Image reconstruction method and device

Abstract

The invention relates to an image reconstruction method and device. The method comprises the following steps: determining a predicted respiratory movement signal of a user according to PET original scanning data of a target part of the user; correcting the predicted respiratory movement signal through the respiratory correction factor to obtain a real respiratory movement signal of the user; the respiration correction factor represents a mapping relation between the real respiration amplitude and the predicted respiration amplitude of the user; and performing image reconstruction according to the real respiratory movement signal and the PET original scanning data to obtain a reconstructed image of the target part. By adopting the method, the quality of the reconstructed image can be improved.

Description

Technical Field

[0001] The present application relates to the field of medical technology, and in particular to an image reconstruction method and device. Background Art

[0002] When a medical device is used to scan a target part of a user, the user's movement may cause motion artifacts in the obtained scanned image, and the motion artifacts may affect the quality of the scanned image.

[0003] Taking the user's movement as breathing as an example, in the related technology, in order to avoid the influence of motion artifacts on the image quality, the user's respiratory movement signal is mainly determined based on the scanning data, and a relatively stable part is selected from the respiratory movement signal for image reconstruction to obtain a reconstructed image.

[0004] However, the related art has the problem of poor reconstructed image quality. Summary of the invention

[0005] Based on this, it is necessary to provide an image reconstruction method and device to improve the quality of the reconstructed image in response to the above technical problems.

[0006] In a first aspect, the present application provides an image reconstruction method, the method comprising:

[0007] Determine a predicted respiratory motion signal of the user based on the PET raw scan data of the target part of the user;

[0008] The predicted respiratory motion signal is corrected by a respiratory correction factor to obtain the user's real respiratory motion signal; the respiratory correction factor represents the mapping relationship between the user's real respiratory amplitude and the predicted respiratory amplitude;

[0009] Image reconstruction is performed based on the real respiratory motion signal and the original PET scan data to obtain a reconstructed image of the target area.

[0010] In one embodiment, the process of obtaining the breathing correction factor includes:

[0011] Obtain sample PET raw scan data;

[0012] Determine a plurality of sample predicted respiratory amplitudes and a plurality of sample actual respiratory amplitudes according to the sample PET raw scan data;

[0013] The predicted breathing amplitude of each sample and the actual breathing amplitude of each sample are regressed to obtain the breathing correction factor.

[0014] In one embodiment, determining a plurality of sample predicted respiratory amplitudes and a plurality of sample actual respiratory amplitudes based on sample PET raw scan data includes:

[0015] Determine the sample predicted respiratory motion signal according to the sample PET raw scan data;

[0016] Divide the sample predicted respiratory motion signal to obtain the sample predicted respiratory motion signal under each respiratory gating;

[0017] Based on the sample predicted respiratory motion signal under each respiratory gating, the sample predicted respiratory amplitude and the sample actual respiratory amplitude under each respiratory gating are determined.

[0018] In one embodiment, determining the predicted respiratory amplitude of each sample under respiratory gating based on the predicted respiratory motion signal of each sample under respiratory gating includes:

[0019] For any respiratory gating, obtain the average value of the sample predicted respiratory motion signal under the respiratory gating;

[0020] Get the difference between the mean and the reference mean;

[0021] The difference is determined as the sample predicted respiratory amplitude under respiratory gating.

[0022] In one embodiment, predicting the respiratory motion signal based on the samples under each respiratory gating and determining the actual respiratory amplitude of the samples under each respiratory gating includes:

[0023] Based on the predicted respiratory motion signal of the samples under each respiratory gating, the sample PET raw scan data under each respiratory gating is determined from the sample PET raw scan data;

[0024] Perform image reconstruction on each sample PET raw scan data under respiratory gating to obtain multiple sample reconstructed images;

[0025] The real respiratory amplitude of each sample under respiratory gating is determined based on the reconstructed image of each sample.

