Detector calibration method and related device

Abstract

The invention discloses a detector calibration method and a related device, and relates to the field of radiation imaging, and the method comprises the steps: obtaining an output voltage signal of a photomultiplier tube for each gamma event of a radioactive source at each test position of the surface of a detector based on a dot matrix labeling method, obtaining an output voltage signal set of each photomultiplier tube, and obtaining the output voltage signal set of each photomultiplier tube; and performing statistical processing on all the output voltage signal sets to obtain statistical characteristic data of the output voltage. And fitting the statistical characteristic data to obtain a relation function between the output voltage of the photomultiplier and the number of photoelectrons. Solving the maximum value of the likelihood function representing the relationship between the number of photoelectrons of each photomultiplier and each test position based on a preset search algorithm to obtain the position with the maximum number of photoelectrons on the surface of the detector; wherein the likelihood function is constructed by an average response function determined by the relation function and the number of photoelectrons of each photomultiplier. Depth information of deposition energy in the crystal can be better reflected, and the imaging quality of the detector is further improved.

Description

Technical Field

[0001] The present application relates to the field of radiological imaging technology, and in particular to a detector calibration method and related devices. Background Art

[0002] SPECT (Single-Photon Emission Computed Tomography) detectors are usually different versions of gamma cameras based on scintillators, and each event will produce an output on the PMT (photomultiplier tube) array. For the output signals of multiple PMTs in the PMT array, a simple centroid algorithm is usually used to process these signals to estimate the two-dimensional spatial position of the event and the deposited energy. Due to the problems of low spatial resolution and edge distortion of the centroid algorithm, when the detector obtained by this calibration method is used for image acquisition, the image quality obtained is poor and can no longer meet the requirements of higher imaging quality. Summary of the invention

[0003] In view of the above problems, this application provides a detector calibration method and related devices to achieve the purpose of improving imaging quality. The specific scheme is as follows:

[0004] The first aspect of the present application provides a detector calibration method, comprising:

[0005] Based on the dot matrix marking method, the output voltage signal of the photomultiplier tube for each gamma event of the radiation source at each test position on the detector surface is obtained to obtain an output voltage signal set of each photomultiplier tube;

[0006] Performing statistical processing on all output voltage signal sets to obtain statistical characteristic data of the output voltage;

[0007] Performing fitting processing on the statistical characteristic data to obtain a relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons;

[0008] Based on a preset search algorithm, the maximum value of the likelihood function characterizing the relationship between the number of photoelectrons of each photomultiplier tube and each test position is solved to obtain the position with the largest number of photoelectrons on the detector surface. The likelihood function is constructed by the average response function determined by the relationship function and the number of photoelectrons of each photomultiplier tube.

[0009] In a possible implementation, performing statistical processing on all output voltage signal sets to obtain statistical characteristic data of the output voltage includes:

[0010] Dividing the output voltage signals in the entire output voltage signal set according to the test positions to obtain a first output voltage divided set;

[0011] Dividing the first voltage division set according to each of the photomultiplier tubes to obtain a second output voltage division set;

[0012] Gaussian fitting is performed on the statistical histogram of the output voltage subset of each photomultiplier tube in the second output voltage division set to obtain the mean and variance of the output voltage of each photomultiplier tube.

[0013] In a possible implementation, the fitting process is performed on the statistical characteristic data to obtain a relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons, including:

[0014] The mean and variance of the output voltage of each photomultiplier tube at each test position are fitted by a least squares polynomial to obtain an action coefficient characterizing the action relationship between the output voltage of the photomultiplier tube and the number of photoelectrons;

[0015] Based on the action coefficient, a relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons is obtained.

[0016] In a possible implementation, the process of determining the average response function includes:

[0017] Based on the relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons, the relationship function between the average output voltage of the photomultiplier tube and the average number of photoelectrons is obtained;

[0018] The average response function is obtained according to the relationship function between the average output voltage of the photomultiplier tube and the average number of photoelectrons.

[0019] In a possible implementation, the process of determining the number of photoelectrons of each photomultiplier tube includes:

[0020] The number of photoelectrons of the photomultiplier tube is calculated based on the action coefficient and the relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons.

