CN102670232B - Positron emission computed tomography device, and method executed through the same - Google Patents

Abstract

A positron emission computed tomography apparatus and a method performed by the positron emission computed tomography apparatus are provided. The positron emission computed tomography apparatus is capable of improved energy resolution. A positron emission computed tomography apparatus according to an embodiment includes a scintillator array, an optical sensor, a determination unit, a storage unit, and a derivation unit. The specifying unit specifies detection positions of interaction events of gamma rays and scintillators in units of regions divided into a number greater than the number of scintillators based on signal values output from the one or more photosensors. The storage unit derives the total signal value, and stores the derived total signal value in the storage unit in association with the detection position. The derivation unit derives a correction value for correcting the energy value for each area unit based on the total signal value and the predetermined energy value stored in the storage unit in association with the detection position.

Description

Positron emission computed tomography apparatus, and methods performed by it

This application claims priority to U.S. Patent Application No. 13/045,610, filed March 11, 2011, and Japanese Patent Application No. 2012-029889, filed February 14, 2012, the entire contents of which are incorporated herein by reference .

technical field

Embodiments relate to a positron emission computed tomography apparatus, and a method performed by a positron emission computed tomography apparatus.

Background technique

In the field of medical imaging, the use of gamma ray detectors, particularly positron emission computed tomography devices, that is, PET (Positron Emission Tomography) devices, is increasing. In PET imaging using a PET device, radiopharmaceuticals are taken into an imaged subject by injection, inhalation, or ingestion. After administration of radiopharmaceuticals, the drugs accumulate in specific parts of the subject's body according to their physical and biomolecular properties. The actual spatial distribution of the agent, the concentration in the area where the agent accumulates, and the dynamics of the process from administration to eventual discharge are all clinically important factors. In this process, a positron emitter attached to a radiopharmaceutical emits positrons according to the physical properties of the isotope such as half-life or branching ratio.

Radionuclides emit positrons. When the emitted positron collides with the electron, a mutual annihilation event occurs, the positron and the electron annihilate. In most cases, two gamma rays (511 keV) are emitted in roughly 180-degree opposite directions by mutual annihilation events.

The PET device can detect two gamma rays detected approximately at the same time, and draw a straight line between these detection positions, that is, LOR (Line-of-response: Line of Response), thereby deriving the position of approximately mutual annihilation with a high probability . The original distribution can be estimated by accumulating a large number of such LORs and executing a tomographic reconstruction process. In addition to the position information of the two annihilation events, when the correct detection timing information (within hundreds of picoseconds) is also available, it is possible to further increase the annihilation along the LOR by calculating the TOF (time-of-flight: time-of-flight) Information about the presumed location of an event. The accuracy of positioning along the LOR is determined by the limit of the temporal resolution of the PET apparatus. In addition, the final spatial resolution of the PET apparatus can be determined by determining the limit at the time of determining the position of the original annihilation event. On the other hand, the inherent properties of the isotope (e.g., positron energy) also become factors that determine (due to the range of the positron until two annihilation gamma rays occur, or the angle between two annihilation gamma rays) PET important reason for the spatial resolution of the device.

prior art literature

Non-Patent Document 1: W.W.Moses, "Time of Flight in PETRevisited", IEEE Transactions on Nuclear Science, Vol.50, No.5, pp.1325-1330

Contents of the invention

(problem to be solved by the invention)

The problem to be solved by the present invention is to provide a positron emission computed tomography apparatus capable of improving energy resolution and a method performed by the positron emission computed tomography apparatus.

(the solution used to solve the problem)

A positron emission computed tomography apparatus according to an embodiment includes a scintillator array, a plurality of photosensors, a determination unit, a storage unit, and a derivation unit. The above scintillator array has a plurality of scintillators. The plurality of photosensors detect scintillation light generated by inputting gamma rays of a predetermined energy value to the scintillator. The determining unit is divided into a unit of a number of regions larger than the number of the scintillators based on signal values detected by the one or more photosensors and output from the photosensors through the diffusion of the scintillation light. Determine the detection location of the above gamma ray interaction event with the scintillator. The storage unit derives a total signal value from the signal value output from the optical sensor, and stores the derived total signal value in a storage unit in association with the detection position. The derivation unit derives a correction value for correcting the energy value of the interaction event for each of the region units based on the total signal value stored in the storage unit in association with the detection position and the predetermined energy value.

(invention effect)

There is an effect of being able to improve the energy resolution.

Description of drawings

FIG. 1 shows an example of nonlinear SiPM (Silicon Photomultiplier: silicon photomultiplier) response when a gamma ray detector is irradiated with gamma rays from various radioactive isotopes.

Figure 2A shows three examples of distributions of light emitted from a 5x5 scintillator array (or array of scintillators) coupled to a thin light guide. In the three examples shown in FIG. 2A , the integrated values of the intensity of light are all the same, but the spread ranges of light are greatly different from each other. FIG. 2A shows the situation where a non-linearity correction of the scintillator based on the identified interactions is insufficient.

FIG. 2B-1 shows an example of approximately calculated signal levels generated roughly in one of the three examples shown in FIG. 2A.

FIG. 2B-2 shows an example of approximately calculated signal levels generated roughly in one of the three examples shown in FIG. 2A.

2B-3 show examples of approximate calculated signal levels generated roughly in one of the three examples shown in FIG. 2A. As shown in FIGS. 2B-1 to 3 , any of the three examples provided an energy value of 511 keV, but the output signal level (signal level of the output signal) was different in each case.

FIG. 3 is a diagram for explaining the configuration of the PET apparatus according to the embodiment.

FIG. 4 is a diagram for explaining the configuration of a central processing unit in the PET apparatus according to the embodiment.

FIG. 5 shows a flood histogram obtained by reading out a scintillator array using an optical system integration process and a four-part photomultiplier tube (PMR (Photomultiplier Tube)).

Figure 6 shows a flood histogram obtained from the readout of a scintillator array using analog data aggregation processing and position sensitive avalanche photodiodes (avalanche photodiodes).

FIG. 7A is a flowchart showing the steps of a method performed by the PET apparatus according to the embodiment.

FIG. 7B shows a flowchart of the steps of a method performed by a PET device according to another embodiment.

Figure 8A shows a typical pulse from a photosensor in response to the incidence of light that occurs through contact of gamma rays with a scintillator array.

FIG. 8B-1 is a diagram in which the time axis of FIG. 8A is changed.

Fig. 8B-2 is a diagram showing energy information corresponding to the pulse shown in Fig. 8A.

FIG. 9A shows the structure of the γ-ray detector according to the embodiment. FIG. 9A shows the positional relationship of the scintillator array, light guide, and photosensor.

FIG. 9B shows the structure of the gamma ray detector according to the embodiment. FIG. 9A shows the positional relationship of the scintillator array, light guide, and photosensor.

Figure 10A shows a flood histogram.

Figure 10B shows the scintillator ID lookup table corresponding to the flood histogram.

(Symbol Description)

100 scintillators

105 scintillator array

130 light guide

135, 140 light sensor

300 scintillators

305 scintillator array

330 light guide

335, 340 light sensor

Detailed ways

Hereinafter, the PET apparatus according to the embodiment and the method performed by the PET apparatus will be described with reference to the drawings.

The PET apparatus according to the embodiment includes a gamma ray detector having a photosensor showing remarkable nonlinearity. The degree of this nonlinearity depends on the spatial distribution of the photon beams on the surface of the photosensor. In order to reduce the number of required electron channels, the PET device according to the embodiment includes a γ-ray detector capable of performing optical system integration processing or analog signal integration processing, and provides an improved method for performing nonlinear correction.

