LABR3 SCINTILLATION DETECTOR AND SPECIFIC EVENT REMOVAL METHOD

Abstract

The present invention identifies ? decay and other events included in the emission of an LaBr.sub.3 scintillator and only collects ? ray events. An LaBr.sub.3 scintillation detector is provided with an LaBr3 scintillator 10, a photomultiplier tube 12, an oscilloscope 14, and a computer 18. The computer 18 detects a peak value Vp and a total charge amount Q.sub.total of a voltage waveform signal and calculates an error propagation expression function for a ratio of the peak value Vp to the total charge amount Q.sub.total. This error propagation expression function is used as a threshold function for identifying and removing ? decay events. The ? decay events are identified from the peak value Vp and total charge amount Q.sub.total, which are measurement values that can be measured in real time.

Claims

1. An LaBr.sub.3 scintillation detector comprising: an LaBr.sub.3 scintillator; a photoelectric converter that converts light emitted from the LaBr.sub.3 scintillator into an electric signal; a waveform signal output unit that converts an output from the photoelectric converter into a voltage waveform signal; a detecting unit that detects a peak value V.sub.p and a total charge amount Q.sub.total of the voltage waveform signal; a calculating unit that calculates an error propagation expression function of a ratio of the peak value V.sub.p and the total charge amount Q.sub.total; and a processing unit that uses the error propagation expression function as a threshold function to specify an event other than ? rays and rejects the event.

2. The LaBr.sub.3 scintillation detector according to claim 1, further comprising: a low-pass filter that removes a high frequency component of the voltage waveform signal output from the waveform signal output unit.

3. The LaBr.sub.3 scintillation detector according to claim 1, wherein the calculating unit calculates a standard deviation by correcting the peak value V.sub.p detected by the detector to become linear with respect to the total charge amount Q.sub.total.

4. The LaBr.sub.3 scintillation detector according to claim 1, wherein the calculating unit calculates the error propagation expression function in an energy range of 1.5 MeV or less.

5. The LaBr.sub.3 scintillation detector according to claim 1, wherein the processing unit uses an error propagation expression function of 3? as the threshold function.

6. The LaBr.sub.3 scintillation detector according to claim 1, wherein the ratio of the peak value V.sub.p and the total charge amount Q.sub.total is V.sub.p/Q.sub.total.

7. The LaBr.sub.3 scintillation detector according to claim 1, wherein the calculating unit corrects the peak value V.sub.p detected by the detector to become linear with respect to the total charge amount Q.sub.total and calculates the error propagation expression function in an energy range of 1.5 MeV or less of V.sub.p/Q.sub.total which is a ratio of the corrected peak value V.sub.p and the total charge amount Q.sub.total, and the processing unit uses an error propagation expression function of 3? as the threshold function.

8. A method of rejecting a specific event of an LaBr.sub.3 scintillator, the method comprising: converting an emission of the LaBr.sub.3 scintillator into a voltage waveform signal and outputting the voltage waveform signal; detecting a peak value V.sub.p and a total charge amount Q.sub.total of the voltage waveform signal; calculating a standard deviation of a ratio of the peak value V.sub.p and the total charge amount Q.sub.total and calculating an error propagation expression function of the standard deviation in an energy range of a predetermined value or less which does not include an event other than ? rays in light emitted from the scintillator; and specifying an event in an energy range of the predetermined value or more using the error propagation expression function as a threshold function and rejecting the event.

Description

BRIEF DESCRIPTION OF DRAWINGS

[0020] FIG. 1 is a configuration diagram of a scintillation detector of an exemplary embodiment.

[0021] FIG. 2 is a waveform signal diagram of self-radioactivity of an LaBr.sub.3:Ce scintillator.

[0022] FIG. 3 is an energy spectrum diagram of the LaBr.sub.3:Ce scintillator.

[0023] FIG. 4 is a plot of V.sub.p/Q.sub.total.

