JPH0462033B2 - - Google Patents

Description

[Detailed Description of the Invention] (a) Field of Industrial Application This invention relates to improvement of a semiconductor radiation position detection device, which is commonly used, for example, in nuclear medicine diagnosis (RI imaging). It is useful for obtaining the distribution of specific RI nuclides by detecting the two-dimensional position of radiation of a specific energy, such as a scintillation camera, or is used in other fields such as science and engineering. Ru.

(b) Prior art In the conventional two-dimensional semiconductor radiation position detection device, for example, when the number of detection pixels is n×n, one of them is
Providing a preamplifier for each pixel would be very expensive and complicated if n is large, so the number of preamplifiers can be reduced by extracting signals from two sides, such as the orthogonal strip type electrode method (Checker-Board Type). . For example, “L. Kaufman et al., IEEE
Transaction on Nuclear Science, vol.NS−
25, No. 1, 1978'',
By connecting a preamplifier to each row on the front side and each column on the back side, the number of preamplifiers is generally 2n. Furthermore, by adopting a charge division type position detection method as disclosed in Japanese Patent Laid-Open No. 53-4389, the number of preamplifiers can be reduced to, for example, four. However, since the semiconductor radiation detector is configured to extract signals from both the rectangular anode and cathode that are arranged perpendicular to the front and back surfaces, it has the disadvantage that the detector is complicated to manufacture and assemble. be. Moreover, since signals are also extracted from the electrode on the radiation incident side, non-uniform radiation scattering tends to occur, and there are also drawbacks such as difficulty in making close contact with the collimator. Furthermore, although a signal can be extracted from one electrode by DC coupling, a signal must be extracted from the other electrode by AC coupling, which increases noise, and in the case of resistive charge splitting, etc., the space in the positional direction relative to the AC coupling side It has the disadvantage of poor resolution.

In addition, in the past, Ge was used as a semiconductor radiation detector.
etc., but this requires low-temperature cooling, complicates production and assembly, and is expensive to maintain.Furthermore, it is difficult to fit tightly with the collimator, making it difficult to move or rotate. There are some points. On the other hand, if a room-temperature compound semiconductor radiation detector such as CdTe or HgI ₂ is used, low-temperature cooling is not required, but on the other hand, the energy resolution is lower than that of Ge etc. due to poor charge (especially hole) collection characteristics. It has the disadvantage of being low.

(c) Purpose This invention is easy to manufacture and assemble, has a small number of preamplifiers, can be made inexpensive and compact, can be used at room temperature, and can further reduce non-uniform radiation scattering and can be in close contact with a collimator, resulting in improved system performance. It is an object of the present invention to provide a semiconductor radiation position detection device that can improve the spatial resolution as described above and also improve the inherent deterioration of the spatial resolution caused by imperfection of charge collection characteristics.

(d) Configuration In the semiconductor radiation position detection device according to the present invention, equipotential electrodes are provided on one surface of the radiation incident side of the radiation detection element made of a compound semiconductor for room temperature use, and arranged in one direction of a two-dimensional coordinate system on the other surface. Detecting sections each having at least two signal extraction electrodes are arranged independently in other directions of the two-dimensional coordinate system. In each of the detection sections, each signal extraction electrode is connected to each connection point of a series-connected impedance group, and both ends of the series-connected impedance group are DC-coupled to a pair of preamplifiers. The signal obtained by symmetrically adding the outputs of a pair of preamplifiers is an energy signal,
The signal obtained by asymmetrically adding the outputs of a pair of preamplifiers is divided and normalized by a signal proportional to the above energy signal, and the position within each detection unit, that is, the position in one direction of the above two-dimensional coordinate system. Obtain a signal representing . In another method, the position of the two-dimensional coordinate system is determined by which of the one-dimensional position detection systems the signal originates from. Furthermore, a pulse height analyzer is provided for determining whether or not the above-mentioned energy signal falls within a predetermined energy range.

