Medical image reconstruction method and PET imaging system

Abstract

The invention relates to a medical image reconstruction method and a PET imaging system. The method comprises the following steps: acquiring empty scanning data and first scanning data of a detection object through a medical scanning device, determining an attenuation coefficient according to the first scanning data and the empty scanning data, performing attenuation correction according to the attenuation coefficient and the first scanning data to obtain correction data, and determining the first scanning data of the detection object according to the correction data and a preset machine learning model. And determining a reconstructed image corresponding to the correction data. By adopting the method, the accuracy of the obtained reconstructed image can be improved.

Description

Technical Field

[0001] The present application relates to the field of medical technology, and in particular to a medical image reconstruction method and a PET imaging system. Background Art

[0002] In the medical field, it is necessary to reconstruct a reconstructed image based on data obtained when a medical scanning device scans the whole body or a certain part of a patient. Taking the medical scanning device as a positron emission tomography (PET) device as an example, the scanning data obtained by the detector can be reconstructed based on a preset reconstruction algorithm to obtain a reconstructed image.

[0003] However, the reconstructed images currently generated suffer from low accuracy. Summary of the invention

[0004] Based on this, it is necessary to provide a medical image reconstruction method and a PET imaging system that can improve accuracy in order to address the above technical issues.

[0005] In a first aspect, the present application provides a method for reconstructing a medical image, comprising:

[0006] Acquire the empty scan data and the first scan data of the detection object by a medical scanning device;

[0007] Determine the attenuation coefficient according to the first scanning data and the empty scanning data;

[0008] Perform attenuation correction according to the attenuation coefficient and the first scanning data to obtain correction data;

[0009] According to the correction data and the preset machine learning model, the reconstructed image corresponding to the correction data is determined.

[0010] In one embodiment, determining the attenuation coefficient according to the first scan data and the empty scan data includes:

[0011] Screening the first scanning data according to the attribute information of the first scanning data to determine the background transmission data in the first scanning data;

[0012] Determine the attenuation coefficient based on the air scan data and background transmission data.

[0013] In one embodiment, determining the attenuation coefficient based on the air scan data and the background transmission data includes:

[0014] Determine first scanning sub-data in the first scanning data that is located in a first time range;

[0015] The first scanning sub-data are screened according to the attribute information of the first scanning sub-data to determine the background transmission data.

[0016] In one embodiment, the method further comprises:

[0017] Screening the first scan data according to the attribute information of the first scan data to determine the second scan data in the first scan data;

[0018] Performing attenuation correction according to the attenuation coefficient and the first scanning data to obtain correction data includes:

[0019] According to the attenuation coefficient, attenuation correction is performed on the second scanning data in the first scanning data to obtain corrected data.

[0020] In one embodiment, performing attenuation correction on the second scan data in the first scan data according to the attenuation coefficient to obtain the corrected data includes:

[0021] Dividing the second scan data to obtain second scan sub-data;

[0022] Attenuation correction is performed on each second scanning sub-data according to the attenuation coefficient to obtain correction data corresponding to each second scanning sub-data.

[0023] In one embodiment, determining a reconstructed image corresponding to the correction data according to the correction data and a preset machine learning model includes:

[0024] Back-projecting the correction data to obtain a back-projection image corresponding to the correction data;

[0025] According to the back-projection image and the machine learning model, a reconstructed image corresponding to the back-projection image is determined.

[0026] In one embodiment, determining the attenuation coefficient based on the air scan data and the background transmission data includes:

[0027] Determine the attenuation coefficient corresponding to each response line in the air scan data and the background transmission data according to the air scan data and the background transmission data;

[0028] Correspondingly, according to the attenuation coefficient, attenuation correction is performed on the second scan data in the first scan data to obtain corrected data, including:

[0029] For each response line, according to the attenuation coefficient corresponding to the response line, attenuation correction is performed on the data corresponding to the response line in the second scanning data to obtain corrected data.

[0030] In one embodiment, the method further comprises:

[0031] Obtaining back-projected image samples and reconstructed image samples corresponding to the back-projected image samples;

[0032] The machine learning model is trained based on the back-projected image samples and the reconstructed image samples.

[0033] In one embodiment, the first scanning data is screened according to the attribute information of the first scanning data to obtain the background transmission data in the first scanning data, including:

[0034] The scanning data whose energy in the first scanning data is not within the preset energy range is used as background transmission data; and / or the scanning data whose flight time in the first scanning data is not within the second time range is used as background transmission data.

[0035] In one embodiment, the first scan data is screened according to the attribute information of the first scan data to obtain the second scan data in the first scan data, including:

[0036] The scanning data whose energy in the first scanning data is within a preset energy range is used as the second scanning data; and / or the scanning data whose flight time in the first scanning data is within a second time range is used as the second scanning data.

[0037] In a second aspect, the present application also provides a medical image reconstruction device, comprising:

[0038] An acquisition module, used for acquiring the air scan data and the first scan data of the detection object through a medical scanning device;

[0039] A first determination module, used for determining an attenuation coefficient according to the first scanning data and the empty scanning data;

[0040] A correction module, used for performing attenuation correction according to the attenuation coefficient and the first scanning data to obtain correction data;

[0041] The second determination module is used to determine the reconstructed image corresponding to the correction data based on the correction data and a preset machine learning model.

[0042] In a third aspect, the present application also provides a PET imaging system, the PET imaging system includes a scanning device and a computer device, the scanning device is connected to the computer device; wherein,

[0043] The computer device includes a memory and a processor, the memory stores a computer program, and the processor executes the computer program to implement the steps of the following method:

[0044] Obtaining empty scan data and TOF-PET imaging data. Empty scan data is acquired when there is no detection object in the scanning cavity of the scanning device; TOF-PET imaging data is acquired when there is a detection object in the scanning cavity of the scanning device;

[0045] Screening TOF-PET imaging data to obtain background transmission data;

[0046] Determine the attenuation coefficient based on background transmission data and air scan data;

[0047] Correcting the TOF-PET imaging data based on the attenuation coefficient, and back-projecting the corrected TOF-PET imaging data to generate a histogram image frame;

[0048] The histogram image frame is reconstructed to obtain a reconstructed image.

[0049] In a fourth aspect, the present application further provides a PET imaging system, the PET imaging system comprising a scanning device and a computer device, wherein the scanning device is connected to the computer device;

[0050] The computer device includes a memory and a processor, the memory stores a computer program, and the processor executes the computer program to implement the steps of the following method:

[0051] Acquire empty scan data and TOF-PET imaging data, wherein the empty scan data is acquired when there is no detection object in the scanning cavity of the scanning device; the TOF-PET imaging data is acquired when there is a detection object in the scanning cavity of the scanning device, and the TOF-PET imaging data is acquired by detecting the detection object in multiple motion phases;

[0052] Screening TOF-PET imaging data in at least one motion phase to obtain background transmission data;

[0053] Determine the attenuation coefficient based on background transmission data and air scan data;

[0054] Correcting TOF-PET imaging data in a plurality of motion phases based on the attenuation coefficient, and back-projecting the corrected TOF-PET imaging data to generate histogram image frames in a plurality of motion phases;

[0055] Histogram image frames within a plurality of motion phases are reconstructed to obtain a dynamically reconstructed image.