The PET apparatus collects a plurality of detection events of mutually annihilated γ-rays, and performs image reconstruction based on the event data, thereby generating an image of the distribution of the administered drug in the subject. Mutual annihilation γ-rays are two γ-rays generated in a direction of approximately 180 degrees by mutual annihilation of positrons generated from a drug administered to a subject and electrons near the point of generation. In the PET apparatus, based on event information detected substantially simultaneously by the detection elements of the gamma ray detector, an LOR that can be histogrammed is formed based on geometrical characteristics. According to the LOR, projection data or sinogram data to be reconstructed is specified. In addition, image data can also be formed by directly adding LORs one by one to the image data.

Therefore, in the PET apparatus, the LOR is a basic element that provides mutual annihilation position information of administered drugs, and the following additional information can be obtained in relation to the mutual annihilation position information. First, it is well known that the ability to draw a point with a PET device varies spatially in the entire image reconstruction area, but becomes higher in the center and gradually decreases toward the periphery. PSF (Point Spread Function: Point Spread Function) is a typical function that expresses this characteristic, and it has been included in the reconstruction process in recent years. Then, the PET device can calculate the probability distribution of the points where mutual annihilation events occur on the LOR based on the TOF, ie, the time difference information of mutual annihilation gamma rays reaching the detector.

The above-mentioned process (imaging process) requires multiple mutual annihilation events. In the current study, it is typical to determine the number of mutual annihilation events required to enable adequate imaging processing, although each imaging event must be resolved individually, but for a typical length of 100cm FDG ( Fluoro-deoxyglucose: checks in fluorodeoxyglucose) need to be repeated hundreds of millions of times. The time required to obtain these event numbers is determined by the injection amount of the drug and the sensitivity and counting capability of the PET apparatus.

A PET imaging system, that is, a PET apparatus uses gamma-ray detectors arranged at positions facing each other in order to detect gamma-rays emitted from a subject. Typically, a PET apparatus uses gamma-ray detectors arranged in a ring to detect gamma-rays coming from various angles. Therefore, typically, the scanner of the PET apparatus has an approximately cylindrical shape in order to capture as much radiation as possible, and of course must be isotropic. In addition, the scanner may have a shape missing a part of the ring, and in this case, the gamma-ray detector can also be rotated to capture the missing angle. However, such a method has a great influence on the sensitivity of the PET apparatus as a whole. In the case of a cylindrical shape, all the gamma rays included in one surface are likely to respond to the gamma ray detector, and increasing the size in the axial direction is extremely effective in the sensitivity and performance of capturing radiation. Therefore, the optimal structure is a spherical structure capable of detecting all gamma rays. Of course, in applications to the human body, spherical structures become very large and expensive. Therefore, in reality, the cylindrical shape in which the axial length of the gamma ray detector is variable is the basis of the structure of the scanner of the latest PET apparatus.

The scanner is composed of a plurality of detector modules, and the detector module is composed of a plurality of scintillators. In order to make the performance of the PET device excellent, it is necessary to configure as many scintillators as possible, and to block as much gamma rays as possible and convert them into light. In order to reconstruct the spatial and temporal distribution of radioactive isotopes, the PET apparatus must obtain information on the energy value (ie, the amount of light generated in the scintillator), position, and timing of each detected event. Most of the latest PET scanners consist of thousands of individual scintillators. These scintillators are configured within the detector module to determine the location of mutual annihilation events. Typically, scintillator elements have a cross-section of approximately 4mm x 4mm. Also possible are smaller or larger sizes, or non-square cross-sections. The length or depth of the scintillator determines the gamma-ray blocking capability, and typically ranges from 10 to 30 mm. The detector module is the main constituent element of the scanner.

The performance of PET imaging based on PET devices depends on the conversion of gamma rays to light based on a high-speed and high-brightness scintillator. The PET device is able to determine where scintillation occurs in the detector (where the gamma ray interacts with the scintillator) and make a temporal pairing of the events (that is, combine two detection events), and calculate the location of mutual annihilation events. In order to perform these operations, a very high-speed detector (detector and electronic equipment) is required, and an excellent signal-to-noise ratio is also required. If high-quality electronics are used, the signal-to-noise ratio is largely determined by the Poisson distribution associated with the detection process. In conclusion, if more photons are detected, the signal-to-noise ratio will be improved, thereby further improving the spatial resolution as well as the temporal resolution. In addition, when scintillation light is missed during detection, even with improved detector design and electronics, it cannot compensate. The ratio of the amount of light that can actually be detected with respect to the light generated in the scintillator is an index suitable for expressing the efficiency of a design component. So whoever wants to maximize the amount of light detected will probably want to place the light sensor as close as possible to the scintillator and avoid reflections and other edge effects. If such a procedure is performed, the gamma ray detector will presumably have to be a detector with a large scintillator array with a short distance between the scintillator and the photosensor.

As mentioned above, PET devices are not just counters. In addition to detecting the existence of a mutual annihilation event, the PET device also needs to identify the location of the mutual annihilation event. For the simplest design to be able to identify the position of each interaction, an independent optical sensor and A/D converter should be designed for each scintillator. In order to limit the physical size of the photosensor, the power required for the A/D converter, and the cost of these, data is generally collected by any device for the purpose of reducing the number of photosensors and the number of channels of the electronic device.

The two most common methods of data collection are the collection processing of optical systems and the collection processing of analog signals. The method of using a light guide to disperse light to four photomultiplier tubes is an example of the collection process of the optical system. By properly recording how light is distributed to multiple photosensors, the location of an interaction event with gamma rays can be calculated for any combination of photosensor responses. In addition, there are so-called position-sensitive avalanche photodiodes and position-sensitive SiPMs as examples of analog signal collection processing.

In the design of the optical system pooling process or the analog data pooling process, Anger logic (centroid calculation) or statistical methods are used to calculate the relative position within the detector frame of each interaction event. Here, the so-called "relative position" refers to the relative position of the actual position (for example, the absolute position in the subject) of the interaction event between the scintillator and the γ-ray, which is determined by calculating the position where the γ-ray is incident on the scintillator "Detection position" of the . In order to determine the gamma-ray interacting scintillators as relative positions, typically the PET device partitions the resulting flood histogram into look-up tables for the scintillators. After the interacting scintillators are identified according to the flood histogram and the position in the look-up table, the energy value and timing of each scintillator are corrected. Non-linear correction can also be performed as correction of the energy value. Typically, the correction is based on measurements using a number of different isotopes (e.g., ²² Na at 511 and 1275 keV, ¹³⁷ Cs at 662 keV, ¹³³ Ba at 356 keV, ⁵⁷ Co at 122 keV, 241 Am at ⁶⁰ keV), for each Both flashes are performed. In addition, as will be described later, the PET apparatus according to the embodiment corrects the energy value not in units of scintillators but in units of regions divided into a number greater than the number of scintillators.

FIG. 1 shows an example of nonlinear SiPM response when a gamma ray detector is irradiated with gamma rays from various radioactive isotopes. The γ-ray detector includes a 3 mm×3 mm×10 mm LYSO (Lutetium Yttrium Oxyorthosilicate: lutetium yttrium silicate) crystal and a 3 mm×3 mm SiPM (manufactured by Hamamatsu Photonics Co., Ltd.) having 3600 cells.

When detecting high-energy gamma rays (for example, 511 keV for PET, etc.), gamma rays may be Compton-scattered by a plurality of scintillators in a gamma-ray detector. Thus, the γ-rays impart energy to a plurality of scintillators. In general, more than 30% of the 511 keV gamma rays interact with multiple scintillators. Even when the total energy given to the scintillators is the same, the spread of light to the photosensor may be quite different depending on the process of interaction or the number of interacting scintillators. For example, when energy of 511 keV is applied to one scintillator and when a total of energy of 511 keV is applied to two adjacent scintillators, it can be seen that there is a large difference in the spread of light passing through the photosensor.