[0024] FIG. 5 is a diagram illustrating a standard deviation ?.sub.Vp/Qtotal of V.sub.p/Q.sub.total and an error propagation expression function.

[0025] FIG. 6 is a diagram in which a threshold function is applied to FIG. 4.

[0026] FIG. 7 is a diagram illustrating a result of rejecting an ? ray event.

[0027] FIG. 8 is a diagram illustrating an accidental rejection rate of an ? ray event.

[0028] FIG. 9 is a histogram diagram (part 1) of a BG subtraction method using an external radiation source and a method of the exemplary embodiment.

[0029] FIG. 10 is a histogram diagram (part 2) of the BG subtraction method using an external radiation source and the method of the exemplary embodiment.

[0030] FIG. 11 is a diagram illustrating a result of rejecting an ? ray event.

[0031] FIG. 12 is an energy spectrum diagram (part 2) of the LaBr.sub.3:Ce scintillator.

[0032] FIG. 13 is a schematic view for explaining ? ray event rejection.

DESCRIPTION OF EMBODIMENTS

[0033] Hereinafter, embodiments of the present disclosure will be described.

[0034] <Overall Configuration>

[0035] FIG. 1 is a configuration diagram of a scintillation detector of an exemplary embodiment. The scintillation detector includes an LaBr.sub.3:Ce scintillator 10, a photomultiplier 12, an oscilloscope 14, a hard disk drive 16, and a computer 18.

[0036] The LaBr.sub.3:Ce scintillator 10 is a scintillator that converts ionizing radiation such as a ? ray into light, and is formed, for example, in a cylindrical shape of 1.5 inch ??1.5 inch. The LaBr.sub.3:Ce scintillator 10 is a scintillator that is excellent in stopping power, energy resolution, and time resolution, but as described above, it always outputs a background signal according to a radionuclide contained therein.

[0037] The photomultiplier tube 12 is connected to the LaBr.sub.3:Ce scintillator 10 so as to convert the light of the LaBr.sub.3:Ce scintillator 10 into an electrical signal corresponding to the intensity and output the electrical signal.

[0038] The oscilloscope 14 converts the detected electric signal into a voltage signal (waveform signal) along the time axis and outputs the voltage signal.

[0039] The hard disk drive 16 is connected to the oscilloscope 14 via a USB interface or the like, and stores the waveform signal (raw waveform signal not subjected to waveform shaping or the like) output from the oscilloscope 14.

[0040] The computer 18 functions as a detecting unit, a calculating unit, and a processing unit in the present exemplary embodiment, so that the waveform signal stored in the hard disk drive 16 is input to the computer 18 and the computer 18 analyzes the waveform signal and outputs the analysis result. The computer 18 has a CPU and a program memory, and a predetermined processing program is stored in the program memory. The computer 18 reads the processing program stored in the program memory and sequentially executes the processing program to analyze the waveform signal. The analysis of the waveform signal in the present exemplary embodiment includes the following processes:

[0041] (a) a process of filtering the waveform signal

[0042] (b) a process of detecting a total integrated value Q.sub.total of charge and a peak value Vp of the voltage and calculating Vp/Q.sub.total

[0043] (c) a process of determining a threshold function dependent on energy

[0044] (d) a process of rejecting an ? ray event using the threshold function

[0045] In FIG. 1, the hard disk drive 16 and the computer 18 may be implemented as one waveform analyzing device, and the oscilloscope 14, the hard disk drive 16, and the computer 18 may be implemented as one waveform analyzing device.

[0046] Further, the computer 18 may have a function of counting the number of events for data in which the ? ray events have been rejected, converting the count value into a radiation dose, and outputting the radiation dose. However, description thereof will be omitted, as this function is well known.

[0047] Next, the above-mentioned processes will be described in order.

[0048] <Filtering Process>

[0049] FIG. 2 is a waveform signal diagram measured at events including a self-radioactive event of the LaBr3:Ce scintillator 10. The waveforms are obtained by the oscilloscope 14, and no external radiation source is used. In the figure, the horizontal axis represents time (ns) and the vertical axis represents voltage (V).

