TW201809722A - Counting method and radiation detection apparatus - Google Patents

Abstract

To provide a counting method for accurately counting radiative rays, etc., over a wide range from the state of a low counting rate to the state of a high counting rate. Provided are: a method for setting a wave-height discriminating threshold value, and counting the frequency of signals exceeding the wave-height discriminating threshold value, wherein, when a counting objective component shows a peak in a wave-height distribution spectrum, an arbitrarily defined value selected from the range of ([mu]-0.4[sigma]) to ([mu]+0.6[sigma]) is set as the wave-height discriminating threshold value, where [sigma] represents the standard deviation and [mu] represents the average wave-height value obtained by fitting a Gaussian function for the peak; and a radiation detection apparatus which processes signals by using the counting method. The counting method and the radiation detection apparatus according to the present invention enable reduction of an error given to a count value due to pile-up, and enable accurate counting of radiative rays, etc., over a wide range from the state of a low radiation dosage rate to the state of a high radiation dosage rate.

Description

Counting method and radiation detection device

The present invention relates to a method for counting radiation and the like and a radiation detection device. Specifically, the present invention relates to a method and a radiation detection device for counting radiation with high accuracy even under a high count rate condition.

In a radiation detection device that combines a radiation detection element such as a scintillator, a semiconductor, and a wave height recognition circuit, a pulse signal corresponding to the radiation energy detected by the radiation detection element is obtained, and its wave height value is read and counted. Pulse signals with wave height values greater than the wave height identification threshold to obtain the dose rate are usually performed extensively.

For example, in the case of obtaining a dose rate neutron detection device for a neutron ray, it is necessary to set the wave height value of the γ-ray that is not included in the background noise as the wave height identification threshold. As a method for optimizing the threshold value of wave height recognition, as described in Patent Document 1, there are μ (average value of wave height values) and σ obtained by fitting a Gaussian function to a peak displayed in a wave height distribution spectrum. (Standard deviation of wave height value) A method of setting the reference.

Prior art literature

Patent literature

[Patent Document 1] Japanese International Publication 2014/192321

[Patent Document 2] Japanese Patent No. 5611357

[Patent Document 3] Japanese Patent Laid-Open No. 2015-227854

[Patent Document 4] Japanese Patent Application Laid-Open No. 2012-136667

[Patent Document 5] Japanese Patent Laid-Open No. 2013-167467

[Patent Document 6] Japanese International Publication 2015/128905

[Patent Document 7] Japanese Special Table 2001-524217

For example, in the radiation treatment of cancer, the environment in which the radiation detecting element is exposed has a very high radiation dose rate. In such a high-dose-rate environment, as described in Patent Documents 2 and 3, the signal waveform overlaps due to the next pulse incident (so-called pile up) immediately after the measurement of a certain pulse, resulting in a count loss ( counting loss) and low counting accuracy.

Patent Literature 2 and Patent Literature 3 disclose a method of correcting a change in wave height value by using a baseline fluctuation while digitizing pulses by an AD converter and separating and counting a plurality of accumulated pulses. However, according to the embodiment, there is a case where the accumulated plurality of pulses cannot be separated and a counting loss occurs. Two pulses (second radiation detection phenomenon) is a case of a radiation detection phenomenon that is processed as one pulse (first radiation detection phenomenon). The wave height distribution spectrum is, for example, as shown in FIG. 1 and is about 2 in the first peak. A double peak value can be observed at a doubled wave height value. In addition, the wave height distribution spectrum of the non-stacked state is shown in Fig. 2 and the second peak cannot be observed.

In addition, when the stacking frequency is further increased (when the dose rate is high or the sensitivity of the detection element is high), there is a radiation detection phenomenon in which three pulses are processed as one pulse, as shown in the wave height distribution in FIG. 3 Spectrum, also The third peak may be observed at a wave height value approximately three times that of the first peak.

In this way, when a plurality of pulses are processed as one pulse, a counting loss occurs and the counting accuracy is low. Therefore, it is necessary to reduce the radiation detection sensitivity by miniaturizing the radiation detection element and the like to reduce the stacking frequency. However, there are other problems caused by reducing the sensitivity of radiation detection, so it must be explored from various aspects.

As an example, as described in Patent Document 4, a radiation detection element having a short decay time (fluorescence lifetime) of a pulse is used.

As an example, work is performed on the circuit to shorten the decay time of the pulse (for example, an amplifier (Amp) that makes the time constant shorter, or using a differential circuit).

As an example, the speed of the AD converter is increased and the sampling points of the data are increased.

However, in this way, there is a case where the problem of low counting accuracy due to accumulation cannot be solved.

In view of the above-mentioned problems, the present inventors conducted intensive studies and found that the method described above is completely different, that is, the optimization of the wave height recognition threshold can be used to suppress the decrease in counting accuracy caused by accumulation, and completed the present invention. invention.

That is, the present invention is a counting method and a radiation detection device, wherein the counting method is a method of setting a wave height recognition threshold value and counting a signal frequency greater than the wave height recognition threshold value, and is characterized in that the purpose component of the counting is in When the wave height distribution spectrum shows a spike, a Gaussian function is performed for the aforementioned spike When the average wave height value obtained by numerical fitting is set to μ and the standard deviation is set to σ, an arbitrary value selected from a range of (μ-0.4σ) to (μ + 0.6σ) is set as the aforementioned wave height identification threshold. value.

According to the present invention, by a simple method of optimizing the threshold value of the wave height recognition, it is possible to suppress a decrease in counting accuracy due to accumulation, and it is possible to accurately perform a wide range from a low radiation count rate to a high status. The radiation is counted well. It is also possible to simplify the circuit and shorten the analysis time.

1‧‧‧ radiation detection element

2‧‧‧wave height recognition circuit

3‧‧‧ scintillator

4‧‧‧ light detector

5‧‧‧ fiber

Fig. 1 is a wave height distribution spectrum (medium count rate condition) obtained when an accelerator neutron is counted by using a Eu: LiCaAlF ₆ crystal neutron detection device.

Fig. 2 is a wave height distribution spectrum (low count rate condition) obtained when an accelerator neutron is counted by using a Eu: LiCaAlF ₆ crystal neutron detection device.

Fig. 3 is a wave height distribution spectrum (high count rate condition) obtained when an accelerator neutron is counted by using a Eu: LiCaAlF ₆ crystal neutron detection device.