[0021] In a possible implementation, the solving of the maximum value of the likelihood function between the number of photoelectrons characterizing each of the photomultiplier tubes and each of the test positions based on a preset search algorithm includes:

[0022] If the detector surface is a square, a preset number of iterations are performed on the detector surface to obtain the maximum value of the likelihood function. When performing the iterations, the position of the connecting line corresponding to the maximum value of the likelihood function value obtained in the previous iteration is used as the center of the first square area of ​​the current iteration;

[0023] In the current iteration, each side of the first square area is equally divided to divide the first square area into a preset number of first square sub-areas, where the side length of the first square sub-area is half of the side length of the second square sub-area obtained in the previous iteration;

[0024] The likelihood function values ​​at the intersection points of the connecting lines constituting the preset number of first square sub-areas are calculated, and the maximum value of this iteration is determined therefrom; in the first iteration, each side of the detector surface is divided into equal parts, and then the detector surface is divided into the preset number of third square sub-areas, and the likelihood function values ​​at the intersection points of the connecting lines constituting the preset number of third square sub-areas are calculated, and the maximum value of the first iteration is determined therefrom.

[0025] In a possible implementation, it also includes: if the detector surface is non-square, then after the detector surface is expanded into a square area, the expanded square area is iterated a preset number of times to obtain the maximum value of the likelihood function, and the likelihood function values ​​of the area in the expanded square area except the detector surface are all 0.

[0026] A second aspect of the present application provides a detector calibration device, comprising:

[0027] An output signal acquisition module is used to obtain the output voltage signal of the photomultiplier tube for each gamma event of the radiation source at each test position on the detector surface based on a dot matrix marking method, and obtain an output voltage signal set of each photomultiplier tube;

[0028] A characteristic data statistics module is used to perform statistical processing on all output voltage signal sets to obtain statistical characteristic data of the output voltage;

[0029] an action relationship determination module, used for performing fitting processing on the statistical characteristic data to obtain a relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons; and

[0030] A likelihood function solving module is used to solve the maximum value of the likelihood function that characterizes the relationship between the number of photoelectrons of each photomultiplier tube and each test position based on a preset search algorithm to obtain the position with the largest number of photoelectrons on the detector surface. The likelihood function is constructed by the average response function determined by the relationship function and the number of photoelectrons of each photomultiplier tube.

[0031] A third aspect of the present application provides a computer program product, comprising computer-readable instructions, which, when executed on an electronic device, enables the electronic device to implement the detector calibration method of the first aspect or any implementation of the first aspect.

[0032] A fourth aspect of the present application provides an electronic device, comprising at least one processor and a memory connected to the processor, wherein:

[0033] The memory is used to store computer programs;

[0034] The processor is used to execute the computer program so that the electronic device can implement the detector calibration method of the above-mentioned first aspect or any implementation manner of the first aspect.

[0035] A fifth aspect of the present application provides a computer storage medium, which carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement the detector calibration method of the above-mentioned first aspect or any implementation of the first aspect.

[0036] By means of the above technical scheme, the detector calibration method provided by the present application obtains the output voltage signal of the photomultiplier tube for each gamma event of the radiation source at each test position on the detector surface based on the dot matrix marking method, obtains the output voltage signal set of each photomultiplier tube, performs statistical processing on all output voltage signal sets, and obtains the statistical characteristic data of the output voltage. The statistical characteristic data is fitted to obtain the relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons. Based on the preset search algorithm, the maximum value of the likelihood function that characterizes the relationship between the number of photoelectrons of each photomultiplier tube and each test position is solved to obtain the position with the largest number of photoelectrons on the detector surface, wherein the likelihood function is constructed by the average response function determined by the relationship function and the number of photoelectrons of each photomultiplier tube. Compared with the traditional centroid method, the position information obtained by the dot matrix marking method and the likelihood function estimation can better reflect the depth information of the energy deposited in the crystal, thereby improving the imaging quality of the detector. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0038] Figure 1 A flow chart of a detector calibration method provided in this application;

[0039] Figure 2 An array arrangement diagram of a photomultiplier tube provided in this application;

[0040] Figure 3 The pan-field diagram of the lattice model provided for this application;

[0041] Figure 4 This is the result diagram of automatic segmentation of the pan-field image provided in this application;

[0042] Figure 5 Average response graph of the photomultiplier tubes provided for this application;

[0043] Figure 6 Example diagram of the maximum likelihood problem provided for this application;

[0044] Figure 7 A schematic diagram of the shrinking grid search algorithm provided for this application;

[0045] Figure 8 A schematic diagram of a directional search algorithm provided for this application;

[0046] Fig. 9 Comparison chart of the effects of the centroid method and maximum likelihood algorithm provided for this application

[0047] Fig.10 A structural diagram of the detector calibration device provided for this application;

[0048] Fig.11 This is a structural diagram of the electronic device provided for this application. DETAILED DESCRIPTION

[0049] The following describes the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation method section of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0050] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0051] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0052] The present application embodiment provides a detector calibration method. The detector calibration method of the present application embodiment is described in detail below with reference to the accompanying drawings.

