Status Monitoring Method and Device for Imaging Equipment

Abstract

The present invention discloses a state monitoring method and device for imaging equipment, which relates to the field of data processing technology, and the main purpose is to solve the problem of poor accuracy of state monitoring of existing imaging equipment. It includes: establishing a system model of imaging equipment, wherein the system model includes multiple sub-models that simulate the complete imaging process of the imaging equipment; acquiring test image data of a scanned target object, and simulating the operation process of the imaging equipment in combination with the system model and the test image data to obtain a simulation result, wherein the simulation process is sequentially executed in linkage according to the input-output coupling relationship between each sub-model; determining the state parameters of the imaging equipment based on the simulation results and the collected equipment data, wherein the state parameters are used to characterize the state evaluation content of the imaging equipment in different dimensions; determining the state monitoring result of the imaging equipment based on the state parameters.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and device for monitoring the status of imaging equipment. Background Art

[0002] Positron emission tomography (PET) is a functional medical imaging device that scans the human body after injecting radioactive nuclides into it to detect the time, space and energy information of the gamma photon pairs emitted by the annihilation of positrons emitted when the nuclides decay in the human body, and then estimate the distribution of the nuclides in the human body. At this time, since the distribution of different nuclides as tracers in the human body is related to the biological metabolic process, the images taken by the PET device can be used to assist doctors in clinical screening and diagnosis of diseases. In this process, since the PET device is easily affected by temperature and humidity changes, electronic system stability and other aspects, its detection accuracy and efficiency in energy, time, space and other aspects will have certain deviations, that is, it will cause image artifacts, thereby reducing image quality. Therefore, it is necessary to monitor the status of the PET device on a daily basis.

[0003] At present, the existing PET device status monitoring is usually completed by pre-scanning monitoring, that is, starting the PET device for simulated scanning and status monitoring before use. However, this pre-scanning monitoring can only ensure whether the PET device status is normal at the time of simulated scanning, and cannot determine whether the PET device is operating normally when the PET device is clinically scanning patients. Therefore, a status monitoring method for imaging equipment is urgently needed to solve the above problems. Summary of the invention

[0004] In view of this, the present invention provides a method and device for monitoring the status of an imaging device, the main purpose of which is to solve the problem of poor accuracy in status monitoring of existing imaging devices.

[0005] According to one aspect of the present invention, a method for monitoring the status of an imaging device is provided, comprising:

[0006] Establishing a system model of an imaging device, wherein the system model includes a plurality of sub-models simulating the complete imaging process of the imaging device;

[0007] Acquire test image data of the scanned target object, and simulate the operation process of the imaging device in combination with the system model and the test image data to obtain a simulation result, wherein the simulation process is sequentially executed in linkage according to the input-output coupling relationship between each sub-model;

[0008] Determine the state parameters of the imaging device based on the simulation results and the collected device data, where the state parameters are used to characterize the state evaluation content of the imaging device in different dimensions;

[0009] Determine the state monitoring result of the imaging device based on the state parameters.

[0010] Further, the system model includes a radionuclide distribution sub-model, an attenuation distribution sub-model, a photon propagator sub-model, a detector sub-model, and an electron sub-model. Establishing the system model of the imaging device includes:

[0011] Establish a radionuclide distribution sub-model that characterizes the spatial distribution of radionuclide activity based on the voxel spatial positions in the test image data;

[0012] Determine the linear attenuation coefficient at each position in the test image data based on the voxel spatial positions in the attenuation image data corresponding to the test image data, and establish an attenuation distribution sub-model that includes the linear attenuation coefficients corresponding to each position where the photons are located;

[0013] Based on the position of the photon in the test image data, as well as the emission probability value and energy absorption probability value of the photon, establish a photon propagator sub-model that includes the energy loss and propagation direction corresponding to the photon at each position;

[0014] Establish a detector sub-model that includes the geometric spatial distribution of the detector crystals based on the detector device data of the imaging device, where the detector device data includes the detector size, thickness, material, detection efficiency, and crystal spatial arrangement;

[0015] Establish an electron sub-model that includes dead time parameters, baseline parameters, and time resolution parameters based on the pulse signals generated when photons hit the detector crystals.

[0016] Further, establishing the detector sub-model that includes the geometric spatial distribution of the detector crystals based on the detector device data of the imaging device includes:

[0017] During the process of establishing the detector sub-model, use the reciprocal of the crystal intrinsic efficiency in the normalization factor to configure the estimated value of the detection efficiency of each detector crystal in the detector sub-model, so that the detector sub-model can detect the pulse signal of the photon hitting the detector crystal.

[0018] Further, establishing the electron sub-model that includes dead time parameters, baseline parameters, and time resolution parameters based on the pulse signals generated when photons hit the detector crystals includes:

[0019] Obtain the radionuclide activity information corresponding to multiple time points;

[0020] Determine the dead time parameter based on the single - event counting rate of the nuclide activity information, and determine the baseline parameter based on the single - event energy loss of the nuclide activity information;

[0021] Determine the reference time resolution based on the single - event counting rate, and perform Gaussian distribution processing based on the reference resolution to determine the deviation corresponding to the random variable;

[0022] Adjust the reference time resolution based on the deviation to obtain the time resolution parameter, so as to establish the electronic sub - model for determining single events and coincidence events based on the dead time parameter, the baseline parameter, and the time resolution parameter.

[0023] Furthermore, the obtaining of the test image data of the scanned target object and the simulation of the operation process of the imaging device by combining the system model and the test image data, and the obtained simulation results include:

[0024] Obtain the test image data of the scanned object;

[0025] Perform simulation processing on the test image data through the nuclide distribution sub - model, determine the pixel points corresponding to the nuclides in the test image data, and determine the positions of the photons as the pixel points;

[0026] Perform simulation processing on the positions through the attenuation distribution sub - model to determine the linear attenuation coefficient of the photons at the positions;

[0027] Perform simulation processing on the positions of the photons through the photon propagator model to determine the propagation paths of the photons. Among them, in the process of processing the positions of the photons by the photon propagator model, determine the probability of particle interaction, energy loss, and propagation direction of the photons through the linear attenuation coefficient to simulate the propagation paths;

[0028] Perform simulation processing on the propagation paths of the photons through the detector sub - model to determine the pulse signals generated when the photons hit the detector crystal;

[0029] Perform simulation processing on the pulse signals based on the dead time parameter, baseline parameter, and time resolution parameter in the electronic sub - model to obtain simulation results, and the simulation results include single events and coincidence events.

[0030] Furthermore, the determining of the state parameters of the imaging device based on the simulation results and the collected device data includes:

[0031] Obtain the first single event and the first coincidence event in the simulation results;

[0032] The state parameters of the first single event, the first conforming event and the device data that match are determined according to the time dimension, the space dimension, the energy dimension and the counting dimension. The device data includes the second single event and the second conforming event. The state parameters include the time dimension state information, the space dimension state information, the energy dimension state information and the counting dimension state information.

[0033] Further, determining the state monitoring result of the imaging device based on the state parameter includes:

[0034] If the state parameter matches the preset state threshold, it is determined that the state monitoring result of the imaging device is a normal monitoring state; or,

[0035] The state parameters and the test image data are classified and labeled based on the trained neural network model to obtain a state monitoring result marked as a normal monitoring state or an abnormal monitoring state, wherein the neural network model is trained based on the labeled state parameters and test image data as sample data.

