JPH0430554B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (a) Field of Industrial Application This invention relates to improvements in scintillation cameras.

(b) Prior art As is well known, scintillation cameras generate scintillation light by injecting radiation into a scintillator, and transmit this light to photomultiplier tubes (hereinafter referred to as photomultiplier tubes) arranged in large numbers on the back of the scintillator.
PMT), the current output from each PMT is converted into a voltage output by each preamplifier, and the output of each preamplifier is input to a position calculation circuit such as a weighting and adding circuit. The image is obtained by calculating the location of the occurrence.

Conventionally, in order to improve the spatial resolution of scintillation cameras, it has been proposed to provide each preamplifier with a so-called threshold type nonlinear characteristic in which the amplification factor for low input signals is zero or sufficiently small (Japanese Patent Publication No. 51) -Refer to Publication No. 29839). However, in this conventional example, the energy signal is also obtained by adding up all the powers of the preamplifier, which has nonlinear characteristics, so there is a decrease in energy resolution due to an apparent decrease in the amount of light emission, and the wave height of the energy signal is directly above the PMT. This results in position dependence such that it is high at certain positions and low at positions between PMTs, and the resulting non-uniformity is promoted.

By the way, in this scintillation camera, if the distance between the scintillator and PMT is narrowed,
Although the spatial resolution improves, spatial distortion (nonlinearity) becomes significant at the position directly above the PMT, resulting in a so-called hot pattern. Therefore, conventionally, so-called suppression type nonlinear circuits have been developed in which the amplification factor for high input signals is reduced to improve spatial distortion while improving spatial resolution by narrowing the spacing between the scintillator and PMT. It has been proposed to connect the following after each preamplifier (see Japanese Patent Publication Nos. 56-51312, 1984-84378, 1984-44998, etc.).
By connecting the nonlinear circuit after the linear preamplifier in this way, the energy resolution, the position dependence of the energy signal wave height, and the nonuniformity caused by this become independent. However, in this conventional example, since the preamplifier is configured to have integral characteristics and is input to the nonlinear circuit after waveform shaping, it is difficult to maintain the suppression characteristics stably and smoothly against temperature changes. In addition, specifically, since the suppression characteristics are obtained by diode clipping, the temperature dependence of the suppression point is sensitive, and the threshold voltage of the diode (0.6 to 0.7 volts at room temperature in the case of silicon) is set so that the nonlinear characteristics are smooth. In comparison, if the maximum value of the input signal wave height is not sufficiently large, the temperature dependence of this suppression nonlinear characteristic will be significant (note that if the input signal wave height is sufficiently larger than the threshold voltage of the diode, smooth nonlinear characteristics will be obtained. The drawback is that it is difficult to achieve a dynamic range from low energy to high energy.

(C) Purpose This invention basically improves the spatial resolution and improves spatial distortion and non-uniformity based on it in a scintillation camera, and further improves the ability of the scintillation camera to improve spatial distortion and non-uniformity based on it. The purpose is to achieve the above-mentioned items even when measuring nuclides, and to stably achieve the above-mentioned items even with temperature changes.

(d) Configuration The scintillation camera according to the present invention has the following features:
a first type high level filter circuit having a characteristic of amplifying signals above a first type threshold level, the threshold level of which is varied by an energy signal obtained for each radiation incident event;
a second type high level filter circuit, which has a characteristic of amplifying signals above a second type threshold level and whose threshold level is changed by an energy signal obtained for each radiation incident event; Each output terminal of each preamplifier is connected in parallel via a delay component, and a position calculation circuit applies a weight corresponding to the position of the photomultiplier tube to the output of each high level filter circuit and the first type high level filter circuit. The signal obtained by multiplying and adding the products and the weights corresponding to each of the second type high level filter circuits is integrated for a certain period of time and then normalized to obtain a position signal.

(E) Embodiment As shown in FIG. 1, a large number of PMTs 3 are arranged on the back surface of a scintillator 1 with light guides 2 interposed therebetween. Although the thickness of the light guide 2 (and the thickness of the scintillator 1) is made thinner to obtain high resolution, the spatial distortion (nonlinearity) resulting from the formation of hot spots directly above each PMT 3 is reduced. The structure is such that the non-uniformity becomes significant.
This spatial distortion and the non-uniformity caused by it are improved by the action of the high-level filter circuit group 7, which will be described later, and as a result, both the spatial resolution and the spatial distortion (and the non-uniformity caused by this) are improved. so that we can improve.