[0053] Reference Figure 1 , Figure 1 A flow chart of a detector calibration method provided in an embodiment of the present application is as follows: Figure 1 As shown, the detector calibration method provided in the embodiment of the present application may include steps 101 to 104, and these steps are described in detail below.

[0054] 101. Based on the dot matrix marking method, the output voltage signal of the photomultiplier tube for each gamma event of the radiation source at each test position on the detector surface is obtained to obtain the output voltage signal set of each photomultiplier tube.

[0055] Specifically, refer to Figure 2 The photomultiplier tube array of a commonly used detector is shown. The detector includes a whole piece of NaI (Tl) crystal (585×470×9.5mm 3 ) and 55 photomultiplier tubes (49 R6233 (Φ76mm) and 6 R6231 (Φ50mm)). The output signal of each PMT is amplified and digitized through the electronic systems such as the front amplifier board, VGA circuit (video graphics array circuit), ADC (analog-to-digital converter), and then passed through FPGA, communication module, etc., and finally the digital voltage signal is collected by the terminal software such as the computer. In the implementation of detector calibration using the likelihood function, the detector average response function needs to be calibrated. In the actual application process, what can be obtained is the voltage signal output by the detector, so the gain of each photomultiplier tube and the gain of the electronics processed by the subsequent circuit also need to be calibrated.

[0056] The commonly used direct calibration method currently uses a collimated source to scan a two-dimensional dot grid on the surface of a scintillation crystal, collects enough gamma events at each point, and then processes the data to obtain the average response of M photomultiplier tubes at that point, combines the average responses of all PMTs at all positions, and finally obtains the average response function of the detector. Although this method is simple and intuitive, it is labor-intensive, time-consuming, and requires dedicated precision mechanical moving equipment.

[0057] In order to overcome the shortcomings of the direct calibration method, the present application adopts the dot matrix calibration method. During the calibration process, a tungsten plate with circular holes evenly distributed on the surface can be placed on the detector surface, and the detector surface can be illuminated by the flood field generated by the metastable technetium isomer, and sufficient gamma event data can be collected to obtain a flood field image.

[0058] For example, a 3mm thick tungsten plate with 55×43 arrays of 3mm circular holes is placed on the plate. The center spacing of the circular holes is 9.86mm horizontally and 9.95mm vertically. The pan-field image obtained by placing it on the detector surface is as follows: Figure 3 The flood field map is automatically segmented using the watershed algorithm in Matlab. The results are as follows: Figure 4 The overall segmentation result is ideal, and only the segmentation errors at the edges need to be manually corrected. Then, the collected gamma event data are divided into 55×43 groups according to the segmentation results. Each group of data corresponds to a hole on the tungsten plate, that is, the spatial position of the gamma event in each group of data is consistent and is located at the position of the corresponding hole on the tungsten plate. Similar to the direct calibration method, the calibration work can be completed by processing the data. The 2365 groups of data obtained by segmenting the data collected by the dot matrix calibration method at one time are equivalent to the data collected by moving the pencil beam 2365 positions in the direct calibration method. Therefore, the dot matrix calibration method greatly reduces the workload and does not require precise and complex mechanical moving devices.

[0059] The data to be processed is that when a gamma event occurs at different locations on the detector surface, all the photomultiplier tubes that respond will output corresponding voltage signals. For each test position R on the detector surface, the detector detects a large number of N gamma events. For each gamma event, some photomultiplier tubes in the detector will have an output signal V m These data can be expressed as {V m (R, i)}, where m is the serial number of the photomultiplier tube, R is the test position, and i is the gamma event serial number (i = 1, 2, 3..., N).

[0060] It is understandable that those skilled in the art can adjust the size of the tungsten plate and the circular hole in the dot matrix calibration method as needed, and no limitation is made here.

[0061] 102. Perform statistical processing on all output voltage signal sets to obtain statistical characteristic data of the output voltage.

[0062] Specifically, the output voltage signals in all output voltage signal sets are divided according to the test positions to obtain a first output voltage division set:

[0063] According to the different incident positions R on the detector surface, the data are divided into multiple groups to obtain a series of {V m (i)}.