[0050] The uppermost waveform signal is a waveform signal of raw data, and the following are obtained:

[0051] Peak value Vp=?0.113 (V)

[0052] Total charge amount (total integrated value of charge) Q.sub.total=4.219

[0053] However, the raw data has high noise, and even though V.sub.p/Q.sub.total is calculated, its accuracy is low.

[0054] The central waveform signal is a waveform signal obtained by performing a moving average process of the raw data with a time width of 2 ns, and following are obtained: V.sub.p=?0.099 (V)

[0055] Total charge amount Q.sub.total=4.213

[0056] The lowermost waveform signal is a waveform signal which is subjected to noise removal by a low-pass filter that removes high frequency noise of 50 MHz or more by FFT and IFFT (inverse FFT), and the following are obtained:

[0057] Vp=?0.099 (V)

[0058] Total charge amount Qtotal=4.214

[0059] The peak value Vp and the total charge amount Q.sub.total may be detected with high accuracy by performing a moving average process or a low pass filter process, particularly a low pass filter process to shape the waveform.

[0060] In comparison of these three sets of V.sub.p and Q.sub.total, for V.sub.p, the same value is obtained in the moving average process and the low-pass filter process, and a value different from that value is obtained in the raw data. On the other hand, however, for Q.sub.total, substantially the same value is obtained in the three sets of data. Q.sub.total is the total charge amount and corresponds to energy, suggesting that there is no major change in the energy distribution between the three sets of data.

[0061] FIG. 3 illustrates an energy spectrum of the lowest waveform signal in FIG. 2; that is, an energy spectrum when high-frequency noise is removed by the low-pass filter of FFT and IFFT. The solid line indicates an energy spectrum of the raw data, and the broken line indicates an energy spectrum when low-pass filtered. It can be seen from this figure that the energy spectrum does not change before and after noise removal. This means that data loss does not occur even when the low pass filter process is performed on the raw data.

[0062] <Calculating Process of Vp/Q.sub.total>

[0063] FIG. 4 illustrates a plotted result of the ratio Vp/Q.sub.total of the peak value Vp and the total charge amount Q.sub.total of the waveform signal (low-pass filtered) for the data of 100,000 events of light emission by the self-radioactivity of the LaBr.sub.3:Ce scintillator 10. In the figure, the horizontal axis represents the total charge amount Q.sub.total and the corresponding energy (MeV), and the vertical axis represents V.sub.p/Q.sub.total.

[0064] As is apparent from part (a. 1) in FIG. 4, two components are present in the energy range of 1.5 to 3 MeV. It is known that the peak value V.sub.p of the waveform signal of ? ray events is larger than the peak value of the waveform signal of ? ray events (see the related art). Accordingly, the upper component in part (a. 1) of FIG. 4 corresponds to the ? ray events and the lower component corresponds to the ? ray events. The low energy portion is considered to be ? ray events by .sup.208Tl 2.6 MeV, which is an environmental radionuclide.

[0065] Further, since each of V.sub.p and Q.sub.total has a linear relationship with respect to energy, the ratio Vp/Q.sub.total should be constant in the entire energy range. However, the ratio Vp/Q.sub.total is not constant and tends to decrease with increasing energy. This is considered to be due to the saturation of the peak value V.sub.p. The present inventors have found that, when plotting the horizontal axis as Q.sub.total and the vertical axis as V.sub.p, the linearity of V.sub.p and Q.sub.total is maintained in the low energy region (1.5 MeV or less), whereas the linearity is not maintained in the high energy region, and V.sub.p tends to be saturated.