This difference in light spread becomes a problem in nonlinear correction of the photosensor. This is due to the degree of non-linearity of the detected photons depending on their spatial distribution. This is a so-called case where SiPM is used. SiPM is a silicon photomultiplier tube (SSPM (Solid-State Photomultiplier) as a solid photomultiplier tube, or GAPD or MAPD) consisting of multiple independent avalanche photodiodes. In addition, SiPM is called a "micro cell" and operates in Geiger mode. In Geiger mode, the SiPM discharges if the microcell detects more than one photon. The charge emitted during discharge is determined by the capacitance and operating voltage of the microcell. The emitted charge is independent of the number of photons that make the discharge. For example, when a pulse of light comes into contact with the SiPM according to an event generated by the interaction of γ-rays in the scintillator, a plurality of microcells are discharged and electric pulses are generated. The amplitude of this electrical pulse is proportional to the number of microcells discharged. When the density of photons is very small, the amplitude of the electrical pulse varies linearly with respect to the number of photons because the probability of multiple photons contacting the same microcell is very low. As the density of photons increases, the probability of multiple photons contacting the same micro-unit increases, so the nonlinearity of the signal becomes stronger.

Figure 2A shows three examples of the distribution of light emitted from a 5 x 5 scintillator array coupled to a thin light guide. FIG. 2A shows the difference in intensity of light predicted by three different events when an energy of 511 keV is given to the scintillator array. In this example, the scintillators within a 5 x 5 array of scintillators are optically separated (i.e., they have reflective material between each) and combined with thin light guides, which are further integrated with the arrayed light sensors or position-sensitive light sensors. The grid lines in FIG. 2A represent the boundaries of the scintillator. In each case, the intensity of the integrated light is the same. In case 1 (left end), all energy is imparted to the central scintillator, and the intensity of the peak light (or photon beam) is highest. In case 2, gamma rays impart 67% of the energy to the central scintillator (based on Compton scattering) and the remaining 33% to nearby scintillators. In Case 3 (right end), the energy of the gamma rays is evenly distributed between the two scintillators. In this case, the resulting maximum intensity of light per scintillator is the lowest. If an optical sensor such as SiPM whose intensity and output of light vary depending on the density of photons is used for reading out the gamma ray detector, the intensity of integrated light is the same, but the spread of light is different, so the output of the optical sensor It is different in the three cases shown here. As a result, energy resolution deteriorates. For example, in the example shown in FIG. 2A , for example, even if an energy of 511 keV is applied to the scintillator in each case, the signal levels corrected by nonlinearity are different in each of the three cases (assuming that the same non-linearity is applied in each case). linear correction). This is a representative example of the mechanism that deteriorates the energy resolution.

Figure 2B represents the approximate calculated signal levels generated by the three cases shown in Figure 2A. Here, consider two scintillators as an example. Depending on the spread of light and the nonlinearity of the SiPM, the integrated signal levels differ in the respective cases even though a total energy of 511 keV is applied to the two scintillators in all three cases.

The above will be described in detail. For example, when gamma rays are incident near the center of a scintillator, the effect of Compton scattering is small, and the spread of light toward the photosensor tends to be small. On the other hand, for example, when γ-rays are incident near the boundary of one scintillator, the effect of Compton scattering becomes larger, and the spread of light toward the photosensor tends to be larger. In the former case, since the spread of light is small, it is considered that the photons touch a small number of microcells (microcells of the SiPM as a photosensor), and in the latter case, since the spread of light is large, therefore, Photons are considered to contact a relatively large number of microcells. Here, when photons are continuously in contact with the same microcell, SiPM sometimes cannot detect all photons due to its hardware limitation. On the other hand, the amplitude of an electric pulse generated by discharging a microcell is proportional to the number of discharged microcells. That is, in the case of the former in which photons are in contact with a relatively small number of microcells, since the probability that photons are continuously in contact with the same microcells becomes high, the probability that SiPM cannot detect all photons also becomes lower. High, there is a tendency for the signal value to become low. On the contrary, in the case of the latter in which photons are in contact with a relatively large number of microcells, since the probability that photons are continuously in contact with the same microcell becomes low, the SiPM can detect all photons and has a signal value change. high tendency. Thus, where in a scintillator the gamma rays are incident has an effect on the signal value. This is because the PET apparatus according to the embodiment corrects the nonlinearity in units smaller than the unit of the scintillator with respect to the signal value output by the photosensor as described below.

Embodiments disclosed in this specification apply non-linearity correction in sub-pixel units or continuously. Here, a sub-pixel unit refers to a unit divided into a number of regions greater than the number of scintillators. If optical system aggregation processing or analog data aggregation processing is used, the scintillator-γ-ray interaction event position in the flood histogram related to the peak value corresponding to each scintillator can be used to obtain the interaction between multiple scintillators Additional information for the interaction. Since the location where flicker occurs in the flood histogram is calculated by Angus logic (centroid calculation), events that energize the two scintillators are placed between the peaks corresponding to the two scintillators. It is self-evident to a person skilled in the art that the calculated interaction positions in these cases do not necessarily correspond to a single physical position, but represent positions of relative signal levels generated by the light sensors.

Also, if the influence of noise is ignored, the correct position between the peaks of the scintillators is determined by the relative energy applied to the two scintillators. Non-linear fluctuations are also determined by relative energies applied to the two scintillators. Therefore, by sub-pixelizing the flood histogram and applying different nonlinear corrections to each sub-pixel region, more optimal nonlinear correction can be performed. Alternatively, a non-linear correction that varies continuously between the peaks of the individual scintillators may also be applied.

FIG. 3 is a diagram for explaining the configuration of the PET apparatus according to the embodiment. It is self-evident to a person skilled in the art that the gamma-ray detector system shown in Fig. 3 forms part of a PET device or a TOF-type PET device. Additional explanations for the PET device and the TOF-type PET device are omitted for brevity. In addition, the explanation about the TOF PET apparatus is given in Non-Patent Document 1, and its entirety is referred to in this specification.

In FIG. 3 , nonlinear photosensors 135 and 140 are disposed on the light guide 130 , and a scintillator array 105 having a plurality of scintillators is disposed below the light guide 130 . It goes without saying to those skilled in the art that the present embodiment can also be applied to a gamma ray detector using any nonlinear optical sensor including SiPM or arrayed SiPM. The second scintillator array 305 having a plurality of scintillators faces the scintillator array 105 and is arranged to overlap the light guide 330 and the photosensors 335 and 340 .

In FIG. 3 , when mutual annihilation γ-rays are emitted from a subject (not shown), they advance in opposite directions about 180 degrees from each other. The mutual annihilation gamma rays are detected by the scintillator 100 and the scintillator 300 approximately simultaneously. And, if gamma rays are detected by the scintillator 100 and the scintillator 300 within a predetermined time limit, an interaction event 110 is determined. In this way, the gamma ray detection system simultaneously detects gamma rays through the scintillator 100 and the scintillator 300 . However, for the sake of brevity, only the gamma-ray detection by the scintillator 100 will be described here. It goes without saying for those skilled in the art that the description about the scintillator 100 is also applicable to the gamma-ray detection by the scintillator 300 .

Returning to FIG. 3 , each of the optical sensor 135 , the optical sensor 140 , the optical sensor 335 , and the optical sensor 340 is connected to the data collection unit 350 or the data collection unit 360 . The data collection unit 350 and the data collection unit 360 generate digitized output values by integrating corresponding waveforms generated by the light sensor 140 , the light sensor 135 , the light sensor 340 , and the light sensor 335 in response to the flicker light.