When radiation enters scintillator 1, scintillation light is generated within this scintillator 1,
This light is guided to each PMT 3 via a light guide 2. In this way, a current signal is generated from each PMT, and each current signal is converted into a voltage signal by the corresponding preamplifier 4 linearly and in a form that is not significantly integrated. After the voltage signals from each preamplifier 4 are added in an adder circuit 5 (the signal after this addition is approximately proportional to the energy of the incident radiation, strictly speaking, the various signals in the scintillator 1, light guide 2, and PMT 3) are The statistical error of
The energy signal wave height position dependence (varies due to non-uniformity) related to the arrangement of the PMTs 3, etc.) is sent to an integrating circuit 94 and integrated for a certain period of time to obtain an energy signal. This energy signal Z is passed through the wave height separator 1
A pulse height analyzer 110 determines whether this energy signal Z is within a predetermined energy window, and if so, an unblank signal is generated from the pulse height analyzer 110. Further, the voltage signals from each preamplifier 4 are also input to each corresponding high level filter circuit group 7, and the obtained multiple output signals are led to a position calculation circuit 8, where three types of signals x, y, and z are obtained. It will be done. The threshold level control voltage supply circuit 6 receives the input signal from the adder circuit 5 (or applies this signal to the integrator circuit 94).
A control voltage for controlling the threshold level in the high level filter circuit group 7 is generated in response to the output signal of the high level filter circuit group 7. signal x,
y, is integrated for a certain period of time in each of the integrating circuits 91 to 93, and the obtained signals x', y',' are passed through the sample and hold circuits 101 to 103 to the dividing circuit 11.
1,112, x'/z' and y'/z' divisions are performed, normalization is performed so that the image size does not depend on radiation energy, and position signals X, Y are obtained.

The high level filter circuit group 7 and the position calculation circuit 8 are constructed as shown in FIG. The high-level filter circuit groups 7 are provided in the same number as the PMTs 3 (for example, 37 or 61), and each of the high-level filter circuit groups 7 includes the same number of high-level filter circuits, which are arranged in parallel. connected, and the number thereof is two or three or more.
This figure 2 shows two examples in particular, but 3
It is likewise possible to have more than one. i-th
The high level filter circuit group 7i that receives the signal of the PMT 3 via the i-th preamplifier 4 includes a delay component 701i, a first type high level filter circuit 71i and a second type high level filter circuit 712i connected in parallel to the output terminal of the delay component 701i. The first type threshold level in the first type high level filter circuit 711i, and the second type high level filter circuit 7
The second type threshold level at 12i is controlled by the first and second control voltages supplied from the threshold level control voltage supply circuit 6. High level filter circuit 711
i, 712i has the characteristic of amplifying only when the input voltage exceeds the threshold level. Although this threshold level is different between the first type and the second type, it is basically the same for each PMT3. However, in order to improve spatial distortion in the peripheral part of the visual field and non-uniformity based on the positional dependence of the energy signal wave height, only the central PMT3 is used for the peripheral part PMT3.
In accordance with adjusting the preamplifier 4 output signal to be larger (or smaller) than that of PMT 3, the threshold level will also be higher (or lower) for PMT 3 at the periphery than for PMT 3 at the center. Therefore, it is preferable to configure the first and second control voltages for threshold level control to be slightly different from each other. The delay component 701i is connected to a high level filter circuit 711 as shown in FIG.
i and 712i may be common, but may be separate as will be described later.