(e) Example Figures 1A and B and Figure 2 show the first example, and in these figures, for the sake of simplicity, it is assumed that the detection elements are arranged in a 4 x 4 matrix. There is. The detection elements 3 are made of a single crystal of a high atomic number room temperature radiation detection compound semiconductor such as CdTe or HgI ₂ , and are arranged in a 4×4 matrix on the insulating plate 1 via equipotential electrodes 2. Ori,
Radiation to be measured, such as gamma rays, is incident from the direction of this insulating plate 1 through a collimator (not shown). This equipotential electrode 2 is a cathode in this embodiment, and is made approximately equipotential by applying a negative bias. Each detection element 3 has signal extraction electrodes (anodes in this embodiment) 411 to 444 on the side facing the cathode 2, respectively. Each of the anodes 411 to 414 arranged in the first row is connected to a predetermined position of a resistor group 51, and both ends of the resistor group 51 are connected to a pair of preamplifiers 61a and 61b. In this embodiment, both connections are direct current connections. The charges generated in the anodes 411 to 414 are divided by the resistor group 51 and sent to the preamplifier 61.
a, 61b, each output V1a, V1b of the preamplifiers 61a, 61b includes position information in the X direction indicating which detection element 3 corresponding to each of the anodes 411 to 414 the radiation has entered. . Although omitted in FIG. 1A, other anodes 421 to 424, 4 in the second, third, and fourth rows
31 to 434 and 441 to 444 are also resistor groups 52 to 444 corresponding to the respective rows, as shown in FIG.
A pair of preamplifiers 6 corresponding through each of 54
2a, 62b, ..., 64a, 64b,
Outputs V2a, V2b,
..., V4a, V4b.

Preamp output V1a, V1b,..., V4a,
Each of V4b has a corresponding arithmetic/shaping amplifier circuit 7.
1 to 74, respectively. The arithmetic/integer amplification circuit 71 in the first row will be explained. first,
V1a and V1b are asymmetrically added (including subtraction and using only one of V1a and V1b) by an asymmetrical addition circuit 8, and the output of this asymmetrical addition circuit 8 is passed through a waveform shaping amplifier 10, such as a Gaussian filter, etc. After being subjected to waveform shaping and amplification, the signal is taken into the sample hold circuit 13 at the timing of the timing signal A1 and held. The output signal X1 is taken into the sample hold circuit 18 at a timing determined by the timing signal E through the analog switch 15 which is opened and closed by the gate signal C1, and is held there.
Further, V1a and V1b are symmetrically added in a symmetrical addition circuit 9, and the output of this symmetrical addition circuit 9 is subjected to a waveform integer and amplification by, for example, a Gaussian filter in a waveform shaping amplifier 11, as described above. ,
Timing signal B1 to sample hold circuit 14
It is captured and retained at the same timing. The output signal z1 is taken into the sample hold circuit 19 at a timing determined by the timing signal F through the analog switch 16 which is opened and closed by the gate signal D1, and is held there. Furthermore, a differential pulse output is generated from the waveform shaping amplifier 11, and this differential pulse output is inputted to a discriminator and pulse generation circuit 12, and when the wave height of the differential pulse output exceeds a certain level, a pulse signal is generated. T1 occurs.

The same goes for the operational/shaping amplifier circuits 72 to 74 in other rows, and the signals Ai, Bi, Ci, Di (i is a subscript indicating the row,
In this case, the output T
Outputs Ti, xi, and zi corresponding to 1, x1, and z1 are obtained, respectively.

The timing control circuit 17 receives pulse signals T1 to
According to each of T4, corresponding signals A1 to A4,
It is configured to output B1 to B4, C1 to C4, D1 to D4, signals E and F, and signals H and H', which will be described later, at appropriate timing, and gate signals C1 to C4 of analog switches,
The timing of D1 to D4 is adjusted so that even if radiation incident events occur simultaneously in different rows, there will be fewer omissions in counting.

One of the signals x1 to x4 and one of the signals z1 to z4 in the corresponding row are selected by the analog switch and input to sample and hold circuits 18 and 19, respectively, from which outputs x and z is obtained. The output z is a signal corresponding to the energy of the incident radiation, and is sent to the division circuit 20 and also to the sample and hold circuit 22, and is taken into this sample and hold circuit 22 at a timing determined by the timing signal H.
is held and an energy signal Z is obtained. Further, this output z is sent to a wave height analyzer 23, and in this wave height analyzer 23, it is determined whether the wave height of the output z falls within one or more energy ranges predetermined according to the nuclide. , an unblank signal and a discrimination signal G are generated. This discrimination signal G is input to the timing circuit 17, and time adjustment with other timing signals is performed. The division circuit 20 outputs a signal proportional to the amount obtained by dividing the output x by the output z, and this output is taken into the sample hold circuit 21 according to the timing signal H and held, thereby obtaining the position signal X.

The aforementioned pulse signals T1 to T4 are sent to the encoder 24.
A code signal corresponding to each row is obtained from the encoder 24, and this code signal is taken into the latch circuit 7 by the timing signal H' through the register and control circuit 25, and is held therein. /A converter 28 to position signal Y
is obtained. At this time, the latch circuit 27 adds the output of the random number generator (for example, composed of a counter, etc.) 26 to the lower bits of the code signal, and
This is designed to make the image smoother. Note that the register and control circuit 25 is a register (for example, a register) controlled so that the position signal Y always corresponds to the position signal
(configured by a FIFO memory, etc.), and is controlled by each signal of the timing control circuit 17.

