The initialization operation is shifted to the accumulation operation when the determination unit 131 detects the start of radiation irradiation to the radiation imaging apparatus 100 based on the output from the detection circuit 120. For example, the operation shifts from the accumulation operation to the read operation in response to a predetermined time elapsed from the start of the accumulation operation.

  The operation of the radiation imaging apparatus 100 will be described with reference to FIGS. 3 and 4. In step S310, the control unit 106 starts an initialization operation. In the initialization operation, the control unit 106 repeats the operation of sequentially setting the drive lines G from the first row to the last row to the active level and setting the reset signal RC to the active level. Here, when the reset signal RC is set to the active level, the integrating amplifier 203 enters a voltage follower state, and the reference potential Vref1 is supplied to the signal line Sig. In this state, the switch T in the row where the drive line G is set to the active level is turned on, and the charge accumulated in the capacitor Cs of the conversion element 201 is initialized. 4, Vg (0), Vg (1), Vg (2),..., Vg (Ys), Vg (Ys + 1),. A drive signal supplied from the row to the drive line G in the last row is shown. That is, Y represents the number of rows of the pixel array 101, that is, the number of drive lines G. Hereinafter, an operation from when the control unit 106 starts supplying an active level signal to any one of the drive lines G to when an active level signal is supplied to the next drive line G is referred to as one initialization operation. A period until an active level signal is supplied to all the drive lines G once is called a frame. In the example of FIG. 4, the initialization operation is performed Y times in one frame, and the pixels in one row are initialized by the initialization operation once.

  In the initialization operation period, the detection circuit 120 detects information correlated with the radiation dose to the pixel array 101 and supplies a detection signal corresponding to the information to the calculation unit 130. In the present embodiment, the detection circuit 120 detects the current I_Vs flowing through the bias line Vs as information correlated with the radiation dose to the pixel array 101, and indicates the current as a detection signal corresponding to the information. The bias current signal VSD is supplied to the calculation unit 130.

  During the initialization operation, in step S320, the determination unit 131 performs a radiation detection process. Specifically, the calculation unit 130 calculates radiation information based on the bias current signal VSD, and the determination unit 131 determines the start of radiation irradiation to the pixel array 101 based on the radiation information. In this embodiment, the determination unit 131 performs one detection process for each initialization operation. However, the determination unit 131 may perform one detection process for a plurality of initialization operations. . Here, the control unit 106 sets the feedback impedance during the detection operation for detecting radiation irradiation to be larger than the feedback impedance during the signal readout operation from the pixel PIX.

  The control unit 106 continues the initialization operation until the start of radiation irradiation to the pixel array 101 is detected (step S370). When the control unit 106 detects the start of radiation irradiation to the pixel array 101 (YES in step S320), the control unit 106 starts an accumulation operation in step S330. That is, when the start of radiation irradiation is detected (indicated as “irradiation start detection” in FIG. 4), the operation proceeds from the initialization operation to the accumulation operation. Details of the detection process in step S320 will be described later.

  During the accumulation operation, the control unit 106 determines the end of radiation irradiation in step S340. The method for determining the end of radiation is not particularly limited. For example, it can be determined that radiation irradiation has ended when a predetermined time has elapsed since the start of the accumulation operation. Alternatively, the control unit 106 can detect the end of radiation irradiation to the pixel array 101 based on the bias current signal VSD.

  The control unit 106 continues the accumulation operation until it is determined that the irradiation of radiation to the pixel array 101 has been completed (step S380). If control unit 106 determines that the irradiation of radiation to pixel array 101 has ended (YES in step S340), it starts a read operation in step S350. That is, when it is determined that the radiation irradiation has ended (indicated as “irradiation end detection” in FIG. 4), the operation shifts from the accumulation operation to the reading operation. In the read operation, signals are sequentially read from the first row pixel to the last row pixel of the pixel array 101.

  FIG. 5 shows an operation example of the radiation imaging apparatus 100 focusing on the timing of shifting from the initialization operation to the accumulation operation. In FIG. 5, Vg (Ys−2), Vg (Ys−1), Vg (Ys), and Vg (Ys + 1) are transferred from the (Ys−2) th row to the (Ys + 1) th driving line G of the pixel array 101. The drive signal supplied is shown. In the present embodiment, the switch element T is high active, and becomes conductive when a high level voltage (on voltage) is applied to the drive line G, and when a low level voltage (off voltage) is applied. Non-conducting state. Alternatively, the switch element T may be low active.

The radiation imaging apparatus 100 can have the following characteristics regarding the bias current flowing through the bias line Vs.
(1) During radiation irradiation, a current proportional to the radiation dose per unit time flows through the bias line Vs. This current is shown as “first signal” in FIG. This current can flow more when the switch element T of the pixel PIX is in the conductive state than when the switch element T is in the non-conductive state, but is shown constant in the figure for simplicity.
(2) When the switch element T of the pixel PIX irradiated with radiation is turned on, a current proportional to the amount of charge accumulated in the conversion element 201 of the pixel PIX before the switch element T is turned on flows in the bias line Vs. This current is shown as “second signal” in FIG.
(3) When the conduction / non-conduction of the switch element T of the pixel PIX is switched, a current flows through the bias line Vs. This current can be called switching noise.
(4) When an impact or magnetic field is applied to the radiation imaging apparatus 100, a current flows through the bias line Vs. This current may be referred to as external noise and is shown as “external noise” in FIG. For example, a current of about 50 to 60 Hz can flow through the bias line Vs due to the influence of an electromagnetic field generated from a commercial power source. Further, when an impact is applied to the radiation imaging apparatus 100, a current of several Hz to several kHz can flow through the bias line Vs.
(5) Even if no magnetic field or impact is applied to the radiation imaging apparatus 100, current flows through the bias line Vs due to electromagnetic waves generated by the radiation imaging apparatus 100 itself, internal noise of the detection circuit 120, or the like. This current can be referred to as system noise. The “bias current” in FIG. 5 shows that the first signal, the second signal, and the external noise are constant over time, but FIG. 5 only conceptually shows at what timing they appear. Yes, and not always constant over time.

  In order to detect radiation irradiation, more specifically, the start of radiation irradiation, the sample value of the bias current signal VSD as a detection signal may be used as it is. However, in the present embodiment, in order to reduce the influence of external noise due to the impact or the influence of a magnetic field, the calculation unit 130 processes the plurality of bias current signals VSD to calculate radiation information, and based on this radiation information The determination unit 131 detects radiation irradiation. For example, the determination unit 131 determines that radiation is being applied when the radiation information or the integrated value of the radiation information exceeds a predetermined threshold.

