FR2467582A1 - Position emission tachograph - by determn. of number of base noise signals due to chance coincidences, subtracted from total coincidences detected (SE 25.5.81) - Google Patents

Abstract

An appts. for carrying out positron emission tachography, i.e. the formation of visible images along one or several planes passing through the body of a patient, incorporates a source of radiations, a number of detectors, first and second coincidence devices and a difference-forming device. The source of radiations directs positrons into the zone being studied. The detectors detect the photons emitted when a positron is annihilated in this area. The first coincidence device determines the number of coincidences in time between photons detected by detectors placed at 180 deg. to one another. The second coincidence device determines simultaneously the number of coincidences in time between photons detected by individual detectors and photons detected with a time lag by detectors placed at 180 deg. The difference-forming device forms the difference between the number of coincidences in time determined by the first coincidence device and the number of coincidences in time detected by the second coincidence device. The difference is a measure of the number of positron annihilations in the zone. The appts. gives a very precise determn. of the number of base noise signals due to chance coincidences and subtracts this base noise from the total number of coincidences detected. Dead time is eliminated.

Description

 The present invention relates generally to positron emission tacography and more specifically to a data acquisition system which allows a very precise determination of the number of background noise signals due to random coincidences and which ensures the subtraction of this noise. background of the total number of coincidences detected, either practically instantaneously or delayed. Input circuits allow operation with virtually no downtime.

 Positron emission tacography is a process of forming visible images along one or more planes passing through the interior of a patient's body. According to this method, radioactive atoms emitting positrons are administered to the patients. Such atoms capable of emitting positrons are for example carbon 11, nitrogen 13, oxygen 15, fluorine 18 and gallium 68, and they emit positrons which travel only a few millimeters before interacting with the material of the patient's body, during a positron annihilation operation. During the operation, the positron interacts with an electron and their mass is transformed into energy in the form of two photons emitted at 1800 from each other, each of the photons having an energy of about 511 keV. positron is detected by determination The sound of the coincidence in time of two photons detected by detectors separated by 1800. The representative drawings of the locations of the atoms emitting the positrons can be restored by a computer, when an arrangement comprising a certain number of detectors surrounding the region to be examined is produced and ensures the detection of coincidences between the photons detected by pairs of detectors placed at the two ends of rectilinear paths.

 A number of systems already implementing this technology are already known. Some of them use detectors intended to determine the interactions of positrons only in a single tomographic plane while others use arrangements of detectors arranged in several planes. Most often, the reconstruction of the positron annihilation positions is ensured by a program computer, according to the number of photons detected coincidentally at the level of detectors separated by line segments, for a certain time. The following documents describe such tomographic systems for diagnosis.

1. Brownell GL, Sweet WH: "LOCALIZATION OF
BRAIN TUMORS WITH POSITRON EMITTERS ". Nucleonics, 11 40-45, 1953.

2. Budinger TF, Gullberg FT: "DRY TRANSVERSE
TION RECONSTRUCTION OF GAMMA-RAY EMITTING RADIONUCLIDES IN
PATIENTS ", Reconstruction in Diagnostic Radiology and Nuclear Medicine.Ter-Pogossian N. #., Phelps ME, Brownell GL

et al, University Park-Press: 315-342, 1977.

3. Cho ZH, Cohen # .B. et-al: "PERFORMANCE AND
EVALUATION OF THE CIRCULAR RING TRANSVERSE AXIAL POSITION
CAMERA (CRTAPC) ". IEE Nucl Sci, NS-24: 532-543, 1977.

4. Muehllehner G, Buchin NP, Dubek GH: "PERFOR MANCIE PARAMETERS OF THE POSITRON IMAGING CAMERA". IEEE
Nucl Sci, NS-23: 528-537, 1976.

5. Phelps NE, Hoffman EJ, Huang SC, and Kuhl
FROM: "ECAT: A NE ## NEW-COMPUTERIZED TOMOGRAPHIC IMAGING
SYSTEM FOR POSITRON-EMITTING RADIOPHARMACEUTICALS ", J Nucl
Med, 19: 635-647, 1978.