Fig. 4 is a schematic diagram of a radiation detection apparatus of the present invention.

Fig. 5 is a signal waveform diagram for explaining the influence caused by the accumulation.

Fig. 6 is a diagram (simulation) illustrating a method of generating a wave height distribution spectrum.

Fig. 7 is a wave height distribution spectrum (simulation) displayed during stacking.

Fig. 8 is a count value (simulation) above the wave height identification threshold value for analyzing the wave height distribution spectrum displayed during stacking.

Fig. 9 is the simulation result of the error caused by the accumulation on the count value.

Fig. 10 is a measurement of the counting error in the most piled-up condition of Example 1. Comparison of values to analog values.

FIG. 11 is a diagram illustrating that the error caused by the stacking count value can be reduced by setting the wave height recognition threshold near μ.

Fig. 12 is a schematic diagram of a neutron detection device used in the embodiment.

FIG. 13 is a graph of the timing of the neutron count rate and the accelerator current value obtained in Example 1. FIG.

FIG. 14 is a graph showing the relationship between the accelerator current value and the neutron count rate in Example 1. FIG.

FIG. 15 is a graph showing the timing of the neutron count rate and the accelerator current value obtained in Comparative Example 1. FIG.

FIG. 16 is a graph showing the relationship between the accelerator current value and the neutron count rate in Comparative Example 1. FIG.

FIG. 17 is a graph of the timing of the neutron count rate and the accelerator current value obtained in Comparative Example 2. FIG.

FIG. 18 is a graph showing the relationship between the accelerator current value and the neutron count rate in Comparative Example 2. FIG.

Forms used to implement the invention

A schematic diagram of a radiation detection apparatus as an example of an apparatus for implementing the present invention is shown in FIG. 4. As shown in FIG. 4 (a), the radiation detection device includes a radiation detection element 1 and a wave height recognition circuit 2 that output pulse-shaped electrical signals upon the incidence of radiation. When a medium such as a scintillator does not output a direct electrical signal but emits light through the incidence of radiation, as shown in FIG. 4 (b), the scintillator 3 and the scintillator 3 emit light. Convert light into electricity A combination of the photodetectors 4 of the signals can be used as the radiation detection element 1. As the photodetector 4 that converts light emitted by the scintillator into an electrical signal, a photomultiplier tube, a photodiode, an avalanche photodiode, and a Geiger mode burst photodiode can usually be used. Photodetectors such as polar bodies.

The wave height recognition circuit 2 is only required to be able to analyze the wave height value of the pulse output from the radiation detection element 1 and count pulses larger than the high recognition threshold. Those who can read the pulse by analog data processing can be applied, and The present invention is applicable to any one of read pulses by digital data processing.

The wave height recognition circuit 2 used in the embodiment of the present invention reads pulses by digital data processing including an amplifier (analog amplifier), an AD converter, a digital data processing device, a display device, and the like. Hereinafter, a device that reads a pulse by digital data processing will be described as an example.

According to the foregoing embodiment, the pulses emitted by the radiation detection element 1 when detecting radiation are amplified and amplified by the analog amplifier along the time axis, and the amplified pulse signals are passed through the AD converter and a certain amount. The sampling interval is converted into digital data, and the digital data is read by a digital data processing device. Then, the wave height distribution spectrum, count values larger than the wave height recognition threshold, and the like are displayed on a display device composed of a computer, software, a display monitor, and the like.

Figure 5 is a signal waveform diagram of the pulses amplified by the analog amplifier to conceptually explain the effect of stacking. Each point in the figure is displayed on the solid analog signal, and it is digitized at a certain time interval. The discrete value obtained at the time.

Fig. 5 (a) shows a case where there is no accumulation. Digital data processing The device is counted in such a way that "the phenomenon corresponding to Pulse Height H is twice".

On the other hand, Fig. 5 (b) shows the state after the accumulation. Originally it should be counted as "three times equivalent to the wave height value H", but immediately after the first pulse, another radiation was incident and the second pulse rose, and the signal value continued to rise when viewed as a discrete value of the digital data. Therefore, the first pulse and the second pulse are indistinguishable from each other and can be identified and counted as "a phenomenon equivalent to H" (≒ 2H). That is, signals that should be counted twice are counted only once. This phenomenon is called count loss.

The third pulse rises before the pulses accumulated by the first pulse and the second pulse return to the baseline. That is, the third pulse-type analog signal rises after acquiring a minimum value. Because the discrete value before the minimum value is used as the basis when calculating the wave height value, the third pulse system is counted as "the phenomenon equivalent to H '(<H) is 1 time" by the apparent baseline displacement. ". This phenomenon is called miniaturization of wave height value. The problem to be solved by the present invention is the counting loss caused by the first pulse and the second pulse. In the following description, for convenience, as shown in FIG. 5 (b), the case where the first pulse overlaps with the second pulse and the plurality of pulses cannot be separated and counted is processed in a stacked manner. Those who are separated from the previous pulse and counted are treated as unstacked. In addition, the problem of reducing the wave height value of the third pulse may cause a count loss. That is, a pulse that should be treated as a wave height value larger than the wave height identification threshold value will become a count loss if it is reduced to a wave height value smaller than the wave height identification threshold value. Regarding this count loss, it is also possible to reduce the count loss by utilizing the Gaussian fitting described later and adjusting the wave height identification threshold value to reduce the wave height value.

In order to simulate the error caused by the above-mentioned stacking to the count value, first, the wave height distribution spectra of the unstacked case and the stacked case are displayed, respectively, and approximated by a Gaussian function of the following formula (1). The maximum count N of the Gaussian function in formula (1) is the count displayed at the average wave height value μ. The X series represents an arbitrary wave height value, and σ is a standard deviation. A diagram illustrating an approximation method of the wave height distribution spectrum is shown in FIG. 6.

Referring to the average wave height value μ and the standard deviation σ of the wave height distribution peaks in the unstacked state shown in FIG. 2, μ and σ of the unstacked first peak are set to μ ₁ = 200ch and σ ₁ = 26ch, respectively. The second spikes appearing when two pulses are stacked are set to μ ₂ = 400ch and σ ₂ = 52ch, and the third spikes appearing when three pulses are stacked are set to μ ₃ = 600ch and σ ₃ = 78ch. After using formula (1) and generating the first to third spikes as independent data, each wave height value is counted and combined to form a wave height distribution spectrum. In addition, as shown in the wave height distribution spectra of Figs. 1 to 3, in fact, the measured spectrum also has a different type of radiation from the counted target component. Noise components are observed on the low wave height side and the When the peaks of the target component partially overlap. However, it is not necessary to consider such noise components in this simulation system.