[0026] In one embodiment, determining the actual respiratory amplitude of each sample under respiratory gating according to the reconstructed image of each sample includes:

[0027] For any respiratory gating, the sample reconstructed image is registered with the reference reconstructed image to obtain the image deformation matrix under the respiratory gating;

[0028] The average value of the image deformation matrix is ​​determined as the true respiratory amplitude of the sample under respiratory gating.

[0029] In one embodiment, determining a predicted respiratory motion signal of a user based on raw PET scan data of a target part of the user includes:

[0030] Obtaining a mask image of the target part in a preset mode;

[0031] Based on the mask image, determining the PET raw scan data of the target site from the PET raw scan data;

[0032] The center of mass movement information of the target part is determined from the PET raw scan data of the target part, and the center of mass movement information is determined as the predicted respiratory motion signal of the user.

[0033] In one embodiment, correcting the predicted respiratory motion signal by the respiratory correction factor to obtain the user's actual respiratory motion signal includes:

[0034] calculating the product of the respiratory correction factor and the predicted respiratory motion signal;

[0035] The product is determined as the user's true respiratory motion signal.

[0036] In one embodiment, image reconstruction is performed based on the real respiratory motion signal and the PET raw scan data to obtain a reconstructed image of the target part, including:

[0037] Acquiring multiple respiratory motion amplitudes of a real respiratory motion signal;

[0038] comparing the amplitude of each respiratory movement with a preset amplitude threshold;

[0039] Determine PET target scan data corresponding to a respiratory motion amplitude that is less than or equal to a preset amplitude threshold from the PET raw scan data;

[0040] The image of the target part is reconstructed based on the PET target scanning data to obtain a reconstructed image of the target part.

[0041] In a second aspect, the present application further provides an image reconstruction device, the device comprising:

[0042] A determination module, used to determine the user's predicted respiratory motion signal based on the PET raw scan data of the user's target part;

[0043] A correction module, used to correct the predicted respiratory motion signal by a respiratory correction factor to obtain a real respiratory motion signal of the user; the respiratory correction factor represents a mapping relationship between the real respiratory amplitude of the user and the predicted respiratory amplitude;

[0044] The reconstruction module is used to reconstruct images based on the real respiratory motion signal and the PET original scanning data to obtain a reconstructed image of the target part.

[0045] In a third aspect, the present application further provides a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the content of any one of the image reconstruction methods in the first aspect is implemented.

[0046] In a fourth aspect, the present application further provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the content of any one of the image reconstruction methods in the first aspect is implemented.

[0047] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the content of any one of the image reconstruction methods in the first aspect when executed by a processor.

[0048] The above-mentioned image reconstruction method and device determine the user's predicted respiratory motion signal based on the PET original scan data of the user's target part; correct the predicted respiratory motion signal through the respiratory correction factor to obtain the user's real respiratory motion signal; perform image reconstruction based on the real respiratory motion signal and the PET original scan data to obtain a reconstructed image of the target part. In this method, the respiratory correction factor represents the mapping relationship between the user's real breathing amplitude and the predicted breathing amplitude. Since the predicted respiratory motion signal is determined based on the PET original scan data, there is a difference between it and the real respiratory motion signal. In order to avoid the influence of this difference on the quality of the reconstructed image, the predicted respiratory motion signal of the user is corrected using the respiratory correction factor, so that the error between the corrected predicted respiratory motion signal and the real respiratory motion signal is small. In this way, the image reconstructed based on the real respiratory motion signal is less affected by artifacts, so that the quality of the reconstructed image will also be higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A diagram showing an application environment of an image reconstruction method in an embodiment;

[0050] Figure 2 is a schematic flow chart of an image reconstruction method in one embodiment;

[0051] Figure 3 is a schematic flow chart of an image reconstruction method in one embodiment;

[0052] Figure 4 is a schematic flow chart of an image reconstruction method in one embodiment;

[0053] Figure 5 is a schematic flow chart of an image reconstruction method in one embodiment;

[0054] Figure 6 is a schematic flow chart of an image reconstruction method in one embodiment;

[0055] Figure 7 is a schematic flow chart of an image reconstruction method in one embodiment;

[0056] Figure 8 is a schematic flow chart of an image reconstruction method in one embodiment;

[0057] Fig. 9 is a schematic flow chart of an image reconstruction method in one embodiment;

[0058] Fig.10 is a schematic flow chart of an image reconstruction method in one embodiment;

[0059] Fig.11 is a schematic flow chart of an image reconstruction method in one embodiment;

[0060] Fig.12 FIG. 4 is a structural block diagram of an image reconstruction device in an embodiment. DETAILED DESCRIPTION

[0061] 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.