The outputs of the high level filter circuits 711i and 712i are transmitted to the respective amplifiers 80i and 806 in the position calculation circuit 8.
i and predetermined amplification factors Gx1i, Gy1i,
The signal is multiplied by each of Gz1i, Gx2i, Gy2i, and Gz2i, is input to one of the adder circuits 811 to 813, and is added to each output of a similar amplifier other than the i-th amplifier, which is omitted in the figure. Addition circuit 811
Each output of ~813 is connected to a corresponding voltage-current converter 8.
21 to 823, the current signals x, y, z
becomes. Each amplification factor G of the above amplifiers 801 to 806
The subscripts x, y, and z correspond to the coordinates x, y and the quantity z that has a positive correlation with the energy signal Z, respectively, and the subscript 1 corresponds to the high level filter circuit 711i,
712i corresponds to whether it is the first type or the second type, and the subscript i corresponds to the i-th PMT3. Then, each of these amplification factors is determined by
It corresponds to the product of the weight determined by the PMT3 position and the weight that differs depending on the type of high-level filter circuit specific to the present invention.

It should be noted that the configuration of the position calculation circuit 8 shown in FIG. It is possible to replace it with a dynamic amplifier, and it is also possible to have the voltage-current converter characteristics in the adder circuits 811 to 813 or the above-mentioned differential amplifier. Further, as shown in FIG. 3, the same type of high-level filter circuits are provided for each type, that is, first-type high-level filter circuits 711i, 711j, etc., and second-type high-level filter circuits 712i, 712j. etc., each of the adder circuits 831 to 836 performs weighted addition according to the position of the PMT 3, and then each of the adder circuits 841 to 843 performs weighted addition according to the type of high-level filter circuit. By reversing the order of the two weighted additions, as shown in FIG.
i, 850j, etc., performs weighted addition according to the type of high-level filter circuit, and then adder circuit 8
A configuration in which weighted addition regarding the position of the PMT 3 is performed in 51 to 853 is possible. In the case of FIG. 4, an open collector current output signal addition circuit may be used for the weighted addition circuit 850 related to the type of high-level filter circuit.

The weights mentioned here generally include positive and negative signs, and weights that differ depending on the type of extra-high level filter circuit always require different signs (details will be described later). Therefore, the delay component 701i of the high-level filter circuit group 7i in FIG. 2 is replaced by a delay circuit 7 using a delay line or the like on one side, as shown in FIG.
21i and the other an inverting amplifier 722i, respectively, so that the delay times of both can be made approximately equal.

Delay component 701i delays the preamplifier output signal so that these threshold levels are stable before the preamplifier output signal is input to high level filter circuits 711i and 712i. That is, these threshold levels are determined by the threshold level control voltage supply circuit 6.
As mentioned above, the output signal of the adder circuit 5 (or the output signal of this signal integrating circuit 94 is added to the threshold level control voltage supply circuit 6). ) is input and the first and second control voltages change according to the energy signal.
The preamplifier output signal is delayed until these control voltages are stabilized. In order to keep the control voltage stable especially when these high-level filter circuits are in operation, the output signal of the adder circuit 5 and the output signal of the integrator circuit 94 are mixed as described above to control the threshold level. It is conceivable to input it to a voltage supply circuit.

FIG. 6 shows an example of the input/output characteristics of the high level filter circuit. The threshold level of the first type high level filter circuit is called the first type threshold level, and the threshold level of the second type high level filter circuit is called the second type threshold level. from the lowest to the 1st, 2nd,...
...says. The solid line shows the case of high energy, and the dashed-dotted line shows the case of low energy, and each threshold level is configured to change continuously or discontinuously with respect to the energy signal. In general, the relationship between the threshold level and the energy signal should be configured to be linearly proportional, but strictly speaking, the higher the energy of the radiation, the deeper the light emitting position in the scintillator 1 will be. Since the tendency for hot spots to occur is likely to be emphasized, a somewhat nonlinear change may be made to correct this tendency. Figures 7A and B show the PMT3 when radiation is absorbed and emitted within the scintillator 1 at a certain position on the central axis of the PMT3.
This figure illustrates the relationship between the output waveform of the preamplifier and each threshold level, where A is a case of low energy and B is a case of high energy.

In this way, since each threshold level is changed for each event of radiation incidence according to the energy signal, spatial distortion is significantly improved regardless of the energy in single-energy measurements. In addition, spatial distortion is significantly improved even in the case of measurements of other nuclides or multispectral measurements.