The wave height distribution spectrum generated by the method described above is set to 0%, 10%, 20%, and 30% in Fig. 7. The wave height distribution spectrum with a stacking rate of 0% has only the data of the first peak. The total count from 0 to 1023 ch of each wave height distribution spectrum is set to double the count of the second peak and The count of the three spikes was set to 3 and the calculations were made to coincide. The accumulation rate is a rate at which pulses cannot be separated due to overlapping pulses. Specifically, when two of the 100 pulses are overlapped and processed as one signal, the accumulation rate is 2%. When 100 pulses are overlapped and 30 pulses are generated as overlapping signals or 3 pulses are overlapped, the stacking rate is 30%. That is, the ratio of the pulses contained in the signals whose pulses are overlapped is called the stacking rate.

Secondly, the wave height identification threshold (Threshold) is set to μ + Xσ, and a count above the wave height identification threshold when X changes in a range of -3 to +3 is obtained from the wave height distribution spectrum. The results are shown in FIG. 8. Furthermore, from the results in FIG. 8, counting errors of 10%, 20%, and 30% of the stacking ratios with respect to the counts above the wave height recognition threshold at the time of stacking of 0% were each calculated. Here, the counting error is 0%, the count at the wave height recognition threshold μ + Xσ is C ₀ , the count is Y%, and the count at the wave height recognition threshold μ + Xσ is C _{For Y} , it is calculated according to 100 × (C _Y -C ₀ ) / C ₀ . The results are shown in Fig. 9. From the results of these Figures 8 and 9, it should be understood that the smaller the X, the larger the negative error caused by the counting loss, and the larger the X, the larger the positive error caused by the excess counting, at a wave height close to μ. When the identification threshold value (that is, around X = 0), the counting error is unlikely to occur due to accumulation.

In addition, in order to set the optimal wave height identification threshold based on the actual measured values, it is based on Example 1 described below. For the most non-stacking situation (accelerator current is 5.7 μA and the wave height distribution spectrum is shown in Figure 2), (Accelerator current is 379.6 μA, wave height distribution spectrum is shown in Figure 3), and the counting error is analyzed. For each of the first spikes in Figure 1 and Figure 2, the Gaussian function is fitted to obtain μ and σ, and the wave height recognition threshold is set based on the obtained μ and σ. μ + Xσ, and the results of counting errors obtained by the same method as the aforementioned simulation are shown in FIG. 10. However, it is necessary to consider the dose rate difference when the wave height distribution spectra of Figs. 2 and 3 are obtained. Therefore, the dose rate difference is regarded as a person showing the same tendency as the accelerator current difference. The counts in Figure 3 are normalized. Further, from the analysis of the wave height distribution spectrum of FIG. 3, it was found that the stacking ratio of the wave height distribution spectrum was about 25%. Fig. 10 shows the simulation value of the count error when the stacking rate is set to 25%.

The counting error caused by the stacking shown in Fig. 10 is somewhat offset for the simulated and measured systems. It is considered that this is because the reduction of the wave height value due to the accumulation is not considered during the simulation, and the reduction of the value due to the accumulation is considered during the actual measurement. In this way, although the simulation and actual measurement systems have some deviations, it should be understood that in actual measurement, the wave height identification threshold close to μ (that is, around X = 0) is not easily affected by the counting loss caused by accumulation. In addition, it was found that when the wave height value of the actual measurement is reduced, it is best to set the wave height recognition threshold to be slightly higher than μ.

Here, the principle that the counting error due to accumulation can be reduced by setting the wave height recognition threshold near μ will be described using FIG. 11.

First, using FIG. 11 (a), it will be described that only two pulses are considered. When the total count of the unstacked first spikes is set to N _s and the total count of the second spikes stacked by two pulses is set to N _d , the number of radiation detected by the radiation detection element can be theoretically said. Is N _s + 2N _d .

Here, the threshold is usually set to be near T ₁ so that all phenomena can be counted. When T _{1 is} set as the threshold for wave height recognition, the count above the threshold becomes N _s + N _d . Compared with the theoretical value N _s + 2N _d, it produces a negative error of N _d , especially the accumulation is significant. When N _d is generated, the counting error becomes extremely large. On the other hand, the counting method of the present invention is characterized in that by setting the wave height recognition threshold to be in the vicinity of μ of the first spike, the counting error can be reduced. For example, when the μ of the first peak is set as the threshold value obtained when the wave height recognition threshold (T ₂ ) is equal to 1 / 2N _s + N _d , compared with the theoretical value N _s + 2N _d , Just count half. In other words, when the count obtained by setting μ of the first spike to the wave height recognition threshold is doubled, a count as a theoretical value can be obtained. At this time, the count obtained by setting the μ of the first peak to the threshold value of the wave height recognition is often half of the theoretical value regardless of the degree of stacking. Therefore, the counting method of the present invention does not generate counting errors due to stacking. .

Next, a description will be given using FIG. 11 (b) to consider the accumulation of two pulses and the accumulation of three pulses. When the total count of the unstacked first spikes is set to N _s , and the total count of the second spikes formed by stacking two pulses is set to N _d , the total count of the third peaks formed by stacking three pulses is set In the case of N _t , the number of radiation detected by the radiation detecting element can be said to be N _s + 2N _d + 3N _t in theory. When the threshold value is set to T _{1 where} all phenomena can be counted, a count above the threshold value obtained is N _s + N _d + N _t . Compared with the theoretical value N _s + 2N _d + 3N _t , a negative error of N _d + 2N _t occurs. On the other hand, a count equal to or more than the threshold value obtained when the μ of the first spike is set as the wave height recognition threshold value (T ₂ ) becomes 1 / 2N _s + N _d + N _t . Compared with 1 / 2N _s + N _d + 3 / 2N _t , which is half of the theoretical value, a negative error of 1 / 2N _t is generated, but when the threshold is set to T ₁ , the system can significantly reduce error.

By setting the wave height identification threshold to be close to μ, the counting method of the present invention to reduce counting errors due to accumulation is based on the theory described above, and the simulation results shown in FIG. 9 also show the present invention. Properness.