[0062] Before introducing the image reconstruction method of the present application, a brief introduction to the background of the present application is first given.

[0063] When a medical device is used to scan a user's target part, the user's movement may cause motion artifacts in the scanned image, which may affect the quality of the scanned image. For example, the user's movement may be breathing, heartbeat, or unconscious body shaking.

[0064] Taking respiratory motion as an example, in the related art, the respiratory motion signal of the user is mainly obtained, and according to the respiratory motion signal of the user, the PET raw scan data with relatively stable respiratory motion is selected from the PET raw scan data scanned by a positron emission tomography (PET) to perform image reconstruction, so as to obtain a reconstructed image with inconspicuous respiratory motion artifacts. The respiratory motion signal of the above-mentioned user is generally detected by an external respiratory detection device, or a certain algorithm is used to directly extract the centroid, total count and other characteristic information of the area affected by the respiratory motion from the PET raw data, and the respiratory motion signal of the user is estimated based on the characteristic information. For example, the external respiratory detection device can be a chest or abdominal strap for pressure detection, a reflective module placed on the chest and abdomen for ray tracing, etc.

[0065] When determining a relatively stable respiratory motion range from the user's respiratory motion signal, the selection basis may be based on the amplitude of the entire respiratory signal, histogram, etc. In general, about 40% of the PET original scan data close to the end-expiratory region may be selected for image reconstruction, so that the obtained reconstructed image does not contain too much respiratory motion information, and the quality of the reconstructed image is not too poor due to the loss of too much PET original scan data.

[0066] Whether the respiratory signal is selected based on the respiratory amplitude or the histogram, only the proportion of the retained PET raw scan data is limited (for example, 40%), but the magnitude of the respiratory motion amplitude remaining in the reconstructed image of the retained PET raw scan data cannot be guaranteed. When the user's respiratory motion amplitude is large, using 40% of the PET raw scan data in the end-expiratory region may still leave obvious respiratory motion artifacts in the reconstructed image. When the user's respiratory motion amplitude is small, using 40% of the PET raw scan data in the end-expiratory region may cause the reconstructed image to lose a lot of important information.

[0067] However, whether it is the respiratory signal obtained by the external detection device or the respiratory signal extracted based on the original PET scan data, it cannot directly reflect the real respiratory amplitude information inside the body. In other words, there is a deviation between the acquired user's respiratory motion signal and the actual respiratory motion signal, which will result in low quality of the reconstructed image.

[0068] In view of the above problems, the present application provides an image reconstruction method, which corrects the predicted respiratory motion signal of the user so that the corrected respiratory motion signal is close to the real respiratory motion signal. In this way, the PET raw scan data with steady breathing is selected from the real respiratory motion signal, and the image is reconstructed based on the PET raw scan data with steady breathing. The obtained reconstructed image can avoid the influence of motion artifacts, thereby improving the quality of the reconstructed image.

[0069] The image reconstruction method provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. For example, the computer device can be a server, a personal computer, a laptop, a smart phone, a tablet computer, a smart mobile phone, etc. The computer device may include a processor, a memory and a network interface connected by a system bus or wirelessly. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device may include 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 data in the image reconstruction process. The network interface of the computer device is used to communicate with an external terminal through a network connection, and the computer program is executed by the processor to implement an image reconstruction method. Among them, the computer device can be implemented by an independent computer device or a computer device cluster composed of multiple computer devices. It should be noted that the memory of the computer device is not limited to the above-mentioned memory, and may also include a high-speed random access memory, a volatile solid-state memory, etc. In addition, the composition architecture of the computer device is not limited to the above-mentioned situation, and some components may also be added or omitted.