[0036] According to another aspect of the present invention, there is provided a state monitoring device for an imaging device, comprising:

[0037] An establishing module, used for establishing a system model of an imaging device, wherein the system model comprises a plurality of sub-models for simulating the complete imaging process of the imaging device;

[0038] An acquisition module is used to acquire test image data of a scanned target object, and simulate the operation process of the imaging device in combination with the system model and the test image data to obtain a simulation result, wherein the simulation process is executed in sequence according to the input-output coupling relationship between each sub-model;

[0039] A first determination module, used to determine the state parameters of the imaging device based on the simulation results and the collected device data, wherein the state parameters are used to characterize the state evaluation content of the imaging device in different dimensions;

[0040] The second determination module is used to determine the status monitoring result of the imaging device based on the status parameter.

[0041] Furthermore, the system model includes a nuclide distribution sub-model, an attenuation distribution sub-model, a photon propagation sub-model, a detector sub-model and an electron sub-model, and the establishment module includes:

[0042] A first establishing unit is used to establish a nuclide distribution sub-model characterizing the spatial distribution of nuclide activity based on the spatial position of voxels in the test image data;

[0043] A second establishing unit, configured to determine the linear attenuation coefficients of each position in the test image data based on the voxel spatial positions in the attenuation image data corresponding to the test image data, and establish an attenuation distribution sub-model including the linear attenuation coefficients corresponding to each position where the photons are located;

[0044] A third establishing unit, configured to establish a photon propagator model including the energy loss and propagation direction corresponding to the photons at each position based on the positions of the photons in the test image data, as well as the emission probability value and energy absorption probability value of the photons;

[0045] A fourth establishing unit, configured to establish a detector sub-model including the geometric spatial distribution of detector crystals based on the detector device data of the imaging device, where the detector device data includes detector size, thickness, material, detection efficiency, and crystal spatial arrangement;

[0046] A fifth establishing unit, configured to establish an electronic sub-model including dead time parameters, baseline parameters, and time resolution parameters based on the pulse signals generated when photons hit detector crystals.

[0047] Further, the fourth establishing unit is specifically configured to, in the process of establishing the detector sub-model, configure the estimated values of the detection efficiencies of the detector crystals in the detector sub-model by using the reciprocal of the crystal solid efficiency in the regularization factor, so that the detector sub-model detects the pulse signals of the photons hitting the detector crystals.

[0048] Further, the fifth establishing unit is specifically configured to obtain the nuclide activity information corresponding to multiple time points; determine the dead time parameter based on the single-event counting rate of the nuclide activity information, and determine the baseline parameter based on the single-event energy loss of the nuclide activity information; determine the reference time resolution based on the single-event counting rate, and perform Gaussian distribution processing based on the reference resolution to determine the deviation corresponding to the random variable; adjust the reference time resolution based on the deviation to obtain the time resolution parameter, so as to establish the electronic sub-model for determining single events and coincidence events based on the dead time parameter, the baseline parameter, and the time resolution parameter.

[0049] Further, the obtaining module includes:

[0050] A first obtaining unit, configured to obtain test image data obtained by scanning a scanned object;

[0051] A first processing unit, configured to perform simulation processing on the test image data through the nuclide distribution sub-model, determine the pixel points corresponding to the nuclides in the test image data, and determine the pixel points as the positions of the photons;

[0052] A second processing unit for performing simulation processing on the position based on the attenuation distribution sub-model to determine the linear attenuation coefficient of the photon at the position;

[0053] A third processing unit for performing simulation processing on the position of the photon through the photon propagator model to determine the propagation path of the photon. In the process of processing the position of the photon by the photon propagator model, the probability of particle interaction, energy loss, and propagation direction of the photon are determined through the linear attenuation coefficient to simulate the propagation path;

[0054] A fourth processing unit for performing simulation processing on the propagation path of the photon based on the detector sub-model to determine the pulse signal generated when the photon hits the detector crystal;

[0055] A fifth processing unit for performing simulation processing on the pulse signal based on the dead time parameter, baseline parameter, and time resolution parameter in the electron sub-model to obtain a simulation result, where the simulation result includes single events and coincidence events.

[0056] Further, the first determination module includes:

[0057] A second acquisition unit for acquiring the first single event and the first coincidence event in the simulation result;

[0058] A first determination unit for determining the state parameters of the first single event, the first coincidence event, and the device data matching according to the time dimension, space dimension, energy dimension, and count dimension. The device data includes a second single event and a second coincidence event, and the state parameters include time dimension state information, space dimension state information, energy dimension state information, and count dimension state information.

[0059] Further, the second determination module includes:

[0060] A third determination unit for determining that the state monitoring result of the imaging device is in a normal monitoring state if the state parameter matches a preset state threshold; or,

[0061] A sixth processing unit for performing classification and marking processing on the state parameter and the test image data based on a trained neural network model to obtain a state monitoring result of a marked normal monitoring state or an abnormal monitoring state, where the neural network model is trained based on the marked state parameter and test image data as sample data.

[0062] According to another aspect of the present invention, a storage medium is provided, wherein at least one executable instruction is stored in the storage medium, and the executable instruction enables a processor to execute operations corresponding to the above-mentioned method for monitoring the status of an imaging device.

[0063] According to another aspect of the present invention, there is provided a terminal, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0064] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the above-mentioned imaging device status monitoring method.

[0065] By means of the above technical solution, the technical solution provided by the embodiment of the present invention has at least the following advantages:

[0066] The present invention provides a state monitoring method and device for an imaging device. Compared with the prior art, the embodiment of the present invention establishes a system model of the imaging device, wherein the system model includes a plurality of sub-models for simulating the complete imaging process of the imaging device; obtains test image data of a scanned target object, and simulates the operation process of the imaging device in combination with the system model and the test image data to obtain a simulation result, wherein the simulation process is executed in sequence in linkage according to the input-output coupling relationship between the sub-models; determines the state parameters of the imaging device based on the simulation result and the collected device data, wherein the state parameters are used to characterize the state evaluation content of the imaging device in different dimensions; determines the state monitoring result of the imaging device based on the state parameters, thereby meeting the state monitoring requirements of the PET device during or before operation, and meeting the consistency requirements of the state monitoring result and the operation state of the PET device, avoiding the situation where there are abnormalities during operation and the monitoring cannot be performed, thereby greatly improving the state monitoring effect of the imaging device.

[0067] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0069] Figure 1 The flowchart of a method for monitoring the state of an imaging device provided by an embodiment of the present invention is shown;

[0070] Figure 2 The schematic diagram of the connection structure between each sub-model provided by an embodiment of the present invention is shown;

[0071] Figure 3 The flowchart of another method for monitoring the state of an imaging device provided by an embodiment of the present invention is shown;

[0072] Figure 4 The schematic diagram of the detector annular structure provided by an embodiment of the present invention is shown;

[0073] Figure 5 The flowchart of a method for photon propagation path and detection provided by an embodiment of the present invention is shown;

[0074] Figure 6 The schematic diagram showing the linear relationship between the single-event counting rate and the radionuclide activity information provided by an embodiment of the present invention is shown;

[0075] Figure 7 The schematic diagram of a monitoring pulse signal provided by an embodiment of the present invention is shown;

[0076] Figure 8 The schematic diagram of another monitoring pulse signal provided by an embodiment of the present invention is shown;

[0077] Figure 9 The flowchart of yet another method for monitoring the state of an imaging device provided by an embodiment of the present invention is shown;

[0078] Figure 10 The schematic diagram of the neural network model training provided by an embodiment of the present invention is shown;

[0079] Figure 11 The block diagram of the composition of a state monitoring device for an imaging device provided by an embodiment of the present invention is shown;

[0080] Figure 12 The schematic diagram of the structure of a terminal provided by an embodiment of the present invention is shown. Detailed implementation manners

[0081] Hereinafter, the exemplary embodiments of the present disclosure will be described in more detail with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be completely conveyed to those skilled in the art.