In addition, the aforementioned simulation system considers the accumulation of two pulses and the accumulation of three pulses, and the generation of several negative errors at the threshold value μ + 0σ is also consistent with this theory. In addition, since the count Nt of the third spike caused by the negative error is sufficiently smaller than the count Ns of the first spike, it is a result that the contribution to the count error due to the accumulation is small. In addition, although the 4th, 5th, and 6th, ‧‧ peaks formed by 4 or more pulses can also be observed, the contribution to the counting error due to the stacking is smaller than that of the 3rd peak, so in most cases The use is not a problem.

According to the present invention, a method for setting a wave height identification threshold and counting a signal frequency greater than the wave height identification threshold is used. When the purpose component of the counting is to display a spike in the wave height distribution spectrum, a Gaussian function simulation is performed for the spike. When the average wave height value obtained by the combination is set to μ and the standard deviation is set to σ, an arbitrary value selected from a range of (μ-0.4σ) to (μ + 0.6σ) is set as the count of the aforementioned wave height identification threshold. This method can suppress a decrease in counting accuracy due to accumulation. The method of setting the wave height identification threshold is not limited by the Gaussian function fitting. When the threshold value of the wave height recognition during the signal processing of the recognition circuit is within the above range, the threshold value set by using other methods is also included in the scope of the present invention.

The purpose component of counting refers to a phenomenon that is the object of counting, and is not particularly limited. For example, as long as a neutron detection device such as a neutron dose rate and a cumulative dose can be provided, the radiation phenomenon of neutrons is a target component for counting. There are various types of radiation, such as X-rays, α-rays, γ-rays, and neutrons. It can also be a radiation detection device that detects more than 2 types of radiation. At this time, the wave height identification threshold value of the present invention can be applied. Any one of the types of radiation to be detected may be directed to two or more types of radiation.

In addition, the so-called counting target component is not limited by the incident phenomenon of radiation, and can also be applied to the detection of electromagnetic waves such as ultraviolet rays, visible rays, infrared rays, and microwaves, and can be applied to various fields.

The counting method of the present invention can be applied to a case where the purpose component for counting shows a sharp peak in the wave height distribution spectrum. For example, in a neutron detection device, the wave height distribution spectrum showing the frequency of each wave height phenomenon is displayed on the vertical axis with the wave height value on the vertical axis, and the phenomenon of detecting neutrons can be detected when the wave height distribution spectrum shows a spike. An arbitrary value selected from a range of (μ-0.4σ) to (μ + 0.6σ) obtained by performing a Gaussian function fitting on the spike is set as the aforementioned threshold value, and counted above the threshold value Phenomenon. When the phenomenon of detecting neutrons does not show a sharp peak in the wave height distribution spectrum, the counting method of the present invention cannot be applied.

As an example of a form in which a spike can be formed by the detection of neutrons, there is a method of combining Eu: LiCaAlF ₆ crystal (neutron scintillator), a photodetector, and a wave height recognition circuit described in Patent Document 5.

The shape of the peak of the wave height distribution is not particularly limited, and there is no problem as long as it follows the shape of the Gaussian distribution, and the effect of the present invention can be obtained. The μ obtained by fitting a Gaussian function according to the method described later is consistent with the wave height value (hereinafter, the wave height value is represented by P) indicating the maximum count among the peaks shown in the wave height distribution spectrum of the target component of counting. It can be expected that a high counting accuracy is statistically natural. When the range of 0.9 μ ≦ P ≦ 1.1 μ is consistent, the counting accuracy can be particularly good.

Before performing Gaussian function fitting on the spike, the wave height distribution spectrum may be previously subjected to processes such as baseline correction, spike separation, and smoothing.

Μ and σ, which are the criteria for determining the threshold value of the wave height identification of the present invention, are among the peaks of the wave height distribution displayed by counting the target components, and the formula (1) is performed for the first peak that is not accumulated. The value obtained by fitting the Gaussian function of is not related to the value obtained for the second spike formed by stacking 2 pulses and the third spike formed by stacking 3 pulses. The so-called μ and σ described below refer to the μ and σ obtained for the first spike. The first spike is usually the spike with the highest count. In addition, when accumulation occurs frequently, the second spike becomes large and separation becomes difficult. At this time, the counting method of the present invention can be applied by using means such as reducing the detection sensitivity to shorten the decay time of the shortened pulse, and setting the conditions such that the shape of the first spike is an approximate Gaussian function.

The data range of the wave height distribution spectrum used in the Gaussian function fitting is the peak value displayed by the wave height distribution spectrum of the target component of the counting, and the wave height value P showing the maximum count is used as a reference. good. In addition, when counting the noise caused by a different component from the target component of the counting, when the analysis values of μ and σ are affected, the lower limit value and / or the upper limit value may be narrowed. Conversely, when there is no noise that affects the analytical values of μ and σ, the range of data used in fitting the Gaussian function can be expanded. For example, the data range of the wave height distribution spectrum used when performing the Gaussian function fitting may be set to 0.75P to 3P. Here, when the second spike appears, the count of the second spike is noisy in fitting. However, the count of the second spike that appears by stacking is sufficient compared to the count of the first spike. The ground is small, so the influence of this noise on the analytical values of μ and σ is limited. The fitting system of the Gaussian function is preferably the least quadratic method, and it is preferable that the maximum count N is also fitted together with μ and σ.

As can be understood from the foregoing description, the essence of the counting method of the present invention is to reduce the counting error due to accumulation by setting the wave height identification threshold near the μ. Moreover, an arbitrary value selected from the range (μ-0.4σ) to (μ + 0.6σ) obtained by performing the Gaussian function fitting described above is set as the aforementioned method of identifying the wave height threshold, which is used to provide The wave height identification threshold is an index appropriately set in the vicinity of μ.

In the present invention, even if analysis using a method using Gaussian function fitting is not performed, for example, the method described below can be used to determine the wave height identification threshold.

In addition, any of the wave height recognition thresholds determined by the method described below is also included in the Gaussian function fitting, and (μ-0.4σ) to (μ + 0.6σ) obtained from the fitting. Range.

It is not necessary to use the first determination method of the wave height recognition threshold of the analytical method of fitting the Gaussian function. For the wave height distribution spike showing the maximum count N 'at the wave height value P, the display will be closest to the following formula (2) And the method of counting the wave height value of the count n is set as the wave height identification threshold.