[0070] In one embodiment, Figure 2 As shown, an image reconstruction method is provided, which is applied to Figure 1 The computer device in the example is used to illustrate, including the following steps:

[0071] S201, determining a predicted respiratory motion signal of a user according to the original PET scan data of a target part of the user.

[0072] The PET raw scan data refers to data obtained by scanning a target part of a user using a positron emission tomography scanner. The target part refers to a part that is affected by the user's breathing movement, for example, the target part may be the lungs, liver, or spleen.

[0073] In this embodiment, the computer device can obtain the PET raw scan data of the user's target part from the scan database according to the identification information of the user's target part. Alternatively, the computer device can also send a scan instruction of the user's target part to the PET device. After receiving the scan instruction, the PET device scans the user's target part. After the scan is completed, the PET device sends the scanned PET raw scan data to the computer device. The device can also be a PET-CT device composed of a PET device and a computed tomography (CT), or a PET-MRI device composed of a PET device and a magnetic resonance imaging (MRI). This embodiment does not limit the method of obtaining the PET raw scan data of the user's target part.

[0074] Further, the computer device can input the PET raw scan data of the user's target part into a preset respiratory motion signal prediction model, and analyze the PET raw scan data through the preset respiratory motion signal prediction model to determine the user's predicted respiratory motion signal when scanning the user's target part. Alternatively, the computer device can use any point of the target part as a reference point, and determine the movement information of the reference point based on the PET raw scan data of the target part. Afterwards, the respiratory motion signal corresponding to the movement information is determined based on the relationship between the movement information and the respiratory motion. This embodiment does not impose any restriction on the method of determining the user's predicted respiratory motion signal based on the PET raw scan data of the user's target part.

[0075] S202, correcting the predicted respiratory motion signal by a respiratory correction factor to obtain a real respiratory motion signal of the user; the respiratory correction factor represents a mapping relationship between the real respiratory amplitude of the user and the predicted respiratory amplitude.

[0076] The respiratory correction factor refers to the mapping relationship between the actual respiratory amplitude and the predicted respiratory amplitude. In this way, the actual respiratory motion signal can be obtained when the predicted respiratory motion signal and the respiratory adjustment factor are known. The actual respiratory amplitude refers to the actual quantitative value of the user's respiratory degree during the scanning process, and the predicted respiratory amplitude is the quantitative value of the respiratory degree during the scanning process estimated based on the PET raw data.

[0077] In this embodiment, the predicted respiratory motion signal refers to the respiratory motion signal of the user within the scanning time period of the target part. Therefore, when the computer device uses the respiratory correction factor to correct the predicted respiratory motion signal, it is necessary to correct all respiratory motion signals within the scanning time period. In this way, the corrected predicted respiratory motion signal is the real respiratory motion signal of the user.

[0078] S203, performing image reconstruction according to the real respiratory motion signal and the original PET scan data to obtain a reconstructed image of the target part.

[0079] In this embodiment, after obtaining the user's real respiratory motion signal, the computer device can determine a relatively stable respiratory motion signal range from the real respiratory motion signal. And determine the PET raw scan data corresponding to the relatively stable respiratory motion signal range from the PET raw scan data. Afterwards, image reconstruction is performed based on the PET raw scan data corresponding to the relatively stable respiratory motion signal range, so that the reconstructed image of the target part will not be affected by artifacts, and the quality of the reconstructed image will be higher.