[0064] The first voltage division set is divided according to each photomultiplier tube to obtain the second output voltage division set:

[0065] {V m (i)} is divided into multiple groups {V(i)} according to different m.

[0066] Gaussian fitting is performed on the statistical histogram of the output voltage subset of each photomultiplier tube in the second output voltage partition set to obtain the mean and variance of the output voltage of each photomultiplier tube:

[0067] Draw a statistical histogram of {V(i)} and use the peak in the Gaussian fit graph to get the mean And variance Var(V), perform the previous step on all group data and get and {Var(V) m (R)}.

[0068] 103. The statistical characteristic data is fitted to obtain the relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons.

[0069] Specifically, the mean and variance of the output voltage of each photomultiplier tube at each test position are fitted with a least squares polynomial to obtain an action coefficient that characterizes the action relationship between the output voltage of the photomultiplier tube and the number of photoelectrons. On this basis, the relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons is obtained based on the action coefficient.

[0070] After grouping the mean and variance of the test positions obtained above according to different m groups, we can get multiple groups: and {Var(V)(R)}, all R corresponding Plotted on a graph, we can see that these points are linearly related, and by performing a least squares polynomial fit, we can get the formula for the fitted line.

[0071] According to the photomultiplier tube output, the electronically amplified signal V m formula:

[0072] V m =G m n m +V m0 (1)

[0073] Combining the above fitting straight line formula, the gain G of each photomultiplier tube can be obtained. m and the gain V of the electronics m0 .

[0074] According to the relationship function between the average output voltage of the photomultiplier tube and the average number of photoelectrons, the average response function is obtained, and then based on the action coefficient and the relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons, the number of photoelectrons of the photomultiplier tube is converted: V output by the photomultiplier tube m The mean of is:

[0075]

[0076] The output signal V m The number of photoelectrons n on the photocathode of the photomultiplier tube m Connect together, Convert to The average response function of the detector can be obtained.

[0077] The MDRF (mean response function) obtained through the calibration process is often displayed as a series of pictures. Each picture corresponds to a photomultiplier tube, and each pixel of the picture corresponds to a spatial position on the detector surface. The pixel value represents the average number of photoelectrons generated on the photocathode of the photomultiplier tube when a gamma event occurs at this position. Figure 5 shown.

[0078] 104. Based on a preset search algorithm, the maximum value of the likelihood function that characterizes the relationship between the number of photoelectrons of each photomultiplier tube and each test position is solved to obtain the position on the detector surface with the largest number of photoelectrons. The likelihood function is constructed by the average response function determined by the relationship function and the number of photoelectrons of each photomultiplier tube.

[0079] Specifically, for the parameter estimation task of detector signal processing, the likelihood function is in the following form:

[0080] pr(g|θ) (3)

[0081] Among them, θ is an N×1 vector, representing N parameters to be estimated, such as spatial position (x, y, z), energy E, time t, etc. g is an M×1 data vector, representing M collected detector signals, such as the output voltage signal of a photomultiplier tube.

[0082] The maximum likelihood estimate of the likelihood function is to choose a The value of makes the likelihood function maximum. The formula of ML (maximum likelihood estimation) is as follows:

[0083]

[0084] For the convenience of calculation, the equivalent form of the ML formula is selected in practical application as follows:

[0085] Given the prior knowledge of θ, we can use this information by using the maximum a posteriori probability estimation (MAP). The MAP formula is as follows:

[0086]

[0087] It can be seen that MAP only has one more prior probability weight compared to ML. When the probability of all values ​​of θ is the same, pr(θ) is a constant, and MAP degenerates into ML.

[0088] Based on the above maximum likelihood estimation theory, the construction process of the statistical model of the scintillator detector signal specifically includes:

[0089] For the 140keV gamma photons emitted by the metastable technetium isomer, there are two possible interactions between them and the NaI(Tl) crystal: photoelectric effect and Compton scattering. Among them, the energy of photoelectrons, X-rays, Auger electrons, etc. generated by the photoelectric effect will be deposited within a scale of 100μm, and converted into scintillation photons. For the spatial resolution of the detector (~mm), 100μm is small enough and can be ignored. When Compton scattering occurs, the energy lost by the gamma photons will be converted into scintillation light, but the scattered photons will move away from the action position and continue to interact with the NaI(Tl) crystal or escape from the NaI(Tl) crystal.