[0082] The status monitoring of PET devices is usually completed by means of pre-scan monitoring, that is, the PET device is started to simulate scanning and status monitoring before use. However, this pre-scan monitoring can only ensure whether the PET device is normal at the moment of simulated scanning, and cannot determine whether the PET device is operating normally when scanning patients clinically. Embodiments of the present invention provide a method for monitoring the status of imaging devices, as Figure 1 shown, the method includes:

[0083] 101. Establish a system model of the imaging device.

[0084] In the embodiments of the present invention, since the PET device needs to perform accurate status monitoring in order to accurately scan the human body image, a system model of the imaging device including multiple sub-models that simulate and simulate the complete imaging process of the imaging device is established. Among them, the system model, as a model for simulating the PET device, may include a radionuclide distribution sub-model, an attenuation distribution sub-model, a photon propagation sub-model, a detector sub-model, and an electron sub-model, so as to simulate and simulate the complete scanning process of the PET device and realize the status detection of the PET device.

[0085] It should be noted that since positron emission tomography scans the human body by injecting a radioactive nuclide into the human body to detect the time, space, and energy information of the γ photon pairs emitted by the annihilation of the positrons emitted during the decay of the nuclide in the human body, and then estimates the distribution of the nuclide in the human body. Therefore, in the embodiments of the present invention, when establishing the system model of the imaging device, each sub-model is constructed separately according to the radionuclide distribution, attenuation distribution, photon propagation, detector, and electron influence parts, that is, the radionuclide distribution sub-model, the attenuation distribution sub-model, the photon propagation sub-model, the detector sub-model, and the electron sub-model. Among them, the radionuclide distribution sub-model is used to describe the distribution of radionuclides in the image scanned by the PET device, the attenuation distribution sub-model is used to describe the linear attenuation coefficient of radionuclides at each position in the image scanned by the PET device, the photon propagation sub-model is used to describe the energy and propagation direction of γ photons, the detector sub-model is used to describe the distribution of each detector crystal in the detector, and the electron sub-model is used to describe the time and energy of each pulse signal obtained when photons hit the detector crystal, so as to simulate the complete scanning process of the PET device based on the radionuclide distribution sub-model, the attenuation distribution sub-model, the photon propagation sub-model, the detector sub-model, and the electron sub-model.

[0086] 102. Obtain the test image data of the scanned target object, and simulate and simulate the operation process of the imaging device in combination with the system model and the test image data to obtain a simulation result.

[0087] In the embodiment of the present invention, the target object can be a human body being scanned by a PET device, or a phantom pre-selected for scanning by the PET device, and test image data of the human body or the phantom is obtained. At this time, since the test image data obtained by scanning the PET device is based on the radionuclide injection into the human body or the phantom, the obtained test image data contains the radionuclide distribution. In addition, since the system model includes a radionuclide distribution sub-model, an attenuation distribution sub-model, a photon propagation sub-model, a detector sub-model, and an electronic sub-model, in order to simulate the operation process of the PET device, that is, the process of γ-photon emission and propagation, so as to obtain the spatial distribution of the radionuclide and the spatial distribution of the material as the state monitoring object affecting the device, a data transmission relationship is established between the various sub-models, such as Figure 2 As shown, in order to complete the simulation process, data flow is performed according to the data transmission relationship between each sub-model in the simulation process. Wherein, the simulation process is executed in sequence according to the input-output coupling relationship between each sub-model. Since the system model includes a nuclide distribution sub-model, an attenuation distribution sub-model, a photon propagation sub-model, a detector sub-model and an electronic sub-model, in the simulation process, there is a coupling relationship between the input and output of each sub-model, that is, the output of the previous sub-model is used as the input of the next sub-model, and the linkage operation of all sub-models is completed in sequence, and the execution of the simulation of all sub-models is completed.

[0088] Specifically, the test image data is first processed as the input parameter of the nuclide distribution sub-model to obtain the pixel point of the nuclide distribution in the voxel space, that is, the position of this pixel point is determined as the position of the gamma photon generated by the annihilation of the positron emitted when the nuclide decays in the human body, which is expressed as a photon pair. Then the position of the gamma photon is used as the input parameter of the attenuation distribution sub-model to determine the linear attenuation coefficient of the gamma photon at this position. The linear attenuation coefficient of the gamma photon at different positions and the position of the gamma photon are used as input parameters of the photon propagation sub-model to simulate the photon propagation and determine the propagation path of the gamma photon. Then the propagation path of the gamma photon is used as the input parameter of the detector sub-model to simulate the gamma photon hitting the detector crystal. The pulse signal generated by the detector crystal that is finally hit is used as the input parameter of the electronic sub-model for simulation to obtain the simulation result of the PET device, that is, the various times and energies of the pulse signal. At this time, the various times and energies of the pulse signal can be characterized as the time and energy of the gamma photon detected in the preset time window and energy window, and described as a single event and a coincidence event.

[0089] 103. Determine the state parameters of the imaging device based on the simulation result and the collected device data.

[0090] In the embodiments of the present invention, since the simulation results include the time and energy at which γ photons are detected within a preset time window and energy window, in order to determine the state of the imaging device PET, state parameters corresponding to the simulation results are determined, that is, the state parameters are used to characterize the state evaluation content of the imaging device in different dimensions. Among them, different dimensions include the time dimension, space dimension, energy dimension, and count dimension, and thus the obtained state parameters include time dimension state information, space dimension state information, energy dimension state information, and count dimension state information.

[0091] 104. Determine the state monitoring result of the imaging device based on the state parameters.

[0092] In the embodiments of the present invention, in order to accurately monitor the state of the imaging device, the state parameters including time dimension state information, space dimension state information, energy dimension state information, and count dimension state information are judged to see if they conform to the normal monitoring state, so as to determine the state monitoring result. Among them, the state monitoring result includes a normal monitoring state and an abnormal monitoring state.

[0093] In another embodiment of the present invention, for further limitation and explanation, as Figure 3 shown, step 101 of establishing the system model of the imaging device includes:

[0094] 1011. Establish a radionuclide distribution sub-model representing the spatial distribution of radionuclide activities based on the voxel spatial positions in the test image data;

[0095] 1012. Determine the linear attenuation coefficient at each position in the test image data based on the voxel spatial positions in the attenuation image data corresponding to the test image data, and establish an attenuation distribution sub-model including the linear attenuation coefficients corresponding to each position where the photons are located;

[0096] 1013. Based on the position of the photon in the test image data, as well as the emission probability value and energy absorption probability value of the photon, establish a photon propagation sub-model including the energy loss and propagation direction corresponding to the photon at each position;

[0097] 1014. Establish a detector sub-model including the geometric spatial distribution of detector crystals based on the detector device data of the imaging device;

[0098] 1015. Establish an electronic sub-model including dead time parameters, baseline parameters, and time resolution parameters based on the pulse signals generated when photons hit the detector crystals.