  As shown in FIG. 5, the drive cycle of the drive circuit 102 is represented by TI. That is, the radiation imaging apparatus 100 performs an initialization operation once every time TI. Of the time TI, the time during which the drive circuit 102 supplies a high level drive signal (hereinafter referred to as ON time) is represented by TH, and the time during which the low level drive signal is provided (hereinafter referred to as OFF time) is represented by TL. In the present embodiment, the control unit 106 controls the drive circuit 102 so that TH = TL. That is, with the start of one initialization operation, the drive circuit 102 switches the drive signal of a certain drive line G from the low level to the high level, returns it to the low level after a lapse of time TH, and further has the same length of time. The next initialization operation is started after TL has elapsed. For example, TH = TL = 16 μsec. The sampling period of the A / D converter 340 of the detection circuit 120 is represented by TS. In this case, the detection circuit 120 supplies one sample value of the bias current signal VSD to the calculation unit 130 every time TS. As shown in FIG. 5, in the present embodiment, the detection circuit 120 performs sampling so that TH = TL = TS. In this case, the detection circuit 120 outputs two sample values of the bias current signal VSD for each initialization operation. Therefore, the sample value output by the detection circuit 120 when the switch element T of the pixel PIX is in a conductive state is called an effective value S, and the detection circuit 120 outputs when the switch element T of the pixel PIX is in a non-conductive state. The sample value is called a noise value N. The calculation unit 130 can remove the external noise by taking the difference between the effective value S and the noise value N, and can extract only the second signal as radiation information.

Since the external noise varies with time, the calculation unit 130 may calculate the radiation signal using the effective value S and the noise value N sampled at times close to each other. For example, the effective value S and the noise value N output in the y-th initialization operation (y is an arbitrary natural number) are set to S (y) and N (y), respectively, and the determination unit 131 in step S320 (FIG. 3). Let X (y) be radiation information used to detect radiation exposure for the y-th initialization operation. In this case, the calculation unit 130 can calculate X (y) by operations such as Expressions (1) to (3). Equations (1) to (3) are expressed as follows: a bias current signal VSD (detection signal) when the switch element T is in a conductive state and a bias current signal VSD (detection signal) when the switch element T is in a nonconductive state. It means the difference calculation.
X (y) = S (y) -N (y) (1)
X (y) = S (y) -N (y-1) (2)
X (y) = S (y)-{N (y) + N (y-1)} / 2 (3)

  The method of reducing external noise as described above is referred to as CDS (correlated double sampling) in the present specification. The CDS calculation is not limited to the above calculation method. For example, the calculation unit 130 may calculate X (y) using non-adjacent sample values such as S (y-1) and N (y-2). The calculation unit 130 may calculate radiation information after performing other four arithmetic operations and differentiation / integration on a plurality of sample values of the bias current signal VSD. Further, as S (y) or N (y) sampled at the yth time, a value obtained by adding a plurality of sample values obtained by sampling a plurality of times (for example, 8 times) during the period may be used.

Next, with reference to FIG. 6, another operation example of the radiation imaging apparatus 100 focusing on the timing of shifting from the initialization operation to the accumulation operation will be described. The sampling period TS of the A / D converter 340 of the detection circuit 120 is different from the example of FIG. Other points may be the same as those in the example of FIG. In the example of FIG. 6, the sampling period TS of the A / D converter 340 is set to be shorter than ½ of the driving period TI of the driving circuit 102. Then, the sample value of the bias current signal VSD sampled in the second half of the on time TH is set as an effective value S, and the sample value of the bias current signal VSD sampled in the first half of the off time TL is set as a noise value N. Similar to the example of FIG. 5, the effective value S and the noise value N output in the y-th initialization operation are set to S (y) and N (y), respectively, and the determination unit 131 performs y in step S320 (FIG. 3). Let X (y) be the radiation information used for detection of radiation exposure for the initialization operation for the second time. In this case, the calculation unit 130 can calculate X (y) by an operation such as Equation (4).
X (y) = S (y) -N (y) (4)
By shortening the sampling period TS, the value of the external noise included in the effective value S and the value of the external noise included in the noise value N can be brought close to each other. Thereby, the influence of the external noise in radiation information can be reduced.

  FIG. 7 shows the experimental results of calculating the noise removal ratio when the sampling period TS is set to various values. In the graph of FIG. 7, the horizontal axis indicates the frequency of external noise, and the vertical axis indicates the noise removal ratio. The on-time TH and off-time TL of the drive signal are each 16 μs. At this time, curves 701, 702, 703, and 704 show the noise removal ratios at the respective frequencies when the sampling period TS is 16 μsec, 10.6 μsec, 8 μsec, and 5.3 μsec. As can be seen from the figure, in the low frequency region, noise is efficiently removed as the sampling period TS is shorter.

  Next, with reference to FIG. 8, another operation example of the radiation imaging apparatus 100 focusing on the timing of shifting from the initialization operation to the accumulation operation will be described. The driving timing (ON time TH, OFF time TL) by the driving circuit 102 and the sampling period TS of the A / D converter 340 of the detection circuit 120 are different from the example of FIG. Other points may be the same as those in the example of FIG. In the example of FIG. 8, the off time TL is longer than the on time TH. For example, the drive circuit 102 operates so that the off time TL is three times the on time TH. Then, the sampling period TS is made the same as the ON time TH. In this case, the detection circuit 120 outputs one effective value S and three noise values N for one reset operation. Therefore, the effective value S and the three noise values N output in the y-th initialization operation (y is an arbitrary natural number) are S (y), N1 (y), N2 (y), and N3 (y), respectively. To do. In step S320 (FIG. 3), the determination unit 131 sets X (y) as radiation information used for detection of radiation irradiation for the y-th initialization operation. However, N1 (y), N2 (y), and N3 (y) are arranged in order from values sampled at an earlier time.

Various methods by which the calculation unit 130 calculates X (y) using the effective value S and the noise value N will be described below. The calculation unit 130 may calculate X (y) according to any of the following formulas (5) to (7), similarly to the example described with reference to FIG.
X (y) = S (y) -N1 (y) (5)
X (y) = S (y) -N3 (y-1) (6)
X (y) = S (y)-{N1 (y) + N3 (y-1)} / 2 (7)
Expressions (5) to (7) express the effective value S (y) obtained in the y-th initialization operation and the noise values N1 (y) and N3 (y−1) sampled before and after this. To calculate X (y).