Equation (2) substitutes the wave height identification threshold μ + Xσ to the arbitrary wave height value x of the Gaussian function of equation (1), and the count n is expressed as a function of X. That is, when the wave height recognition threshold is set to P or less, an arbitrary value selected from the range of -0.4 to 0.0 is set to X, and the count closest to the count n obtained in accordance with equation (2) is displayed. The wave height value below P can be set as the wave height identification threshold. When the threshold value is set to be P or more, an arbitrary value selected from the range of 0.0 to 0.6 is set to X, and the wave height value of P or more that is closest to the count of the count n obtained in accordance with equation (2) is displayed. It can be set as the wave height identification threshold.

It is not necessary to use the second determination method of the wave height recognition threshold value of the analysis method of the Gaussian function fitting method, which is a method of determining the wave height recognition threshold value using the half-value width FWHM instead of the standard deviation σ as a reference. In the Gaussian distribution, μ is the same as the wave height value P, and the standard deviation σ and the half-value width FWHM have a relationship of the following formula (3).

Therefore, since the range of (μ-0.4σ) to (μ + 0.6σ) is the same range as (P-0.28FWHM) to (P + 0.42FWHM), the wave height value P and the half-value width FWHM analysis can also determine the wave height identification threshold according to the present invention.

It is not necessary to use the third determination method of the analytical wave height identification threshold value of the method of fitting the Gaussian function, and the shape of the wave height distribution spectrum is determined from the wave height value P showing the maximum count of the wave height distribution peak, and the wave height value P is near The wave height value of is set as the wave height recognition threshold.

In the above, the first to third determination methods of the wave height identification threshold value without using the analysis of the Gaussian function fitting method have been exemplified. However, the wave height identification threshold value may be determined by using a method different from these. The wave height identification threshold determined by any method is included in the range (μ-0.40) to (μ + 0.6σ) analyzed by Gaussian fitting, and the effect of the present invention can be exerted.

As described above, in the present invention, the wave height identification threshold is set near the maximum count of the first spike of the wave height distribution. Hereinafter, there will be described threshold value in the present invention, the set of T _A is the case. In the counting method of the present invention, the frequency of pulses larger than T _A is counted. In the counting method of the present invention, for the purpose of component count than closer to the high peak value of the noise object side of the removal, also provided together with the high T _A wave peak value in the other side of the peak-identification threshold (T _B) to count is greater than T _a but less than T _B, T _a or T _B above and the following pulse. At this time, in order not to lose the count of the second and third spikes, the threshold value T _B is preferably 3 μ or more, and more preferably 5 μ or more. In addition, while setting the wave height recognition threshold to count the first target component, it is also possible to set the wave height recognition threshold to count the second target component.

According to the present invention, by setting an arbitrary value selected from a range of (μ-0.4σ) to (μ + 0.6σ) as a wave height recognition threshold, it is possible to suppress a decrease in counting accuracy due to accumulation. According to the actual measurement value of the counting error caused by the stacking shown in Figure 10, when counting the pulse signal larger than the wave height recognition threshold, the counting error due to the stacking can be suppressed to approximately ± 10% or less. Counting errors can achieve highly accurate measurement in most applications.

However, depending on the stacking frequency and the degree of miniaturization caused by the stacking described above, there may be an error greater than ± 10%. According to the theory described using FIG. 9, it is possible to deviate from the scope of the present invention. If the wave height recognition threshold is used, it can achieve significantly better counting accuracy. According to the actual measurement value of the counting error caused by stacking shown in Figure 10, when the wave height recognition threshold is set to a wave height value lower than (μ + 0.3σ), the wave height recognition threshold is lowered, and the The more the counting loss affects the counting error on the negative side, the more easily it will expand. When the wave height recognition threshold is set to a wave height value higher than (μ + 0.3σ), the more Increasing the threshold value of the wave height recognition, the increase in the counts of the second and third spikes is more affected, so that the counting error on the positive side is easy to expand.

The preferred wave height recognition threshold is an arbitrary value selected from the range of (μ + 0.0σ) to (μ + 0.5σ), and the better wave height recognition threshold is from (μ + 0.2σ) to (μ + 0.4σ). Any value selected in the range of) can suppress counting errors of about ± 5% and ± 3%.

The counting method of the present invention can be suitably applied to a radiation detection device in which a scintillator 3, a photodetector 4 and a wave height recognition circuit 2 are combined.

The present invention provides a radiation detection device, which is a radiation detection device using a combination of a scintillator 3, a photodetector 4, and a wave height recognition circuit 2. The wave height recognition circuit 2 is provided with amplifying the signal of the light detector 4. Amplifier, an AD converter that converts the analog signal output of the aforementioned amplifier into a digital signal, and a wave height recognition circuit of a digital data processing device that outputs the digital signal of the aforementioned AD converter for signal processing, and the wave height recognition circuit Signal processing is performed by the aforementioned counting method.

As the wave height identification circuit 2, as described in Patent Documents 2 and 3, an AD converter including an amplifier that amplifies the signal of the photodetector 4 and an analog signal output of the amplifier is converted into a digital signal. , And a digital data processing device that outputs the digital signals of the AD converter and performs signal processing. Compared with the wave height recognition circuit that reads the pulse signal by analog data processing, the wave height recognition circuit that reads the pulse signal by digital data processing can reduce the wave height value caused by the accumulation described above at low cost. The circuit structure of the miniaturization and the correction processing of the wave height value can be simplified. The effect of the present invention can be obtained with higher accuracy by reducing the miniaturization due to accumulation.

In the present invention, the aforementioned digital data processing device is preferably installed to prepare a wave height distribution spectrum at a predetermined time interval, and to fit the peaks displayed for the counted target component to a Gaussian function fitting in the prepared wave height distribution spectrum, and successively A function for setting the threshold value of the wave height recognition based on the obtained μ and σ as the reference. By installing such a function, it is possible to automatically update the wave height recognition threshold value following the miniaturization caused by the accumulation described above, and it is possible to prevent an increase in the counting error due to the miniaturization.

In addition, when using a wave height recognition circuit with a small value due to accumulation, the function of fitting a Gaussian function and the function of sequentially setting the wave height recognition threshold value may not be installed, and the wave height recognition threshold value according to the present invention may be analyzed in advance. The value is counted from the start of counting to the end based on the threshold of the wave height recognition.