[0090] NaI(Tl) crystal can convert the energy deposited by γ photons into scintillation light. Assuming that the response of scintillation crystal to the deposited energy is linear, the average number of scintillation photons is:

[0091]

[0092] Among them, Q sc is the luminous efficiency of the scintillator, that is, the average number of scintillation photons emitted per unit of deposited energy. E is the energy value deposited by the gamma photons.

[0093] For scintillators, it is usually assumed that the specific number of scintillation photons obeys Poisson distribution, but in fact many scintillators, especially NaI(Tl) crystals, respond nonlinearly to energy.

[0094] Since the average number of scintillation photons is also a random variable, the average number of scintillation photons follows a doubly random Poisson process.

[0095] On the basis that the average number of scintillation photons obeys the Poisson distribution, the visible light transport process and the photoelectric conversion process on the photocathode of the photomultiplier tube are both binomial processes. According to the basic principle of statistics: the binomial choice of a Poisson random variable is still a Poisson random variable, so the number of photoelectrons generated on the photocathode of each photomultiplier tube is also a Poisson random variable, and the average number of photoelectrons is:

[0096]

[0097] Among them, the vector R = (x, y, z) is the spatial coordinate where the gamma photon acts, n m is the number of photoelectrons in the photocathode of the mth photomultiplier tube, η m represents the quantum efficiency of the photocathode, f m (R) represents the fraction of scintillation photons reaching the photocathode of the mth photomultiplier tube.

[0098] According to the Poisson distribution formula, n is observed in the case of R, E m The probability of a photoelectron is:

[0099]

[0100] For n observed by different photomultiplier tubes m are independent of each other, so for M photomultiplier tubes, {n m ,m=1,...,M} photoelectrons are:

[0101]

[0102] The above formula is a likelihood function that characterizes the relationship between the number of photoelectrons of each photomultiplier tube and each test position.

[0103] The nonlinearity of the scintillator energy response causes the number of scintillation photons to not obey the Poisson distribution. Since the number of scintillation photons is a relatively large value, the quantum efficiency and photon transport issues η m f m (R) is a relatively small fraction. Whether the scintillation photons can reach the photocathode of the photomultiplier tube and generate photoelectrons can be regarded as a binomial distribution process. According to the law in statistics: for binomial distribution, if the probability is small and the number of experiments is large, its probability distribution can be approximately regarded as Poisson distribution. Therefore, whether the number of scintillation photons obeys Poisson distribution or not, n m The assumptions of Poisson distribution are all valid. Therefore, the likelihood function obtained above is valid.

[0104] Using the above formula (2) Convert to The detector average response function can be obtained. According to formula (1), the voltage signal output by the photomultiplier tube is converted into the number of photoelectrons, and then it is substituted into formula (9) to solve the maximum likelihood problem.

[0105] For the current maximum likelihood problem, its characteristic is that there is only one maximum value on the rectangular domain, such as Figure 6 As shown. The shrinking grid search algorithm can be used to solve it. The specific process is as follows:

[0106] If the detector surface is a square, a preset number of iterations are performed on the detector surface to obtain the maximum value of the likelihood function. When performing the iterative processing, the position of the connecting line corresponding to the maximum value of the likelihood function value obtained in the previous iteration is used as the center of the first square area of ​​the current iteration.

[0107] In the current iteration, each side of the first square area is equally divided to divide the first square area into a preset number of first square sub-areas, and the side length of the first square sub-area is half of the side length of the second square sub-area obtained in the previous iteration.

[0108] The likelihood function values ​​at the intersection points of the connecting lines constituting the preset number of first square sub-areas are calculated, and the maximum value of this iteration is determined therefrom; in the first iteration, each side of the detector surface is divided equally, and then the detector surface is divided into a preset number of third square sub-areas, and the likelihood function values ​​at the intersection points of the connecting lines constituting the preset number of third square sub-areas are calculated, and the maximum value of the first iteration is determined therefrom.

[0109] If the detector surface is non-square, the detector surface is expanded into a square area, and the expanded square area is iterated a preset number of times to obtain the maximum value of the likelihood function, wherein the likelihood function values ​​of the area in the expanded square area except the detector surface are all 0.

[0110] For details, please refer to Figure 7 An example diagram of the shrinking grid search algorithm shown in FIG. For example, the search domain is a 78×78 dot matrix. For the shrinking grid search algorithm, in order to facilitate the execution of the algorithm, the 78×78 dot matrix of the detector must first be expanded to 128×128, and the likelihood function value of the expanded point is fixed to 0. Then, the center of the detector surface is taken as the initial value of the center point of the first iteration, and the grid width is 32, such as the 4×4 grid in FIG (a), and the likelihood function values ​​at these 16 points are calculated. The coordinates of the point with the largest likelihood function value are taken as the center point of the next iteration, and half of the grid spacing in the current iteration is taken as the grid spacing of the next iteration. After a fixed number of iterations, the algorithm will search for the location of the maximum value of the likelihood function. For the case where the detector surface is a 78×78 grid, the number of iterations of the shrinking grid search algorithm is 6.