6. Phelps ME, Hoffman EJ, Mullani N., Higgins
CS and Ter-Pogossian MM: "DESIGN AND HERFORMANCE
CHARACTERISTICS OF A WHOLE-BODY TRANSAXIAL TOMOGRAPH (PETT III) ". IEEE Nucl Sci, NS-23: 516-522, 1976.

7. Hoffman EJ, Phelps ME, Mullani N. et al "DESIGN AND PERFORMANCE CHARACTERISTICS OF A WHOLE-BODY
TRANSAXIAL TOMOGRAPHY ", J Nucl Med, 17: 493-502, 1976.

8. Ter-Pogossian MM: "BASIC PRINCIPLES OF COMPU
TED AXIAL TOMOGRAPH ", Sem Nucl Med, 7: 109-128, 1977.

9. Ter-Pogossian MM., Mullani NA, Hood J.,
Higgins CS and Currie MC: "A NUIsTI-SLICE POSITRON
COMPUTED TOMOGRAPH (PETT-IV) YIELDING TRANSVERSE
AND LONGITUDINAL IMAGES ", Radiology, 128: 477-484, 1978.

10. Derenzo SE, Budinger TF, Cahoon JL et al: "HIGH RESOLUTION COMPUTED TOMOGRAPHY OF POSITRON
EMITTERS ", IEEE Nucl Sci, NS-24: 554-558, 1977.

11. Budinger TF, Dorenzo SE et al: "EMISSION
COMPUTED AXIAL TOMOGRAPHY ", J Compt Assisted Tomography 1: 31-45, 1977.

12. Todd-Pokropek A., Plummer D., Pizer S.M.

"MODULARITY AND COL SND LANGUAGES IN MEDICAL COMPUTING" in A Review of Information Processing in Medical Imaginq,
Proceedings of the Fifth International Conference. Shiny
BA, Price RR, McClain WJ and Landay MW, ORNL / BCTIC-2,
Oak Ridge National Laboratory, Oak Ridge, TN 426-455, 1977, and 13. U.S. Patent No. 3,984,689 issued October 5, 1976 under the title "SCINTILLATION
CAMERA FOR HIGH ACTIVITY SOURCES ", to Roger E. Arseneau.

 Improving measurement accuracy requires determining as precisely as possible the number of accidental coincidences between separate detectors along line segments. For a given resolution period, set as representing the coincidence resolution time, a number of coincidences of photons appears which is only due to the random distribution of photons and not to real events, these being defined as the annihilation of positrons for which the two photons are detected by the detector arrangement. A method of determining this random background noise comprises moving the system relative to the coincidence and measuring the number of total and random occurrences observed, then subtracting the background noise thus determined from the total number of recorded coincidences. The implementation of this background noise determination process requires the use of considerable time which could also be available for the measurement of true events.

The longer the time used for the determination of the background noise and the more precise this determination, but the more the time available for the counting of the sources decreases and necessitates either an increase in the intensity of the radiation or the acceptance of a less precision.

On the other hand, if the measurement time of the background noise is reduced and introduces an imprecision in the determination of this component, the difference between the total number and the number corresponding to the background noise represents an imprecise value. However, it is this difference which is representative of true measurement events. Various attempts have been made on intermediate measurements such that the recorded number corresponding to the background noise is determined first, before the total number, and before the number corresponding to the background noise, according to a process which also compensates for the variation in the number corresponding to the background noise as a function of time. However, the time spent measuring the background noise is necessarily a time which is not used for the measurement of the sources and consequently the accuracy is reduced below that which could be obtained if all the time was spent on measurement.

 In addition, since the measurements are not carried out simultaneously, sources of error, such as a movement of the patient, can cause a disturbance of the results.