Instead of this, the calculation unit 130 may calculate X (y) according to the following equation (8).
X (y) = S (y)-{3 * N1 (y) -3 * N2 (y) + N3 (y)} (8)
In this equation (8), the noise value N (N1 (y) to N3 (y) in the above example) is weighted. That is, a different coefficient is given to the noise value N. Expression (8) is obtained by repeatedly subtracting adjacent values for one effective value S and a plurality of noise values N obtained in the same initialization operation. Specifically, the calculation unit 130 first subtracts adjacent sample values of the same initialization operation, sets the difference between the effective value S and the noise value N as a new effective value S, and Let the difference be a new noise value N. The calculation unit 130 repeats the above calculation until one value is obtained. Accordingly, when the noise value N obtained by one initialization operation is n, the calculation unit 130 repeats the above calculation n times. This process is expressed as follows. Here, the updated effective value and noise value are represented by “′”.

First stage calculation:
S ′ (y) = S (y) −N1 (y)
N1 ′ (y) = N1 (y) −N2 (y)
N2 ′ (y) = N2 (y) −N3 (y)
Second stage calculation:
S ″ (y) = S ′ (y) −N1 ′ (y)
N1 ″ (y) = N1 ′ (y) −N2 ′ (y)
Third stage calculation:
X (y) = S ″ (y) −N1 ″ (y)

In Expression (8), the calculation unit 130 calculates radiation information using one effective value S and three noise values N obtained successively thereafter. Instead, the calculation unit 130 calculates radiation information using one effective value S and three noise values N obtained successively before that, as in the following Expression (9). May be.
X (y) = S (y)-{3 * N3 (y-1) -3 * N2 (y-1) + N1 (y-1)} (9)
This equation (8) is also obtained by the same calculation as equation (7).

In yet another embodiment, as shown in the following equation (10), the calculation unit 130 averages the radiation information obtained by equation (8) and the radiation information obtained by equation (9). Radiation information may be calculated.
X (y) = S (y)-{3 * N1 (y) -3 * N2 (y) + N3 (y) + 3 * N3 (y-1) -3 * N2 (y-1) + N1 (y-1 )} / 2 (10)
In Expression (10), the calculation unit 130 calculates radiation information based on one effective value S and six noise values N obtained successively before and after the effective value S.

  In any of the equations (8) to (10), the noise value N sampled near the timing of switching between the effective value S and the noise value N is given a large weight. Here, the large weight means that the influence on the radiation information is large, and specifically, it can mean that the absolute value of the coefficient of the noise value N is large. The same applies to the following examples.

  FIG. 9 shows the experimental results of calculating the noise removal ratio when the radiation information X (y) is calculated by various methods. In the graph of FIG. 9, the horizontal axis indicates the frequency of the external noise, and the vertical axis indicates the noise removal ratio. The on time TH of the drive signal is 8 μsec, and the off time TL is 24 μsec. At this time, the curves 801, 802, 803, and 804 respectively remove noise at each frequency when the radiation information X (y) is calculated by the equations (5), (7), (8), and (10). Indicates the ratio. For comparison, a curve 701 in FIG. 7 is also shown in FIG. As can be seen from the figure, in the low frequency region, noise is efficiently removed by calculating radiation information with weights assigned to a plurality of noise values N.

  The radiation information of Expression (8) to Expression (10) may be calculated by sequentially calculating the subtraction between adjacent sample values, or adding / subtracting the sample values multiplied by a coefficient according to these expressions May be calculated. The same applies to the case where the radiation information is calculated according to the formula described later. Further, the external noise that flows when pressure or impact is applied to the radiation imaging apparatus 100 can take a value 10 to 100 times the current that flows through the bias line Vs during radiation irradiation. Therefore, the dynamic range of the detection circuit 120 can be set so that the external noise can be sampled without being saturated. When the dynamic range of the detection circuit 120 is expanded, the radiation information is calculated only by addition / subtraction / multiplication by multiplying the radiation information and the threshold by an integer to reduce the occurrence of quantization error due to division. In addition, the control unit 106 may be configured.

  Next, with reference to FIG. 10, still another operation example of the radiation imaging apparatus 100 focusing on the timing of shifting from the initialization operation to the accumulation operation will be described. The sampling period TS of the A / D converter 340 of the detection circuit 120 is different from the example of FIG. Other points may be the same as those in the example of FIG. In the example of FIG. 10, the off time TL and the on time TH are equal. Then, the sampling period TS is made shorter than the ON time TH. For example, the sampling period TS is set to 1/3 of the on time TH. In this case, the detection circuit 120 outputs three effective values S and three noise values N for one reset operation. Therefore, the three effective values S and three noise values N output in the y-th initialization operation (y is an arbitrary natural number) are converted into S1 (y), S2 (y), S3 (y), and N1 (y ), N2 (y), and N3 (y). In step S320 (FIG. 3), X (y) is radiation information used by the control unit 106 to detect radiation irradiation for the y-th initialization operation. However, S1 (y), S2 (y), S3 (y), N1 (y), N2 (y), and N3 (y) are arranged in order from the value sampled at an earlier time.

Various methods by which the calculation unit 130 calculates X (y) using the effective value S and the noise value N will be described below. For example, the calculation unit 130 may calculate X (y) according to any of the following formulas (11) to (13).
X (y) = S1 (y) + S2 (y) + S3 (y)-{N1 (y) + N2 (y) + N3 (y)} (11)
X (y) = S1 (y) + S2 (y) + S3 (y)-{N1 (y-1) + N2 (y-1) + N3 (y-1)} (12)
X (y) = 2 * S1 (y) + 2 * S2 (y) + 2 * S3 (y)-{N1 (y) + N2 (y) + N3 (y) + N1 (y-1) + N2 (y-1) + N3 (Y-1)} (13)
In Expressions (11) to (13), the calculation unit 130 calculates three effective values S (y) obtained in the y-th initialization operation and three noise values N or six sampled before and after this. X (y) is calculated using the noise value N. Here, for the reason described above, the calculation unit 130 calculates the radiation information only by adding / subtracting the effective value / noise value, and does not use division.

Instead of this, the calculation unit 130 may calculate X (y) according to the following equation (14).
X (y) = {11 * S1 (y) + 5 * S2 (y) + 2 * S3 (y)} / 18- {11 * N3 (y-1) + 5 * N2 (y-1) + 2 * N1 (y -1)} / 18 (14)
In this equation (14), the effective value S and the noise value N (in the above example, S1 (y) to S3 (y), N1 (y-1) to N3 (y-1)) are weighted. . That is, different coefficients are given to the effective value S and the noise value N. Equation (14) is expressed as follows: k effective values S among the three effective values S obtained in the y-th initialization operation and the three noise values N obtained in the immediately preceding y-1 initialization operation. It is obtained by averaging the average difference between k = 1-3. This process is expressed as follows. Here, the difference between the average of the effective values S and the average of the noise values N is represented by D1 to D3.