The position signals X, Y and the unblank signal are input to a display device (for example, a CRT) (not shown) through a suitable driver circuit, or are digitized and stored in a memory matrix to produce a two-dimensional image. is obtained.

In the above configuration, the signal x corresponding to the position in the X direction is converted to the signal z corresponding to the energy for each event.
Since it is standardized by dividing by , the following problem is solved. Generally, in charge-splitting position detectors, the spatial resolution tends to be proportional to the energy resolution, and in room-temperature compound semiconductor radiation detectors, the energy resolution is lower than that of low-temperature Ge detectors due to imperfection in charge collection. The problem is that it is generally inferior to

In general, the energy resolution △E of a conductor detector
is expressed as △E=√(△) ² + (△) ² + (△) ² . Here, △En is the circuit noise of the detector-preamplifier system (including leakage current, etc.), △Ec is a term based on incomplete charge collection, and △Ef is a statistical error in the number of charges generated in the semiconductor (Fano effect). including)
It is a term based on , and usually △Ef≪△En, △Ec. Furthermore, in the case of a Ge detector, △Ec≪△En, but in the case of a compound semiconductor detector, it varies depending on the thickness of the sensitive layer and the bias conditions, and γ of several tens of keV or more
In applications that detect lines, △Ec>~△En is often the case.
As a result, the energy resolution is inferior to that of Ge detectors.

Here, the present inventor discovered that by performing the above-mentioned normalization, it is possible to eliminate the deterioration in spatial resolution based on ΔEc.

Note that it is obvious that the deterioration in spatial resolution based on ΔEn cannot be improved even if normalization is performed.
Therefore, the characteristics are different from the normalization by division described in JP-A-53-4389. In other words, in the above publication, as shown in the description in the lower right column of page 38, the standardization is particularly effective when measuring multiple spectra or multiple nuclides.
In the case of normal single energy measurements, there is no need to divide by the energy signal for each event, and it is sufficient to adjust (attenuate) the amplification degree of one signal depending on the energy value to be measured. On the other hand, in the configuration of the present invention described above, it is essential to perform division by the energy signal for each event even when measuring a single energy. In addition, since the one-dimensional radiation position detection system that detects the position in the X direction using a charge decomposition method is arranged in parallel in the Y direction to perform two-dimensional position detection, a preamplifier is provided for each detection pixel. With a smaller number of preamplifiers compared to the previous one, it is possible to create a configuration in which signals can be extracted only from either the anode or cathode side. Therefore, it is possible to choose not to take out signals through AC coupling, thereby avoiding the increase in noise caused by AC coupling, and also to arrange equipotential electrodes on the radiation incidence side and avoid placing electrodes for signal extraction. .

These eliminate the need for low-temperature cooling, and
Since no signal is extracted from the radiation entrance electrode, uneven scattering of radiation by the entrance electrode is improved, and the entrance electrode and collimator can be brought into close contact, improving the resolution of the system. The result is an RI imaging device with higher energy and spatial resolution than scintillation cameras.

In addition, the above shows one example,
Various structural changes are possible. That is, for example, the filters in waveform shaping circuits 10 and 11 in FIG. The system may have a partially different configuration. Further, although one division circuit 20 is provided above, it may be provided for each arithmetic/shaping amplifier circuit, or may be divided into a plurality of groups, and one division circuit may be provided for each group. It is also possible to adopt a configuration in which the outputs of the sample and hold circuits 18 and 19 are A/D converted and subsequent circuits perform digital calculations.

Furthermore, for convenience of explanation, the case of a 4×4 matrix has been described above, but it goes without saying that the present invention can be similarly applied to cases of many other arrays.

Furthermore, in FIG. 1, each of the detection elements 3 is provided with one signal extraction electrode 411 to 444, but as shown in FIGS. An electrode 4 may also be provided. Further, in the above embodiment, only one charge decomposition type one-dimensional position detection system is formed in the X direction as shown in FIG. 1B, but for example, as shown in FIG. A plurality of them may be formed.

Although the resistor group 51 is used in the above embodiment as a means for charge division, it is also possible to use other impedances such as a capacitor.

Regarding the arrangement of the detection elements and signal extraction electrodes, an arrangement other than the usual matrix as in the above embodiment is also possible, and a coordinate system other than the X-Y orthogonal coordinate system can be used.

In addition, in CdTe, HgI _2, etc., holes generally have a smaller product of mobility μ and average lifetime τ than electrons, so the hole collection characteristics are poor, so they are usually treated with radiation incident as described above. Although the equipotential electrode 2 disposed on the side serves as a cathode, not only a negative bias but also a positive bias may be applied to the equipotential electrode 2.