The data collection unit 350 and the data collection unit 360 include an analog-to-digital converter such as a sigma delta (sigma delta) converter operating at a sampling rate of 1 GHz to 5 GHz. Alternatively, the data collection unit 350 and the data collection unit 360 may include a multi-threshold sampler that samples the optical sensor waveform using a voltage threshold as a trigger instead of a constant sampling rate. It is obvious to those skilled in the art that other sampling methods and data collection units can be included without departing from the scope of the present embodiment. For example, it is also possible to use respective other channels according to the energy value and timing. At this time, typically, in the energy channel, a shaping filter and an analog-to-digital converter with a lower sampling rate are used. Additionally, the timing channel typically sums the signals from multiple photosensors. The aggregated timing signal is then input to a comparator and a time-to-digital converter is used to generate time stamps for each arrival of each event.

If the output value is obtained, the output value is sent to the calculation unit 370, and the energy level of the scintillator and the interaction event is determined according to the method described below. The output value and time of arrival are then stored in electronic storage device 375 and can be displayed on display 385 . The interface 380 may be used for both or one of the configuration and control of the computing device 370 , and for both or one of issuing additional commands to the computing unit 370 .

It goes without saying to those skilled in the art that the display 385 may also be a CRT (Cathode Ray Tube: Cathode Ray Tube) or an LCD (Liquid Crystal Display: Liquid Crystal Display) or the like. The interface 380 may be a known device for connecting a keyboard, a mouse, a trackball, a microphone, a touch screen, etc. to the central processing unit to function. Likewise, it is self-evident to those skilled in the art that the electronic storage device 375 can also be a hard disk drive, CD-ROM, DVD disk, flash memory, or other central processing devices. In addition, the electronic storage device 375 may be detachable or separated from the calculation unit 370, or may not be mounted thereon. Since the electronic storage device 375 is connected to the computing device via a network, it may be installed in another place where the computing unit 370 is associated, such as another room or a building.

FIG. 4 is a diagram for explaining the configuration of a central processing unit in the PET apparatus according to the embodiment. The computing unit 370 includes a processing unit 480 that processes data and commands stored in both or one of the main memory 440 and the ROM (Read Only Memory) 450 . In addition, processing unit 480 may process information stored on magnetic disk 410 or CD-ROM 420 . An exemplary processing unit 480 may be a Xenon processor (registered trademark) manufactured by Intel Corporation of the United States or an Opteron processor (registered trademark) manufactured by AMD Corporation of the United States. It goes without saying to those skilled in the art that the processing device 480 may also be a Pentium processor (registered trademark) or a Core 2Duo processor (registered trademark), or the like. Thus, commands corresponding to methods for the gamma ray detector may also be stored in any one of the magnetic disk 410 , CD-ROM 420 , main memory 440 , or ROM 450 .

In addition, the calculation unit 370 may include a network interface 475 such as an Intel Ethernet PRO network interface card (registered trademark) manufactured by Intel Corporation of America in order to connect to a network such as the Internet or a personal network through an interface. The display control unit 430 may be an NVIDIA G-Force GTX graphics adapter (graphics adapter) (registered trademark) manufactured by NVIDIA Corporation of the United States for connecting to the display 385 and functioning. In addition, the computing unit 370 may also include an I/O interface 490 for connecting the keyboard 295 , the pointing device 285 , or other general-purpose interfaces such as a microphone, a trackball, a joystick, and a touch screen to function.

Disk control unit 460 connects disk 410 , CD-ROM 420 or DVD drive, and bus 470 to each other. Disk 410 may be a hard drive or a flash drive. The bus 470 can also be an industry standard architecture (ISA (Industry Standard Architecture:)), an extended industry standard architecture (EISA (Extended IndustryStandard Architecture:)), a video screen electronic device standard association (VESA (VideoElectronics Standards Association:)), peripheral Device interconnection (PCI (Peripheral Component Interconnect:)), etc., is a bus for interconnecting all components of the calculation unit 370 . Since the general functions and functionality of the components of the computing unit 370 are well known, descriptions are omitted for brevity. Of course, Freescale ColdFire, I.MX (registered trademark), and ARM processor (registered trademark) manufactured by Freescale Corporation of the United States, and other processing devices or hardware manufacturers and products well known in this technical field can also be used.

In addition, the exemplary computing unit 370 can also be installed in FPGA (Field Programmable Gate Array: Field Programmable Gate Array), special ASIC (Application Specific Integrated Circuit: Application Specific Integrated Circuit), microcontroller, PLD (Programmable Logic Device: Programmable Logic Devices), or computer-readable media such as CD-ROMs. In addition, an exemplary calculation unit 370 is a hardware platform of a computing device such as a PC (Personal Computer), and the processing unit 480 is an arbitrary processing device known in the art such as an Intel Pentium processor (registered trademark). The computer-readable commands stored in any one of the main memory 440, ROM 450, magnetic disk 410, or CD-ROM 420 are provided as application programs, background programs, or components of the operating system, or a combination thereof, and the processing unit 480, and systems well known to those skilled in the art such as Microsoft Windows Vista (registered trademark), UNIX (registered trademark), Solaris (registered trademark), Linux (registered trademark), and Apple Mac OS (registered trademark) are executed in conjunction.

Both or one of the main memory 440 and the ROM 450 supports the recording of the calculation unit 370 and the same function. Therefore, the main memory 440 may be RAM (Random Access Memory), flash memory, EPROM (Electrically Erasable Programmable Read Only Memory: Electrically Erasable Programmable Read Only Memory) memory, and the like. On the other hand, ROM 450 is a read-only memory such as PROM. In addition, since such memories are well known, descriptions of the main memory 440 and the ROM 450 are omitted for brevity.

5 and 6 show two flood histograms that can clearly identify Compton scattering between scintillators. FIG. 5 shows a flood histogram obtained by reading out a scintillator array using four-divided photomultiplier tubes as an optical system integration process. In FIG. 5 , the 36 peaks respectively represent 36 scintillators included in the scintillator array. In addition, lines connecting the nearest peaks in the scintillator array can be seen, and interaction events at positions between the peaks are generated by Compton scattering events in the scintillator array.

FIG. 6 shows a flood histogram obtained by readout of a scintillator array using a position-sensitive avalanche photodiode as an analog data collection process. In FIG. 6 , 64 peaks respectively represent 64 scintillators included in the 8×8 scintillation array. In addition, the interaction events between the sharp peaks are generated by Compton scattering events in the scintillator array. In Figure 6, the pattern of lines connecting the nearest peaks can be clearly seen. This is based on the Compton interaction that energizes the closest pair of scintillators.

As a specific example, consider the case shown in FIG. 5 . In this example, a 6×6 scintillator array consisting of 36 individual scintillators is read out using photomultiplier tubes divided into four. Assume a case in which the scintillator array is read out using a four-divided photomultiplier tube, but is read out using a four-divided SiPM array. If the conventional method is used, the flood histogram in FIG. 5 is divided into 36 regions, and each region represents one scintillator. Energy correction coefficients may contain non-linear corrections, presumably for each scintillator. In one embodiment, the flood histogram is partitioned into a number of sub-pixels (eg, 900). Then, according to the calibration data (calibration date), different nonlinear corrections are obtained and applied to each sub-pixel area.

In another embodiment, after the corrections for each sub-region have been found from the calibration data, a mathematical function is derived that varies the nonlinear correction continuously as a function of position within the flood histogram. At this point, the math function is seen as a method of interpolating between the centers of the multiple regions used by the calibration.