[0056] In a fifth aspect, the present application further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any of the above methods when executing the computer program.

[0057] In a sixth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of any of the above methods when executed by a processor.

[0058] In a seventh aspect, the present application also provides a computer program product, comprising a computer program, which implements the steps of any of the above methods when executed by a processor.

[0059] The above-mentioned medical image reconstruction method and PET imaging system obtains the air scan data and the first scan data of the detection object through the medical scanning device. Since the attenuation coefficient can be determined based on the first scan data and the air scan data, the air scan result of the medical scanning device is taken into account in the process of determining the attenuation coefficient, that is, the radiation and response of the medical scanning device itself are taken into account. In this way, after attenuation correction is performed according to the attenuation coefficient and the first scan data, more accurate correction data can be obtained. Furthermore, according to the correction data and the preset machine learning model, the corresponding reconstructed image can be efficiently obtained. On the one hand, compared with the process of determining the reconstructed image in combination with the results of the CT device scan in the related art, this embodiment does not require CT scanning, and the process of obtaining the reconstructed image is faster and more efficient; on the other hand, the radiation and response of the medical scanning device itself are taken into account, which improves the accuracy of the reconstructed image. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0061] FIG1 is a diagram showing an application environment of a medical image reconstruction method according to an embodiment;

[0062] Figure 2 is a schematic flow chart of a medical image reconstruction method in one embodiment;

[0063] Figure 3 A schematic diagram of a process for determining an attenuation coefficient in an embodiment;

[0064] Figure 4 is a schematic diagram of a process for determining an attenuation coefficient in another embodiment;

[0065] Figure 5 A schematic diagram of a process for obtaining correction data in an embodiment;

[0066] Figure 6 A schematic diagram of a process of determining a reconstructed image in an embodiment;

[0067] Figure 7 A schematic diagram of a process for obtaining a machine learning model in one embodiment;

[0068] Figure 8 is a schematic diagram of a process of a medical image reconstruction method in one embodiment;

[0069] Fig. 9 is another process schematic diagram of a medical image reconstruction method in one embodiment;

[0070] Fig.10 is a structural block diagram of a medical image reconstruction device in one embodiment;

[0071] Fig.11 FIG. 1 is a schematic diagram of a PET imaging system according to an embodiment. DETAILED DESCRIPTION

[0072] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0073] FIG. 1 is an application environment diagram of a medical image reconstruction method in an embodiment. In an exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be shown in FIG. 1. The computer device includes a processor, a memory, an input / output interface (I / O for short) and a communication interface. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store relevant data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a medical image reconstruction method is implemented.

[0074] Those skilled in the art will understand that the structure shown in FIG. 1 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0075] This embodiment uses the method applied to a server as an example for illustration. It is understandable that the method can also be applied to a terminal, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. The terminal can be, but is not limited to, various personal computers, laptops, smart phones, and tablet computers. The server can be implemented as an independent server or a server cluster consisting of multiple servers.

[0076] Figure 2 FIG. 1 is a flow chart of a method for reconstructing a medical image in an embodiment. In an exemplary embodiment, Figure 2 As shown, a medical image reconstruction method is provided, which is described by taking the method applied to the computer device in FIG. 1 as an example, and includes the following S201 to S204.

[0077] S201, acquiring empty scan data and first scan data of a detection object through a medical scanning device.

[0078] In this embodiment, the medical scanning device may be a PET device, a PET-CT (Computed Tomography) device, a PET-MR (Magnetic Resonance Imaging) device, or other scanning devices including PET devices. The following uses the medical scanning device as an example of a PET device.

[0079] The first scan data refers to the scan data obtained by the medical scanning device when scanning the test object under normal working conditions. For example, the test object enters the PET device for scanning, and the detector of the PET device obtains the first scan data during the scanning process. The first scan data and the empty scan data can both be list-mode data.

[0080] Empty scan data refers to scan data obtained by a medical scanning device when performing an empty scan. Empty scan is a scanning method in which there is no bed, object or any source in the field of view of the medical scanning device. It can be understood that the empty scan data is used to reflect the radiation or response of the medical scanning device itself. For example, a scan can be performed when there is no bed, object or any source in the field of view of the PET device, and the empty scan data of the empty scan is acquired by the detector.

[0081] Furthermore, the computer device can obtain the empty scan data and the first scan data of the detected object through the medical scanning device. Optionally, the computer device can obtain the first scan data and the empty scan data sent by the medical scanning device, or obtain the first scan data and the empty scan data stored in advance from the storage, and this embodiment is not limited to this.

[0082] S202, determining an attenuation coefficient according to the first scanning data and the empty scanning data.

[0083] In this embodiment, after the computer device acquires the first scan data and the empty scan data, it can determine the attenuation coefficient, wherein the attenuation coefficient can represent the attenuation degree of the radioactive tracer when passing through the body tissue of the detection object.

[0084] Optionally, both the first scan data and the empty scan data include data corresponding to each response line, and the attenuation coefficient may be the attenuation coefficient corresponding to each response line. Optionally, the attenuation coefficient may also be obtained based on the attenuation image, for example, the pixel value corresponding to each pixel point in the attenuation image is the attenuation coefficient corresponding to the pixel point, and the pixel points in the attenuation image have a corresponding relationship with the pixel points of the first scan data.

[0085] Further optionally, the computer device may use a first preset algorithm to determine the attenuation coefficient based on the first scan data and the empty scan data.

[0086] Among them, the first preset algorithm may include but is not limited to a direct attenuation coefficient calculation method and various attenuation image generation methods, such as a maximum likelihood activity and attenuation image synchronous reconstruction (MLAA) algorithm based on the maximum likelihood function, a maximum likelihood attenuation correction factors (MLACF) algorithm based on the maximum likelihood function, and a maximum likelihood attenuation correction factors - Transmission (MLACF-TX) algorithm based on the maximum likelihood function.

[0087] In some implementations, the computer device may also determine the attenuation coefficient by a difference between the first scan data and the blank scan data.

[0088] In some embodiments, the computer device may also utilize an attenuation calculation model to input the first scan data and the empty scan data into a trained attenuation calculation model, and the attenuation calculation model may output an attenuation coefficient.

[0089] S203, performing attenuation correction according to the attenuation coefficient and the first scanning data to obtain correction data.

[0090] Further, after obtaining the attenuation coefficient, the computer device can perform attenuation correction according to the attenuation coefficient and the first scan data to obtain correction data. Optionally, the computer device can perform attenuation correction on the first scan data using the attenuation coefficient to obtain correction data. Exemplarily, assuming that the attenuation coefficient is the attenuation coefficient corresponding to each response line, the computer device can use the attenuation coefficient corresponding to each response line to respectively correct the data corresponding to each response line in the first scan data to obtain correction data.

[0091] S204: Determine a reconstructed image corresponding to the correction data according to the correction data and a preset machine learning model.