[0066] Therefore, in order to correct the saturation of V.sub.p, a saturation curve of V.sub.p is defined as follows:

[00001]$\begin{array}{cc}{V}_p=\frac{?.Math..Math.Q_{\mathrm{total}}}{1+?.Math..Math.Q_{\mathrm{total}}}.Math.\left(?=\mathrm{Const}.,?=\mathrm{Const}.\right)& \left[\mathrm{Eq}..Math.1\right]\end{array}$

[0067] Here, ?Q.sub.total is a linear term, and 1+?Q.sub.total is a saturation term. Then, assuming that the corrected V.sub.p (taken as V.sub.pCorr) should be proportional to Q.sub.total; that is, V.sub.pCorr=?Q.sub.total, V.sub.p is corrected to V.sub.pCorr by the following equation:

[00002]$\begin{array}{cc}{V}_{\mathrm{pCorr}}=\frac{V_p}{1-\left(?/?\right).Math.V_p}& \left[\mathrm{Eq}..Math.2\right]\end{array}$

[0068] In part (a.2) of FIG. 4, a plotted result of V.sub.p/Q.sub.total when corrected V.sub.p is used is illustrated. The obtained result is that V.sub.p/Q.sub.total using the corrected V.sub.p becomes substantially constant in the entire energy range. Coefficients ? and ? in the above equation may be determined experimentally.

[0069] In FIG. 4, the ? ray events of 1.5 MeV or less have larger variations as the energy is lower, but this is theoretically derived from error propagation. That is, standard deviation ?V.sub.p/Q.sub.total (hereinafter referred to as ?) of V.sub.p/Q.sub.total is:

[00003]$\begin{array}{cc}{?}_{V_p/Q_{\mathrm{total}}}=\sqrt{{\left(\frac{1}{Q_{\mathrm{toal}}}.Math.{?}_{V_p}\right)}^2+{\left(\frac{V_p}{Q_{\mathrm{total}}^2}.Math.{?}_Q\right)}^2}& \left[\mathrm{Eq}..Math.3\right]\end{array}$

[0070] The above-described equation may be approximated to the following equation:

?.sub.V.sub.p.sub./Q.sub.total=kQ.sub.total.sup.?1+l(k=const.,l=const.)[Eq. 4]

[0071] As is apparent from this equation, the standard deviation ?V.sub.p/Q.sub.total of V.sub.p/Q.sub.total increases with decreasing Q.sub.total; that is, with decreasing energy.

[0072] <Determining Process of Threshold Function>

[0073] In FIG. 4, the threshold function for identifying the ? ray events and the ? ray events may be determined by using the standard deviation a of pure ? ray events of 1.5 MeV or less.

[0074] FIG. 5 illustrates ?, 2?, and 3? calculated for V.sub.p/Q.sub.total using the corrected V.sub.p, and its error propagation expression function. In the figure, the horizontal axis represents energy, which is an energy range of 1.4 MeV or less (i.e., an energy range of only the ? ray events), and the vertical axis represents the standard deviation a of V.sub.p/Q.sub.total. From this figure, it can be seen that the variation of V.sub.p/Q.sub.total is very well reproduced by the error propagation expression function. Therefore, the ? ray events present in the energy range of 1.5 MeV or more may be clearly identified by determining the threshold function from this error propagation expression function.

[0075] FIG. 6 is a diagram in which the threshold function is applied to the plotting of V.sub.p/Q.sub.total using the corrected V.sub.p illustrated in part (a.2) of FIG. 4. The threshold function is inversely proportional to Q.sub.total and represents energy dependence. The threshold functions of ?, 2?, and 3? are illustrated in the figure, but the ? ray events and the ? ray events may be clearly identified, especially by using the threshold function of 3?. Therefore, it is possible to reject the ? ray events using the threshold function.

[0076] It should be noted that the threshold functions of ? to 3? are unambiguously and objectively determined from the data group of V.sub.p/Q.sub.total at 1.5 MeV or less.