 Another method used includes the determination of coincidences which take place outside the field of vision of the positron emitting system, but around the ~ detectors which delimit this field of vision, with simultaneous measurement of all coincidences within the field of vision. The simultaneous measurement of coincidences in this external field and the use of the measurement as representative of the noise allow the determination of the number corresponding to the background noise at the same time as the measurement of the total number corresponding to the sources and to the background noise is carried out in the field of vision corresponding to the positrons. The resulting number corresponding to the background noise arrives at a memory, as well as the total number, and they then arrive at a suitably programmed computer which ensures the subtraction. This system, although it has the advantage of using efficiently the measurement time, has the disadvantage of determining the background noise outside the field of vision and, if there is a spatial variation in the photonic background noise, the measurement lacks precision.

This characteristic can be very important because there can be a photonic background noise directly related to the radiation in the field of vision.

 The invention relates generally to a positron emission tacography apparatus in which the background noise due to random coincidences of photons is determined from the lines of sight on which the positron annihilations are located. The circuit is constructed so that this background noise can be subtracted almost simultaneously from the measurement of the total coincidences of the photons or can be stored in a temporary memory for subsequent subtraction. In the apparatus, a suitable coincidence resolution time, for example 12.5 ns, is chosen, and the coincidences of the photons detected by detectors arranged at 1800 during this resolution time are recorded as a global number It should be noted that this number corresponds both to the photons which coincide in time because they are produced by an annihilation of a positron, and to the photons which appear randomly and which therefore coincide accidentally. Thus, this total number corresponds to a number of sources <true events) and to a number forming a background noise <random coincidences).

 According to the invention, the number of background noise is determined by measuring the photons detected for the same sets of photon detectors and with the same coincidence resolution period, the signals of a set of detectors being transmitted with a greater delay. at this resolution period. The coincidences thus determined cannot represent true events since they are not simultaneous. Thus, these delayed coincidences constitute a suitable measurement of the level of the random coincidences caused by photons which are in the field of measurement followed the lines of sight.

 The background noise signal formed as described above can either be directly subtracted from the number representing the sources and the background noise, when the measurement is in progress, by using a coder-down arrangement, or subtracted after the end of the measurement, the overall number and the number of background noise being stored in a suitable memory for later calculation. In such a device, the numbers entered in the manner indicated above are transmitted to a programmed computer so that it reproduces the image of the region from which the positrons were emitted and that it presents the image of a classic way in tacographic display devices.

 A problem encountered during the production of tacographic transmission equipment of this type is the loss of signals due to dead time in the circuit.

As positron emissions are due to radioactive decays, they are characterized by a random distribution over time and, for a determined concentration of radioactive atoms, the minimum time between the detected events depends on the system and can be much lower than the average time of separation of these same events. The data acquisition circuit, so that it does not lose numbers of events appearing in a dead time, must have a high speed and a short time resolution corresponding to these minimum separatios In the device according to the invention, the problem is solved by using a fast buffer circuit of the first in-first out type in which the resolution over the acquisition time of the input data corresponding to a short time allowing the detection of very close events while the transfer time of the output data is fixed by the transfer frequency of the memory. The circuit to which these signals are transferred can be characterized by a much longer resolution time, without loss of data due to dead time in the case of high frequency bursts.

Other characteristics and advantages of the invention will emerge more clearly from the description which follows, given with reference to the appended drawings in which
- Figure 1 is a block diagram of a data acquisition device implementing the invention
- Figure 2 is a block diagram of the discrimination and coincidence circuit of the apparatus of Figure 1
- Figure 3 is a block diagram of an input processor portion of the apparatus of Figure 1; and
FIG. 4 is a more detailed block diagram of parts of the processor of FIG. 3.

 The overall configuration of a tacography apparatus implementing detection of coincidence of annihilation is represented in FIG. 1. The whole of the apparatus generally comprises three sub-assemblies, the detection sub-assembly, the data acquisition subset 26 and the calculation subset 27. The invention relates essentially to the data acquisition sub-assembly 26 and to its particular interaction with the detection sub-assembly 25. As previously indicated, the role of the device shown in Figure 1 is the determination of the location and intensity of the radioactive atoms emitting positrons, inside the body of a patient. The basis of this determination is the detection of photons emitted simultaneously at 1800 relative to one another due to the annihilation of the positrons.