[0099] In order to simulate the complete processing process of the PET device and thus conduct comparative monitoring on the state of the PET device, a radionuclide distribution sub-model, an attenuation distribution sub-model, a photon propagation sub-model, a detector sub-model, and an electronic sub-model are established respectively.

[0100] For the nuclide distribution sub-model, a model representing the nuclide activity distribution is specifically established by testing the voxel spatial positions in the image data. Among them, since the test image data scanned by the PET device can truly reflect the distribution of nuclides, and the voxel spatial positions are constructed for the test image data, the pixel points where the nuclides are located in the voxel spatial positions can be determined. Since the emission probability of nuclide radioactive decay is proportional to the pixel value at the voxel spatial position where the nuclide is located, therefore, the position corresponding to the determined pixel point is the position where photons are generated, that is, it can be expressed as a pair of γ photons. In addition, the voxel space is the space corresponding to the divided grid image with a fixed resolution. In the voxel space, each pixel point is represented in the form of a grid, so that the position where the nuclide radioactive decay emits a pair of γ photons can be determined.

[0101] For the attenuation distribution sub-model, specifically, the attenuation image data corresponding to the test image data scanned by the PET device is obtained, and the linear attenuation coefficient at each position in the test image data is determined according to the voxel spatial positions in this attenuation image data, so as to establish an attenuation distribution sub-model of the linear attenuation coefficient of photons at each position. Among them, since photons will experience energy attenuation in different directions during propagation, so that the photons move and propagate in the direction with the least energy attenuation during the propagation process. Therefore, the linear attenuation coefficient of photons at each position can be determined based on the attenuation image data, and the propagation path of photons can be determined according to the continuously obtained positions. At the same time, the linear attenuation coefficient is proportional to the probability value of particle interaction per unit distance of γ photon propagation, and the embodiments of the present invention do not make specific limitations.

[0102] For the photon propagation sub-model, a photon propagation sub-model is specifically established by the position of the photon in the test image data, as well as the emission action probability value and energy absorption probability value of the photon. Among them, since photons will undergo Compton scattering, Rayleigh scattering, and photoelectric absorption during propagation, the energy absorption probability values of Compton scattering, Rayleigh scattering, and photoelectric absorption can be found based on the photon energy table. At the same time, combining the linear attenuation coefficient of photons at each position can determine the emission action probability value of photons, and a photon propagation sub-model including the energy loss and propagation direction corresponding to photons at each position is obtained by physical simulation with this probability value. Among them, the annihilation position of positrons can be randomly generated according to the probability distribution by the nuclide distribution sub-model, and the propagation direction of γ photons is isotropic in all directions. Therefore, when determining the propagation direction, the emission action probability value and energy absorption probability value are calculated in any direction of the annihilation position randomly generated in combination with the probability distribution, so as to obtain the photon propagation sub-model.

[0103] For the detector sub-model, a detector sub-model including the geometric distribution of all detector crystals is established specifically through detector device data including detector size, thickness, material, detection efficiency, and crystal spatial arrangement. Among them, the detector device data includes detector size, thickness, material, detection efficiency, and crystal spatial arrangement. The detector size, thickness, and crystal spatial arrangement are used to characterize the spatial geometric information of the detector, and the material is used to characterize the detector density and atomic sequence information. The detector efficiency is the calculated intrinsic efficiency of each crystal, that is, an intrinsic efficiency attribute can be set for each crystal as the detector efficiency. The detector efficiency is a probability value between 0 and 1. As Figure 4 shown, when establishing the detector sub-model as a geometric model, the detector can be a cylindrical detector ring including an inner edge and an outer edge, including a radial direction and an axial direction. A plurality of detector crystals are distributed on the ring edge according to the crystal spatial arrangement method in the radial direction. The axial direction is the direction of the cylindrical axis. At the same time, several detector crystals are also arranged along the axial direction. The embodiments of the present invention do not make specific limitations. In order to determine the detector crystal hit by the photon passing through the propagation path in the detector, a coordinate system is established in the detector sub-model, and the origin is the detector center, so as to simulate and obtain the hit detector crystal in combination with the propagation path of the photon.

[0104] For the electron sub-model, an electron sub-model is established specifically through the pulse signal generated when the photon hits the detector crystal. Among them, since the simulation results obtained by the electron sub-model include single events and coincidence events describing various times and energies of the pulse signal, the electron sub-model includes dead time parameters, baseline parameters, and time resolution parameters describing the pulse pile-up effect, so as to determine single events and coincidence times based on the dead time parameters, baseline parameters, and time resolution parameters. Among them, a single event is used to characterize a pulse signal. At the same time, in order to describe the time and energy of the generated pulse signal, that is, the time and energy when the photon is detected by the detector, by judging whether the time difference and energy difference between two pulse signals meet the set time window and energy window, that is, the time difference and energy difference between two photons (two single events) in the photon pair reaching the detector meet the set time window and energy window, it is characterized as a coincidence event. Of course, if the number of single events judged within a time window or energy window is greater than two, it is defined as a multi-coincidence event, that is, relative to a coincidence event with two single events within a time window and energy window, a multi-coincidence event is that the number of single events satisfying a time window and energy window is greater than two.

[0105] In addition, in the embodiments of the present invention, the photon propagator model is used to simulate the photon propagation path. Among them, when the photon moves at each position, the physical model is used to simulate the photon movement direction and energy. Specifically, if there is a γ photon with an energy of α (in units of mc2, the corresponding value for 511 keV is 1) traveling in a free electron cloud, then the γ photon has a probability of g(x,α)dxdl of colliding within a distance of dl, and its energy becomes α': where cosθ = 1 / α = x / α + 1 is the cosine value of the scattering angle, α0 = 2.81833×10 -13 cm is the classical radius of the electron, n is the number of electrons per cubic centimeter, and α' = xα is the energy of the γ photon after the collision. In addition, when σ(α) / n is the scattering cross section. The probability of any scattering collision of the γ ray within a distance of dl is σ(α)dl. Thus, g(α,x) = nσ(α)f(α,x), where f(α,x) is the probability density distribution of the γ ray with an energy of α that undergoes a collision and the energy ratio coefficient before and after the collision is x, and K(α) is the normalization coefficient. In the embodiments of the present invention, the propagation direction can be determined by random sampling in this probability distribution. At this time, the position of the new γ photon is equal to the position of the previous step plus the propagation direction of the previous position multiplied by the step size, as Figure 5 shown, so as to simulate the complete propagation path of the photon and the process of being detected by the detector crystal.

[0106] In another embodiment of the present invention, for further limitation and explanation, the step of establishing a detector sub-model including the geometric spatial distribution of detector crystals based on the detector device data of the imaging device includes: during the process of establishing the detector sub-model, the reciprocal of the crystal intrinsic efficiency in the regularization factor is used to configure the estimated value of the detector efficiency of each detector crystal in the detector sub-model, so that the detector sub-model detects the pulse signal of the photon hitting the detector crystal.