[0092] In this embodiment, the machine learning model may be a supervised learning model, a semi-supervised learning model, an unsupervised learning model, or the like, for obtaining a reconstructed image. Exemplarily, the machine learning model may include, but is not limited to, a convolutional neural network (CNN) model, a recurrent neural network (RNN), a fully convolutional neural network (FCN) model, a generative adversarial network (GAN) model, a back-propagation (BP) machine learning model, a radial basis function (RBF) model, a deep belief network (DBN) model, an Elman model, or at least one of a combination thereof.

[0093] Optionally, the machine learning model can be trained based on different correction data samples and reconstructed image samples corresponding to the correction data samples. Exemplarily, the computer device can input the correction data samples into the machine learning model, and the machine learning model outputs the predicted reconstructed image corresponding to the correction data samples, and calculates the loss based on the predicted reconstructed image corresponding to the correction data samples and the reconstructed image samples, and trains the prediction model based on the loss until the calculated loss is less than the preset loss, then stops training to obtain the machine learning model for subsequent use.

[0094] In this way, the machine learning model has the ability to obtain a reconstructed image based on the correction data. Optionally, the computer device can input the correction data into a trained machine learning model to obtain a reconstructed image output by the machine learning model. In some embodiments, the computer device can also pre-process the correction data and input the pre-processed correction data into a trained machine learning model to obtain a reconstructed image output by the machine learning model. Among them, pre-processing includes but is not limited to noise reduction processing, filtering processing, smoothing processing, etc.

[0095] Further optionally, different parts may correspond to different machine learning models, for example, the machine learning model used by the head may be different from the machine learning model used by the body. In some embodiments, different parts may correspond to the same machine learning model, for example, the head and the body use the same machine learning model.

[0096] In this way, the computer device can reconstruct the image of the detection object according to the correction data and the reconstruction algorithm to obtain a reconstructed image, that is, a reconstructed medical image.

[0097] In the above-mentioned medical image reconstruction method, the medical scanning device acquires the air scan data and the first scan data of the detection object. Since the attenuation coefficient can be determined based on the first scan data and the air scan data, the air scan result of the medical scanning device is taken into account in the process of determining the attenuation coefficient, that is, the radiation and response of the medical scanning device itself are taken into account. In this way, after attenuation correction is performed according to the attenuation coefficient and the first scan data, more accurate correction data can be obtained. Furthermore, according to the correction data and the preset machine learning model, the corresponding reconstructed image can be efficiently obtained. On the one hand, compared with the process of determining the reconstructed image in combination with the results of the CT device scan in the related art, this embodiment does not require CT scanning, and the process of obtaining the reconstructed image is faster and more efficient; on the other hand, the radiation and response of the medical scanning device itself are taken into account, which improves the accuracy of the reconstructed image.

[0098] Figure 3 FIG. 1 is a flow chart of determining an attenuation coefficient in an embodiment. In an exemplary embodiment, as shown in FIG. Figure 3 As shown, S202 includes S301 to S302.

[0099] S301, screening the first scan data according to attribute information of the first scan data, and determining background transmission data in the first scan data.

[0100] In this embodiment, the first scanning data includes the second scanning data and the background transmission data. In other words, since the medical scanning device still has its own radiation or response in the process of obtaining the first scanning data, the first scanning data includes two parts, one part is the second scanning data obtained based on the scanning of the detection object, and the other part is the background transmission data obtained by the radiation or response of the medical scanning device itself. In other words, the second scanning data is the data emitted from the detection object, that is, the scanning data corresponding to the tracer; the background transmission data is the data emitted from the detector of the medical scanning device. For example, the crystal based on LSO (Lantanium Oxyorthosilicate) / LYSO (lutetium-yttrium oxyorthosilicate) is a common component of the detector, wherein Lu177 (thulium-177) in the LSO material will decay by itself and emit radiation, and these radiations can be detected by the medical scanning device itself like the radiation in the detection object, but the emission position of the radiation is different. Therefore, the radiation from the detector itself is also called background radiation, that is, the background transmission data in the first scanning data.

[0101] Furthermore, the computer device will screen the first scanning data according to the attribute information of the first scanning data to obtain the background transmission data in the first scanning data.

[0102] Further optionally, the computer device may also filter the first scan data according to the attribute information of the first scan data to obtain the second scan data in the first scan data.

[0103] The attribute information includes but is not limited to energy, time of flight (TOF), and position. Energy refers to the energy of gamma rays detected by the medical scanning device in the process of scanning to obtain the first scanning data, which is used to characterize the energy of the corresponding photon itself. Time of flight refers to the time difference from the collision of gamma rays detected by the medical scanning device in the process of scanning to obtain the first scanning data to the time when the gamma rays are received by the detector. Position refers to the detection position of the gamma rays detected by the medical scanning device in the process of scanning to obtain the first scanning data on the detector.

[0104] Optionally, the first scan data may include attribute information. That is, the medical scanning device will also record the corresponding attribute information in the first scan data during the process of scanning and obtaining the scan data. Therefore, after the computer device determines the first scan data, it can obtain the attribute information of the first scan data from the first scan data.

[0105] Further optionally, the computer device may filter the first scan data by the energy in the attribute information. For example, the computer device may use the data in the first scan data whose energy is greater than a preset energy threshold as the second scan data, and use the data in the first scan data whose energy is not greater than the preset energy threshold as the background transmission data.

[0106] Optionally, the computer device may also filter the first scan data by the flight time in the attribute information. For example, the computer device may also use the data in the first scan data whose flight time is greater than a preset flight time threshold as the second scan data, and use the data in the first scan data whose flight time is not greater than the preset flight time threshold as the background transmission data.

[0107] In some embodiments, the computer device may also filter the first scan data by location or other information in the attribute information.

[0108] In some embodiments, the computer device may also filter the first scan data by using multiple information in the attribute information. For example, the computer device may also use the data in the first scan data whose energy is greater than a preset energy threshold and whose flight time is greater than a preset flight time threshold as the second scan data, and the rest as background transmission data.

[0109] S302, determining the attenuation coefficient according to the air scan data and the background transmission data.

[0110] The computer equipment can then determine the attenuation coefficient based on the air scan data and the background transmission data.

[0111] Optionally, the computer device may determine the attenuation coefficient according to the air scan data and the background transmission data based on a first preset algorithm.

[0112] In some embodiments, the computer device may also segment the air scan data and the background transmission data to determine the attenuation coefficient in each segment.

[0113] In the above embodiment, since the first scanning data can be screened according to the attribute information of the first scanning data to determine the background transmission data in the first scanning data, a more accurate attenuation coefficient can be determined based on the empty scanning data and the background transmission data while taking into account the radiation and response of the medical scanning device itself.

[0114] Figure 4 FIG. 1 is a flow chart of another method for determining an attenuation coefficient in an embodiment. In an exemplary embodiment, as shown in FIG. Figure 4 As shown, S302 includes S401 to S402.

[0115] S401, determining first scanning sub-data in a first scanning data that is located in a first time range.

[0116] In this embodiment, the first time range can be determined according to demand. Optionally, the computer device can determine the first time range in response to the user's selection. The computer device can also preset a time range as the first time range. It is understandable that the first time range belongs to the time range corresponding to the first scan data.