[0111] In addition, you can also use Figure 8 The directional search algorithm shown in the figure can be divided into multiple iterations. In the first iteration, the search algorithm calculates the likelihood function value on a 5×5 grid of equally spaced points, and selects the point with the largest likelihood function as (x0, y0), as shown above. Figure 8 As shown in (a). In the second iteration, the search algorithm calculates the likelihood function values ​​of the 8 points in the neighborhood of (x0, y0), and takes the coordinates of the point with the maximum likelihood function as (x1, y1). The second iteration process is repeated until the search algorithm converges to the maximum value of the likelihood function (xk, yk).

[0112] For the directional search algorithm, the speed of the algorithm is closely related to the selection of the initial point (x0, y0). If the initial point is very close to the final extreme point, the number of iterations of the algorithm will be very small. In the directional search algorithm, the initial point is determined by using a 5×5 grid calculation. The accuracy of the initial point calculated in this way is related to the number of grid points. The more grid points there are, the closer it is to the extreme point, but more probability values ​​of points need to be calculated. Using the result of the Anger algorithm as the initial point of the directional search algorithm is a good choice. Not only is the result very close to the extreme point, but the calculation speed is also fast.

[0113] In addition, the most direct and simple exhaustive method can be used to calculate the likelihood function values ​​of all positions and take the maximum value to get θ. This method gives accurate results but is the slowest.

[0114] Compared with the exhaustive method, the directed search algorithm is much faster and the results are more accurate. The number of iterations required to reach convergence for each calculation is uncertain and increases significantly as the search domain expands. The combination of the Anger algorithm and the directed search algorithm is an optimization based on the directed search algorithm.

[0115] The shrinking grid search algorithm is fast, and the number of iterations for calculating convergence is fixed, making the algorithm structure simpler and easier to execute on various platforms. Compared with other algorithms, the shrinking grid search algorithm is more suitable for situations with larger search domains. As the number of points in the search domain increases sharply, the amount of calculation for other search algorithms will also increase greatly, but the shrinking grid search algorithm only needs one or two more iterations (that is, dozens of more point likelihood function calculation processes), and the rate at which the amount of calculation increases is much lower than that of other algorithms.

[0116] This detector calibration method uses the maximum likelihood estimation algorithm to estimate parameters such as spatial position with excellent performance and no distortion, and combines the dot matrix labeling method to calibrate the detector, which can retain more depth information of the detector, thereby making the obtained detector image higher in quality.

[0117] And refer to Fig. 9 As shown, the left side is the projection image of the detector surface determined by the centroid method, and the right side is the projection image of the detector surface estimated according to the likelihood function of the present application. It can be seen that compared with the traditional centroid method, the gamma photon position determined by the present application is clearer and has less edge distortion.

[0118] A detector calibration method provided in an embodiment of the present application is introduced above, and a device for executing the above detector calibration method will be introduced below.

[0119] See also Fig.10 , Fig.10This is a schematic diagram of the structure of a detector calibration device provided in an embodiment of the present application. Fig. 9 As shown, the detector calibration device comprises:

[0120] The output signal acquisition module 1001 is used to obtain the output voltage signal of the photomultiplier tube for each gamma event of the radiation source at each test position on the detector surface based on the dot matrix marking method, and obtain the output voltage signal set of each photomultiplier tube;

[0121] The characteristic data statistics module 1002 is used to perform statistical processing on all output voltage signal sets to obtain statistical characteristic data of the output voltage;

[0122] The function relationship determination module 1003 is used to perform fitting processing on the statistical characteristic data to obtain a relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons; and

[0123] The likelihood function solving module 1004 is used to solve the maximum value of the likelihood function characterizing the relationship between the number of photoelectrons of each photomultiplier tube and each test position based on a preset search algorithm to obtain the position with the largest number of photoelectrons on the detector surface. The likelihood function is constructed by the average response function determined by the relationship function and the number of photoelectrons of each photomultiplier tube.