The PET apparatus according to the embodiment described below includes a scintillator array, a plurality of photosensors, a determination unit, a storage unit, and a derivation unit. The scintillator array has a plurality of scintillators. The plurality of photosensors detect scintillation light generated by interaction of gamma rays of a predetermined energy value with the scintillator. The determining unit determines gamma rays and scintillation in units of areas divided into a number larger than the number of scintillators based on signal values that are detected by one or more photosensors and output from the photosensors through the diffusion of scintillation light. The detection position of the interaction event of the organ. The storage unit derives a total signal value from the signal value output from the photosensor, and stores the derived total signal value in association with the detection position in the storage unit. The derivation unit derives a correction value for correcting an energy value of an interaction event for each area unit based on the total signal value and the predetermined energy value stored in the storage unit in association with the detection position.

In addition, in the PET apparatus according to the embodiment, the specifying unit may specify the detection position for each of the plurality of interaction events based on the signal value output from the optical sensor. The storage unit may derive total signal values for each of the plurality of interaction events, and may store the derived total signal values in association with the detection positions in the storage unit. The derivation unit may further derive an average value for each area unit based on a plurality of total signal values stored in the storage unit in association with the detection position, and calculate a value for each area unit based on the derived average value and a predetermined energy value. A correction value that corrects the energy value of the interaction event is derived.

In addition, in the PET apparatus according to the embodiment, when there are a plurality of optical sensors outputting signal values, the determination unit may determine the position of each optical sensor based on the weighted average of the signal values output by each optical sensor, thereby determining the position of the optical sensor. Location. In addition, the derivation unit may derive the ratio of the predetermined energy value to the total signal value as a unit of correction value for each region. In addition, the light sensor may also comprise at least one silicon photomultiplier tube. In addition, the PET device may also be a TOF type.

In addition, each unit such as a specifying unit, a saving unit, and a deriving unit is equipped in, for example, the computing unit 370 described using FIG. 3 .

In addition, a PET apparatus according to another embodiment includes a scintillator array, a plurality of photosensors, a specifying unit, and a correcting unit. The scintillator array has a plurality of scintillators. The plurality of photosensors detect scintillation light generated by inputting gamma rays of a predetermined energy value to the scintillator. The determination unit determines the interaction of gamma rays in units of regions divided into a number larger than the number of scintillators based on signal values that are detected by one or more photosensors and output from the photosensors through the diffusion of scintillation light. The detection position of the action event. The correction unit derives a total signal value from the signal value output from the optical sensor, and corrects the total signal value based on a correction value corresponding to the detection position by referring to the calibration data using the derived total signal value and the detection position determined by the determination unit. . The calibration data is data defining a correction value for correcting an energy value of an interaction event for each area unit.

In addition, in the PET apparatus according to another embodiment, when there are a plurality of photosensors that output signal values, the determination unit may perform a weighted average on the positions of the respective photosensors based on the signal values output from the respective photosensors to determine Determine the detection location. In addition, the correction unit may correct the total signal value by multiplying the total signal value by the correction value. In addition, the PET device may also be a TOF type.

In addition, each unit, such as a specifying unit and a correcting unit, is equipped, for example, in the computing unit 370 described using FIG. 3 .

In addition, a PET apparatus according to another embodiment includes a scintillator array, a plurality of photosensors, a determination unit, a storage unit, and a derivation unit. The scintillator array has a plurality of scintillators. The plurality of photosensors detect scintillation light generated by inputting gamma rays of a predetermined energy value to the scintillator. The determination unit determines the interaction of gamma rays in units of regions divided into a number larger than the number of scintillators based on signal values that are detected by one or more photosensors and output from the photosensors through the diffusion of scintillation light. The detection position of the action event. The storage unit derives a total signal value from the signal value output from the photosensor, and stores the derived total signal value in association with the detection position in the storage unit. The derivation unit derives a mathematical function that continuously changes a correction value for correcting an energy value of an interaction event as a function of the detection position, based on the total signal value and the predetermined energy value stored in the storage unit in association with the detection position.

In addition, each unit such as a specifying unit, a saving unit, and a deriving unit is equipped in, for example, the computing unit 370 described using FIG. 3 .

In addition, a PET apparatus according to another embodiment includes a scintillator array, a plurality of photosensors, a determination unit, a storage unit, and an energy window derivation unit. The scintillator array has a plurality of scintillators. The plurality of photosensors detect scintillation light generated by inputting gamma rays of a predetermined energy value to the scintillator. The determination unit determines the interaction of gamma rays in units of regions divided into a number larger than the number of scintillators based on signal values that are detected by one or more photosensors and output from the photosensors through the diffusion of scintillation light. The detection position of the action event. The storage unit derives a total signal value from the signal value output from the photosensor, and stores the derived total signal value in association with the detection position in the storage unit. The energy window derivation unit derives the energy window for each region based on the total signal value and the predetermined energy value stored in the storage unit in association with the detection position.

FIG. 7A is a flowchart showing steps of a method performed by the PET apparatus according to the embodiment. FIG. 7A shows a method of determining a set of correction values (also referred to as "correction coefficients"). This correction value is used to determine the energy value of the interaction event detected by the gamma ray detector. In addition, this γ-ray detector has one or more nonlinear optical sensors arranged on a scintillator array composed of scintillator elements. In addition, the gamma ray detector uses optical system collection processing or analog data collection processing. By the method shown in FIG. 7A , a set of correction values for determining an accurate energy value is determined, and the gamma ray detector is calibrated by determining the set of correction values. In other words, for example, as shown in FIG. 1, the purpose of the method shown in FIG. 7A is to determine and integrate the The calculated nonlinear relationship between the signal and the energy value of the interaction event. The number of cell positions is greater than the number of scintillators contained in the scintillator array.

In step S710, the PET apparatus according to the embodiment generates gamma rays having a first energy value (for example, 511 keV) from a radiation source, and a detector receives gamma rays. The gamma rays are generated from a ²² Na, ⁶⁸ Ge, or ¹⁸ F radiation source, and typically, the radiation dose of the radiation source ranges from tens of microcuries to several millicuries. In order to obtain a sufficiently high count rate, that is, to collect the required data within an appropriate time (tens of minutes to several hours), the radiation dose of the radiation source is selected. In addition, because the count rate is too high, the interaction events overlap, and the result will produce significant errors. In order to avoid this situation, the radiation dose of the radiation source is selected.

In step S720, the gamma rays interact with the scintillator array composed of scintillator elements to generate scintillation light, and the generated scintillation light is detected by more than one photosensor. The data collection unit according to the embodiment collects, for each photosensor, a signal value (arbitrary unit) output from the photosensor as an energy value triggered by an interaction event. For example, FIG. 8A shows a typical pulse from a photosensor in response to the incidence of light that occurs by contacting a scintillator array with 511 keV gamma rays. 8B-1 shows a diagram in which the time axis of FIG. 8A is changed, and FIG. 8B-2 shows a diagram of energy information corresponding to the pulse shown in FIG. 8A . As shown in Fig. 8B-1, timing information is obtained from the rising edge of the pulse, and energy information is obtained by integrating the pulse.

In step S730, the PET apparatus according to the embodiment determines the relative position of the scintillator event and the total signal value based on the signal value output from each photosensor. 9A and 9B show the configuration of the gamma ray detector according to the embodiment. In addition, FIG. 9A and FIG. 9B show a method for determining the relative positions of interaction events. 9A and 9B are two diagrams of a γ-ray detector, showing the positional relationship of a scintillator, a light guide, and an optical sensor.