[0117] For example, taking the first time range as the first 3 minutes corresponding to the first scan data as an example, assuming that the PET device obtains the first scan data after scanning the object for 60 minutes, the computer device can use the data corresponding to the first 3 minutes of the first scan data from the start time point as the first scan sub-data.

[0118] S402, screening the first scanning sub-data according to the attribute information of the first scanning sub-data to determine the background transmission data.

[0119] Continuing with the above example, the computer device may filter the first scan sub-data of the first 3 minutes in the first scan data to obtain background transmission data.

[0120] Further optionally, the computer device may also screen the first scan sub-data according to the attribute information of the first scan sub-data to determine the second scan data.

[0121] It should be noted that the process of screening the first scan sub-data according to the attribute information of the first scan sub-data may refer to the process of screening the first scan data according to the attribute information of the first scan data, which will not be described in detail here.

[0122] In the above embodiment, since the first scanning sub-data located in the first time range in the first scanning data can be determined, the first scanning sub-data can be screened according to the attribute information of the first scanning sub-data, so that the background transmission data and the second scanning data can be determined quickly and efficiently, which is beneficial to reducing the workload of calculating the attenuation coefficient later.

[0123] In an exemplary embodiment, optionally, the above S403 may be implemented in the following manner:

[0124] According to the attenuation coefficient, attenuation correction is performed on the second scanning data in the first scanning data to obtain corrected data.

[0125] In this embodiment, after obtaining the attenuation coefficient, the computer device performs attenuation correction on the second scan data in the first scan data to obtain correction data. The process of performing attenuation correction is to apply the attenuation coefficient to the second scan data to obtain the corrected correction data.

[0126] Optionally, the computer device may apply the attenuation coefficient corresponding to each pixel point in the attenuation image to the corresponding pixel point in the second scan data to obtain the correction data. For example, the computer device may multiply the attenuation coefficient corresponding to each pixel point in the attenuation image with the data of the corresponding pixel point in the second scan data to obtain the correction data.

[0127] Optionally, the computer device may also perform attenuation correction on data corresponding to the response line in the second scanning data according to the attenuation coefficient corresponding to the response line for each response line, so as to obtain corrected data.

[0128] It should be noted that the computer device can perform attenuation correction on all the second scanning data in the first scanning data according to the attenuation coefficient to obtain correction data corresponding to all the second scanning data, or it can segment the second scanning data and then correct the segmented second scanning data separately according to the attenuation coefficient to obtain correction data corresponding to each segmented second scanning data. The present embodiment is not limited to this.

[0129] In the above embodiment, since the second scanning data in the first scanning data can be attenuated corrected according to the attenuation coefficient, more accurate correction data can be obtained. In this way, a more accurate reconstructed image can be obtained by using the correction data and a preset machine learning model.

[0130] The following describes a process for segmenting the second scan data. Figure 5 FIG. 1 is a flow chart of obtaining correction data in an embodiment. In an exemplary embodiment, as shown in FIG. Figure 5 As shown, the above-mentioned “performing attenuation correction on the second scanning data in the first scanning data according to the attenuation coefficient to obtain correction data” includes S501 to S502.

[0131] S501, dividing the second scan data into second scan sub-data.

[0132] In this embodiment, the computer device may divide the second scan data to obtain each second scan sub-data. The division method includes but is not limited to time division. For example, the computer device may divide the second scan data using a preset time window. Assuming that the second scan data is data obtained after scanning for 60 minutes, the computer device may divide the second scan data into one second scan sub-data every 5 minutes to obtain 12 second scan sub-data, which are respectively recorded as second scan sub-data 1 to second scan sub-data 12.

[0133] Optionally, the computer device may preset a dynamic frame divider, and the dynamic frame divider automatically divides the second scan data to obtain each second scan sub-data. The computer device may also respond to the user's division operation, and the user may define the division of the second scan data to obtain each second scan sub-data, but this embodiment is not limited thereto.

[0134] S502, performing attenuation correction on each second scanning sub-data according to the attenuation coefficient to obtain correction data corresponding to each second scanning sub-data.

[0135] Furthermore, after the computer device determines the attenuation coefficient, it can perform attenuation correction on each second scan sub-data according to the attenuation coefficient to obtain correction data corresponding to each second scan sub-data. Taking the second scan sub-data 1 as an example, the computer device performs attenuation correction on the second scan sub-data 1 according to the attenuation coefficient to obtain correction data 1 corresponding to the second scan sub-data 1. Similarly, the computer device can also perform correction data 2 to correction data 12 corresponding to the second scan sub-data 2 to the second scan sub-data 12, respectively.

[0136] The process of performing attenuation correction on the second scanning sub-data may refer to the process of performing attenuation correction on each second scanning data, which will not be described in detail here.

[0137] In the above embodiment, since the second scanning data can be divided to obtain each second scanning sub-data, and each second scanning sub-data is attenuated and corrected according to the attenuation coefficient, the correction data corresponding to each second scanning sub-data can be obtained, which is beneficial to obtaining multiple reconstructed images in subsequent dynamic scanning.

[0138] Figure 6 FIG. 1 is a flow chart of determining a reconstructed image in an embodiment. In an exemplary embodiment, as shown in FIG. Figure 6 As shown, S204 includes S601 to S602.

[0139] S601, back-projecting the corrected data to obtain a back-projected image corresponding to the corrected data.

[0140] In this embodiment, optionally, after obtaining the correction data, the computer device may back-project the correction data based on a second preset algorithm to obtain a back-projection image corresponding to the correction data. The second preset algorithm includes but is not limited to a direct point projection method and a time-of-flight-based filtered back-projection method.

[0141] In some embodiments, if the correction data includes correction data corresponding to each second scanning sub-data, the computer device will back-project the correction data corresponding to each second scanning sub-data to obtain a back-projection image of the correction data corresponding to each second scanning sub-data. Exemplarily, the computer device back-projects correction data 1 to obtain back-projection image 1, back-projects correction data to obtain back-projection image 2 corresponding to the correction data, and so on.

[0142] S602: Determine a reconstructed image corresponding to the back-projection image according to the back-projection image and the machine learning model.

[0143] In this embodiment, the machine learning model can be trained based on the back-projection image samples and the reconstructed image samples corresponding to the back-projection image samples. Exemplarily, the computer device can input the back-projection image samples into the machine learning model, and the machine learning model outputs the predicted reconstructed image corresponding to the back-projection image samples, and calculates the loss based on the predicted reconstructed image and the reconstructed image samples corresponding to the back-projection image samples, and trains the prediction model based on the loss until the calculated loss is less than the preset loss, then stops training to obtain the machine learning model used subsequently.

[0144] After that, the machine learning model has the ability to obtain a reconstructed image based on the back-projected image corresponding to the correction data. Optionally, the computer device can input the back-projected image into the trained machine learning model to obtain the reconstructed image output by the machine learning model. In some embodiments, the computer device can also pre-process the back-projected image and input the pre-processed back-projected image into the trained machine learning model to obtain the reconstructed image output by the machine learning model.