[0107] Since the detection effect of the PET device is affected by changes in external environmental factors, during the establishment of the detector sub-model, the estimated value of the detector efficiency of each detector crystal in the detector sub-model is configured using the reciprocal of the crystal intrinsic efficiency in the normalization factor. Since each detector efficiency is e (0 < e < 1), whether a pulse signal generated by a photon hitting a detector crystal is detected is determined by a random number x that follows a uniform distribution from 0 to 1. If x < e, the pulse signal is detected as a single event; otherwise, this single event is lost. Therefore, in order to improve the detector efficiency, the reciprocal of the crystal intrinsic efficiency in the normalization factor of the PET device is configured as the estimated value of the detector efficiency of each detector crystal in the detector sub-model, thereby greatly increasing the probability that the pulse signal is detected as a single event, and thus improving the simulation effect of the detector sub-model.

[0108] In another embodiment of the present invention, for further limitation and illustration, the step of establishing an electronic sub-model including dead time parameters, baseline parameters, and time resolution parameters based on the pulse signal generated by a photon hitting a detector crystal includes: obtaining the nuclide activity information corresponding to multiple time points; determining the dead time parameter based on the single event counting rate of the nuclide activity information, and determining the baseline parameter based on the single event energy loss of the nuclide activity information; determining a reference time resolution based on the single event counting rate, and performing Gaussian distribution processing based on the reference resolution to determine the deviation corresponding to the random variable; adjusting the reference time resolution based on the deviation to obtain the time resolution parameter, so as to establish the electronic sub-model for determining single events and coincidence events based on the dead time parameter, the baseline parameter, and the time resolution parameter.

[0109] Since the dead time parameter, baseline parameter, and time resolution parameter are greatly affected by factors such as temperature, humidity, and radionuclide activity, in order to improve the simulation effect of the electronic sub-model, during the establishment of the electronic sub-model, the dead time parameter, baseline parameter, and time resolution parameter are optimized respectively. Specifically, by injecting nuclide F18 with an activity concentration of a into a cylindrical hollow phantom, so that when the nuclide activity decays over time, the nuclide activity information at multiple time points is collected to determine the single event counting rate and single event energy loss corresponding to the nuclide activity information. Among them, since the single event counting rate has a linear relationship with the nuclide activity information, as Figure 6 shown. Among them, as Figure 7 shown, the PET system determines the arrival of a single event by detecting a photon. As Figure 8 shown, when the PET system is in the detection process, when a single event arrives, no second single event will be recorded again within the time τ of detecting the pulse signal, resulting in the omission of the second single event. At this time, the time τ is the dead time.

[0110] In the embodiments of the present invention, the dead time parameter is determined by the single event counting rate, that is, calculated through a dead time estimation function, and the dead time estimation function is: where n i and m i respectively represent the ideal single event counting rate and the actual collected single event counting rate of the nuclide activity information corresponding to the i-th acquisition.

[0111] Meanwhile, when collecting the nuclide activity information, the energy value of each single event is statistically analyzed to generate a local energy distribution map, such as the energy distribution map of the detector crystal, and the peak position of each energy distribution map is calculated. The deviation △E between the calculated peak position and the preset peak position of the PET device is calculated, and the energy value for calculating the baseline parameter is compensated based on the △E deviation, so as to ensure that the baseline parameter in the electronic sub-model is more consistent with the actual baseline parameter of the PET device. For example, compensation is performed by an addition method. where A(t) is the nuclide activity information at time t, and the embodiments of the present invention do not make specific limitations.

[0112] In addition, for the time resolution parameter, specifically including: 1. Determine the reference time resolution T according to the determined single event counting rate res ; 2. Generate a Gaussian distribution function G = N(0,σ), where σ = T res / 2.3548; 3. Generate a random variable p uniformly distributed in the interval [-3σ, 3σ]; 4. Generate a random variable q uniformly distributed in the interval; 5. If G(p) < q, then use p as the deviation caused by the time resolution, and adjust the reference time resolution based on the deviation to obtain the time resolution parameter, otherwise return to step 3.

[0113] It should be noted that in order to simulate the change of nuclide activity, the activity of F18 decays with time. The nuclide activity information at several time points is collected, and the nuclide activity concentration ai in each time period i is calculated, so as to estimate the dead time parameter, baseline parameter, and time resolution parameter under each activity concentration for fitting, so as to obtain the dead time parameter, baseline parameter, and time resolution parameter of any activity in the interval (0, a]. Among them, the formula where A is the average activity starting from time T and scanning for a duration of T acq , A cal and T cal are the starting activity and time for monitoring respectively, and T 1 / 2 is the nuclide half-life.

[0114] In another embodiment of the present invention, for further limitation and illustration, such as Figure 9As shown, in step 102, test image data of the scanned target object is obtained, and the operation process of the imaging device is simulated and emulated by combining the system model and the test image data. The simulation and emulation results include:

[0115] 1021. Obtain test image data of the scanned object;

[0116] 1022. Perform simulation and emulation processing on the test image data through the radionuclide distribution sub-model, determine the pixel points corresponding to the radionuclides in the test image data, and determine the positions of the photons as the pixel points;

[0117] 1023. Perform simulation and emulation processing on the positions through the attenuation distribution sub-model, and determine the linear attenuation coefficient of the photons at the positions;

[0118] 1024. Perform simulation and emulation processing on the positions of the photons through the photon propagator model to determine the propagation paths of the photons. During the processing of the positions of the photons by the photon propagator model, the probability of particle interaction, energy loss, and propagation direction of the photons are determined through the linear attenuation coefficient to simulate the propagation paths;

[0119] 1025. Perform simulation and emulation processing on the propagation paths of the photons based on the detector sub-model, and determine the pulse signals generated when the photons hit the detector crystal;

[0120] 1026. Perform simulation and emulation processing on the pulse signals based on the dead time parameter, baseline parameter, and time resolution parameter in the electron sub-model to obtain the simulation and emulation results.

[0121] In the embodiments of the present invention, when simulating and emulating test image data based on a system model, a radionuclide is injected into a human body or a phantom, and scanning is performed by devices including but not limited to PET devices, photon sensors, etc. to obtain the scanned test image data. In this process, it is applicable to the normal operation scenario of imaging devices and also to the test scenario before the imaging devices operate. The embodiments of the present invention do not make specific limitations. After obtaining the test image data, the test image data is used as an input parameter of the radionuclide distribution sub-model for simulation and emulation processing to determine the pixel points corresponding to the radionuclide in the test image data. Since this pixel point is the position of the pair of γ photons generated by the annihilation of the positrons emitted during the decay of the radionuclide in the human body, the pose of this pixel point is determined as the position of the photon. At the same time, the position of this γ photon is used as an input parameter of the attenuation distribution sub-model for simulation and emulation processing to determine the linear attenuation coefficient of the γ photon at this position. Then, the linear attenuation coefficient combined with the position of the γ photon pair is used as an input parameter of the photon propagator sub-model for simulation and emulation processing, so as to determine the probability of photon-particle interaction, energy loss, and propagation direction of the photon through the linear attenuation coefficient. Among them, the probability of photon-particle interaction is the probability value of possible photon-particle interaction at this position, which can be generated according to a random probability distribution based on the linear attenuation coefficient. At the same time, when the photon propagator sub-model is simulating and emulating, the position of the photon is used to simulate and emulate the probability of photon-particle interaction, energy loss, and propagation direction of the photon based on the photon propagation model to obtain the next position. At this time, by re-obtaining the linear attenuation coefficient of the next position for iterative simulation and emulation, the probability of photon-particle interaction, energy loss, and propagation direction of the γ photon at the next position are determined, and so on, to obtain multiple positions of the γ photon, thereby simulating and emulating the propagation path. The propagation path of the γ photon is used as an input parameter of the detector sub-model for simulation and emulation processing to determine the pulse signal generated when the γ photon hits the detector crystal. At this time, since a pair of γ photons can generate two pulse signals. Finally, the pulse signal is used as an input parameter of the electron sub-model for simulation and emulation processing under the calculation of dead time parameters, baseline parameters, and time resolution parameters to obtain the simulation and emulation results including single events and coincidence events.