Difference between one piece:
D1 = S1 (y) -N3 (y-1)
Difference between two averages:
D2 = {S1 (y) + S2 (y)} / 2- {N3 (y-1) + N2 (y-1)} / 2
Difference between three averages:
D3 = {S1 (y) + S2 (y) + S3 (y)} / 3- {N3 (y-1) + N2 (y-1) + N1 (y-1)} / 3
Difference average:
X (y) = (D1 + D2 + D3) / 3

In Expression (14), the calculation unit 130 calculates radiation information using the three effective values S and the three noise values N obtained successively before that. Instead, the calculation unit 130 may calculate radiation information using the three effective values S and the three noise values N obtained successively thereafter, as in the following equation (15). Good.
X (y) = {11 * S3 (y) + 5 * S2 (y) + 2 * S1 (y)} / 18- {11 * N1 (y) + 5 * N2 (y) + 2 * N3 (y)} / 18 ... (15)
This equation (15) is also obtained by the same calculation as the equation (14). In yet another embodiment, the calculation unit 130 may calculate the radiation information by averaging the radiation information obtained by Expression (14) and the radiation information obtained by Expression (15). In this case, the calculation unit 130 uses the three effective values S obtained by the y-th initialization operation and the six noises acquired adjacent to the y-th initialization operation as radiation information about the y-th initialization operation. The calculation is based on the value N weighted. Both Expression (14) and Expression (15) give a large weight to the noise value N sampled near the timing of switching between the effective value S and the noise value N.

  FIG. 11 shows the experimental results of calculating the noise removal ratio when the radiation information X (y) is calculated by various methods. In the graph of FIG. 11, the horizontal axis indicates the frequency of the external noise, and the vertical axis indicates the noise removal ratio. The on-time TH of the drive signal is 16 μsec, and the off-time TL is 16 μsec. The sampling time TS is set to 5.3 μsec. At this time, the curves 1101 and 1102 indicate the noise removal ratios at the respective frequencies when the radiation information X (y) is calculated by the equation (14) and the average of the equations (14) and (15). For comparison, a curve 701 in FIG. 7 is also shown in FIG. As can be seen from the figure, in the low frequency region, noise is efficiently removed by calculating radiation information with weights assigned to a plurality of noise values N.

  Next, with reference to FIG. 12, another operation example of the radiation imaging apparatus 100 focusing on the timing of shifting from the initialization operation to the accumulation operation will be described. The sampling period TS of the A / D converter 340 of the detection circuit 120 and the off time TL are different from the example of FIG. Other points may be the same as those in the example of FIG. In the example of FIG. 12, the driving cycle TI of the driving circuit 102 is equal to the on time TH, and the off time TL is zero. That is, in one initialization operation, the drive circuit 102 maintains the drive signal for the target drive line G at an active level. As a result, the drive signal to a certain row is switched from the high level to the low level, and at the same time, the drive signal to the next row is switched from the low level to the high level. Further, the sampling period TS is made shorter than the ON time TH. For example, the sampling period TS is set to 1/3 of the on time TH. In this case, the detection circuit 120 outputs three valid values S for one reset operation.

  In general, it is known that when the switch element T is switched between conduction and non-conduction, a current flows through the bias line Vs. This current can be referred to as switching noise. As shown in FIG. 12, the switching noise is canceled by overlapping the falling edge of the driving signal to the switching element T in a certain pixel row and the rising edge of the driving signal to the switching element T in another pixel row. Can do. Therefore, the example of FIG. 12 is effective when the switching noise is large.

  Three effective values S output in the y-th initialization operation (y is an arbitrary natural number) are assumed to be S1 (y), S2 (y), and S3 (y), respectively. In step S320 (FIG. 3), the determination unit 131 sets X (y) as radiation information used for detection of radiation irradiation for the y-th initialization operation. However, S1 (y), S2 (y), and S3 (y) are arranged in order from values sampled at an earlier time.

In this case, the calculation unit 130 may calculate X (y) according to the following equation (16).
X (y) = {11 * S1 (y) + 5 * S2 (y) + 2 * S3 (y)} / 18- {11 * S3 (y-1) + 5 * S2 (y-1) + 2 * S1 (y -1)} / 18 (16)
This equation (16) is similar to the equation (14) in that the three effective values S obtained in the y-th initialization operation and the three effective values S obtained in the y−1-th initialization operation. Of these, the average difference between adjacent k pieces is obtained by averaging up to k = 1 to 3. This process is expressed as follows. Here, the difference between the average of the effective values S obtained by the y-th initialization and the average of the effective values N obtained by the y−1-th initialization is represented by D1 to D3.

Difference between one piece:
D1 = S1 (y) -S3 (y-1)
Difference between two averages:
D2 = {S1 (y) + S2 (y)} / 2- {S3 (y-1) + S2 (y-1)} / 2
Difference between three averages:
D3 = {S1 (y) + S2 (y) + S3 (y)} / 3− {S3 (y−1) + S2 (y−1) + S1 (y−1)} / 3
Difference average:
X (y) = (D1 + D2 + D3) / 3

  Since Expression (16) performs the same calculation as Expression (14) for external noise, it can have the same noise removal characteristics as Expression (14). The radiation information X (y) obtained by the equation (16) is a value obtained by weighting a plurality of effective values S obtained by successive initialization operations and taking a difference. The determination unit 131 detects radiation irradiation when this value exceeds a threshold value.

  Next, with reference to FIG. 13, another operation example of the radiation imaging apparatus 100 focusing on the timing of shifting from the initialization operation to the accumulation operation will be described. The drive timing (ON time TH, OFF time TL) by the drive circuit 102 is different from the example of FIG. Other points may be the same as those in the example of FIG. In the example of FIG. 13, the drive circuit 102 overlaps the falling and rising edges of two drive signals supplied before and after. For example, the drive circuit 102 overlaps the fall of the drive signal in the odd-numbered initialization operation with the rise of the drive signal in the even-numbered initialization operation. In other words, the drive circuit 102 sets the off time TL in the odd-numbered initialization operation to zero. As a result, the driving cycle TI is different between the odd number and the even number. In the example of FIG. 13, the fall of the drive signal supplied to the drive line G in the Ys-1th row overlaps with the rise of the drive signal supplied to the drive line G in the Ys row.

  When the sampling period TS of the A / D converter 340, the ON time TH, and the OFF time TL are equal to each other, the detection circuit 120 outputs one effective value S in the odd-numbered reset operation, and the even-numbered reset operation. Outputs one effective value S and one noise value N. The effective value S output in the y-th initialization operation (y is an arbitrary odd number) initialization operation is S (y), and the effective value S and the noise value N output in the immediately subsequent y + 1 initialization operation are S. Let (y + 1) and N (y + 1). In step S320 (FIG. 3), the determination unit 131 sets X (y) as radiation information used for detection of radiation irradiation for the y-th initialization operation.