 In FIG. 1 which only represents an example of configuration according to the invention, the field of vision of the assembly for detecting the presence of positrons according to their annihilation is marked by the circle 30. An arrangement of detectors of photons, forming groups A to F, surround the field of vision 30. Each group has 11 detectors, for example of the NaI (Tl) type (sodium iodide doping.

The output signal from each of the detectors corresponds to a photon reaching the detection region and creating a flash of light which is transformed into an electrical signal. The signals from the detectors reach the data acquisition sub-assembly 26 which ensures the discrimination in time of events, in each of the detectors and which determines the coincidence relationships of the events detected in the opposite groups.

The computer thus has not only a quantitative measure of the number of coincidences between each pair of detectors of the arrangement but also and at the same time of the number of coincidences which are due to random events. The computer can then determine, for each pair of detectors, what is the frequency of true coincidences due to positron annihilations. In a variant, this function can be fulfilled in the data acquisition sub-assembly 26, transmitting the real or resulting number to the computer practically in real time. In both arrangements, the computer renders the image in the field of vision of the detectors in a manner known for such devices.

 We now consider the detection sub-assembly 25. As indicated above, in the device under consideration, 66 NaI (Tl) detectors form groups of 11 detectors Each detector is mounted in an individual shielding intended to protect it against radiation from outside the field of vision. The latter bears the reference 30 in FIG. 1. A suitable distance between the groups opposite the arrangement of detectors is 100 cm. Under these conditions, the field of vision 30 is a circular zone whose diameter is approximately 50 cm. Each detector is individually shielded in a lead block. A partial lead shield which can be placed in front of each group of detectors, has individual rectangular holes allowing each detector to see all the detectors of the group placed opposite only along lines of aimed. The whole arrangement of the detectors is placed between two lead discs of annular shape having an internal diameter of 60 cm and an external diameter such that they protrude from the sensitive parts of the individual detectors. For example, each lead disc can have a thickness of 38 mm.

In an advantageous configuration, the detectors of
NaI (Tl) have a diameter of 3.8 cm and a length of 7.5 cm. When the holes of the partial shields have a width of 2.3 or 1.5 cm and a length of 3.8 cm in the axial direction, the average resolution of the pairs of detectors in the plane, for a value corresponding to half the maximum , has an overall width of about 1.1 to 0.8 cm. With this configuration, there are 121 lines of sight at 1800 for each pair of groups, i.e. 363 lines of sight in total. As indicated above, the coincidences must be detected between the photons emitted simultaneously at 1800 from each other so that an annihilation of positrons is determined. In the apparatus described, the coincidence of the detected photons must be determined for each of the 363 detector appearances.

 Although a particular hexagonal planar arrangement of detectors has been described, it should be noted that there are numerous other suitable geometric configurations according to the invention, both in a single plane and in several. The essential criterion is that the vsion field is delimited by several pairs of detectors separated by 1800.

 We now consider the data acquisition subset. This receives the signals from each of the individual detectors and processes them so that they form, at the output of the computer, a quantitative measure of the number of coincidences between each pair of detectors in the arrangement, the determination of the coincidences dences being the simultaneous appearance of signals detected during a time of resolution of coincidence fixed in a determined manner. This apparatus also ensures the determination of coincidences which must be attributed to true events and coincidences which are due to random photons, appearing accidentally during the resolution period in the time fixed for the determination of coincidences. these accidental or random coincidences are immediately subtracted from the total number of coincidences and form a measure of the coincidences of the true annihilations of positrons, and it is these only last signals corresponding to each pair of detectors which are transmitted to the computer.

In the data acquisition device shown in FIG. 1, the complete circuit is shown only for groups A and D of detectors. It should be noted that a similar complete circuit is used for groups C and F forming a pair and for groups B and
E forming another pair. This circuit, which is not shown, transmits an output signal shown in FIG. 1 in the form of 24 channels coming from groups B and E and 24 channels coming from groups C and F.