Scintillation light from a scintillator is diffused through a light guide to multiple photosensors. The PET apparatus according to the embodiment can calculate the relative position of the interaction event as follows based on the ratio of signals received by different optical sensors.

x＝(∑x _i Signal _i )/∑Signal _i …(1)

y＝(∑y _i Signal _i )/∑Signal _i …(2)

Here, Signal _i is a signal value output from the i-th photosensor, and x _i and y _i are the positions of the center of the i-th photosensor. In addition, PET devices can also use other algorithms to determine the relative positions of interaction events detected by multiple light sensors. In addition, the PET apparatus according to the embodiment can calculate the total amount of energy values of the interaction event from the sum of the signal values (total signal value) of the respective photosensors that received the scintillation light of the interaction event.

In one embodiment, the relative position is determined as a position corresponding to one of a predetermined number of cell positions. Here, the predetermined number of cell (sub-pixel) positions is greater than the number of scintillator elements in the scintillator array composed of scintillator elements.

In step S740 , the PET apparatus according to the embodiment stores the total signal value related to the interaction event in the storage unit in association with the relative position of the interaction event determined in step S730 .

In step S750, the PET apparatus according to the embodiment repeats steps S710 to S740 for a plurality of interaction events. Figure 10A shows a flood histogram. In addition, FIG. 10B shows a scintillator ID lookup table corresponding to the flood histogram. As shown in FIG. 10A , conventionally, a flooded histogram representing the frequency of interaction events is generated for each interaction event from stored data. Flooded histograms represent individual peaks for each scintillator. In addition, by arranging a scintillator, an optical sensor, etc., a unique distorted pattern is generated. As shown in FIG. 10B , the PET device can use the flood histogram to determine the scintillator that is believed to have an interaction event by segmenting the flood histogram into regions. All interaction events that fit the same region are assigned to the same scintillator. In this way, a look-up table is generated and the PET device is able to convert the relative positions of interaction events within the flood histogram to scintillator positions.

In one embodiment, each area in the flood histogram shown in FIG. 10A is divided into multiple units (sub-pixels) again. In the PET apparatus according to the embodiment, a correction value for correcting an energy value of an interaction event is determined for each sub-pixel.

In step S760 , the PET apparatus according to the embodiment determines an average value of total signal values related to a plurality of interaction events for each relative position (sub-pixel unit position) based on the stored data of the interaction events. In addition, the PET apparatus stores the average value of the total signal value in the storage unit in association with the first energy value of the gamma rays used to generate the interaction event.

In step S770 , the PET apparatus according to the embodiment determines a correction value for each relative position based on the average value of the total signal values determined for each relative position and the first energy value. For example, the PET apparatus sets a ratio of a predetermined energy value (for example, 511 keV) to an average value (arbitrary unit) of total signal values for a sub-pixel cell position as a correction value for the cell position.

In step S780, the PET apparatus according to the embodiment also repeats steps S710 to S770 for other γ-rays having predetermined energy values, and determines correction values corresponding to each predetermined energy value and each cell position. In this way, after step S77 , at the end of the calibration process, the PET device according to the embodiment performs each unit (sub-pixel) position, for example, a group with data values {S _i , E _i }. Here, S _i is a total signal value (arbitrary unit), and E _i is an energy value (keV). Using this set of data values, the PET device generates, for each cell position, a non-linear curve, such as that shown in FIG. 1 , relating signal values to corrected energy values. Also, the PET device is able to convert the signal value of the interaction event into an energy value using the non-linear curve generated at each cell location, and can compensate for the non-linearity of the SiPM response.

As described above, the PET apparatus according to another embodiment derives a mathematical function that changes continuously as a function of position in a flooded histogram after deriving a correction value for each cell position from calibration data as a nonlinear correction. At this time, the mathematical function is regarded as a method of interpolating the centers of the multiple regions used for calibration with each other.

In another embodiment, a radiation source of gamma rays having a certain energy is used, and while the analysis is being performed, an analysis is also performed for another energy at the same time. In this case, PET devices can use feature points such as Compton edges or backscatter peaks where known energies can be found in the acquired spectrum to calibrate the non-linear response of the gamma ray detector.

FIG. 7B is a flowchart showing the steps of a method performed by the PET apparatus according to another embodiment. As shown in FIG. 7B , a PET apparatus according to another embodiment corrects the energy value of an interaction event using the correction value obtained by the method shown in FIG. 7A .

In step S715 , the detector of the PET apparatus according to the embodiment receives gamma rays generated from a certain radiation source.

In step S725, if the gamma rays come into contact with the scintillator array composed of scintillator elements, scintillation light is generated, and the generated scintillation light is detected by one or more photosensors. The data collection unit according to the embodiment collects signal values (arbitrary units) output from the photosensors for each photosensor as energy values triggered by interaction events with γ-rays.

In step S735 , the PET apparatus according to the embodiment determines the relative position of the interaction event with the gamma ray and the total signal value based on the signal value output from each photosensor as described above. The relative position is determined as a position corresponding to one of the predetermined number of cell positions. The relative position is converted to a corresponding unit position of the plurality of unit positions from which the calibration data was taken. The predetermined number of cell (sub-pixel) positions is greater than the number of scintillator elements in a scintillator array composed of scintillator elements.

In step S745, the PET apparatus according to the embodiment calculates the corrected energy value for the interaction event based on the total signal value (value determined in step S735) stored for each cell position corresponding to the relative position and the calibration data . Here, the calibration data is data defining a nonlinear relationship between a signal value and an energy value at each cell position. For example, calibration data essentially defines a non-linear curve representing the relationship between signal value and energy value. In one embodiment, the PET apparatus according to the embodiment obtains a corrected energy value by multiplying the total signal value determined in step S735 by a correction value using a nonlinear curve. It is self-evident for a person skilled in the art that, prior to applying the embodiments described herein, additional gain or offset corrections can also be applied to the signal without departing from the scope of the present embodiments.

An advantage over conventional systems is that, according to the embodiment described here, better nonlinear correction can be performed and energy resolution can be improved. For PET, an improvement in energy resolution can in other words reduce the scatter fraction and, ultimately, improve image quality.

In addition, the PET apparatus according to the embodiment also includes an energy window derivation unit that derives an energy window for each region. The energy window derivation unit derives the energy window for each region based on the total signal value and the predetermined energy value stored in the storage unit in association with the detection position. Here, the PET apparatus may also be provided with an energy window, and the signal value only takes the signal within the energy window as an object of processing. This energy window is operated by setting a fixed upper limit value, lower limit value, etc. in advance, for example. As described above, the position where the γ-rays enter one scintillator affects the signal value. Therefore, the PET apparatus according to the embodiment may determine, for example, an energy window of a predetermined width centered on the total signal value on a per-area basis based on the total signal value stored in each area unit.

In addition, the energy window deriving unit may generate an energy distribution curve based on the total signal value stored in the storage unit in association with the detection position, and perform a predetermined ratio from the central area to the entire area under the generated energy distribution curve. Taken together, the energy window is derived.