[0145] Similarly, if the correction data includes correction data corresponding to each second scanning sub-data, the computer device can determine the reconstructed image corresponding to each back-projection image based on the back-projection image corresponding to each correction data and the machine learning model, that is, determine the reconstructed image corresponding to each second scanning sub-data.

[0146] Continuing with the above example, the computer device can input the correction data 1 to correction data 12 obtained in S502 into the machine learning model to obtain the reconstructed image 1 corresponding to the correction data 1, the reconstructed image 2 corresponding to the correction data 2, and so on and so forth.

[0147] It should be noted that the computer device can input correction data 1 to correction data 12 into the machine learning model at the same time, or can input correction data 1 to correction data 12 into the machine learning model separately, and this embodiment does not impose any restrictions.

[0148] In the above embodiment, since the corrected data can be back-projected to obtain a back-projected image corresponding to the corrected data, and the reconstructed image can be determined based on the back-projected image and the machine learning model, the corresponding reconstructed image can be efficiently determined using the back-projected image and the machine learning model.

[0149] In an exemplary embodiment, optionally, the above S302 may be implemented in the following manner:

[0150] According to the air scan data and the background transmission data, the attenuation coefficient corresponding to each response line in the air scan data and the background transmission data is determined.

[0151] In this embodiment, both the air scan data and the background transmission data include data corresponding to each response line. Thus, the attenuation coefficient corresponding to each response line can be obtained according to the air scan data and the background transmission data.

[0152] Optionally, the computer device may use a first preset algorithm to determine the attenuation coefficient corresponding to each response line in the air scan data and the background transmission data.

[0153] In some embodiments, the computer device may perform segmented calculations, for example, the air scan data and the background transmission data may be segmented to determine the attenuation coefficient of each response line in each segment.

[0154] Correspondingly, the above-mentioned “performing attenuation correction on the second scan data in the first scan data according to the attenuation coefficient to obtain correction data” can be implemented in the following manner:

[0155] For each response line, according to the attenuation coefficient corresponding to the response line, attenuation correction is performed on the data corresponding to the response line in the second scanning data to obtain corrected data.

[0156] Furthermore, the second scanning data also includes data corresponding to each response line. In this way, after the computer device obtains the attenuation coefficient corresponding to each response line, it can perform attenuation correction on the data corresponding to the response line in the second scanning data to obtain corrected data.

[0157] Exemplarily, assuming that the air scan data, the background transmission data, and the second scan data all include data corresponding to the response line A, the response line B, and the response line C, the computer device first obtains the attenuation coefficient A corresponding to the response line A according to the data corresponding to the response line A in the air scan data and the data corresponding to the response line A in the background transmission data, and uses the attenuation coefficient A corresponding to the response line A to perform attenuation correction on the data corresponding to the response line A in the second scan data to obtain the correction data A corresponding to the response line A. Similarly, the computer device can also obtain the correction data B corresponding to the response line B and the correction data C corresponding to the response line C, that is, obtain the correction data. Among them, the correction data includes the correction data corresponding to each response line, for example, the correction data includes the above-mentioned correction data A to correction data C.

[0158] In some embodiments, if the attenuation coefficient of each response line in each segment is calculated, the attenuation coefficient of each response line in each segment can be used to perform corresponding attenuation correction on the data corresponding to each response line in different segments in the second scanning data to obtain corrected data.

[0159] In the above embodiment, since the attenuation coefficient corresponding to each response line in the air scan data and the background transmission data can be determined based on the air scan data and the background transmission data, for each response line, attenuation correction can be performed on the data corresponding to the response line in the second scan data based on the attenuation coefficient corresponding to the response line to obtain more accurate correction data.

[0160] Figure 7 FIG. 1 is a flow chart of obtaining a machine learning model in an embodiment. In an exemplary embodiment, as shown in FIG. Figure 7 As shown, the above-mentioned medical image reconstruction method also includes S701 to S702.

[0161] S701, obtaining a back-projection image sample and a reconstructed image sample corresponding to the back-projection image sample.

[0162] In this embodiment, the back-projection image samples are a large number of back-projection images acquired in advance by the computer device. The reconstructed image samples are reconstructed images obtained based on the back-projection image samples. Generally, in order to improve the effect of the machine learning model, the quality index of the reconstructed image samples can be higher than the preset quality index. The quality index can include but is not limited to a clarity index, a contrast index, a signal-to-noise ratio, a resolution, and the like. Further optionally, the back-projection image samples can be back-projection image samples corresponding to different parts.

[0163] The computer device may obtain back-projection image samples and corresponding reconstructed image samples specified by a user, may receive back-projection image samples and corresponding reconstructed image samples sent by other devices, and may generate back-projection image samples and corresponding reconstructed image samples.

[0164] Optionally, the computer device may acquire a large number of correction data samples, and back-project the correction data samples to obtain back-projection image samples. Afterwards, the computer device may reconstruct the back-projection image samples to obtain reconstructed image samples. In some embodiments, the user may modify the result after the back-projection image samples are reconstructed to improve the accuracy of the reconstructed image samples.

[0165] Exemplarily, the computer device may use the attenuation-corrected histogram image frame (histo-image) as the back-projection image sample of the model, and use the reconstructed data obtained based on the traditional ordered subsets expectation maximization (OSEM) as the reconstructed image sample.

[0166] S702: Train a machine learning model based on the back-projected image samples and the reconstructed image samples.

[0167] In this embodiment, optionally, the computer device can input the back-projection image samples into the machine learning model, and the machine learning model outputs the predicted reconstructed image corresponding to the back-projection image samples, and calculates the loss according to the predicted reconstructed image and the reconstructed image samples corresponding to the back-projection image samples, and obtains the machine learning model according to the loss training. Further optionally, the computer device can perform back propagation according to the calculated loss to update the model parameters of the machine learning model, and repeat the above process until the calculated loss is less than the preset loss, and the machine learning model at this time is put into subsequent use.

[0168] In the above embodiment, since it is possible to obtain back-projection image samples and reconstructed image samples corresponding to the back-projection image samples, and train a machine learning model based on the back-projection image samples and the reconstructed image samples, the machine learning model has the ability to obtain the reconstructed image based on the back-projection image. In this way, the reconstructed image can be determined efficiently and accurately based on the back-projection image and the machine learning model.

[0169] In an exemplary embodiment, optionally, the above-mentioned “screening the first scan data according to the attribute information of the first scan data to obtain the second scan data in the first scan data” can be implemented by at least one of the following:

[0170] (1) Scanning data whose energy in the first scanning data is within a preset energy range is used as the second scanning data; (2) Scanning data whose flight time in the first scanning data is within a second time range is used as the second scanning data.

[0171] In an exemplary embodiment, optionally, the above-mentioned “screening the first scan data according to the attribute information of the first scan data to obtain the background transmission data in the first scan data” can be implemented by at least one of the following:

[0172] (1) The scanning data whose energy in the first scanning data is not within the preset energy range is used as the background transmission data; (2) The scanning data whose flight time in the first scanning data is not within the second time range is used as the background transmission data.