[0122] In another embodiment of the present invention, for further limitation and illustration, step 103 of determining the state parameters of the imaging device based on the simulation and emulation results and the collected device data includes: obtaining the first single event and the first coincidence event in the simulation and emulation results; determining the state parameters of the first single event, the first coincidence event, and the device data matching in terms of time dimension, space dimension, energy dimension, and count dimension.

[0123] In order to accurately monitor the status of the imaging device PET and achieve matching with the real-time status of the PET device, before determining the status parameters of the imaging device, the device data of the PET device running in real time is collected. At this time, the device data and the simulation results are the same object, that is, the device data also includes single events and coincidence events. Thus, the time dimension status information, space dimension status information, energy dimension status information, and count dimension status information of the matching between the first single event, the first coincidence event and the second single event, the second coincidence event are determined according to the time dimension, space dimension, energy dimension, and count dimension. That is, the difference is calculated for the status parameters corresponding to the first single event, the first coincidence event and the second single event, the second coincidence event according to the four dimensions, and thus used as the status parameters for final monitoring. Among them, the time dimension is used to describe the time difference of coincidence events corresponding to a preset time interval, and the corresponding time dimension status information is the distribution of the time difference of coincidence events within this preset time interval, which can be represented in the form of a matrix, histogram, etc. Since two single events form a coincidence event and each detector crystal is arranged on the detector according to the spatial layout, the space dimension is used to describe the number of detector crystals participating as single events. At this time, the space dimension status information is to characterize the radial ring detector crystal C AD and the axial ring detector crystal C RD The total number of single events detected at a preset time interval, represented in the form of a C AD *C RD matrix. The energy dimension is used to describe the energy distribution of the detector crystals. Since the energy distribution of the detector crystals approximates a Gaussian distribution, at this time, the peak position and full width at half maximum of the energy distribution of each detector crystal can be calculated. Therefore, the energy dimension status information is the peak position and full width at half maximum of the energy distribution when each detector crystal detects photons within a preset time interval. The count dimension is used to describe the counting of coincidence events. Since the position of positron annihilation is within the scanned human body or phantom, at this time, the projection of the attenuation distribution map is used to determine the area outside the object, which is completely composed of random coincidences and scatter coincidences at this time. Therefore, the count dimension status information is the percentage of the coincidence events of the PET device to the random coincidence events, scatter coincidence events and true coincidence events, which can be represented by an array of length 4. Among them, the first element is the count of the coincidence events of the PET device, and the last three are the percentages of the random coincidence events, scatter coincidence events and true coincidence events respectively. The embodiments of the present invention do not make specific limitations.

[0124] It should be noted that in the embodiments of the present invention, the first single event and the first coincidence event obtained by simulation are compared with the second single event and the second coincidence event collected from the PET device in terms of time dimension, space dimension, energy dimension, and counting dimension to find differences, so as to obtain the state information of the time dimension, space dimension, energy dimension, and counting dimension, thereby realizing the state monitoring of the PET device in the time dimension, space dimension, energy dimension, and counting dimension, greatly improving the accuracy of the device state monitoring, and thus realizing the efficiency of the PET device state monitoring.

[0125] In another embodiment of the present invention, for further limitation and explanation, step 104 of determining the state monitoring result of the imaging device based on the state parameter includes: if the state parameter matches the preset state threshold, determining that the state monitoring result of the imaging device is in a normal monitoring state; or, classifying and labeling the state parameter and the test image data based on a trained neural network model to obtain a state monitoring result marked as a normal monitoring state or an abnormal monitoring state, where the neural network model is trained based on the labeled state parameter and test image data as sample data.

[0126] In the embodiments of the present invention, in order to achieve the purpose of automatically monitoring the PET device, the state parameter is judged to determine the state monitoring result. Specifically, the state threshold can be set for the state parameter. If the state parameter matches the preset state threshold, it is determined that the state monitoring result of the imaging device is in a normal monitoring state. It is also possible to classify and label the state parameter and the test image data based on a trained neural network model to obtain a state monitoring result marked as a normal monitoring state or an abnormal monitoring state, where the neural network model is trained based on the labeled state parameter and test image data as sample data. Specifically, since the state parameter is the parameter content of four dimensions, namely the state information of the time dimension, space dimension, energy dimension, and counting dimension, therefore, such as Figure 10As shown, the corresponding four-dimensional indicators and reconstructed images are generated. Each time the parameters of the PET device or the parameters of each sub-model are changed, the four-dimensional indicators and reconstructed images are generated as a sample data. Among them, the reconstructed image needs to be manually judged whether it is qualified, and the main influencing dimension that causes the image to be the most unqualified is marked as the label of the sample data. At the same time, a neural network model is established, with the sample data as input, the qualified label and the main image dimension that causes the unqualified as output, and the neural network model is trained to fit the relationship between the indicators of the four dimensions and the qualified test image data, so as to determine the main imaging factors that cause the unqualified. Among them, the information of each dimension can be used as a parameter in the form of a vector or matrix, so it can be compressed by an autoencoder first to highlight the feature information, and then the results of each autoencoder are connected with the fully connected network to perform feature classification and estimation of the main cause of unqualified, thereby balancing the relationship between multiple dimensions in the monitoring process and improving the accuracy of state monitoring.

[0127] The embodiment of the present invention provides a state monitoring method for an imaging device. Compared with the prior art, the embodiment of the present invention establishes a system model of the imaging device, wherein the system model includes multiple sub-models that simulate the complete imaging process of the imaging device; obtains test image data of a scanned target object, and simulates the operation process of the imaging device in combination with the system model and the test image data to obtain a simulation result, wherein the simulation process is executed in sequence according to the input-output coupling relationship between each sub-model; determines the state parameters of the imaging device based on the simulation result and the collected device data, and the state parameters are used to characterize the state evaluation content of the imaging device in different dimensions; determines the state monitoring result of the imaging device based on the state parameters, thereby meeting the state monitoring requirements of the PET device during or before operation, and meeting the consistency requirements of the state monitoring results and the operation state of the PET device, avoiding the situation where there are abnormalities in the operation process and the monitoring cannot be performed, thereby greatly improving the state monitoring effect of the imaging device.

[0128] Furthermore, as a response to the above Figure 1 The embodiment of the present invention provides a device for monitoring the state of an imaging device, such as Figure 11 As shown, the device comprises:

[0129] Establishing module 21, used for establishing a system model of an imaging device, wherein the system model includes a plurality of sub-models simulating the complete imaging process of the imaging device;

[0130] The acquisition module 22 is used to acquire the test image data of the scanned target object, and simulate the operation process of the imaging device in combination with the system model and the test image data to obtain a simulation result, wherein the simulation process is executed in sequence according to the input-output coupling relationship between each sub-model;

[0131] A first determination module 23, used to determine the state parameters of the imaging device based on the simulation results and the collected device data, wherein the state parameters are used to characterize the state evaluation content of the imaging device in different dimensions;

[0132] The second determination module 24 is used to determine the status monitoring result of the imaging device based on the status parameter.