[0080] In the above-mentioned image reconstruction method, the predicted respiratory motion signal of the user is determined according to the PET original scan data of the target part of the user; the predicted respiratory motion signal is corrected by the respiratory correction factor to obtain the real respiratory motion signal of the user; image reconstruction is performed according to the real respiratory motion signal and the PET original scan data to obtain a reconstructed image of the target part. In this method, the respiratory correction factor represents the mapping relationship between the real breathing amplitude and the predicted breathing amplitude of the user. Since the predicted respiratory motion signal is determined according to the PET original scan data, there is a difference between it and the real respiratory motion signal. In order to avoid the influence of this difference on the quality of the reconstructed image, the predicted respiratory motion signal of the user is corrected by the respiratory correction factor, so that the error between the corrected predicted respiratory motion signal and the real respiratory motion signal is small. In this way, the image reconstructed based on the real respiratory motion signal is less affected by artifacts, so that the quality of the reconstructed image will also be higher.

[0081] Based on the above embodiments, this embodiment is to Figure 2 The relevant contents of the process of obtaining the breathing correction factor in step S202 are introduced and explained. Figure 3 As shown, as a non-limiting example, the process of obtaining the above-mentioned breathing correction factor may include the following contents:

[0082] S301, obtaining sample PET raw scan data.

[0083] The sample PET raw scan data is raw data used to determine the respiratory correction factor.

[0084] In this embodiment, the computer device may obtain PET raw scan data of one user from the scan database, and determine the PET raw scan data of the user as the sample PET raw scan data. It should be noted that, in order to ensure the accuracy of the breathing correction factor, the computer device may also obtain PET raw scan data of multiple users from the scan database, and determine the PET raw scan data of the multiple users as the sample PET raw scan data.

[0085] S302, determining a plurality of sample predicted respiratory amplitudes and a plurality of sample actual respiratory amplitudes according to the sample PET original scan data.

[0086] In this embodiment, when the sample PET raw scan data refers to the PET raw scan data of a user, the computer device can determine the sample predicted respiratory motion signal based on the sample PET raw scan data. Since the sample predicted respiratory motion signal is a signal within a scanning time, the computer device can divide and process the sample predicted respiratory motion signal to obtain multiple groups of sample predicted respiratory motion signals. According to each group of sample predicted respiratory motion signals, the corresponding sample predicted respiratory amplitude and sample actual respiratory amplitude are determined.

[0087] When the sample PET raw scan data refers to the PET raw scan data of multiple users, for the PET raw scan data of any one user, the sample predicted respiratory motion signal of the user is determined based on the PET raw scan data of the user. Afterwards, the computer device can determine the difference between the maximum predicted respiratory amplitude and the minimum predicted respiratory amplitude of the user based on the sample predicted respiratory motion signal, and determine the difference as the sample predicted respiratory amplitude of the user. And, based on the sample predicted respiratory motion signal, determine the difference between the maximum real respiratory amplitude and the minimum real respiratory amplitude of the user, and determine the difference as the sample real respiratory amplitude of the user. Afterwards, the above method is used to obtain the sample predicted respiratory amplitude and the sample real respiratory amplitude corresponding to each user, and multiple sample predicted respiratory amplitudes and multiple sample real respiratory amplitudes are obtained.

[0088] S303, performing regression processing on the predicted breathing amplitude of each sample and the actual breathing amplitude of each sample to obtain a breathing correction factor.

[0089] In this embodiment, after obtaining multiple sample predicted breathing amplitudes and multiple sample real breathing amplitudes, the computer device can perform regression analysis on each sample predicted breathing amplitude and each sample real breathing amplitude, obtain a mapping relationship between the predicted breathing amplitude and the real breathing amplitude, and determine the mapping relationship as a breathing correction factor. The regression analysis can be a linear regression analysis, a polynomial regression analysis, or other regression analysis methods, or an analysis method based on machine learning or deep learning.

[0090] In the above-mentioned image reconstruction method, sample PET original scan data is obtained; based on the sample PET original scan data, multiple sample predicted respiratory amplitudes and multiple sample real respiratory amplitudes are determined; regression processing is performed on each sample predicted respiratory amplitude and each sample real respiratory amplitude to obtain a respiratory correction factor. This method can determine multiple sample predicted respiratory amplitudes and multiple sample real respiratory amplitudes based on the sample PET original scan data; thus, regression analysis can be performed based on multiple sample predicted respiratory amplitudes and multiple sample real respiratory amplitudes. Since the number of samples used in the regression analysis is large, the respiratory correction factor obtained by the regression analysis is more accurate.