Since the signals passing through the high level filter circuit group 7 are weighted and added in the position calculation circuit 8 and then integrated for a certain period of time in the integration circuits 91 to 93, the following effects are obtained. First, for a certain energy, the input/output characteristics of each high-level filter circuit are as shown in FIG. When the effect of addition is removed, the relationship between the outputs of the integrating circuits 91 to 93) and the preamplifier output is as shown in FIG. 8B. The preamplifier output waveform is not a square wave (Fig. 7 A, B)
Therefore, as shown in FIG. 8B, the preamplifier output has smooth characteristics within a range that is slightly larger than the threshold level. Next, a case in which weighted addition is added will be explained. For convenience of explanation, as shown in FIG. 4, each output of the high-level filter circuit group 7i is first processed by a weighting addition circuit 850i related to the type of high-level filter circuit.
This will be explained assuming that it is input. In this way the first
type, second type high level filter circuit 711i, 7
Since each output of 12i (shown in FIG. 8A) is added by the weighted addition circuit 850i, the addition circuit 85
The output of 0i is as shown in FIG. 8C. In addition, the first
Generally, the weight for the high level filter circuit 711i has a larger absolute value and has a different sign than the weight for other high level filter circuits 712i, etc. In FIG. 8C, the solid line shows the case where only the first type high level filter circuit and the second type high level filter circuit are included, and the dotted line shows the case where the third type high level filter circuit is further included. And this figure 8C
When the output of the adder circuit 850i is integrated for a certain period of time, the relationship between the integrated value and the preamplifier output is as follows.
As shown in Figure D, it becomes smooth for the same reason as in Figure 8B. Also in FIG. 8D, the solid line indicates the case where only the first type high level filter circuit and the second type high level filter circuit are included, and the dotted line indicates the case where the third type high level filter circuit is further included. In the end, as shown in FIG. 8D, it can be seen that a smooth threshold effect and a smooth suppression effect can be obtained. Since the suppression effect is smooth in this way, it is usually sufficient to use only one type of high-level filter circuit for suppression (here, the second type high-level filter circuit 712i), and spatial distortion is significantly improved. No artifacts appear either. Even in special cases, that is, when the scintillator 1 and the light guide 2 are made extremely thin, sufficient effects can be obtained with two types of high-level filter circuits (here, the second and third types of high-level filter circuits). In addition,
For convenience of explanation, the configuration of FIG. 4 has been explained, but
The same applies to other configurations as long as only the weighted addition and integration related to the type of high-level filter circuit are taken out.

The threshold effect by the first type high level filter circuit 711i improves the spatial resolution, but it also results in further aggravating spatial distortion such as a hot spot at a position directly above the PMT3. It is necessary to further deepen the suppression effect of the type high level filter circuit 712i (or even the third type high level filter circuit).

By using a current-switching threshold circuit or the like as the high-level filter circuit, stable characteristics with sharp characteristics near the threshold level and almost no dependence on temperature changes can be achieved, as shown in Figure 8A. A high level filter circuit can be easily constructed. Therefore, as shown in FIGS. 8B and 8D, it is easy to make the characteristics after smooth integration stable with respect to temperature.

Note that in this embodiment, the integrating circuit 91 in FIG.
-94 integration times are provided for different lengths and short lengths so that they can be selected (circuits related to timing are omitted in the block diagram of FIG. 1). and,
Select long-time integration when measuring low coefficient rates, and select short-time integration when measuring high coefficients. In this way, it is possible to obtain high resolution etc. when the coefficient rate is low, and when measuring a high coefficient rate, it is possible to obtain excellent coefficient rate characteristics and relatively good resolution etc. for the high coefficient rate. By the way, in the conventional example mentioned at the beginning (the above-mentioned publications), the integration time is generally constant, so if the integration time is increased to maintain high resolution and high energy resolution at low coefficient rates, the coefficient rate characteristics will change. On the other hand, if it was kept short, the spatial resolution and energy resolution would deteriorate (Also, there is a configuration that detects peaks without integrating, but the spatial resolution is inferior).