[0077] <Rejection Process of ? Ray Events>

[0078] FIG. 7 is a diagram illustrating the result of rejecting the ? ray events using three kinds of threshold functions (threshold functions of ?, 2?, and 3?). In the figure, part (a. 1) is a result when the threshold function of a is used, part (a.2) is a result when the threshold function of 2? is used, and part (a.3) is a result when the threshold function of 3? is used. In the figure, the horizontal axis represents energy (MeV), and the vertical axis represents event counts. In the case of the energy range of 1.5 MeV or more, the event counts are magnified by a factor often times. In the energy range of 1.5 MeV and more, the solid line indicates event counts after the ? ray events are rejected, and the ? ray events are rejected in all the threshold functions.

[0079] FIG. 8 is a diagram illustrating an accidental rejection rate of the ? ray events: that is, a rejection rate of the ? ray events that should not be rejected, in each of the three kinds of threshold functions. In the figure, the horizontal axis represents energy in a range of 0.3 MeV to 1.5 MeV; that is, an energy range in which only the ? ray events occur. It is understood that as the threshold function becomes larger in the manner of ?, 2?, and 3?, the accidental rejection rate decreases drastically, so that only the ? ray events are correctly rejected. Specifically, when the threshold function of 3? is used, the accidental rejection rate is approximately 1% or less in the whole region of 1.5 MeV or less, and as a whole, the accidental rejection rate of about 0.716% is obtained. In other words, in this exemplary embodiment, it is not meant that the ? ray events are not eliminated at all, but it is meant that the ? ray events may be somewhat rejected.

[0080] As described above, it is possible to determine a threshold function for identifying ? ray events using only the self-radioactivity of the LaBr.sub.3:Ce scintillator 10. In addition, since the threshold function has a mathematical basis and does not include parameters to be set artificially, it may be determined unambiguously and objectively. Specifically, when the threshold function is determined as a function of energy, it is possible to dramatically improve the accuracy of identifying the ? ray events. Furthermore, the accidental rejection rate of the ? ray events of 1.5 MeV or less may be set to about 0.7% by using the threshold function of 3?.

[0081] Next, descriptions will be made on a case where a measurement is performed using an external radiation source, in order to confirm whether or not the ? ray events can be rejected correctly.

[0082] For example, Ge/Ga-68 (.sup.68Ga 1.883 MeV) is used as the external radiation source. At this time, since contribution from environmental radiation (.sup.208Tl 2.61 MeV, etc.) may exist, the ? ray events and the ? ray events are mixed in at 1.5 MeV to 3 MeV. Therefore, after the ? ray events are rejected using the 3? threshold function in the above-described method, an evaluation is conducted in order to determine whether or not the correct number of ? ray events (.sup.68 Ga 1.883 MeV), which is previously known, remains.

[0083] Specifically, a result obtained by the highly reliable background (BG) subtraction method is compared with the result obtained by the exemplary embodiment, and an evaluation is conducted as to whether or not there is a difference between the two results. That is, by comparing distribution of ? rays of 1.883 MeV remaining after subtracting the measurement result in the absence of an external radiation source from the measurement result in the presence of an external radiation source (background BG subtraction method) and distribution of ? rays of 1.883 MeV remaining after using V.sub.p/Q.sub.total and the threshold function of the exemplary embodiment in the presence of an external radiation source, it is evaluated whether the ? ray events can be rejected correctly by the method of the exemplary embodiment.

[0084] FIG. 9 also illustrates energy spectra with and without an external radiation source. In part (a), the solid line indicates an energy spectrum in the case where an external radiation source is present, the broken line indicates an energy spectrum in the case where an external radiation source is not present, and the vertical axis represents normalized event counts. The event counts of .sup.68Ga 1.883 MeV may be measured by subtracting the latter from the former.

[0085] Meanwhile, part (b) of FIG. 9 illustrates energy spectra in the state where an external radiation source is provided and after the ? ray events are rejected by the method of the exemplary embodiment. In part (b), the solid line indicates the energy spectrum in the state where an external radiation source is present, and the broken line indicates the energy spectrum after the ? ray events are rejected by the threshold function of 3?. As illustrated in part (b), even when the ? ray events are rejected by the threshold function of 3?, 1.883 MeV components remain without being rejected.