The amplifier can be a preamp, a waveform shaping amplifier (Shaping amp), etc., or a combination of these. Also, it may be a circuit configuration in which a differential circuit is combined. In addition to the circuit configurations described in Patent Documents 2 and 3, the circuit configurations described in Patent Documents 6 and 7 are also suitable.

The scintillator 3 is capable of using LiCaAlF ₆ crystals, LiSrAlF ₆ crystals, LiF / CaF ₂ co-crystals, LiF / SaF ₂ co-crystals, and NaI crystals, without limitation, depending on the application, without activating agents such as Eu, Ce, Tl, and Na. , Bi ₄ Ge ₃ O ₁₂ crystals, Gd ₃ Al ₂ Ga ₃ O1 ₂ crystals, CeF ₃ crystals, and other inorganic scintillators; and organic compounds composed of organic phosphors such as anthracene, fluorene, and diphenyloxazole Scintillator; and a plastic scintillator containing polystyrene, polyvinyltoluene, polyethylene terephthalate, etc. of the organic phosphor; and a plastic scintillator containing toluene, xylene, etc. of the organic phosphor. Liquid scintillators; and gas scintillators such as helium, argon, xenon, krypton, etc.

In the present invention, the scintillator 3 is preferably LiCaxSr _1-x AlF ₆ crystal (x is 0 to 1). Specifically, LiCaAlF ₆ crystals and LiSrAlF ₆ crystals containing additives such as Eu and Ce as an activating agent can be suitably used. The scintillator crystals are neutron scintillators, and the Li-6 system contained in the scintillator crystals reacts with neutrons to generate alpha rays and thorium. The alpha rays and thorium can provide 4.8 MeV to the scintillator crystals. energy of. Then, the activator contained in the crystal of the scintillator can emit light corresponding to the energy provided. Because the LiCa‧AlF ₆ crystal and the LiSrAlF ₆ crystal can detect the neutron phenomenon, they can display spikes consistent with Gaussian distribution with an accuracy of 0.95μ ≦ P ≦ 1.05μ, so they are particularly suitable for use in The scintillator of the radiation detection device of the present invention.

Eu: LiCaAlF ₆ crystals used in the examples of the present invention, such as LiCaxSr _1-x AlF ₆ crystals containing Eu as an activator, are particularly suitable for use in the present invention because their luminescence is particularly high and the signal / noise ratio is relatively high. invention. In addition, the LiCaxSr _1-x AlF ₆ crystal containing Eu as an activator has a fluorescence lifetime of 1.6 μsec, because it has a long fluorescence lifetime among ordinary scintillator materials, and although it is prone to deposit, it is easy to produce by using this The counting method of the invention enables accurate counting even if accumulation occurs.

On the other hand, a LiCaxSr _1-x AlF ₆ crystal containing Ce as an activating agent has a particularly short fluorescence lifetime of tens of nsec, and therefore can be suitably used in the present invention. LiCaxSr _1-x AlF ₆ crystals containing Ce as an activator, such as Ce: LiCaAlF ₆ crystals and Ce: LiSrAlF ₆ crystals, have a short fluorescence lifetime and are not prone to build up. Therefore, when the counting method of the present invention is applied, it can be used. Measurement is performed at a very high count rate.

The adjustment of the neutron sensitivity of the neutron scintillator containing Li-6 can be adjusted by the size of the scintillator and the adjustment of the Li-6 isotope ratio. As a substance that adjusts the ratio of Li-6 and Li-7 of Li's stable isotope, there are a substance obtained by concentrating Li-6, a substance obtained by natural ratio, and a substance obtained by concentrating Li-6 and Li-7. The higher the Li-6 isotope ratio, the higher the neutron sensitivity. Moreover, the neutron sensitivity can also be adjusted by such mixing. By adjusting the neutron sensitivity as described above, it is possible to provide a radiation detection device in accordance with the dose rate of the counting object.

In the present invention, the means for transmitting light from the scintillator 3 to the photodetector 4 is not particularly limited, and the scintillator 3 and the photodetector 4 may be directly connected as shown in FIG. 4 (b), or may be The connection shown in FIG. 12 is indirectly connected using a light guide such as an optical fiber 5. When the scintillator 3 and the photodetector 4 are directly connected, because the photodetector 4 and the scintillator 3 simultaneously expose the radiation field, the radiation is incident on the photodetector 4, and the counting accuracy is observed due to the influence of noise. The possibility of a drop or a malfunction of the photodetector 4. Therefore, when counting radiation in a high-dose-rate environment, the scintillator 3 is set in the high-dose-rate environment of the counting object and the photodetector 4 is set far away from the dose rate sufficiently to use optical fiber 5 or the like The light guide preferably connects the scintillator 3 and the light detector 4 indirectly.

Examples of high-dose-rate radiation fields include radiation therapy devices for cancer, especially for boron neutron capture therapy (BNCT). The neutron dose rate is known to be very high at 1 × 10 ⁹ n / cm. ² / sec. Previously, BNCT treatment was performed using atomic furnace neutrons. In recent years, because cyclotrons and electrostatic accelerators can also be used to obtain high dose rate neutrons, an accelerator BNCT treatment device that can also be installed in hospitals has been developed.

The counting method and the radiation detection device of the present invention are advantageous It is used in a variety of radiation fields and is particularly suitable for the detection of the aforementioned accelerator neutrons. It is especially suitable for the detection of neutrons in boron neutron capture therapy with very high neutron dose rate.

When detecting neutrons for boron neutron capture therapy, the best form of implementing the present invention is a radiation detection device that combines a neutron scintillator 3, a photodetector 4, and a wave height recognition circuit 2 as a neutron scintillator. The device 3 is any one of Eu: LiCaAlF ₆ crystal, Ce: LiCaAlF ₆ crystal, Eu: LiSrAlF ₆ crystal, and Ce: LiSrAlF ₆ crystal, and an optical fiber 5 is used to connect the neutron scintillator 3 and the photodetector 4 Connected between.