In this case, the calculation unit 130 may calculate X (y) according to the following equation (16).
X (y) = [S (y) + S (y + 1) − {N (y + 1) −N (y−1)} / 18] (17)
In this equation, the amount of radiation signal per unit time can be increased.

  Next, another example of the sampling timing of the detection circuit 120 will be shown with reference to FIG. The current (second signal) that flows through the bias line Vs in proportion to the amount of charge accumulated in the conversion element 201 of the pixel PIX until the switch element T of the pixel PIX irradiated with radiation is turned on has the following properties. Can have. That is, this current is a spike-like current that begins to flow (at time t0) with the rise of the drive signal, peaks when the switch element T becomes conductive (at time t1), and then ends (at time t2). The time from when the drive signal rises until the switch element T becomes conductive is determined by the time constant of the resistance and parasitic capacitance of the drive line G.

  Therefore, in the example shown in FIG. 14, the detection circuit 120 detects that the delay time TD has elapsed (at time t1) after the drive circuit 102 raises the drive signal (at time t0) in one initialization operation. The current flowing through the bias line Vs is sampled. In this case, the sample value output from the detection circuit 120 is the effective value S. In addition, the detection circuit 120 detects a current flowing through the bias line Vs after the same delay time TD (at time t4) after the drive circuit 102 has lowered the drive signal (at time t3) in one initialization operation. Sampling. In this case, the sample value output from the detection circuit 120 is the noise value N. As a method for calculating radiation information from the effective value S and the noise value N, any of the above-described methods may be used. It is known that the switching noise of the switch element T starts to flow at the same time as the drive signal rises and falls, and stops flowing after the switch element T is switched between the conductive state and the non-conductive state. Therefore, when the detection circuit 120 performs sampling at the above-described timing, the influence of switching noise can be reduced from the radiation signal.

  The delay time TD is set so that sampling is performed at the timing of the peak of the current flowing through the bias line Vs (for example, time t1). For example, the delay time TD is determined based on the time constant of the drive line G and the time constant of the bias line Vs. Further, the delay time TD may be determined further based on the processing time of the circuit preceding the A / D converter 340 in the detection circuit 120.

  An example of a method for determining the delay time TD will be described with reference to FIG. The control unit 106 pre-samples the current flowing through the bias line Vs at a time interval shorter than the sampling interval of FIG. The manufacturer of the radiation imaging apparatus 100 irradiates the radiation imaging apparatus 100 with test radiation while pre-sampling is being performed. Then, the control unit 106 delays the time between the rise time of the drive signal in one initialization operation and the sampling time when the sample value of the bias current signal VSD becomes the maximum for this initialization operation. Store as TD.

  Instead of using the test radiation, a current flowing in the bias line Vs due to the dark charge accumulated in the pixel PIX may be used. The control unit 106 maintains the drive signal supplied from the drive circuit 102 at a low level for a predetermined period, and waits for dark charge to be accumulated in the pixel PIX. Thereafter, the initialization operation shown in FIG. 15 is started. When the switch element T becomes conductive, a current corresponding to the dark charge flows through the bias line Vs, so the control unit 106 determines the delay time TD based on this current.

  The time constant of the drive line G, the time constant of the bias line Vs, and the processing time of the detection circuit 120 may change depending on the aging deterioration after the product shipment, the temperature of the use environment, and the like. Therefore, the method for determining the delay time TD described above may be performed even after product shipment. For example, the radiation imaging apparatus 100 may be configured to be able to switch between an adjustment mode in which the above pre-sampling is performed and a normal mode in which normal sampling is performed. In the adjustment mode, the user of the radiation imaging apparatus 100 irradiates the radiation imaging apparatus 100 with test radiation. Instead of this, the radiation imaging apparatus 100 may shift to an operation mode in which dark charges are accumulated. Then, the control unit 106 delays the time between the rise time of the drive signal in one initialization operation and the sampling time when the sample value of the bias current signal VSD becomes the maximum for this initialization operation. Store as TD. When the radiation imaging apparatus 100 is changed to the normal mode, the detection circuit 120 performs sampling according to the stored delay time TD.

  As described above, by performing sampling at the timing of the peak of the current flowing through the bias line Vs, the S / N ratio is improved, and the accuracy of detection of radiation irradiation is improved. Although FIG. 14 is described based on the drive timing of the drive circuit 102 shown in FIG. 5, sampling may be performed after the elapse of the delay time similarly in the drive timings of FIGS. 12 and 13.

  Next, another operation example of the radiation imaging apparatus 100 will be described with reference to FIG. The sampling period TS of the A / D converter 340 of the detection circuit 120 is different from the example of FIG. Other points may be the same as those in the example of FIG. In the example of FIG. 16, the on time TH = the off time TL = 16 μsec, and the sampling period TS = 1 μsec. In this case, 32 bias current signals VSD are output from the detection circuit 120 for one initialization operation.

  In the y-th initialization operation, it is assumed that the drive signal supplied to the drive line G in the Ys row switches to a high level at time t0. Then, it is assumed that the influence on the analog signal input to the A / D converter 340 of the detection circuit 120 due to the switching of the drive signal to the high level appears at time t1. In this case, the 16 sample values output from the detection circuit 120 from time t1 to time t5 correspond to the case where the switch element T in the y-th initialization operation is in a conductive state. Of the 16 sample values, the calculation unit 130 sets the total value of the middle 8 sample values output from time t3 to time t4 as an effective value S (y) for the y-th initialization operation. Similarly, 16 sample values output from the detection circuit 120 from time t5 to time t9 correspond to the case where the switch element T in the y-th initialization operation is in a non-conductive state. Of the 16 sample values, the calculation unit 130 sets the total value of the middle 8 sample values output from time t7 to time t8 as the noise value N (y) for the y-th initialization operation. And the calculation part 130 calculates radiation information according to above-mentioned Formula (1)-(3), for example.