 As shown in the block diagram of the circuits of groups A and D, a discriminator 40 to 11 channels transmits it data output signals to a processor 45 of input and simultaneously feeds a line of signal signals which transmits a signal each time the any of the 11 detectors in group A has a signal to coincidence circuits 47 and 49. Similarly, the output signal from each of the detectors in group D is in the form of an 11-signal input signal arriving at a second 11-channel discriminator 42 which transmits an 11-channel output data signal to processor 45 and a union signal indicating that any of the 11 detectors in group D is transmitting a signal, as the second input signal of the coincidence circuit 49, and also via a delay circuit, in the form of a signal arriving at the second input terminal of the coincidence circuit 47. As shown in FIG. 1, the output signal from circuit 47 east in the form of a random sampling signal arriving at the processor 45 while the output signal from the coincidence circuit 49 is in the form of a total coincidence signal transmitted to the input processor 45.

 Thus, for the pair of groups A and D, there are 11 signals leaving the discriminator 40 of group A, 11 signals leaving the discriminator 42 of group D, one for the random sampling signal of circuit 47 and one for the signal total circuit sampling 49 There are therefore a total of 24 signals for the pair of groups A and D e Similarly, each pair of groups B and E on the one hand and C and F on the other hand transmits 24 signals to input processor 45.

 During the operation of the whole apparatus, each photon detected by any of the individual photomultipliers of any of the groups A to F causes the formation of an output signal which is amplified and transmitted to the discriminator at 11 channels associated with the group in which the photomultiplier is located. In the discriminator, this analog signal undergoes an energy discrimination which makes it possible for example to separate the photons having an energy lower than 100 keV and is put in the form of a logic pulse. The pulses reach the input processor 45 directly.

Thus, the latter has 11 separate lines of data which come from the discriminator 40, a signal in any one of these lines being representative of a photon detected in the corresponding detector. Each of the 11 channel discriminators also has an output transmitting a meeting signal indicating that one of the group's photomultipliers receives a signal. The meeting signals from the opposite groups, for example from groups A and
D, arrive at a coincidence circuit (for example circuit 49 in FIG. 1) which transmits a total sampling signal each time it receives the meeting signals of two opposite groups, during the coincidence period. This is for example set to 12.5 ns.

 The output meeting gate of the discriminator 42 of group D is also connected to a second coincidence circuit 47 by means of a delay circuit 50 and the other input of this circuit 47 directly receives the output signal of the discriminator 40 from the other group A of detectors. For example, the delay introduced can be 40 ns.

 The coincidence circuit 47 determines the coinidence between the signals coming from the discriminator 40, indicating the detection of a photon in one of the detectors of group A, and the delayed union signal from the discriminator 42 indicating the detection of a signal. of one of the photomultipliers in group D, a # some time before. This coincidence cannot be attributed to positron annihilation pulses since these events occur simultaneously in real time. These delayed coin- cidences constitute a suitable measure of the random coincidences between the photons which are in the field of vision 30 of the detector arrangement. Thus, the output signal of the coincidence circuit 47 represents the frequency of the accidental background noise or random while the signal of the non-delayed coincidence circuit 49 represents the totality of the random coincidences and of the photons formed by annihilation.

 FIG. 1 shows in a more detailed form the discriminator arrangement and forming a coincidence circuit. Each of the 11 output signals from a group of detectors reaches, via a preamplifier 60, a discriminator 62 of constant fraction. This discriminator has a threshold which can be adjusted over a wide range of voltages, for example between 2.5 and 300 mV, so that it allows the variation of the gain of the photomultipliers. The discriminator is further characterized by a time excursion, varying with the amplitude, of approximately - 250 ps over a dynamic range of 100/1. However, in such a measurement system, the dynamic range is only 5/1, between 100 and 511 keV.