[0173] The preset energy range can also be understood as an energy window of a certain length. Since the second scanning data and the background transmission data have different energies, the computer device can distinguish the second scanning data and the background transmission data in the first scanning data by whether the energy is within the preset energy range, so that the scanning data in the first scanning data whose energy is within the preset energy range is used as the second scanning data, and the scanning data whose energy is not within the preset energy range is used as the background transmission data.

[0174] The second time range may be a range determined by the user based on experience. Similarly, since the flight time of the second scan data and the background transmission data are different, the computer device may distinguish the second scan data and the background transmission data in the first scan data by whether the flight time is within the second time range, so as to use the scan data whose flight time is within the second time range in the first scan data as the second scan data, and the scan data whose flight time is not within the second time range as the background transmission data.

[0175] It can be understood that in some embodiments, the computer device can also distinguish the second scanning data and / or background transmission data in the first scanning data by whether the energy is within a preset energy range and whether the flight time is within a second time range.

[0176] In the above embodiment, by using the preset energy range and / or the second time range, the first scanning data can be screened according to the attribute information of the first scanning data to efficiently obtain the background transmission data and / or the second scanning data.

[0177] Figure 8 FIG. 1 is a schematic diagram of a process of a medical image reconstruction method in an embodiment. Figure 8As shown, the computer device first obtains the first scan data and the air scan data, and screens the first scan data according to the attribute information of the first scan data through the screening module to determine the background transmission data and the second scan data in the first scan data. Then, the computer device determines the attenuation coefficient according to the air scan data and the background transmission data through the attenuation coefficient generation module. Furthermore, the computer device performs attenuation correction on the second scan data in the first scan data according to the attenuation coefficient through the maximum likelihood position projector to obtain the corrected data, and back-projects the corrected data to obtain the back-projection image corresponding to the corrected data. Finally, the computer device can obtain the reconstructed image based on the back-projection image and the machine learning model.

[0178] Among them, the attenuation coefficient generation module can determine the attenuation coefficient corresponding to each response line in the air scan data and the background transmission data according to the air scan data and the background transmission data, and the maximum likelihood position projector can perform attenuation correction on the data corresponding to the response line in the second scan data according to the attenuation coefficient corresponding to the response line for each response line to obtain corrected data.

[0179] Fig. 9 FIG. 4 is another process schematic diagram of a medical image reconstruction method in an embodiment. Fig. 9 As shown, the computer device first acquires the first scan data and the air scan data, and determines the first scan sub-data in the first scan data that is located in the first time range through the screening module, and screens the first scan sub-data according to the attribute information of the first scan sub-data to determine the background transmission data and the second scan data. Then, the computer device determines the attenuation coefficient according to the air scan data and the background transmission data through the attenuation coefficient generation module. In addition, the computer device also divides the second scan data through a dynamic frame divider to obtain each second scan sub-data. Further, the computer device performs attenuation correction on each second scan sub-data according to the attenuation coefficient through a maximum likelihood position projector to obtain the correction data corresponding to each second scan sub-data, and back-projects each correction data to obtain multiple back-projection images. Finally, the computer device can obtain multiple reconstructed images based on each back-projection image and the machine learning model.

[0180] Exemplarily, assuming that the medical scanning device dynamically scans for 60 minutes to obtain medical scanning data, the computer device can extract the first scanning sub-data of the first n minutes of the first scanning sub-data of the dynamic scanning, and use it to screen the first scanning sub-data to determine the background transmission data and the second scanning data, and determine the attenuation coefficient based on the empty scan data and the background transmission data to generate the attenuation coefficient. Afterwards, the computer device can obtain the second scanning sub-data of the i-th frame based on the medical scanning data, and perform attenuation correction and back-projection on the second scanning sub-data of the i-th frame based on the attenuation coefficient of the first n minutes to obtain the back-projection image histo-image of the i-th frame, and use the back-projection image histo-image of the i-th frame as the input of the reconstructed image to quickly obtain the reconstructed image of the i-th frame. Wherein, n and i can both be integers greater than or equal to 1.

[0181] It can be seen that this embodiment can provide an attenuation coefficient estimation based on the background radiation of the LSO / LYSO detector, and a fast imaging technology based on TOF and background transmission data, so as to achieve effective attenuation correction and fast imaging of the collected data, and reduce or avoid CT scanning.

[0182] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0183] Based on the same inventive concept, the embodiment of the present application also provides a medical image reconstruction device for implementing the medical image reconstruction method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiments of one or more medical image reconstruction devices provided below can refer to the limitations of the medical image reconstruction method above, and will not be repeated here.

[0184] Fig.10 FIG. 1 is a block diagram of a medical image reconstruction device in an embodiment. In an exemplary embodiment, Fig.10 As shown, a medical image reconstruction device 1000 is provided, comprising: an acquisition module 1001, a first determination module 1002, a correction module 1003 and a second determination module 1004, wherein:

[0185] The acquisition module 1001 is used to acquire the air scan data and the first scan data of the detection object through a medical scanning device.

[0186] The first determination module 1002 is used to determine the attenuation coefficient according to the first scanning data and the empty scanning data.

[0187] The correction module 1003 is used to perform attenuation correction according to the attenuation coefficient and the first scanning data to obtain correction data.

[0188] The second determination module 1004 is used to determine the reconstructed image corresponding to the correction data according to the correction data and a preset machine learning model.

[0189] In the above-mentioned medical image reconstruction device, the medical scanning device acquires the empty scan data and the first scan data of the detection object. Since the attenuation coefficient can be determined based on the first scan data and the empty scan data, the empty scan result of the medical scanning device is taken into account in the process of determining the attenuation coefficient, that is, the radiation and response of the medical scanning device itself are taken into account. In this way, after attenuation correction is performed according to the attenuation coefficient and the first scan data, more accurate correction data can be obtained. Furthermore, according to the correction data and the preset machine learning model, the corresponding reconstructed image can be efficiently obtained. On the one hand, compared with the process of determining the reconstructed image in combination with the results of the CT device scan in the related art, this embodiment does not require CT scanning, and the process of obtaining the reconstructed image is faster and more efficient; on the other hand, the radiation and response of the medical scanning device itself are taken into account, which improves the accuracy of the reconstructed image.

[0190] Optionally, the first determining module 1002 includes:

[0191] The screening unit is used to screen the first scanning data according to the attribute information of the first scanning data, and determine the background transmission data in the first scanning data.

[0192] The first determination unit is used to determine the attenuation coefficient according to the air scan data and the background transmission data.

[0193] Optionally, the first determining unit includes:

[0194] The first determining subunit is used to determine first scanning sub-data in the first scanning data that is located in a first time range.

[0195] The second determining subunit is used to screen the first scanning sub-data according to the attribute information of the first scanning sub-data to determine the background transmission data.

[0196] Optionally, the screening unit is further configured to screen the first scanning data according to attribute information of the first scanning data to determine the second scanning data in the first scanning data.

[0197] The correction module 1003 is used to perform attenuation correction on the second scanning data in the first scanning data according to the attenuation coefficient to obtain corrected data.