[0124] In a possible implementation, the characteristic data statistics module 1002 performs statistical processing on all output voltage signal sets to obtain statistical characteristic data of the output voltage, including:

[0125] Dividing the output voltage signals in the entire output voltage signal set according to the test positions to obtain a first output voltage divided set;

[0126] Dividing the first voltage division set according to each of the photomultiplier tubes to obtain a second output voltage division set;

[0127] Gaussian fitting is performed on the statistical histogram of the output voltage subset of each photomultiplier tube in the second output voltage division set to obtain the mean and variance of the output voltage of each photomultiplier tube.

[0128] In a possible implementation, the process of the action relationship determination module 1003 performing fitting processing on the statistical characteristic data to obtain the relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons includes:

[0129] The mean and variance of the output voltage of each photomultiplier tube at each test position are fitted by a least squares polynomial to obtain an action coefficient characterizing the action relationship between the output voltage of the photomultiplier tube and the number of photoelectrons;

[0130] Based on the action coefficient, a relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons is obtained.

[0131] In a possible implementation, the process of determining the average response function in the likelihood function solving module 1004 includes:

[0132] Based on the relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons, the relationship function between the average output voltage of the photomultiplier tube and the average number of photoelectrons is obtained;

[0133] The average response function is obtained according to the relationship function between the average output voltage of the photomultiplier tube and the average number of photoelectrons.

[0134] In a possible implementation, the process of determining the number of photoelectrons of each photomultiplier tube in the likelihood function solving module 1004 includes:

[0135] The number of photoelectrons of the photomultiplier tube is calculated based on the action coefficient and the relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons.

[0136] In a possible implementation, the likelihood function solving module 1004 solves the likelihood function representing the number of photoelectrons of each photomultiplier tube and each test position for the maximum value based on a preset search algorithm, including:

[0137] If the detector surface is a square, a preset number of iterations are performed on the detector surface to obtain the maximum value of the likelihood function. When performing the iterations, the position of the connecting line corresponding to the maximum value of the likelihood function value obtained in the previous iteration is used as the center of the first square area of ​​the current iteration;

[0138] In the current iteration, each side of the first square area is equally divided to divide the first square area into a preset number of first square sub-areas, where the side length of the first square sub-area is half of the side length of the second square sub-area obtained in the previous iteration;

[0139] The likelihood function values ​​at the intersection points of the connecting lines constituting the preset number of first square sub-areas are calculated, and the maximum value of this iteration is determined therefrom; in the first iteration, each side of the detector surface is divided into equal parts, and then the detector surface is divided into the preset number of third square sub-areas, and the likelihood function values ​​at the intersection points of the connecting lines constituting the preset number of third square sub-areas are calculated, and the maximum value of the first iteration is determined therefrom.

[0140] In one possible implementation, the likelihood function solving module 1004 is also used to, if the detector surface is non-square, expand the detector surface into a square area, perform iterative processing on the expanded square area a preset number of times to obtain the maximum value of the likelihood function, and the likelihood function values ​​of the area in the expanded square area except the detector surface are all 0.

[0141] The present application also provides an electronic device in an embodiment. Fig.11 As shown, it shows a schematic diagram of the structure of an electronic device suitable for implementing the embodiments of the present application. The electronic device in the embodiments of the present application may include but is not limited to fixed terminals such as mobile phones, notebook computers, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Fig.11 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0142] like Fig.11 As shown, the electronic device may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 1101, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1102 or a program loaded from a storage device 1108 to a random access memory (RAM) 1103. When the electronic device is powered on, various programs and data required for the operation of the electronic device are also stored in the RAM 1103. The processing device 1101, the ROM 1102, and the RAM 1103 are connected to each other via a bus 1104. An input / output (I / O) interface 1105 is also connected to the bus 1104.

[0143] Typically, the following devices may be connected to the I / O interface 1105: an input device 1106 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 1107 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 1108 including, for example, a memory card, a hard disk, etc.; and a communication device 1109. The communication device 1109 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although Fig.11 An electronic device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0144] An embodiment of the present application also provides a computer program product including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements any detector calibration method provided in the embodiment of the present application.

[0145] A computer-readable storage medium is also provided in an embodiment of the present application. The storage medium carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement any detector calibration method provided in the embodiment of the present application.