 Output pulses of the discriminator 62 of constant fraction arrive directly with the signals of the input lines II in the processor 45 and arrive at a meeting gate 64. The output signal of the latter arrives at an input of an intersection gate 66 and at an input of a second intersection gate 68. These latter gates form the coincidence circuits of FIG. 1. The signal from gate 68 constitutes the total sampling signal whereas that from gate 66 constitutes the signal for random sampling. The other input of gate 68 is supplied by another channel for combining the preamplifier discriminator of the type shown in FIG. 2 but for the signals created in the opposite group of detectors. The other input signal from door 66 is however formed by the output signal from the meeting door of a second channel, via a delay circuit.

 As described above with reference to FIG. 1, the signals from the coincidence circuits reach an input processor 45. The signals arriving at the latter then include 11 data channels corresponding to the 11 detectors of each group and, for each pair of opposite groups, a random sampling signal representing the coincidences between the photons detected in a group and the delayed signals coming from photons detected in the opposite group. Furthermore, a total sampling signal representing total coincidences, without delay, between photons detected in a group coincident with photons detected in the opposite group, is formed.

 In addition to the aforementioned signals, the processor 45 receives routing and triggering signals ensuring control of its operation.

 FIG. 3 represents a block diagram of the input processor This includes a temporary memory 70 which directly receives the input signals indicated above The output signals from this memory 79 reach an encoder 72 which itself powers the processor 74 of data, also forming a memory, this processor being connected by a circuit 76 for controlling and coupling to the computer 80 of the sub-assembly 27. The main function of the processor 45 is the accumulation, for each of the 363 rectilinear paths crossing the field of vs ion, of a number corresponding to the total of the coincidences during each path, and of a number corresponding to the random coincidences for each path. In an operating mode, the input processor quickly determines the difference between the numbers for each line of sight and transmits it to the set 27. In another operating mode and in a variant #, the processor can transmit the two values, that is, the total number of coincidences and the number of random coincidences, to the computer which performs other calculations.

 The temporary memory 70 constitutes a buffer circuit for the input data. Although it has a burst coincidence determination frequency of 8 MHz, the data from the temporary memory is transferred to the encoder to memory 74 with a frequency of 700 kHz. Thus, the temporary memory allows processing without dead time because very close events can be received in the temporary memory because of the coincidence frequency of 8 MHz of the latter while the transfer frequency in the mass memory 74 can be 700 kHz only, that is to say much higher than the analysis frequency which is generally of the order of 20 kHz.

Figure 4 shows the input processor in more detail. The 11 input signals from the group
A and the 11 signals coming from group D arrive at the flip-flop 90 of the pair of groups AD. Similarly, the 11 signals of groups B and E reach a second flip-flop 92 and those of groups C and F reach a third flip-flop 94. Each of the sampling signals, both random and total, of the coincidence circuits reaches an input sampling and synchronization flip-flop circuit 98.

 The circuit 98 transmits separately pairs of signals to each of the flip-flops 90, 92 and 94. Depending on the sampling signal, random or total, coming from the coincidence circuits associated with a particular pair of groups, the circuit 98 transmits first a clock pulse to the associated flip-flop then a validation pulse. The clock pulse has for example a duration of 150 ns. Circuit 98 also transmits separate output signals to a first-in-first-out register 100, indicating whether the input sampling signal is a normal random or total signal and indicating which pair of groups it is from.

 The output signals of flip-flops 90, 92 and 94 reach two 11-channel output lines, one 11-channel line transmitting the data signal representative of a photon of sufficient energy detected by any of the detectors of groups A, B and C. The channel through which the signal appears is obviously representative of the position of the particular detector in the group.

Thus, a signal from channel n01 in this output indicates that a photon is detected in the first detector position of any of the groups A, B and C. Similarly, a second 11-channel output transmits the signals of the opposite groups D, E and F. The two 11-channel signals reach the inputs of register 100 with the sampling identification signals coming from circuit 98.