In this state, in order to improve the transmission efficiency of light from the neutron scintillator 3 to the optical fiber 5, it is better to optically bond the neutron scintillator 3 and the front end of the optical fiber 5. The method of performing optical bonding is not particularly limited. The method uses an organic adhesive and an inorganic adhesive with good transmittance and good radiation resistance at the emission wavelength of the scintillator 3, or temporarily melts the front end portion of the optical fiber 5 and The method of welding the scintillator 3 is preferable. In addition, in order to improve the light collection efficiency from the neutron scintillator 3 to the optical fiber 5, it is preferable to cover the periphery of the neutron scintillator 3 with a reflective material. The reflecting material is not particularly limited, and white materials such as tetrafluoroethylene, polyethylene, titanium oxide, and barium sulfate are preferred. In addition, the optical fiber 5 can be made of quartz fiber, plastic fiber, or the like, and preferably has a large number of openings (NA) and a small transmission loss. Specifically, it is preferable that NA is 0.45 or more and transmission loss is 200 dB / km or less. By using a larger NA and a smaller transmission loss, when a sufficient amount of light is transmitted to the photodetector 4, because the phenomenon of neutrons is detected, it can be formed with an accuracy of 0.95μ ≦ P ≦ 1.05μ. Spikes with consistent Gaussian distribution are particularly good.

In the case of the aforementioned configuration, it is not necessary to arrange the photodetector 4 at a high level. In the dose rate environment, a peak in the phenomenon of neutron detection by the scintillator 3 can be formed in the wave height distribution spectrum. In addition, when the average wave height value obtained by fitting a Gaussian function to the spike is set to μ and the standard deviation is set to σ, an arbitrary value selected from a range of (μ-0.4σ) to (μ + 0.6σ) is used. When the wave height recognition threshold is set, and the count is greater than the wave height recognition threshold, it can provide a small decrease in counting accuracy due to accumulation and a wide range from low dose rate to high dose rate. Radiation detection device that counts neutrons with a good degree.

As can be clearly understood from the above description, one aspect of the counting method of the present invention is a method of setting a wave height recognition threshold and counting signal frequencies greater than the wave height recognition threshold. The purpose component of the counting is displayed on the wave height distribution spectrum. For spikes, set any value selected from the range (μ-0.4σ) ~ (μ + 0.6σ) obtained by performing Gaussian function fitting on the spikes, but μ is the average wave height value and σ is the standard deviation. Identify the threshold for the aforementioned wave height.

The counting method of other aspects of the present invention is a method for setting a wave height recognition threshold value and counting signal frequencies greater than the wave height recognition threshold value when the wave height distribution spectrum shows a plurality of spikes. The threshold is in the range of (μ-0.4σ) ~ (μ + 0.6σ). However, μ and σ are average wave height values μ and standard deviation σ when the largest peak among the plurality of peaks is approximated by a Gaussian function.

In still another aspect of the present invention, the counting method is based on a wave height distribution spectrum obtained by plotting a pulse signal originating from radiation or electromagnetic waves based on its wave height value and a count of each wave height value. A method for setting a wave height recognition threshold and counting signal frequencies greater than the wave height recognition threshold, and approximates the largest spike among the aforementioned plurality of spikes with a Gaussian function When the obtained average wave height value is set to μ and the standard deviation is set to σ, the counting method of the aforementioned wave height identification threshold is in a range of (μ-0.4σ) to (μ + 0.6σ).

Examples

Hereinafter, specific experimental examples will be given to explain the embodiments of the present invention in more detail, but the present invention is not limited to these.

Example 1

First, a neutron detection device that combines a neutron scintillator 3, a photodetector 4, and a wave height recognition circuit 2 is manufactured as follows. The manufactured neutron detection device is shown schematically in FIG. 12.

The neutron scintillator 3 is made of Eu: LiCaAlF ₆ crystal, which is made by using Li raw material which concentrates the Li-6 isotope ratio to 95%. Using an epoxy resin adhesive, Eu: LiCaAlF ₆ crystal optics of 0.6 mm × 0.6 mm × 0.6 mm were processed. Then, the front end of the optical fiber 5 made of quartz with a core diameter of 1 mm and a length of 10 m was set (the reverse end was set as the back end). . The optical fiber 5 uses a NA of 0.48, and a transmission loss at a luminous wavelength (370 nm) of Eu: LiCaAlF ₆ crystal of 100 dB / km. The optical bonding surfaces of the scintillator 3 and the optical fiber 5 composed of Eu: LiCaAlF ₆ crystal are each subjected to optical polishing. Furthermore, by covering the scintillator 3 made of Eu: LiCaAlF ₆ crystal with a reflecting material made of barium sulfate, the improvement of the light-concentrating efficiency toward the optical fiber is improved, and the optical fiber 5 is covered by using a light-shielding material to cover Room light is not incident.

The rear end of the optical fiber 5 following the scintillator 3 composed of Eu: LiCaAlF ₆ crystal is optically connected to the photodetector 4. The photodetector 4 is electrically connected to the wave height identification circuit 2. As the photodetector 4, a photomultiplier is used.

The wave height recognition circuit 2 is used to increase the signal of the photodetector. Amplifiers, AD converters that convert analog signal output from amplifiers into digital signals, digital data processing devices that output digital signals from AD converters for signal processing, and display devices that display the results of digital data processing.

The display device is a computer, software, or monitor. The display is capable of displaying the wave height distribution spectrum and the count value of pulses which are larger than the wave height recognition threshold.

In addition, the aforementioned digital data processing device is installed to produce a wave height distribution spectrum every 10 seconds, and performs a Gaussian function fitting on the peaks displayed by the counted target components in the created wave height distribution spectrum, and sequentially sets to obtain the The function of μ and σ as the reference wave height identification threshold. In this example, as the stacking frequency is increased, the peak of the wave height distribution is gradually shifted to the low wave height value side, and compared with the situation where the accumulation frequency is the smallest, the μ system is reduced by about 14% in the situation where the accumulation frequency is the largest. However, because the wave height identification threshold is automatically updated in accordance with the peak position at this point, the count loss caused by the minimization of the wave height value is slight. Using the manufactured neutron detection device, as described below, the accelerator neutrons were counted and the error caused by the accumulation to the count value was analyzed.

A polyethylene block is disposed near the neutron irradiation port using the electrostatic accelerator neutron generating device, and the tip of the optical fiber 5 followed by the scintillator 3 is inserted into the insertion hole provided in the polyethylene block. And fixed. The photodetector 4 and the wave height identification circuit 2 are arranged by extending the optical fiber 5 away from the neutron irradiation port.