[0146] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0147] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0148] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0149] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

Claims

1. A detector calibration method, characterized in that: include: Based on the dot matrix marking method, the output voltage signal of the photomultiplier tube for each gamma event of the radiation source at each test position on the detector surface is obtained to obtain an output voltage signal set of each photomultiplier tube; Performing statistical processing on all output voltage signal sets to obtain statistical characteristic data of the output voltage; Performing fitting processing on the statistical characteristic data to obtain a relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons; Based on a preset search algorithm, the maximum value of the likelihood function characterizing the relationship between the number of photoelectrons of each photomultiplier tube and each test position is solved to obtain the position with the largest number of photoelectrons on the detector surface. The likelihood function is constructed by the average response function determined by the relationship function and the number of photoelectrons of each photomultiplier tube.

2. The detector calibration method according to claim 1, characterized in that: The statistical processing of all output voltage signal sets to obtain statistical characteristic data of the output voltage includes: Dividing the output voltage signals in the entire output voltage signal set according to the test positions to obtain a first output voltage divided set; Dividing the first voltage division set according to each of the photomultiplier tubes to obtain a second output voltage division set; Gaussian fitting is performed on the statistical histogram of the output voltage subset of each photomultiplier tube in the second output voltage division set to obtain the mean and variance of the output voltage of each photomultiplier tube.

3. The detector calibration method according to claim 2, characterized in that: The fitting process is performed on the statistical characteristic data to obtain a relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons, including: The mean and variance of the output voltage of each photomultiplier tube at each test position are fitted by a least squares polynomial to obtain an action coefficient characterizing the action relationship between the output voltage of the photomultiplier tube and the number of photoelectrons; Based on the action coefficient, a relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons is obtained.

4. The detector calibration method according to claim 3, characterized in that: The process of determining the average response function includes: Based on the relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons, the relationship function between the average output voltage of the photomultiplier tube and the average number of photoelectrons is obtained; The average response function is obtained according to the relationship function between the average output voltage of the photomultiplier tube and the average number of photoelectrons.

5. The detector calibration method according to claim 3, characterized in that: The process of determining the number of photoelectrons of each photomultiplier tube comprises: The number of photoelectrons of the photomultiplier tube is calculated based on the action coefficient and the relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons.

6. The detector calibration method according to claim 1, characterized in that: The method of solving the maximum value of the likelihood function between the number of photoelectrons characterizing each photomultiplier tube and each test position based on a preset search algorithm includes: If the detector surface is a square, a preset number of iterations are performed on the detector surface to obtain the maximum value of the likelihood function. When performing the iterations, the position of the connecting line corresponding to the maximum value of the likelihood function value obtained in the previous iteration is used as the center of the first square area of ​​the current iteration; In the current iteration, each side of the first square area is equally divided to divide the first square area into a preset number of first square sub-areas, where the side length of the first square sub-area is half of the side length of the second square sub-area obtained in the previous iteration; The likelihood function values ​​at the intersection points of the connecting lines constituting the preset number of first square sub-areas are calculated, and the maximum value of this iteration is determined therefrom; in the first iteration, each side of the detector surface is divided into equal parts, and then the detector surface is divided into the preset number of third square sub-areas, and the likelihood function values ​​at the intersection points of the connecting lines constituting the preset number of third square sub-areas are calculated, and the maximum value of the first iteration is determined therefrom.

7. The detector calibration method according to claim 6, characterized in that: Also includes: If the detector surface is non-square, the detector surface is expanded into a square area, and the expanded square area is iteratively processed a preset number of times to obtain the maximum value of the likelihood function, and the likelihood function values ​​of the area in the expanded square area except the detector surface are all 0.

8. A detector calibration device, characterized in that: include: An output signal acquisition module is used to obtain the output voltage signal of the photomultiplier tube for each gamma event of the radiation source at each test position on the detector surface based on a dot matrix marking method, and obtain an output voltage signal set of each photomultiplier tube; A characteristic data statistics module is used to perform statistical processing on all output voltage signal sets to obtain statistical characteristic data of the output voltage; An action relationship determination module is used to perform fitting processing on the statistical characteristic data to obtain a relationship function between the output voltage of the photomultiplier tube and the number of photoelectrons; as well as, A likelihood function solving module is used to solve the maximum value of the likelihood function that characterizes the relationship between the number of photoelectrons of each photomultiplier tube and each test position based on a preset search algorithm to obtain the position with the largest number of photoelectrons on the detector surface. The likelihood function is constructed by the average response function determined by the relationship function and the number of photoelectrons of each photomultiplier tube.

9. An electronic device, characterized in that: The method comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the electronic device can implement the detector calibration method as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: The storage medium carries one or more computer programs, and when the one or more computer programs are executed by an electronic device, the electronic device can implement the detector calibration method as described in any one of claims 1 to 7.

Images