Furthermore, with HgI _2, the polarization effect due to hole capture is a problem, but as a countermeasure, it is sufficient to temporarily reverse the bias applied to the equipotential electrode 2 from time to time. That's easy.

As shown in FIG. 5, the detection elements 3 are arranged diagonally,
It is also possible to configure the radiation to be incident on the equipotential electrode 2 from a direction other than perpendicular to the equipotential electrode 2 in order to improve the relationship between the detection efficiency and the charge collection characteristics related to the thickness of the detection element 3. It is also possible to improve the relationship between detection efficiency and charge collection characteristics by stacking the two-dimensional array of detection elements 3 in layers as shown in Figure 1 or Figures A and B in Figure 3 to create a three-dimensional array. It is.

Furthermore, it is also possible to configure an Emission Computed Tomograph device having a plurality of slice layers by arranging the two-dimensional position detecting devices described above in a ring shape, for example.

(f) Effects According to the present invention, since the equipotential electrode is provided on the radiation incident side and the signal extraction electrode is formed only on the surface other than the radiation incident side, manufacturing and assembly are easy and miniaturization is possible. In addition, the scattering of non-uniform radiation is improved, and it is possible to make close contact with the collimator, which improves the resolution of the system.In addition, the signal can be extracted with DC coupling, and the noise increase (but not with AC coupling) is improved. (deterioration of energy resolution and spatial resolution) can be improved. Furthermore, since a charge division method is adopted for position detection in one direction of the two-dimensional coordinate system, the number of preamplifiers can be reduced and the cost can be reduced. In addition, since a compound semiconductor for room temperature is used as the radiation detection element, it can be used at room temperature, is convenient to use, as it can be easily moved and rotated, and there is no maintenance cost. To deal with incomplete charge collection that occurs when room-temperature compound semiconductors are used, the unidirectional position signal obtained by the charge decomposition method is divided by the energy signal in real time for each event. Since this is normalized and handled, it is possible to prevent a decrease in spatial resolution due to incomplete charge collection.

[Brief explanation of drawings]

Figures 1A, B, and 2 show an embodiment of the present invention. Figure 1A is a schematic diagram showing the spatial arrangement relationship, and Figure 1B is the connection relationship between the signal extraction electrode and the preamplifier. Fig. 2 is a block diagram of the signal system, Fig. 3 A and B are schematic diagrams showing other examples of spatial arrangement, and Fig. 4 is a connection relation between the signal extraction electrode and the preamplifier. A wiring diagram showing another example, and FIG. 5 is a schematic diagram showing still another example of the spatial arrangement relationship. 1... Insulating plate, 2... Equipotential electrode, 3... Detection element,
4,411-444...Signal extraction electrode, 51-
54..., 5n, 51'-54',..., 5n'...resistance group, 61a-64a,..., 6na, 61a'-64
a',...,6na',61b~64b,...,6nb,6
1b'~64b',...,6nb'...Preamplifier, 71~
74... Arithmetic/shaping amplifier circuit, 8... Asymmetrical addition circuit, 9... Symmetrical addition circuit, 10, 11... Waveform shaping amplifier, 12... Pulse generation circuit, 13, 14, 1
8, 19, 21, 22...sample hold circuit,
15, 16... Analog switch, 17... Timing control circuit, 20... Division circuit, 23... Wave height analyzer, 24... Encoder, 25... Register and control circuit, 26... Random number generator, 27... Latch circuit,
28...D/A converter.

Claims (1)

[Claims] 1 Radiation detection elements made of room-temperature compound semiconductors are arranged continuously or in isolation in one direction of a two-dimensional coordinate system, and equipotential electrodes that are maintained at approximately equal potential are placed on one surface of the elements on the radiation incident side. At the same time, on the other surface of this element, radiation one-dimensional position detecting sections each having at least two or more signal extraction electrodes arranged in the above-mentioned direction are arranged independently of each other in the other direction of the above-mentioned two-dimensional coordinate system, and furthermore, these For each of the radiation one-dimensional position detection sections, each signal extraction electrode is connected to each line connection point of the series-connected impedance group, and both ends of the series-connected impedance group are DC-coupled to a pair of preamplifiers, The signal obtained by symmetrically adding the respective outputs of the pair of preamplifiers is used as an energy signal, and the signal obtained by adding the respective outputs of the pair of preamplifiers asymmetrically is divided by a signal proportional to the above energy signal and normalized.
A one-dimensional position detection system is formed to obtain a position signal in one direction of the two-dimensional coordinate system, and a position signal in another direction of the two-dimensional coordinate system is obtained depending on which of the one-dimensional position detection systems the signal is generated from. 1. A semiconductor radiation position detection device characterized by comprising: a pulse height analyzer for determining whether or not the energy signal falls within a predetermined energy range.