[0086] FIG. 10 is a diagram comparing the case of subtracting background (BG) with the case of the exemplary embodiment. The horizontal axis represents energy in the energy range around 1.883 MeV. The vertical axis represents normalized event counts. Part (C. 1) is a result of the BG method which is obtained by subtracting the case of not setting an external radiation source from the case of setting an external radiation source, part (c.2) is a result of the present exemplary embodiment, and part (c.3) is a result of comparing the two.

[0087] The peak count numbers (including measured value and fitting value), average energy, full width at half maximum (FWHM), and difference in each case are as follows.

[0088] Peak Count Number (Measured Value)

[0089] BG method: 139.6?28.54

[0090] Exemplary embodiment: 138.6?19.22

[0091] Difference: 0.716%

[0092] Peak Count Number (Fitting Value)

[0093] BG method: 129.9?6.34

[0094] Exemplary embodiment: 127.7?4.46

[0095] Difference: 1.694%

[0096] Average Energy (keV)

[0097] BG method: 1885.9?0.7334

[0098] Exemplary embodiment: 1885.5?0.5281

[0099] Difference: 0.021%

[0100] FWHM (keV)

[0101] BG method: 33.1?3.62 (1.755%)

[0102] Exemplary embodiment: 36.2?3.31 (1.920%)

[0103] Difference: 0.165%

[0104] From the above results, it is understood that in the exemplary embodiment the same result as in the BG method may be obtained with higher accuracy.

[0105] In the exemplary embodiment, there may be a counting loss of certain events due to the dead time required to acquire the waveform signal. For example, there may be a counting loss of a .sup.215Po short-lived daughter nucleus that decays in a cascade manner from .sup.219Rn and the like. However, since the rejection of the measured events may be reliably executed, this counting loss does not contribute to the accuracy evaluation.

[0106] Further, in the exemplary embodiment, since .sup.208Tl 2.6 MeV, which is an environmental radionuclide, may be contained as described above, nuclides of the same series may be mixed as well. FIG. 11 is a diagram illustrating a result when ? ray events are rejected using V.sub.p/Q.sub.total and the threshold function of 3? by further increasing the statistical number. Very few events of .sup.212, 214Bi have been identified. In this regard, the method of the exemplary embodiment also has the effect of being able to remarkably detect even a small number of ? ray events.

[0107] As described above, according to the exemplary embodiment, only ? ray events may be collected by identifying ? ray events using a measured value that can be actually measured in real time such as a peak voltage of the signal and a total charge amount, and rejecting the ? ray events. Further, in the exemplary embodiment, it is possible to obtain the same result as that in the background (BG) subtraction method with higher accuracy. Further, in the exemplary embodiment, since measurement may be performed independently of the S/N ratio of the measurement target and the ? decay background, it is also suitable for detection of very small signals. In the exemplary embodiment, attention is paid particularly to the ? decay events included in the self-radioactivity, but it is applicable not only to self-radioactivity but also to ? rays and heavy particle rays incident from the outside. That is, the exemplary embodiment is not necessarily limited to self-radioactivity and may be applied to the rejection of specific events that may exist in a specific energy range.

[0108] In the exemplary embodiment, since data are saved over several seconds for each event, the dead time tends to increase, but the dead time may be reduced by using a high-speed ADC for flash analog to digital converter (FADC) that acquires a signal waveform at high speed.

[0109] It is also possible to extract only the ? ray spectrum in real time by installing a processing program for implementing a process in the exemplary embodiment in a field programmable gate array (FPGA).

[0110] In the exemplary embodiment, ? ray events are identified using V.sub.p/Q.sub.total, but it goes without saying that Q.sub.total/V.sub.p, which is the reciprocal thereof, may be used.

REFERENCE SIGNS LIST

[0111] 10 LaBr.sub.3:Ce scintillator, 12 photomultiplier tube, 14 oscilloscope, 16 hard disk, 18 computer.