As described above, in one embodiment, a method for determining a correction coefficient used for determining the energy of an event is provided. Events are detected by a gamma ray detector. The gamma ray detector has at least one nonlinear optical sensor arranged on a scintillator array composed of scintillator elements. Gamma ray detectors use either optical system collection processing or analog data collection processing. The above method includes (a) the step of generating gamma rays having a first predetermined energy value. Additionally, the method described above includes the step of (b) obtaining a corresponding signal value generated by at least one nonlinear optical sensor. at least one nonlinear optical sensor generating corresponding signal values in response to the arrival of a generated gamma ray indicative of the occurrence of an event and in response to receiving scintillation light emanating from at least one scintillator in the array of scintillator elements, respectively . In addition, the above-mentioned method includes (c) the step of determining the relative position of the event and the total signal value based on the signal values respectively obtained by at least one nonlinear optical sensor. The relative position is one of a predetermined number of cell positions greater than the number of scintillator elements in the array of scintillator elements. In addition, the above-mentioned method includes (d) a step of storing the total signal value in association with the determined cell position. In addition, the above-mentioned method includes (e) the step of repeating the acquisition step, the determination step, and the storage step for a plurality of events in order to generate the stored event data. In addition, the above-mentioned method includes (f) a step of determining, for each cell position, an average total signal value for the first predetermined energy value based on the stored event data. In addition, the above-mentioned method includes (g) a step of determining a correction coefficient for each cell position based on the determined average total signal value and the first predetermined energy value.

According to one aspect of the above-mentioned embodiment, the above-mentioned method further includes (1) a step of generating a second γ-ray having a second predetermined energy value. In addition, the above method includes (2) repeating steps (a) to (g) in order to determine a second correction coefficient corresponding to a second predetermined energy value for each cell position. In addition, the above-mentioned method includes (3) a step of storing each determined correction value in association with a corresponding predetermined energy value for each cell position.

According to another mode, the above method further includes the step of determining, for each unit position, a nonlinear relationship between the signal value and the energy value based on the stored correction coefficients.

In addition, according to another aspect, the above-mentioned step of determining the relative position includes a step of determining one of a predetermined number of cell positions in order to correspond to the relative position.

In addition, according to another aspect, the above-mentioned step of determining the relative position includes (1) a step of calculating a weighted average based on the respective x-y positions of at least one nonlinear optical sensor receiving the flicker light corresponding to the event. Averaging provides a weighted value by accepting the corresponding acquired signal values of the at least one non-linear light sensor for the flicker light corresponding to the event. In addition, the above-mentioned step of determining the relative position includes (2) a step of determining a total signal value by totaling the respective acquired signal values of at least one nonlinear optical sensor.

In addition, according to another aspect, the above-mentioned step of determining the correction value includes setting the correction value corresponding to the first predetermined energy value for each cell position as the difference between the first predetermined energy value and the total signal value determined for the cell position. The steps to get it.

In addition, according to another mode, at least one nonlinear light sensor comprises at least one silicon photomultiplier tube.

In another embodiment, a computer-readable medium storing a computer program for execution by a computer is provided. The computer determines the correction factor used to determine the energy of the event. Events are detected by a gamma ray detector. The gamma ray detector has at least one nonlinear optical sensor arranged on a scintillator array composed of scintillator elements. Gamma ray detectors use optical system collection processing or analog data collection processing. Through the above-mentioned program, the above-mentioned computer executes the step of (1) receiving a corresponding signal value generated by at least one nonlinear optical sensor. at least one non-linear optical sensor responsive to the arrival of a generated gamma ray indicative of the occurrence of an event and responsive to receiving scintillation light emanating from at least one scintillator of the array of scintillator elements to generate a corresponding signal value. In addition, the computer executes (2) the step of determining the relative position of the event and the total signal value based on the signal values respectively obtained from at least one nonlinear optical sensor. The relative position is one of a predetermined number of cell positions greater than the number of scintillator elements in the array of scintillator elements. In addition, the computer executes (3) the step of storing the total signal value in association with the determined cell position. Also, the computer executes (4) repeating the acquiring step, determining step, and storing step for a plurality of events in order to generate the stored event data. In addition, the computer executes (5) the step of determining the average total signal value for the first predetermined energy value based on the stored event data for each cell position. In addition, the computer executes (6) the step of determining a correction coefficient for each cell position based on the determined average total signal value and the first predetermined energy value.

In another embodiment, a method for determining a correction factor to use in determining the energy of an event is provided. Events are detected by a gamma ray detector. The gamma ray detector has at least one nonlinear optical sensor arranged on a scintillator array composed of scintillator elements. The above method includes (1) the step of generating a plurality of gamma rays. In addition, the above method includes (2) the step of obtaining a corresponding signal value generated by at least one nonlinear optical sensor. at least one non-linear optical sensor responsive to the arrival of a generated gamma ray indicative of the occurrence of an event and responsive to receiving scintillation light emanating from at least one scintillator of the array of scintillator elements to generate a corresponding signal value. In addition, the above-mentioned method includes (3) the step of determining the relative position of the event and the total signal value based on the signal values respectively obtained from at least one nonlinear optical sensor. The relative position is one of a predetermined number of cell positions greater than the number of scintillator elements in the array of scintillator elements. In addition, the above-mentioned method includes (4) a step of calculating a corrected energy value for an event based on the summed signal value for the cell position corresponding to the determined relative position and the stored calibration data. The stored calibration data defines the non-linear relationship between the signal value and the energy value at each cell location.

In another embodiment, a computer-readable medium storing a computer program executed by a computer is provided. A computer determines corrected energy values corresponding to events detected by a gamma ray detector having at least one nonlinear photosensor disposed on a scintillator array of scintillator elements. Through the above-mentioned program, the above-mentioned computer executes the step of (1) receiving a corresponding signal value generated by at least one nonlinear optical sensor. at least one non-linear optical sensor responsive to the arrival of a generated gamma ray indicative of the occurrence of an event and responsive to receiving scintillation light emanating from at least one scintillator of the array of scintillator elements to generate a corresponding signal value. In addition, the computer executes (2) the step of determining the relative position of the event and the total signal value based on the signal values respectively obtained from at least one nonlinear optical sensor. The relative position is one of a predetermined number of cell positions greater than the number of scintillator elements in the array of scintillator elements. In addition, the computer executes (3) a step of calculating a corrected energy value for an event based on the total signal value for the cell position corresponding to the determined relative position and the stored calibration data. The stored calibration data defines the non-linear relationship between the signal value and the energy value at each cell location.

In another embodiment, a method for determining a correction factor to use in determining the energy of an event is provided. Events are detected by a gamma ray detector. The gamma ray detector has at least one nonlinear optical sensor arranged on a scintillator array composed of scintillator elements. The above method includes (a) the step of generating gamma rays having a first predetermined energy value. Additionally, the method described above includes the step of (b) obtaining a corresponding signal value generated by at least one nonlinear optical sensor. at least one non-linear optical sensor responsive to the arrival of a generated gamma ray indicative of the occurrence of an event and responsive to receiving scintillation light emanating from at least one scintillator of the array of scintillator elements to generate a corresponding signal value. In addition, the above-mentioned method includes (c) the step of determining the relative position of the event and the total signal value based on the signal values respectively obtained by at least one nonlinear optical sensor. The relative position is one of a predetermined number of cell positions greater than the number of scintillator elements in the array of scintillator elements. In addition, the above method includes (d) a step of storing the total signal value in association with the determined cell position. In addition, the above-mentioned method includes (e) the step of repeating the acquisition step, the determination step, and the storage step for a plurality of events in order to generate the stored event data. In addition, the method described above includes (f) the step of determining parameters of a continuously varying mathematical correction function as a function of position within an array of scintillator elements based on the stored event data and the first prescribed energy value . The mathematical correction function represents the spatial variation in the aggregated signal value of the acquired events.

In addition, the above-mentioned method of this embodiment includes, for each event, the step of determining the energy of the event using the determined mathematical correction function and the determined relative position of the event.

According to the positron emission computed tomography apparatus of at least one embodiment described above, and the method performed by the positron emission computed tomography apparatus, energy resolution can be improved.

Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the present invention. These embodiments can be implemented in other various forms. Various omissions, substitutions, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope or gist of the invention, and are also included in the invention described in the claims and its equivalent scope.