  Next, another example of the operation of the radiation imaging apparatus 100 will be described with reference to FIG. The timing of the drive signal that the drive circuit 102 supplies to the drive line G in the initialization operation is different from the example of FIG. Since other points may be the same as the example of FIG. 4, overlapping description is omitted. The drive circuit 102 sequentially supplies high-level drive signals to odd-numbered drive lines G counted from one side of the pixel array 101 in the initialization operation of one frame, and then high-level to even-numbered drive lines G. Are sequentially supplied. Despite the fact that the pixel PIX has accumulated the radiation information from when radiation is actually applied until the irradiation is detected and the storage operation is started, the radiation imaging apparatus 100 applies the pixel PIX to the pixel PIX. The initialization operation will be performed. Therefore, a part of the radiation information is lost from the pixel row initialized during this period. As in the example of FIG. 17, by preventing the adjacent pixel rows from being reset continuously in time, the pixel rows in which a part of the radiation information is lost can be dispersed in the image data. Thus, it becomes easy to correct the image data by complementing the lost radiation information using the information of the pixel row from which the radiation has been normally acquired.

  In the example of FIG. 17, the drive circuit 102 supplies a high level drive signal to the two drive lines G simultaneously. Thus, in one initialization operation, a current proportional to the charge accumulated in the pixels for two rows flows through the bias line Vs. As a result, the S / N ratio of the bias current signal can be improved. In the example of FIG. 17, the two drive lines G are driven simultaneously, but a drive line group including an arbitrary number of two or more drive lines G may be used. Further, unlike FIG. 17, the drive lines G that are driven simultaneously may be adjacent to each other. The radiation imaging apparatus 100 may be configured to be able to change the number of drive lines that are driven simultaneously by an input by a user. In this case, as the number of drive lines that are driven simultaneously increases, the switching noise also increases. Therefore, the control unit 106 may switch the gain of the detection circuit 120 according to the number of drive lines.

  As described above, in the present embodiment, radiation information is acquired by various drive timings, sampling cycles, and calculation formulas. However, the present invention is not limited to the above-described embodiment. For example, the on time TH may be longer than the off time TL. Further, a plurality of effective values S and one noise value N may be sampled in one initialization operation. Further, the calculation unit 130 calculates radiation information using one or more of the plurality of effective values S obtained in one initialization operation and one or more of the plurality of noise values N. May be. Further, the calculation unit 130 does not have to calculate radiation information using both the effective value S and the noise value N, and may calculate radiation information using a plurality of effective values S, or may calculate a plurality of noises. Radiation information may be calculated using the value N. As described with reference to FIG. 5, the first signal of the bias current flows not only when the switch element T is turned on / off, but also when the pixel array 101 is irradiated with radiation. Therefore, the calculation unit 130 may use, as radiation information, the difference between the effective value S when radiation is irradiated and the effective value S when radiation previously acquired is not irradiated. Similarly, the calculation unit 130 may use, as radiation information, the difference between the noise value N when radiation is irradiated and the noise value N when radiation acquired before that is not irradiated. In general, the calculation unit 130 calculates radiation information based on a sample value acquired a plurality of times of the bias current signal VSD supplied from the detection circuit 120, and the determination unit 131 compares the radiation information with a threshold value to thereby determine radiation. The presence or absence of irradiation can be determined. In addition, the weights to the effective value S and the noise value N are not limited to the values determined by the above-described method, and arbitrary weights are assigned to each of the one or more effective values S and the one or more noise values N. May be attached.

[Second Embodiment]
The overall configuration of a radiation imaging apparatus 1800 according to another embodiment of the present invention will be described with reference to FIG. The radiation imaging apparatus 1800 is different from the radiation imaging apparatus 100 of FIG. 1 in that it further includes a memory 132 and an adder 133. Other configurations may be the same as those in the first embodiment, and thus redundant description is omitted.

  The memory 132 can store the radiation information output from the calculation unit 130 for one frame (Y in the above example), and outputs it to the adder 133 in the order of storage in the FIFO manner. The adder 133 subtracts the radiation information output from the memory 132 from the radiation information output from the calculation unit 130 and supplies the result to the determination unit 131. For example, when the radiation information X (y) obtained for the y-th initialization operation is output from the calculation unit 130, X (y) −X (y−Y) is supplied to the determination unit 131. That is, the determination unit 131 is supplied with a difference from the radiation information for one frame before.

  The inventors have found that although the amount of switching noise can be different for each drive line G, the same drive line G is highly reproducible. Therefore, as described above, by subtracting the radiation information previously obtained by driving the same drive line G from the radiation information obtained by driving a certain drive line G, switching noise is effectively obtained. Can be removed. This is called frame correction.

  In the above example, the radiation information of one frame before is subtracted from the current radiation information, but the radiation information of k frames (k is 2 or more) may be subtracted, or a plurality of frames for the same drive line G may be subtracted. You may subtract the average value (simple average or weighted average) of radiation information. Further, the memory 132 is arranged not between the calculation unit 130 and the determination unit 131 but between the detection circuit 120 and the calculation unit 130, and subtraction is performed at the stage of the bias current signal VSD supplied to the calculation unit 130. May be. In this way, when taking a difference between frames, switching noise is effectively removed by making the drive timing of the drive signal and the sample / hold timing of the A / D converter 340 uniform for each frame. sell. Therefore, a clock may be supplied from the common clock generator to the drive circuit 102 and the A / D converter 340.

  The radiation detection apparatus 1800 may include an integration circuit 500 illustrated in FIG. 21 between the adder 133 and the determination unit 131. The integration circuit 500 can include a shift register 501, adders 502a and 502b, and registers 503a and 503b that hold integration values. The determination unit 131 can include comparators 504a and 504b and an OR circuit 505. In FIG. 21, the latest radiation information output from the adder 133 is described as X.

First, the value X [n] stored in the shift register 501 and the integral value Sum [m] held in the registers 503a and 503b are initialized. This is called integrator reset. Next, every time a clock (not shown) is supplied to the shift register 501, X [n] stored in the shift register 501 is shifted. That is, the shift register 501 performs processing represented by the following formula.
X [n] = X [n-1] (n> 1)
X [n] = X (n = 0)

The adders 502a and 502b and the registers 503a and 503b for holding integrated values perform cumulative addition (integration) every time a clock (not shown) is given. That is, adders 502a and 502b and registers 503a and 503b perform processing represented by the following expression. The adder 502a and the register 503a constitute one integrator, and the adder 502b and the register 503b constitute another one integrator.
Sum [m] = X + Sum [m] −X [W [m]]

For example, assuming that the radiation information Xk is k clocks after the reset, the value of Sum [m] when W [m] = 4 changes as follows. Sum [m] is a value obtained by integrating the radiation information X in the integration interval W [m].
Sum [m] = 0 (immediately after reset)
Sum [m] = X1 (k = 1)
Sum [m] = X2 + X1 (k = 2)
Sum [m] = X3 + X2 + X1 (k = 3)
Sum [m] = X4 + X3 + X2 + X1 (k = 4)
Sum [m] = X5 + X4 + X3 + X2 (k = 5)
Sum [m] = X6 + X5 + X4 + X3 (k = 6)
...
Sum [m] = XK + XK-1 + XK-2 + XK-3 (k = K)

  That is, the integration value can be calculated in the integration interval determined by the tap (reading position) of the shift register 501. Since such calculation of the integral value is completed in one clock, the time required for the irradiation start determination can be greatly shortened.