[0133] Furthermore, the system model includes a nuclide distribution sub-model, an attenuation distribution sub-model, a photon propagation sub-model, a detector sub-model and an electron sub-model, and the establishment module includes:

[0134] A first establishing unit is used to establish a nuclide distribution sub-model characterizing the spatial distribution of nuclide activity based on the spatial position of voxels in the test image data;

[0135] A second establishing unit, configured to determine a linear attenuation coefficient at each position in the test image data based on a voxel spatial position in the attenuation image data corresponding to the test image data, and establish an attenuation distribution sub-model including a linear attenuation coefficient corresponding to each position where the photon is located;

[0136] A third establishing unit is used to establish a photon propagation sub-model including the energy loss and propagation direction of the photon at each position based on the position of the photon in the test image data, the emission action probability value and the energy absorption probability value of the photon;

[0137] A fourth establishing unit is used to establish a detector sub-model including a geometric spatial distribution of detector crystals based on the detector device data of the imaging device, wherein the detector device data includes detector size, thickness, material, detection efficiency, and crystal spatial arrangement;

[0138] The fifth establishing unit is used to establish an electronic sub-model including dead time parameters, baseline parameters, and time resolution parameters based on the pulse signal generated by the photon hitting the detector crystal.

[0139] Furthermore, the fourth establishing unit is specifically used to configure the estimated value of the detector efficiency of each detector crystal in the detector sub-model using the inverse of the crystal intrinsic efficiency in the regularization factor during the process of establishing the detector sub-model, so that the detector sub-model can detect the pulse signal of the photon hitting the detector crystal.

[0140] Further, the fifth establishing unit is specifically configured to obtain nuclide activity information corresponding to multiple time points; determine a dead time parameter based on the single-event counting rate of the nuclide activity information, and determine a baseline parameter based on the single-event energy loss of the nuclide activity information; determine a reference time resolution based on the single-event counting rate, and perform Gaussian distribution processing based on the reference resolution to determine the deviation corresponding to the random variable; adjust the reference time resolution based on the deviation to obtain a time resolution parameter, so as to establish the electronic sub-model for determining single events and coincidence events based on the dead time parameter, the baseline parameter, and the time resolution parameter.

[0141] Further, the obtaining module includes:

[0142] The first obtaining unit is configured to obtain test image data obtained by scanning a scanned object;

[0143] The first processing unit is configured to perform simulation processing on the test image data through the nuclide distribution sub-model, determine the pixel points corresponding to the nuclides in the test image data, and determine the positions of the photons as the pixel points;

[0144] The second processing unit is configured to perform simulation processing on the positions through the attenuation distribution sub-model to determine the linear attenuation coefficient of the photons at the positions;

[0145] The third processing unit is configured to perform simulation processing on the positions of the photons through the photon propagator model to determine the propagation paths of the photons. During the processing of the positions of the photons by the photon propagator model, the probability, energy loss, and propagation direction of the photons generating particle interactions are determined through the linear attenuation coefficient to simulate the propagation paths;

[0146] The fourth processing unit is configured to perform simulation processing on the propagation paths of the photons based on the detector sub-model to determine the pulse signals generated when the photons hit the detector crystal;

[0147] The fifth processing unit is configured to perform simulation processing on the pulse signals based on the dead time parameter, the baseline parameter, and the time resolution parameter in the electronic sub-model to obtain a simulation result, where the simulation result includes single events and coincidence events.

[0148] Further, the first determining module includes:

[0149] The second obtaining unit is configured to obtain the first single event and the first coincidence event in the simulation result;

[0150] A first determination unit is used to determine the state parameters of the first single event, the first conforming event and the device data according to the time dimension, the space dimension, the energy dimension and the counting dimension, the device data includes the second single event and the second conforming event, and the state parameters include the time dimension state information, the space dimension state information, the energy dimension state information and the counting dimension state information.

[0151] Furthermore, the second determining module includes:

[0152] a third determining unit, configured to determine that the state monitoring result of the imaging device is a normal monitoring state if the state parameter matches a preset state threshold; or

[0153] The sixth processing unit is used to classify and label the state parameters and the test image data based on the trained neural network model to obtain a state monitoring result marked as a normal monitoring state or an abnormal monitoring state, wherein the neural network model is trained based on the labeled state parameters and test image data as sample data.

[0154] An embodiment of the present invention provides a state monitoring device for an imaging device. Compared with the prior art, the embodiment of the present invention establishes a system model of the imaging device, wherein the system model includes multiple sub-models that simulate the complete imaging process of the imaging device; obtains test image data of a scanned target object, and simulates the operation process of the imaging device in combination with the system model and the test image data to obtain a simulation result, wherein the simulation process is executed in sequence according to the input-output coupling relationship between each sub-model; determines the state parameters of the imaging device based on the simulation result and the collected device data, and the state parameters are used to characterize the state evaluation content of the imaging device in different dimensions; determines the state monitoring result of the imaging device based on the state parameters, thereby meeting the state monitoring requirements of the PET device during or before operation, and meeting the consistency requirements of the state monitoring results and the operation state of the PET device, avoiding the situation where there are abnormalities during operation and the monitoring cannot be performed, thereby greatly improving the state monitoring effect of the imaging device.

[0155] According to an embodiment of the present invention, a storage medium is provided, wherein the storage medium stores at least one executable instruction, and the computer executable instruction can execute the state monitoring method of the imaging device in any of the above method embodiments.

[0156] Figure 12 A schematic diagram of the structure of a terminal provided according to an embodiment of the present invention is shown. The specific embodiment of the present invention does not limit the specific implementation of the terminal.

[0157] like Figure 12As shown in the figure, the terminal may include: a processor 302, a communications interface 304, a memory 306, and a communication bus 308.

[0158] Among them: The processor 302, the communications interface 304, and the memory 306 complete their mutual communication through the communication bus 308.

[0159] The communications interface 304 is used to communicate with network elements of other devices such as clients or other servers.

[0160] The processor 302 is used to execute the program 310, and specifically can execute the relevant steps in the embodiment of the method for monitoring the state of the above imaging device.

[0161] Specifically, the program 310 may include program code, and the program code includes computer operation instructions.

[0162] The processor 302 may be a central processing unit CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present invention. One or more processors included in the terminal may be of the same type of processor, such as one or more CPUs; or may be of different types of processors, such as one or more CPUs and one or more ASICs.

[0163] The memory 306 is used to store the program 310. The memory 306 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory.

[0164] The program 310 is specifically used to cause the processor 302 to perform the following operations:

[0165] Establish a system model of the imaging device, where the system model includes multiple sub-models that simulate and simulate the complete imaging process of the imaging device;

[0166] Obtain test image data of the scanned target object, and combine the system model and the test image data to simulate and simulate the operation process of the imaging device to obtain a simulation result, where the simulation process is sequentially linked and executed according to the input-output coupling relationship between the sub-models;

[0167] Based on the simulation result and the collected device data, determine the state parameters of the imaging device, where the state parameters are used to characterize the state evaluation content of the imaging device in different dimensions;

[0168] Determine the status monitoring result of the imaging device based on the status parameter.

[0169] Obviously, those skilled in the art should understand that the various modules or steps of the present invention described above can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. Optionally, they can be implemented by program codes executable by the computing device. Thus, they can be stored in a storage device and executed by the computing device. And in some cases, the steps shown or described can be executed in a sequence different from that here, or they can be separately fabricated into individual integrated circuit modules, or multiple modules or steps among them can be fabricated into a single integrated circuit module for implementation. In this way, the present invention is not limited to any specific combination of hardware and software.