[0198] Optionally, the correction module 1003 includes:

[0199] The dividing unit is used to divide the second scanning data into second scanning sub-data.

[0200] The correction unit is used to perform attenuation correction on each second scanning sub-data according to the attenuation coefficient to obtain correction data corresponding to each second scanning sub-data.

[0201] Optionally, the first determination unit is used to determine the attenuation coefficient corresponding to each response line in the air scan data and the background transmission data according to the air scan data and the background transmission data.

[0202] Correspondingly, the correction module 1003 is used to perform attenuation correction on the data corresponding to the response line in the second scanning data according to the attenuation coefficient corresponding to the response line for each response line, so as to obtain corrected data.

[0203] Optionally, the second determining module 1004 includes:

[0204] The back-projection unit is used to back-project the correction data to obtain a back-projection image corresponding to the correction data.

[0205] The second determination unit is used to determine a reconstructed image corresponding to the back-projection image according to the back-projection image and the machine learning model.

[0206] Optionally, the medical image reconstruction device 1000 includes:

[0207] The acquisition module is used to acquire the back-projection image samples and the reconstructed image samples corresponding to the back-projection image samples.

[0208] The training module is used to train a machine learning model based on the back-projected image samples and the reconstructed image samples.

[0209] Optionally, the screening unit is configured to perform at least one of the following:

[0210] Using scanning data whose energy is within a preset energy range in the first scanning data as second scanning data;

[0211] The scanning data whose flight time is within the second time range in the first scanning data is used as the second scanning data.

[0212] Optionally, the screening unit is configured to perform at least one of the following:

[0213] Using the scanning data whose energy is not within the preset energy range in the first scanning data as background transmission data;

[0214] The scanning data whose flight time is not within the second time range in the first scanning data is used as background transmission data.

[0215] Each module in the medical image reconstruction device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0216] Fig.11 FIG. 1 is a schematic diagram of a PET imaging system in one embodiment. Fig.11 As shown, in one embodiment, a PET imaging system 1100 is provided. The PET imaging system 1100 includes a scanning device 1101 and a computer device 1102 . The scanning device 1101 is connected to the computer device 1102 .

[0217] The computer device 1102 includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, it is used to implement the following steps S1 to S5:

[0218] S1, acquiring air scan data and TOF-PET imaging data.

[0219] In this embodiment, the empty scan data is acquired when there is no detection object in the scanning cavity of the scanning device 1101; the TOF-PET imaging data is acquired when there is a detection object in the scanning cavity of the scanning device 1101. It can be understood that the TOF-PET imaging data is also the first scanning data in the above embodiment. This step can refer to the above S201 and will not be repeated here.

[0220] S2, screening the TOF-PET imaging data to obtain background transmission data.

[0221] This step may refer to the above-mentioned S301 and will not be described in detail here.

[0222] S3, determining the attenuation coefficient according to the background transmission data and the air scan data.

[0223] This step may refer to the above-mentioned S302 and will not be described in detail here.

[0224] S4, correcting the TOF-PET imaging data based on the attenuation coefficient, and back-projecting the corrected TOF-PET imaging data to generate a histogram image frame.

[0225] In this embodiment, after the attenuation coefficient is obtained, the TOF-PET imaging data can be corrected using the attenuation coefficient to obtain corrected TOF-PET imaging data. The corrected TOF-PET imaging data is also the correction data in the above embodiment. This step can refer to the above S203 and will not be repeated here.

[0226] Optionally, the attenuation coefficient may be used to correct the second scanning data in the TOF-PET imaging data to obtain corrected TOF-PET imaging data.

[0227] Further, after obtaining the corrected TOF-PET imaging data, the corrected TOF-PET imaging data is back-projected, that is, the corrected TOF-PET imaging data is back-projected to obtain a histogram frame. The histogram frame is also the back-projected image in the above embodiment. This step can refer to the above S601 and will not be repeated here.

[0228] S5, reconstructing the histogram image frame to obtain a reconstructed image.

[0229] After obtaining the histogram image frame, the histogram image frame can be reconstructed to obtain a reconstructed image. The histogram image frame can be reconstructed using a reconstruction algorithm to obtain a reconstructed image, and the reconstruction algorithm includes but is not limited to an iterative reconstruction algorithm, a Fourier reconstruction algorithm, or a direct reconstruction algorithm. The reconstructed image can also be obtained by using a machine learning model in the manner referred to in S602 above, but the present embodiment is not limited thereto.

[0230] In the above PET imaging system, since the air scan data is acquired when there is no detection object in the scanning cavity of the scanning device, and the TOF-PET imaging data is acquired when there is a detection object in the scanning cavity of the scanning device, after acquiring the air scan data and the TOF-PET imaging data, screening the TOF-PET imaging data to obtain the background transmission data, and determining the attenuation coefficient based on the background transmission data and the air scan data, the air scan result of the medical scanning device can be taken into account in the process of determining the attenuation coefficient to determine a more accurate attenuation coefficient. Furthermore, by correcting the TOF-PET imaging data based on the attenuation coefficient, and back-projecting the corrected TOF-PET imaging data to generate a histogram frame, and reconstructing the histogram frame to obtain a reconstructed image, the corresponding reconstructed image can be obtained efficiently, and the accuracy of the reconstructed image can be improved.

[0231] Please continue to refer to Fig.11In one embodiment, a PET imaging system 1100 is also provided. The PET imaging system 1100 includes a scanning device 1101 and a computer device 1102 . The scanning device 1101 is connected to the computer device 1102 .

[0232] The computer device 1102 includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, it is used to implement the following steps S1 to S5:

[0233] S1, acquiring air scan data and TOF-PET imaging data.

[0234] In this embodiment, the empty scan data is acquired when there is no detection object in the scanning cavity of the scanning device 1101; the TOF-PET imaging data is acquired when there is a detection object in the scanning cavity of the scanning device 1101, and the TOF-PET imaging data is acquired by detecting the detection object in multiple motion phases.

[0235] The motion phase may be a specific period divided according to different motion stages and characteristics. For example, the TOF-PET imaging data may be imaging data obtained after scanning the test object for 60 minutes, and imaging data of every 5 minutes is regarded as a motion phase, that is, the TOF-PET imaging data obtained within 12 motion phases are obtained.

[0236] S2, screening the TOF-PET imaging data in at least one motion phase to obtain background transmission data.

[0237] Exemplarily, the TOF-PET imaging data of the first 5 minutes in the TOF-PET imaging data may be screened to obtain background transmission data.

[0238] The TOF-PET imaging data in at least one motion phase can be understood as the first scanning sub-data. The process of screening the TOF-PET imaging data in at least one motion phase can refer to the above S402 and will not be repeated here.

[0239] S3, determining the attenuation coefficient according to the background transmission data and the air scan data.

[0240] This step can refer to the above embodiment and will not be described again here.

[0241] S4, correcting the TOF-PET imaging data in multiple motion phases based on the attenuation coefficient, and back-projecting the corrected TOF-PET imaging data to generate histogram image frames in multiple motion phases.