  By arranging M integrators as described above, the m-th integrated value Sum [m] in the m-th integration interval W [m] can be obtained (m = 1 to M). Also, M comparators such as comparators 504a and 504b are arranged. The comparator compares the mth threshold value T [m] with the mth integral value Sum [m]. When the logical sum circuit 505 satisfies Sum [m]> T [m] in any of the comparators, it can be determined that radiation irradiation has started. If Sum [m]> T [m] is not satisfied in any of the comparators, it is determined that no radiation is applied.

  In the configuration example shown in FIG. 22, an adder and a comparator are provided for each register that holds an integral value, but this is only an example. For example, one adder and comparator may be shared by a plurality of registers. The shift register 501 can be configured by a single block, but may be divided into a plurality of blocks. The plurality of blocks constituting the shift register 501 can be mounted in different memory partitions on the FPGA, for example. The configuration shown in FIG. 22 may be realized by software.

  The determination unit 131 determines that radiation is being applied when the radiation information or the integrated value of the radiation information exceeds a predetermined threshold. Further, as shown in FIG. 10, a configuration in which CDS is performed on the added value of a plurality of effective values S and the added value of a plurality of noise values N can be used. These operations are out of order. However, in order to reduce the amount of data to be handled, signal processing may be performed in the order of addition, CDS, frame correction, and integration, and then comparison with a predetermined threshold value may be performed.

[Third Embodiment]
The overall configuration of a radiation imaging apparatus 1900 according to another embodiment of the present invention will be described with reference to FIG. The radiation imaging apparatus 1900 is different from the radiation imaging apparatus 100 of FIG. 1 in that a bias line Vs is arranged for each pixel column and a plurality of detection circuits 120 and reference bias potential generation circuits 126 are provided. A detection circuit 120 and a reference bias potential generation circuit 126 are provided for each bias line Vs. Other configurations may be the same as those in the first embodiment, and thus redundant description is omitted. Further, the radiation imaging apparatus 1900 may include a memory 132 between the calculation unit 130 and the determination unit 131 as in the second embodiment. Further, the radiation imaging apparatus 1900 has the detection circuit 120 and the reference bias potential generation circuit 126 for each of two or more bias lines Vs instead of having the detection circuit 120 and the reference bias potential generation circuit 126 for each bias line Vs. You may have.

  As described above with reference to FIG. 14, the time from when the drive signal rises until the second signal of the bias current reaches a peak is the time constant of the portion of the drive line G from the drive circuit 102 to the switch element T. Depends on. Therefore, in the present embodiment, the bias line Vs is connected in common to the pixels PIX arranged in the column direction in which the lengths of the drive lines G from the drive circuit 102 to the switch element T are equal to each other, and for each bias line Vs. A detection circuit 120 and a reference bias potential generation circuit 126 are provided. In this case, since the position of the peak of the second signal of the bias current may be different for each detection circuit 120, the delay time TD is determined for each detection circuit 120 by the above-described method, and sampling is performed. The bias current signals VSD output from each of the plurality of detection circuits 120 are added and supplied to the calculation unit 130. Since the subsequent processing is the same as that of the first embodiment, description thereof is omitted. In this embodiment, the S / N ratio of the bias current signal can be further improved.

[Fourth Embodiment]
Since the configuration of the radiation imaging apparatus according to the present embodiment may be the same as any of the radiation imaging apparatuses according to the first to third embodiments described above, the radiation imaging apparatus 100 will be described as an example. In the present embodiment, the radiation imaging apparatus 100 measures noise generated when radiation is not irradiated, and changes its own operation setting based on the amount of this noise. Here, the operation setting refers to the above-described on-time TH, off-time TL, sampling period TS, mathematical formula for calculating radiation information, the number of drive lines G to be driven simultaneously, and a threshold value that the determination unit 131 compares with radiation information. The setting related to the radiation detection operation of FIG.

  With reference to FIG. 20, the operation example of the radiation imaging device 100 in this embodiment is demonstrated. In step S2010, the control unit 106 starts noise measurement. The control unit 106 may start noise measurement in response to an input from the user of the radiation imaging apparatus 100, or may automatically start noise measurement after the radiation imaging apparatus 100 is turned on.

  In step S2020, the control unit 106 determines whether the noise measurement end condition is satisfied. The noise measurement end condition may be, for example, that a predetermined time has elapsed from the start of noise measurement, or that there is an input indicating the end from the user. When the noise measurement end condition is not satisfied (NO in S2020), the control unit 106 continues the initialization operation in S2030. The initialization operation here is the same as the initialization operation S370 in FIG. 3, but the determination unit 131 stores the radiation information supplied from the calculation unit 130 instead of determining the presence or absence of radiation irradiation. For example, the determination unit 131 can store the latest one frame of radiation information. Then, the process returns to S2020, and the control unit 106 determines whether the noise measurement end condition is satisfied.

  When the noise measurement end condition is satisfied (YES in S2020), in S2040, the control unit 106 determines whether or not the radiation information stored in the determination unit 131 is greater than a predetermined value. In the noise determination process of FIG. 20, since the radiation imaging apparatus 100 is not irradiated with radiation, the radiation information represents noise that has not been removed by the above-described CDS operation. In a state where the noise is large, even if the processing of FIG. Therefore, when the noise is larger than the predetermined value (YES in S2040), the control unit 106 changes the operation setting in S2050. For example, the control unit 106 may increase the threshold used by the determination unit 131 for comparison with the radiation information in S320, or may reduce the influence of external noise according to the detected frequency characteristics of the noise. The driving cycle of 102 may be changed. When the noise is smaller than the predetermined value (NO in S2040), the control unit 106 proceeds to S310 in FIG. 3 and starts an imaging standby start state. In this case, the control unit 106 may change the operation setting. For example, the control unit 106 may decrease the threshold used by the determination unit 131 for comparison with radiation information in S320.

  After the operation setting is changed in S2050, the process may return to S2020 and noise measurement may be repeated. In S2040, when the noise value greatly exceeds the predetermined value, the control unit 106 may present this result to the user. In this case, the control unit 106 may prohibit the transition to the imaging standby state.