[0170] The foregoing is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for monitoring the state of an imaging device, characterized in that, include: Establishing a system model of an imaging device, wherein the system model includes a plurality of sub-models simulating the complete imaging process of the imaging device; Acquire test image data of the scanned target object, and simulate the operation process of the imaging device in combination with the system model and the test image data to obtain a simulation result, wherein the simulation process is sequentially executed in linkage according to the input-output coupling relationship between each sub-model; Determine the state parameters of the imaging device based on the simulation results and the collected device data, wherein the state parameters are used to characterize the state evaluation content of the imaging device in different dimensions; Determine a status monitoring result of the imaging device based on the status parameter; The system model includes a nuclide distribution sub-model, an attenuation distribution sub-model, a photon propagation sub-model, a detector sub-model and an electron sub-model, and the system model for establishing the imaging device includes: A nuclide distribution sub-model is established based on the spatial position of voxels in the test image data to characterize the spatial distribution of nuclide activity; Determine the linear attenuation coefficient of each position in the test image data based on the spatial position of the voxel in the attenuation image data corresponding to the test image data, and establish an attenuation distribution sub-model including the linear attenuation coefficient corresponding to each position where the photon is located; Based on the positions of the photons in the test image data, as well as the emission action probability value and energy absorption probability value of the photons, a photon propagation sub-model including the energy loss and propagation direction of the photons at each position is established; Establishing a detector sub-model including the geometric spatial distribution of detector crystals based on the detector device data of the imaging device, wherein the detector device data includes detector size, thickness, material, detection efficiency, and crystal spatial arrangement; Based on the pulse signal generated by photons hitting the detector crystal, an electronic sub-model including dead time parameters, baseline parameters and time resolution parameters is established.

2. The method according to claim 1, characterized in that, The step of establishing a detector sub-model including the geometric spatial distribution of detector crystals based on the detector device data of the imaging device comprises: In the process of establishing the detector sub-model, the estimated value of the detector efficiency of each detector crystal in the detector sub-model is configured using the inverse of the crystal intrinsic efficiency in the regularization factor, so that the detector sub-model can detect the pulse signal of the photon hitting the detector crystal.

3. The method according to claim 1, characterized in that, The electronic sub-model including dead time parameters, baseline parameters, and time resolution parameters is established based on the pulse signal generated by photons hitting the detector crystal, including: Obtain the nuclide activity information corresponding to multiple time points; determining a dead time parameter based on a single event count rate of the nuclide activity information, and determining a baseline parameter based on a single event energy loss of the nuclide activity information; Determining a reference time resolution based on the single event count rate, and performing Gaussian distribution processing based on the reference time resolution to determine a deviation corresponding to a random variable; The reference time resolution is adjusted based on the deviation to obtain a time resolution parameter, so as to establish the electronic sub-model for determining single events and coincident events based on the dead time parameter, the baseline parameter and the time resolution parameter.

4. The method according to claim 1, characterized in that The acquiring of the test image data of the scan target object and simulating the operation process of the imaging device in combination with the system model and the test image data to obtain the simulation result includes: Acquire test image data of scanning the scanning object; The test image data is simulated by the nuclide distribution sub-model to determine the pixel point corresponding to the nuclide in the test image data, and the pixel point is determined as the position of the photon; Performing simulation processing on the position based on the attenuation distribution sub-model to determine the linear attenuation coefficient of the photon at the position; The photon propagation sub-model is used to simulate the position of the photon to determine the propagation path of the photon, wherein, in the process of processing the position of the photon by the photon propagation sub-model, the probability of the photon having a particle effect, the energy loss and the propagation direction are determined by the linear attenuation coefficient to simulate the propagation path; Simulating the propagation path of the photon based on the detector sub-model to determine the pulse signal generated by the photon hitting the detector crystal; The pulse signal is simulated based on the dead time parameter, baseline parameter and time resolution parameter in the electronic sub-model to obtain a simulation result, which includes a single event and a coincidence event.

5. The method according to claim 4, wherein Determining the state parameters of the imaging device based on the simulation results and the collected device data includes: Obtaining a first single event and a first matching event in the simulation result; The state parameters of the first single event, the first conforming event and the device data that match are determined according to the time dimension, the space dimension, the energy dimension and the counting dimension. The device data includes the second single event and the second conforming event. The state parameters include the time dimension state information, the space dimension state information, the energy dimension state information and the counting dimension state information.

6. The method according to any one of claims 1-5, characterized in that, Determining the state monitoring result of the imaging device based on the state parameter includes: If the state parameter matches the preset state threshold, it is determined that the state monitoring result of the imaging device is a normal monitoring state; or, The state parameters and the test image data are classified and labeled based on the trained neural network model to obtain a state monitoring result marked as a normal monitoring state or an abnormal monitoring state, wherein the neural network model is trained based on the labeled state parameters and test image data as sample data.

7. A state monitoring device for an imaging device, characterized in that, include: An establishing module, used for establishing a system model of an imaging device, wherein the system model comprises a plurality of sub-models for simulating the complete imaging process of the imaging device; An acquisition module is used to acquire test image data of a scanned target object, and simulate the operation process of the imaging device in combination with the system model and the test image data to obtain a simulation result, wherein the simulation process is executed in sequence according to the input-output coupling relationship between each sub-model; A first determination module, configured to determine state parameters of the imaging device based on the simulation result and the collected device data, where the state parameters are used to characterize state evaluation contents of the imaging device in different dimensions; A second determination module, configured to determine a state monitoring result of the imaging device based on the state parameters; Wherein, the system model includes a radionuclide distribution sub-model, an attenuation distribution sub-model, a photon propagation sub-model, a detector sub-model, and an electronics sub-model, and the establishment module includes: A first establishment unit, configured to establish a radionuclide distribution sub-model characterizing the spatial distribution of radionuclide activities based on the voxel spatial positions in the test image data; A second establishment unit, configured to determine the linear attenuation coefficient at each position in the test image data based on the voxel spatial positions in the attenuation image data corresponding to the test image data, and establish an attenuation distribution sub-model including the linear attenuation coefficients corresponding to each position where the photons are located; A third establishment unit, configured to establish a photon propagation sub-model including the energy loss and propagation direction corresponding to the photons at each position based on the position of the photons in the test image data, and the emission action probability value and energy absorption probability value of the photons; A fourth establishment unit, configured to establish a detector sub-model including the geometric spatial distribution of detector crystals based on the detector device data of the imaging device, where the detector device data includes detector size, thickness, material, detection efficiency, and crystal spatial arrangement; A fifth establishment unit, configured to establish an electronics sub-model including dead time parameters, baseline parameters, and time resolution parameters based on the pulse signals generated when photons hit the detector crystals.

8. A storage medium storing at least one executable instruction, where the executable instruction causes a processor to perform operations corresponding to the state monitoring method of the imaging device according to any one of claims 1-6.

9. A terminal, comprising: A processor, a memory, a communication interface, and a communication bus, where the processor, the memory, and the communication interface complete communication with each other through the communication bus; The memory is used to store at least one executable instruction, and the executable instruction causes the processor to perform operations corresponding to the state monitoring method of the imaging device according to any one of claims 1-6.

Images