Claims (16)

1. A positron emission computed tomography device, characterized in that it possesses: a scintillator array having a plurality of scintillators; a plurality of photosensors for detecting scintillation light generated by inputting gamma rays of a predetermined energy value to the scintillator; The specifying unit is configured to divide the scintillation light into units of areas larger than the number of the scintillators based on signal values detected by the one or more optical sensors and output from the optical sensors. Determining the detection position of the above-mentioned gamma ray interaction event with the scintillator; a storage unit that derives a total signal value based on the signal value output from the photosensor, and stores the derived total signal value in a storage unit in association with the detection position; and The derivation unit derives a correction value for correcting the energy value of the interaction event for each of the region units based on the total signal value stored in the storage unit in association with the detection position and the predetermined energy value. 2. The positron emission computed tomography apparatus according to claim 1, characterized in that, The determination unit determines the detection position based on the signal value output from the optical sensor for each of the plurality of interaction events, The storage unit derives the total signal value for each of the plurality of interaction events, and stores the derived total signal value in the storage unit in association with the detection position, The derivation unit further derives an average value per unit of the region from the plurality of total signal values stored in the storage unit in association with the detection position, and calculates a value for each The above-mentioned units of area lead to correction values that correct the energy values of the above-mentioned interaction events. 3. The positron emission computed tomography apparatus according to claim 1 or 2, characterized in that, When there are a plurality of optical sensors outputting the signal value, the specifying unit determines the detection position by weighting the positions of the optical sensors based on the signal values output from the individual optical sensors. 4. The positron emission computed tomography apparatus according to claim 1, characterized in that, The derivation unit derives the ratio of the predetermined energy value to the total signal value as the correction value for each unit of the region. 5. The positron emission computed tomography apparatus according to claim 1, characterized in that, The photosensor described above comprises at least one silicon photomultiplier tube. 6. The positron emission computed tomography apparatus according to claim 1, characterized in that, The positron emission computed tomography apparatus is a TOF type, that is, a time-of-flight type. 7. A method performed by a positron emission computed tomography apparatus having a gamma ray detector having: a scintillator array having a plurality of scintillators; a plurality of light A sensor for detecting scintillation light generated by inputting gamma rays of a predetermined energy value to the scintillator, the method is characterized by comprising: A collection step of detecting gamma rays of a predetermined energy value by one or more of the above-mentioned photosensors through the diffusion of the above-mentioned scintillation light, and collecting signal values output from the above-mentioned photosensors; a determining step of determining a detection position of the γ-ray interaction event in units of regions divided into a number greater than the number of the scintillators based on the signal value output from the photosensor; A storing step of deriving a total signal value based on the signal value output from the optical sensor, and storing the derived total signal value in a storage unit in association with the detection position; Repeating the process, repeating the above-mentioned collection process, the above-mentioned determination process, and the above-mentioned preservation process; an average value deriving step of deriving a unit average value for each of the above-mentioned regions based on the total signal value stored in the storage unit in association with the above-mentioned detection position; and The correction value deriving step derives a correction value for correcting the energy value of the interaction event for each of the region units based on the average value and the predetermined energy value. 8. The method of claim 7, wherein, It further includes a second correction value deriving step of repeating the generating step, the collecting step, the determining step, the storing step, the repeating step, the The average value derivation step and the correction value derivation step derive a correction value corresponding to the second energy value for each of the region units. 9. A positron emission computed tomography device, characterized in that it has: a scintillator array having a plurality of scintillators; a plurality of photosensors for detecting scintillation light generated by inputting gamma rays of a predetermined energy value to the scintillator; The specifying unit is configured to divide the scintillation light into units of areas larger than the number of the scintillators based on signal values detected by the one or more optical sensors and output from the optical sensors. Determining the detection location of the above gamma ray interaction event; A correcting unit derives a total signal value based on the signal value output from the photosensor, and uses the derived total signal value and the detection position determined by the determination unit to correct the energy value of the interaction event with reference to the unit definition for each of the above regions. Calibration data of the correction value, and correct the above-mentioned total signal value according to the correction value corresponding to the above-mentioned detection position. 10. Positron emission computed tomography apparatus according to claim 9, characterized in that, When there are a plurality of optical sensors outputting the signal value, the specifying unit determines the detection position by weighting the positions of the optical sensors based on the signal values output from the individual optical sensors. 11. Positron emission computed tomography apparatus according to claim 9 or 10, characterized in that The correction unit corrects the total signal value by multiplying the total signal value by the correction value. 12. The positron emission computed tomography apparatus of claim 9, wherein The positron emission computed tomography apparatus is TOF type. 13. A method performed by a positron emission computed tomography apparatus having a gamma ray detector having: a scintillator array having a plurality of scintillators; a plurality of light A sensor for detecting scintillation light generated by inputting gamma rays of a predetermined energy value to the scintillator, the method is characterized by comprising: The determining step is performed in units of areas divided into a number greater than the number of scintillators based on signal values detected by one or more of the photosensors and output from the photosensors through the diffusion of the scintillation light. Determining the location of detection of the aforementioned gamma-ray interaction event; and a correction step of deriving a total signal value from the signal value output from the above-mentioned photosensor, using the derived total signal value and the detection position determined by the above-mentioned determination process, and correcting the energy of the above-mentioned interaction event with reference to the unit definition for each of the above-mentioned regions The calibration data of the correction value of the value corrects the above-mentioned total signal value based on the correction value corresponding to the above-mentioned detection position. 14. A positron emission computed tomography device, characterized in that it has: a scintillator array having a plurality of scintillators; a plurality of photosensors for detecting scintillation light generated by inputting gamma rays of a predetermined energy value to the scintillator; The specifying unit is configured to divide the scintillation light into units of areas larger than the number of the scintillators based on signal values detected by the one or more optical sensors and output from the optical sensors. Determining the detection location of the above gamma ray interaction event; a storage unit that derives a total signal value based on the signal value output from the photosensor, and stores the derived total signal value in a storage unit in association with the detection position; and The derivation unit derives a mathematical function that continuously changes a correction value for correcting the energy value of the interaction event as a function of the detection position based on the total signal value stored in the storage unit in association with the detection position and the predetermined energy value. . 15. A method performed by a positron emission computed tomography apparatus having a gamma ray detector comprising: a scintillator array having a plurality of scintillators; a plurality of light A sensor for detecting scintillation light generated by inputting gamma rays of a predetermined energy value to the scintillator, the method is characterized by comprising: The determining step is performed in units of areas divided into a number greater than the number of scintillators based on signal values detected by one or more of the photosensors and output from the photosensors through the diffusion of the scintillation light. Determining the detection location of the above gamma ray interaction event; A saving step of deriving a total signal value from the signal value output from the photosensor, and storing the derived total signal value in a storage unit in association with the detection position; and The derivation step derives a mathematical function that continuously changes a correction value for correcting the energy value of the interaction event as a function of the detection position based on the total signal value stored in the storage unit associated with the detection position and the predetermined energy value. 16. A positron emission computed tomography device, characterized in that it has: a scintillator array having a plurality of scintillators; a plurality of photosensors for detecting scintillation light generated by inputting gamma rays of a predetermined energy value to the scintillator; The specifying unit is configured to divide the scintillation light into units of areas larger than the number of the scintillators based on signal values detected by the one or more optical sensors and output from the optical sensors. Determining the detection location of the above gamma ray interaction event; A storage unit that derives a total signal value based on the signal value output from the photosensor, and stores the derived total signal value in a storage unit in association with the detection position; and The energy window deriving unit derives an energy window for each of the above-mentioned region units based on the total signal value stored in the storage unit in association with the above-mentioned detection position and the above-mentioned predetermined energy value.