 During operation, the signals of the individual discriminators of the detector groups are transmitted to the input of the flip-flops of the pairs of groups but do not enter the flip-flop unless a sampling signal appears. After a sampling signal appearing at the input of circuit 98, the appropriate output clock pulse is transmitted and this signal ensures the introduction of the data signal into the flip-flop of the pair of groups. At the end of the clock pulse, a validation signal is created in circuit 98 and controls the transfer of the data signal from the corresponding flip-flop to register 100. As circuit 98 simultaneously transmits output signals to register 100 indicating the type of sampling and the pair from which the signal comes, the multi-bit signal of the register is representative of the location of the two # detectors which. have detected the photon at the origin of the coincidence, - in the pair of groups, of the position of the detectors and of the fact that the coincidence is random or total.

 It should be noted that, in this apparatus, more than two lines of data can be excited at the same time, at the entry of a rocker of a pair of groups so that, in this case, it is not possible to identify which straight path is involved in the coincidence.

 In this case, a signal is transmitted to the memory circuit 104 and indicates a multiple coincidence.

 The register 100 is for example of the first-in-first-out type having at least 28 bit positions and it has a length of 16 registers in the clockwise direction. As indicated previously, the flip-flops of the pairs of groups are controlled by a clock pulse of 150 ns corresponding to a burst frequency acceptance of approximately 8 MHz, and the signals of these flip-flops are introduced into the register at about this frequency. This register is however intended to extract the data from the shift register to the encoder 102 at a much lower clock frequency, for example 700 kHz.

As the memory capacity of register 100 is sufficient taking into account the average frequency of separation of events, the transfer frequency is sufficiently high, although the frequency of entry of the register must allow the resolution of events having a minimum separation, being given the random nature of radioactive decay.

 The 28-bit output signal from register 100 reaches encoder 102 which encodes them in the form of a 12-bit address signal transmitted to memory 104.

 As mentioned earlier, there are 121 lines of sight for each pair of detector groups, so 121 separate addresses are required in memory. A 7-bit binary code can then represent all these addresses, 7 addresses remaining free. Since there are three pairs of detector groups in the embodiment considered, a two-bit code can identify the pair of groups in which the sight lines are located, one bit remaining free. This can be used for the indication that the coincidence detected corresponds to a random or total sampling. The additional bits from encoder 102 can then be used for the delivery of control and instruction signals.

 Such a control instruction determines the operating mode of the entire apparatus. There are three main modes. In one mode, the memory accumulates only the total samples constituting a measure of the number. In a second mode of operation, the total sampling signals are accumulated in the memory and the random sampling signals also, the difference being calculated later. In a third mode, the total coincidences are transmitted to the memory in the form of an algebraically positive signal while the random coincidences are transmitted in the form of an algebraically negative signal so that the accumulated numbers represent the algebraic sum of the signals total sampling and random sampling.

In this latter operating mode, the memory contains, at any time and practically immediately, the resulting number which is representative of the coincidences due to the positrons. The signals transmitted by the memory to the encoders can also control various household functions of the device such as erasing and restarting, and the transmission of an unloading clock signal regulating the reading speed of register 98 in the encoder 102.

 Although a particular example of a circuit using a hexagonal arrangement of three pairs of detector groups has been described, the invention also applies to variants 7 not only relating to the geometric configuration of the arrangement of detectors, which can be placed in one or more planes, but also relating to the configurations of the circuits giving the results indicated.

Claims (10)