[0242] In this embodiment, after the attenuation coefficient is obtained, the TOF-PET imaging data in multiple motion phases can be corrected using the attenuation coefficient to obtain corrected TOF-PET imaging data. Continuing with the above example, the TOF-PET imaging data obtained in the above 12 motion phases can be corrected using the attenuation coefficient to obtain 12 corrected TOF-PET imaging data.

[0243] Optionally, the attenuation coefficient may be used to correct the second scan data in the TOF-PET imaging data in multiple motion phases to obtain corrected TOF-PET imaging data. This step may refer to the above S502 and will not be described in detail here.

[0244] Further, after obtaining the corrected TOF-PET imaging data in multiple motion phases, the corrected TOF-PET imaging data can be back-projected, that is, the corrected TOF-PET imaging data can be back-projected to obtain histogram image frames in multiple motion phases. Continuing the above example, the 12 corrected TOF-PET imaging data can be back-projected to obtain histogram image frames in 12 motion phases.

[0245] S5, reconstructing the histogram image frames in a plurality of motion phases to obtain a dynamic reconstructed image.

[0246] The reconstruction process can refer to the above embodiment. It can be understood that, since the reconstruction is a histogram of image frames in multiple motion phases, multiple reconstructed images in motion phases are reconstructed, that is, dynamic reconstructed images are obtained. Continuing with the above example, 12 reconstructed images in motion phases are obtained, which are recorded as reconstructed image 1 to reconstructed image 12.

[0247] In the above-mentioned PET imaging system, since the air scan data is acquired when there is no detection object in the scanning cavity of the scanning device, the TOF-PET imaging data is acquired when there is a detection object in the scanning cavity of the scanning device, and the TOF-PET imaging data is acquired by detecting the detection object in multiple motion phases, therefore, the air scan data and the TOF-PET imaging data are acquired, the TOF-PET imaging data in at least one motion phase is screened to acquire background transmission data, and the attenuation coefficient is determined based on the background transmission data and the air scan data, so that a more accurate attenuation coefficient can be determined. In this way, the TOF-PET imaging data in multiple motion phases can be corrected based on the attenuation coefficient, and after the corrected TOF-PET imaging data is back-projected to generate histogram frames in multiple motion phases, the histogram frames in multiple motion phases can be reconstructed to efficiently and accurately acquire dynamic reconstructed images, which is conducive to viewing the dynamic process of the detection object.

[0248] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0249] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0250] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0251] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0252] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0253] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for reconstructing a medical image, characterized in that: The method comprises: Acquire the empty scan data and the first scan data of the detection object by a medical scanning device; Determining an attenuation coefficient according to the first scanning data and the empty scanning data; Perform attenuation correction according to the attenuation coefficient and the first scanning data to obtain correction data; According to the correction data and a preset machine learning model, a reconstructed image corresponding to the correction data is determined.

2. The method according to claim 1, characterized in that: The step of determining the attenuation coefficient according to the first scanning data and the empty scanning data includes: screening the first scanning data according to the attribute information of the first scanning data to determine background transmission data in the first scanning data; The attenuation coefficient is determined according to the air scan data and the background transmission data.

3. The method according to claim 2, characterized in that The screening of the first scanning data to determine the background transmission data in the first scanning data includes: Determine first scanning sub-data in the first scanning data that is located in a first time range; The first scanning sub-data are screened according to the attribute information of the first scanning sub-data to determine the background transmission data.

4. The method according to claim 2, characterized in that: The method further comprises: screening the first scan data according to the attribute information of the first scan data to determine the second scan data in the first scan data; The performing attenuation correction according to the attenuation coefficient and the first scanning data to obtain correction data includes: According to the attenuation coefficient, attenuation correction is performed on the second scanning data in the first scanning data to obtain the corrected data.

5. The method according to claim 4, characterized in that The step of performing attenuation correction on the second scan data in the first scan data according to the attenuation coefficient to obtain the corrected data includes: Dividing the second scanning data to obtain second scanning sub-data; Attenuation correction is performed on each of the second scanning sub-data according to the attenuation coefficient to obtain correction data corresponding to each of the second scanning sub-data.

6. The method according to claim 5, characterized in that Determining the reconstructed image corresponding to the correction data according to the correction data and a preset machine learning model includes: Back-projecting the correction data to obtain a back-projection image corresponding to the correction data; According to the back-projection image and the machine learning model, a reconstructed image corresponding to the back-projection image is determined.

7. The method according to claim 6, characterized in that The step of determining the attenuation coefficient according to the air scan data and the background transmission data comprises: Determine, according to the air scan data and the background transmission data, an attenuation coefficient corresponding to each response line in the air scan data and the background transmission data; Correspondingly, performing attenuation correction on the second scan data in the first scan data according to the attenuation coefficient to obtain the corrected data includes: For each of the response lines, attenuation correction is performed on the data corresponding to the response line in the second scanning data according to the attenuation coefficient corresponding to the response line to obtain the corrected data.

8. The method according to any one of claims 2 to 6, characterized in that: The step of screening the first scanning data according to the attribute information of the first scanning data to determine the background transmission data in the first scanning data includes: The scanning data whose energy in the first scanning data is not within the preset energy range is used as background transmission data; and / or the scanning data whose flight time in the first scanning data is not within the second time range is used as background transmission data.

9. A PET imaging system, characterized in that: The PET imaging system includes a scanning device and a computer device, wherein the scanning device is connected to the computer device; wherein, The computer device comprises a memory and a processor, the memory stores a computer program, and the processor executes the computer program to implement the following method steps: Acquire empty scan data and TOF-PET imaging data, wherein the empty scan data is acquired when there is no detection object in the scanning cavity of the scanning device; and the TOF-PET imaging data is acquired when there is a detection object in the scanning cavity of the scanning device; Screening the TOF-PET imaging data to obtain background transmission data; Determine an attenuation coefficient according to the background transmission data and the air scan data; Correcting the TOF-PET imaging data based on the attenuation coefficient, and back-projecting the corrected TOF-PET imaging data to generate a histogram image frame; The histogram image frame is reconstructed to obtain a reconstructed image.

10. A PET imaging system, characterized in that: The PET imaging system includes a scanning device and a computer device, wherein the scanning device is connected to the computer device; wherein, The computer device comprises a memory and a processor, the memory stores a computer program, and the processor executes the computer program to implement the following method steps: Acquire empty scan data and TOF-PET imaging data, wherein the empty scan data is acquired when there is no detection object in the scanning cavity of the scanning device; the TOF-PET imaging data is acquired when there is a detection object in the scanning cavity of the scanning device, and the TOF-PET imaging data is acquired by detecting the detection object in multiple motion phases; Screening TOF-PET imaging data in at least one motion phase to obtain background transmission data; Determine an attenuation coefficient according to the background transmission data and the air scan data; Correcting TOF-PET imaging data in multiple motion phases based on the attenuation coefficient, and back-projecting the corrected TOF-PET imaging data to generate histogram image frames in multiple motion phases; A plurality of histogram image frames in the motion phases are reconstructed to obtain a dynamic reconstructed image.

Images