  FIG. 22 is a diagram showing an example in which the radiation imaging apparatus according to the present invention is applied to an X-ray diagnostic system (radiation imaging system). The radiation imaging system includes a radiation imaging apparatus 6040 (corresponding to the radiation imaging apparatuses 100, 1800, and 1900) and an image processor 6070 that processes a signal output from the radiation imaging apparatus 6040. X-rays 6060 generated by an X-ray tube (radiation source) 6050 pass through the chest 6062 of the patient or subject 6061 and enter the radiation imaging apparatus 6040. This incident X-ray includes information inside the body of the subject 6061. The image processor (processor) 6070 processes a signal (image) output from the radiation imaging apparatus 6040, and can display an image on the display 6080 of the control room based on the signal obtained by the processing, for example.

  Further, the image processor 6070 can transfer a signal obtained by the processing to a remote place via the transmission path 6090. Accordingly, an image can be displayed on a display 6081 disposed in a doctor room or the like in another place, or an image can be recorded on a recording medium such as an optical disk. The recording medium may be a film 6110. In this case, the film processor 6100 records an image on the film 6110.

  Some operations of the above-described embodiments can be realized by a computer executing a program. Also, means for supplying a program to a computer, for example, a computer-readable recording medium such as a CD-ROM recording such a program, or a transmission medium such as the Internet for transmitting such a program is also applied as an embodiment of the present invention. Can do. The above program can also be applied as an embodiment of the present invention.

DESCRIPTION OF SYMBOLS 100 Radiation imaging device, 101 Pixel array, 102 Drive circuit, 105 Signal processing circuit, 106 Control part

Claims (18)

A pixel array in which a plurality of pixels including a conversion element that converts radiation into electric charge and stores and a switch element that connects the conversion element to a signal line are arranged to form a plurality of rows and a plurality of columns;
A bias line for applying a bias potential to the conversion elements of the plurality of pixels;
A plurality of drive lines connected to control terminals of the switch elements of the plurality of pixels;
Switching a driving unit for supplying a driving signal to one or more drive lines your capital, each drive signal from the off-voltage to the switching element non-conductive, the ON voltage of the switching element to a conducting state , A drive unit that repeats the initialization operation to return to the off-voltage,
A current flows through the bias line every driving cycle from when a driving signal to one or more driving lines is switched to the on voltage to when a driving signal to the next one or more driving lines is switched to the on voltage. An acquisition unit that samples a signal representing a current with a cycle shorter than the drive cycle, acquires a signal sampled after a predetermined delay time has elapsed since switching the drive signal to the on-voltage, and obtains a drive signal An acquisition unit for acquiring a signal sampled after a delay time has elapsed since switching to the off-voltage as a noise value;
A calculation unit for calculating radiation information based on the effective value and the noise value ;
A radiation imaging apparatus comprising: a determination unit that determines presence / absence of radiation irradiation to the pixel array based on the radiation information. The driver, after returning the drive signal to have more than one drive line to the off-voltage, at a time interval, then switch the driving signal to the next one or more drive lines to the on-voltage The radiation imaging apparatus according to claim 1. The calculating unit, based an the effective value obtained for driving cycle in, in at least one of said noise values obtained for driving the immediately preceding cycle of the noise value and the driving cycle obtained for the drive cycle the radiation imaging apparatus according to claim 1 or 2, characterized in that to calculate the radiation information. The calculation unit calculates the radiation information based on a difference between a value calculated from one or more effective values and a value calculated from one or more noise values,
The determination unit, the radiation imaging apparatus according to any one of claims 1 to 3, characterized in that the determination by comparing the radiation information with the threshold value. The acquisition unit acquires the noise value a plurality of times for each driving cycle,
The radiation imaging apparatus according to claim 4, wherein the calculation unit calculates a value calculated from the one or more noise values by adding a weight to the plurality of noise values.   The radiation imaging apparatus according to claim 5, wherein the calculation unit assigns a greater weight to a noise value acquired at a time closer to an effective value used for calculating the radiation information. The acquisition unit acquires the effective value a plurality of times for each driving cycle,
The radiation imaging according to claim 4, wherein the calculation unit calculates a value calculated from the one or more effective values by adding a weight to the plurality of effective values. apparatus. The said calculation part calculates the said radiation information based on the said effective value and the noise value acquired at the time interval for acquisition of the said effective value, The any one of Claim 1 thru | or 7 characterized by the above-mentioned. The radiation imaging apparatus according to item 1. The drive unit switches a drive signal to one or more drive lines to the off voltage, and switches a drive signal to the next one or more drive lines to the on voltage.
The said calculation part calculates the said radiation information based on the said effective value acquired about a certain driving cycle, and the said effective value acquired about the driving cycle different from the said driving cycle. The radiation imaging apparatus described in 1. The radiation imaging according to any one of claims 1 to 9, wherein the predetermined delay time is determined based on at least one of a time constant of the drive line and a time constant of the bias line. apparatus. A plurality of the bias lines, and the acquisition unit for each of the plurality of bias lines,
The radiation imaging apparatus according to claim 10 , wherein the predetermined delay time is determined for each acquisition unit. A memory for storing the radiation information;
The determination unit calculates the radiation information calculated for a certain driving cycle and one or more driving lines to which the on-voltage is supplied in the driving cycle, in a driving cycle prior to the driving cycle, the radiation imaging apparatus according to any one of claims 1 to 11, characterized in that to determine the presence or absence of irradiation of radiation to the pixel array based on the difference between the radiation information stored in the memory. The acquisition unit includes an A / D converter,
Said effective value and said noise value, claims 1 to 12, characterized in that said A / D converter analog signal values representative of the current flowing through the bias line is a digital signal value obtained by sampling The radiation imaging apparatus according to any one of the above. The driving unit is configured such that one or more drive lines to which the on-voltage is supplied in a certain driving cycle and one or more driving lines to which the on-voltage is supplied in the next driving cycle are not adjacent to each other. the radiation imaging apparatus according to any one of claims 1 to 13 and supplying a driving signal. The drive unit sequentially switches the drive signal to the odd-numbered drive lines counted from one side of the pixel array to the on-voltage, and then sends the drive signal to the even-numbered drive lines. the radiation imaging apparatus according to any one of claims 1 to 14, wherein the switching to sequentially the on-voltage. The drive unit, the radiation imaging apparatus according to any one of claims 1 to 15, characterized in that for supplying a driving signal every two or more drive lines. When the determination unit determines that radiation to the pixel array is illuminated, any one of claims 1 to 16 wherein the drive unit is characterized in that to end the switching to the on-voltage of said drive signal The radiation imaging apparatus according to item 1. A radiation imaging apparatus according to any one of claims 1 to 17 ,
A radiation imaging system comprising: a processor that processes a signal output from the radiation imaging apparatus.