Neutrons are generated by an electrostatic accelerator, and the neutrons are detected by a neutron detection device. Analyze the detected neutron pulses and count pulses greater than the wave height identification threshold every 1 second to obtain the neutron count rate [cps] per unit time. The threshold value of the wave height recognition in Example 1 is set to (μ + 0.3σ). Furthermore, the neutron dose rate [n / cm ² / sec] can be obtained by multiplying the obtained neutron count rate [cps] by a factor for conversion into a neutron dose rate. The neutron dose rate of the environment in which the detection element is placed at the maximum output of the accelerator is about 1 × 10 ⁸ n / cm ² / sec.

The change with time of the count rate [cps] obtained in Example 1 is shown in FIG. 13 (left vertical axis). In FIG. 13, the current value [μA] of the accelerator is also displayed (right vertical axis), and the current value is shown in the experience of the accelerator output in Example 1. The scale on the right vertical axis of FIG. 13 is for a neutron count rate showing a portion below 6 kcps, and the accelerator current value is adjusted so as to overlap the neutron count rate on the graph. The wave height distribution spectrum when the accelerator current value is about 5 μA is the lowest, as shown in FIG. 2. The wave height distribution spectrum when the accumulation is small is about 45 μA, and the wave height distribution spectra when about 380 μA are shown in FIG. 1 and 3 respectively. The figure shows that the higher the accelerator current value, that is, the higher the neutron dose rate, the more the packing frequency increases.

Regarding the data in FIG. 13, a graph in which the average value of the accelerator current value and the neutron count rate is calculated and compared for each period in which the accelerator current value is stable is shown in FIG. 14. The regression line in FIG. 14 is obtained from a portion with a low neutron count rate and a small effect of stacking (about 4 points of 1 kcps to 5 kcps). Compared to the expected neutron count rate from the regression line, the error of the neutron count rate can actually be obtained as shown in Table 1. The maximum error is only -2.1%, which is a good result.

[Table 1]

Comparative Example 1

Using the same neutron detection device as in Example 1, the accelerator neutron was counted in the same manner as in Example 1, except that the wave height recognition threshold was set to (μ-1.0σ).

About the neutron count rate and the accelerator current value obtained in the comparative example 1, the analysis result by the method similar to Example 1 is shown in FIG.15 and FIG.16. In Fig. 15, in the part where the neutron counting rate is greater than 40 kcps, the neutron counting rate becomes significantly lower than the accelerator current value, which shows the result of counting loss due to accumulation. The difference between the neutron counting rate and the regression line in each accelerator current value is as shown in Table 1, which produces a counting error greater than -10%.

Comparative Example 2

Using the same neutron detection device as in Example 1, the accelerator neutron was counted in the same manner as in Example 1, except that the wave height recognition threshold was set to (μ + 0.7σ).

Regarding the neutron count rate and accelerator current value obtained in Comparative Example 2, the results of analysis by the same method as in Example 1 are shown in Figures 17 and 18. In Figure 17, the neutron count rate is significantly higher than the accelerator current value in the part where the neutron count rate is greater than 15 kcps. This shows that the accumulation of the second and third spikes is affected by the accumulation. the result of. The difference between the neutron count rate and the regression line in each accelerator current value is as shown in Table 1, which produces a count error greater than + 10%.

In the above embodiment, even under the condition of a high counting rate where the stacking frequency becomes high, by adopting the counting method of the present invention, it is shown that it is possible to suppress the decrease in counting accuracy due to the stacking. When Eu: LiCaAlF ₆ crystal with Li-6 isotope ratio set to the natural ratio and reduced in size to 0.3mm × 0.3mm × 0.2mm is used as the neutron scintillator 3, because the neutron detection made in this example The device can become neutron sensitivity below 1/100, so if this Eu: LiCaAlF ₆ crystal neutron detection device is used, even if the neutron dose rate is 1 × 10 ⁹ n / cm ² / sec in BNCT treatment, It is possible to count neutrons with good accuracy. With this neutron detection device, even at a higher dose rate of 1 × 10 ¹⁰ n / cm ² / sec, neutrons can be counted with good accuracy.

Claims (7)

A counting method is a method of setting a wave height recognition threshold and counting signal frequencies greater than the wave height recognition threshold. When the target component of the counting is to show a spike in the wave height distribution spectrum, it is selected from the Gaussian function simulation for the aforementioned spike. Arbitrary values in a range of (μ-0.4σ) to (μ + 0.6σ) obtained in combination are set as the aforementioned wave height identification threshold, where μ is the average wave height value and σ is the standard deviation. A counting method is a method for setting a wave height recognition threshold and counting a signal frequency greater than the wave height recognition threshold when the wave height distribution spectrum shows a plurality of peaks. The wave height recognition threshold is at ( μ-0.4σ) to (μ + 0.6σ), where μ and σ are average wave height values μ and standard deviation σ when a Gaussian function approximation is performed on the largest peak among the plurality of peaks. A counting method is to set the wave height identification threshold when there are multiple spikes in the wave height distribution spectrum obtained by plotting a pulse signal derived from radiation or electromagnetic waves based on its wave height value and the count of each wave height value. And the method of counting signal frequencies greater than the threshold value of the wave height identification, the average wave height value obtained by approximating the largest peak of the plurality of peaks with a Gaussian function is set to μ, and the standard deviation is set to σ, the wave height is identified The threshold is in the range of (μ-0.4σ) ~ (μ + 0.6σ). A radiation detection device is a radiation detection device used in combination of a scintillator, a photodetector, and a wave height recognition circuit. The wave height recognition circuit is provided with the following wave height recognition circuit: an amplifier, which amplifies the signal of the light detector. ; An AD converter converts the analog signal output of the aforementioned amplifier into a digital signal; and a digital data processing device performs signal processing on the digital signal output of the aforementioned AD converter; Apply the counting method described in any one of the first to third patent scopes for signal processing. The radiation detection device as described in item 4 of the scope of patent application, wherein the scintillator is LiCa _x Sr _1-x AlF ₆ crystal, and x is 0 ~ 1. The radiation detection device according to any one of items 4 or 5 of the scope of patent application, wherein the radiation detected by the scintillator is an accelerator neutron. The radiation detection device according to any one of claims 4 to 6, wherein the radiation detected by the scintillator is a neutron for boron neutron capture therapy.