What must be noted here is that a plurality of high-level filter circuits are connected in parallel in the high-level filter circuit group 7. If, as in the conventional case (see the first publication), the threshold circuit and the suppression circuit (usually two stages) are connected in series, then weighted addition regarding the PMT position is performed.
When performing voltage-current conversion and integration, the signal waveform becomes double when passing through the series circuit of the threshold circuit and the suppression circuit, making it impossible to achieve a short integration time. By connecting high level filter circuits in parallel, the waveform becomes less rounded and it becomes possible to shorten the integration time.

Although circuits related to timing are omitted in the block diagram shown in FIG. A delay circuit before the integrating circuit 94 may be provided so that the timing of each integration time with the integrating circuit 94 is the same, or only the integrating circuit 94 may be integrated in advance, and the wave height component may be performed earlier ( In the latter case, by shortening the integration time of the integrating circuit 94 compared to the integrating circuits 91 to 93 and appropriately adjusting the timing, it is possible to obtain high resolution considering the high coefficient rate characteristic.

(f) Effects According to the present invention, firstly, it is possible to improve spatial resolution and improve spatial distortion and non-uniformity based thereon, and secondly, it is possible to achieve these improvements with respect to radiation in a wide energy range, multispectral and It can be satisfied even when measuring multiple nuclides, and thirdly, the first and second functions can be realized stably even against temperature changes.

[Brief explanation of drawings]

Figure 1 is a block diagram of one embodiment of this invention;
The figure is a block diagram showing the main parts of this embodiment.
Figures 4 and 5 are block diagrams showing the main parts of the modified example, Figure 6 is a graph showing the input/output characteristics of the high level filter circuit, and Figures 7A and B are the preamplifier output waveforms and thresholds. Graphs showing the relationship with the Jord level, Figure 8 A, B, C, and D are graphs for explaining the operation. Figure 8 A is a graph showing the input/output characteristics of the high level filter circuit, and Figure 8 B is the graph showing the input/output characteristics of the high level filter circuit. A graph showing the relationship between the integral value of the high level filter circuit output and the preamplifier output. Figure 8C is a graph showing the relationship between the output when the high level filter circuit output of Figure 8A passes through the adder circuit and the preamplifier output. ,
FIG. 8D is a graph showing the relationship between the integral value of the output of the adder circuit shown in FIG. 8C and the preamplifier output. 1...Scintillator, 2...Light guide, 3
...PMT, 4...Preamplifier, 5,811~8
13...Addition circuit, 6...Threshold level control voltage supply circuit, 7, 7i...High level filter circuit group, 8...Position calculation circuit, 91-94...
Integrating circuit, 101 to 103... Sample hold circuit, 110... Wave height analyzer, 111, 112...
...Division circuit, 701i, 721i...delay component,
711i, 711j...First type high level filter circuit, 712i, 712j...Second type high level filter circuit, 7722i...Inverting amplifier, 801
i~806i...Amplifier, 831~836,84
1-843, 850i, 850j, 850-85
3... Weighting addition circuit, 821-823... Voltage-current converter.

Claims (1)

[Claims] 1. A scintillator, a large number of photomultiplier tubes arranged on the back side of this scintillator, a preamplifier that converts the current signal output from each photomultiplier tube into a voltage signal, and a An adder circuit that obtains an energy signal by adding, and a wave height analysis that determines whether the wave height of the energy signal obtained from this adder circuit is within a preset energy window and generates an unblank signal when it is within a preset energy window. A scintillation camera that has a position calculation circuit that calculates a scintillation position based on the output of each of the preamplifiers, and shortens the distance between the scintillator and a large number of photomultiplier tubes. a first type of high level filter circuit, which has a characteristic of amplifying a signal higher than the threshold level, and whose threshold level is changed by the energy signal obtained for each radiation incident event; A second type of high level filter circuit, which has a characteristic of amplifying a signal above the threshold level and whose threshold level is changed by the energy signal obtained for each radiation incident event, is connected to the output terminal of each of the preamplifiers. The position calculation circuit applies a weight corresponding to the photomultiplier tube position to the output of each of the high level filter circuits, and the first type high level filter circuit and the second type high level filter circuit. A scintillation camera characterized in that a position signal is obtained by integrating a signal obtained by multiplying each of the seed height level filter circuits by a weight corresponding to each other and adding the signals for a certain period of time, and then normalizing the signals to obtain a position signal.