1. Tacography apparatus, of the type which includes  a source of radiation directing positrons in an area to be analyzed,  several photon detectors emitted during a positron annihilation in this area, and  a first coincidence device intended to determine the number of coincidences in time between photons detected by detectors located at 1800 practically from one another, said apparatus being characterized in that it comprises  a second coincidence device (47) for simultaneously determining the number of coincidences in time between photons detected by individual detectors and photons detected with a time offset by detectors having an offset of practically 1800, and  a device (45) for forming the difference between the number of coincidences in time determined by the first coincidence device (49) and the number of coincidences in time detected by the second coincidence device (47), in the form of a measure of the number of positron annihilations in said zone (30). 2. Apparatus according to claim 2, characterized in that the device (45) intended to form the difference effects this difference practically at the time when the coincidences take place. 3. Apparatus according to claim 1, characterized in that it further comprises a memory (74) which receives the number of coincidences in time from the first and the second coincidence device (49, 47) so that the memory accumulates separately the numbers from the first coincidence device (49) and the second coincidence device (47), the difference forming device being connected to the memory so that it forms the difference after the accumulation of the coincidence numbers #. 4. Apparatus according to claim 1, characterized in that it further comprises a memory (70) to which are connected the first and the second coincidence device (49, 47) and a device for coupling the detectors to the memory, this coupling device transmitting to the memory signals representative of the pair of detectors which detected the photons which caused each of the coincidences. 5. Tacography apparatus, of the type which includes  a source that emits positrons to an area to be analyzed,  several detectors of the photons emitted during the annihilation of the positrons in said zone, the detectors forming an annular arrangement, and  a first coincidence circuit having a first and a second input connection and at least one output connection, the first coincidence circuit being intended to transmit a signal by the output connection each time the signals present at the first and at the second input connection coincide in time with each other, the detectors being coupled to the input connections of the first coincidence circuit so that a first group of detectors is connected to the first input and to a second group, placed opposite the first group, to the second input connection, said apparatus being characterized in that it comprises  a second coincidence circuit (47) having a first and a second input connection and an output connection, this second coincidence circuit being intended to transmit a signal through the output connection each time the signals arriving at the first and at the second input connection coincide with each other in time, the detectors being coupled to the first input connection of the second coincidence circuit,  a delay circuit (50) mounted between the detectors and the second input connection of the second coincidence circuit (47), and  a device (45) for determining the difference between the number of signals appearing at the output of the second coincidence circuit and the number of signals appearing at the output of the first coincidence circuit (49), the difference being an indication of the number of posit positively annihilated in said area (30). 6. Apparatus according to claim 5, characterized in that the detectors form an arrangement of groups (A to F) of detectors, each group being paired with a parallel group placed opposite, and, in the first coincidence circuit (49) coincidences in time are determined between the signals of the detectors of the paired groups whereas, in the second coincidence circuit (47), the first input receives the signals of the detectors of a group while the second input receives the signals delayed group detectors paired. 7. Apparatus according to claim 5, characterized in that the difference between the output signals of the two coincidence circuits (49, 47) is determined practically at the same time as these signals are produced at the output of the coincidence circuits. 8. Apparatus according to claim 5, characterized in that it further comprises  a memory (74),  a device intended to transmit the output signals of the first and second coincidence circuits (49, 47) to the memory, and  a control circuit (98) for determining the difference between the signals stored in the memory and originating from the first coincidence circuit (49) and the signals retained and originating from the second coincidence circuit (47) at a time subsequent to that of the introduction of these signals into the memory. 9. Apparatus according to claim 7, characterized in that the device (45) for determining the difference comprises a buffer circuit (70) for storage having a lower input resolution time by at least one factor. equal to 100 at the maximum expected average frequency of the data. 10. Tacography apparatus, of the type which includes  a source of positrons directing positrons in an area (30) to be analyzed,  several photon detectors placed around the periphery of said zone, the detectors transmitting output signals when they receive photons, and  a first coincidence device (49) intended to transmit output signals representative of the number of coincidences in time of the photons detected by detectors placed at 1800 from one another, said apparatus being characterized in that it comprises  a second coincidence device (45) for transmitting output signals representative of the number of coincidences between photons detected by pairs of detectors set at 1800, with a fixed separation in time between the detection of photons from a detector the pair and the detection in the other detector of the pair,  an encoder (72), and  a device for coupling the output signals of the detectors and the output signals of the coincidence devices to the encoder, the latter transmitting output signals representative of the location of the detectors transmitting the signals which caused the signal of the coincidence device to form, and the coincidence device that formed the signal.