[0091] Based on the above embodiments, this embodiment is to Figure 3 The relevant contents of step S302 in "determining the predicted respiratory amplitudes of multiple samples and the actual respiratory amplitudes of multiple samples according to the sample PET original scan data" are introduced and explained. Figure 4 As shown, as a non-limiting example, the above step S302 may include the following content:

[0092] S401, determining a sample predicted respiratory motion signal according to the sample PET original scan data.

[0093] In this embodiment, the sample PET raw scan data refers to the PET raw scan data of a user. The computer device can use a centroid-based respiratory signal analysis method to analyze the sample PET raw scan data to determine the movement information of the centroid of the target part of the user within the scanning time period, and use the movement distance of the centroid as the amplitude change unit of the sample predicted respiratory motion signal. Based on the movement information of the centroid, the sample predicted respiratory motion signal of the user is determined. Among them, the movement information of the centroid can be the movement information of the real centroid movement direction, or the movement information of the centroid in a certain direction, or the centroid movement information processed using a fitting method or a deep learning method.

[0094] S402, dividing the sample predicted respiratory motion signal to obtain each sample predicted respiratory motion signal under respiratory gating.

[0095] In this embodiment, the computer device can divide the sample predicted respiratory motion signals according to the size of the respiratory amplitude, and divide the sample predicted respiratory motion signals corresponding to similar respiratory amplitudes into the same respiratory gating, so as to obtain the sample predicted respiratory motion signals under each respiratory gating. Alternatively, the computer device can also divide the sample predicted respiratory motion signals according to the amount of photons in the scanning process, and divide the sample predicted respiratory motion signals with equal photon amounts into the same respiratory gating, so as to obtain the sample predicted respiratory motion signals under each respiratory gating. Alternatively, the computer device can also divide the scanning time into multiple time periods, and divide the sample predicted respiratory motion signals based on the multiple time periods, and divide the sample predicted respiratory motion signals in the same time period into the same respiratory gating, so as to obtain the sample predicted respiratory motion signals under each respiratory gating. This embodiment does not limit the division method.

[0096] S403, based on the sample predicted respiratory motion signal under each respiratory gating, determine the sample predicted respiratory amplitude and the sample actual respiratory amplitude under each respiratory gating.

[0097] In this embodiment, after obtaining the sample predicted respiratory motion signal under each respiratory gating, for any respiratory gating, the computer device can calculate the difference between the maximum and minimum values ​​of the predicted respiratory signal under the respiratory gating, and determine the difference as the sample predicted respiratory amplitude. In addition, the computer device can determine the real predicted respiratory motion signal under the respiratory gating based on the sample predicted respiratory motion signal under the respiratory gating. Afterwards, the difference between the maximum and minimum values ​​of the real respiratory signal under the respiratory gating can be calculated, and the difference can be determined as the sample real respiratory amplitude. The sample predicted respiratory amplitude and the sample real respiratory amplitude under each respiratory gating can be obtained by the above method.

[0098] Alternatively, the computer device can also obtain the predicted respiratory motion signal under each respiratory gating, calculate the average predicted respiratory amplitude under each respiratory gating, and use the minimum average predicted respiratory amplitude as the reference respiratory amplitude, calculate the difference between each average predicted respiratory amplitude and the reference respiratory amplitude, and determine the difference as the sample predicted respiratory amplitude under each respiratory gating. In this way, the real respiratory amplitude of the sample under each respiratory gating can also be obtained.

[0099] In the above-mentioned image reconstruction method, the sample predicted respiratory motion signal is determined according to the sample PET original scan data; the sample predicted respiratory motion signal is divided to obtain the sample predicted respiratory motion signal under each respiratory gating; based on the sample predicted respiratory motion signal under each respiratory gating, the sample predicted respiratory amplitude and the sample true respiratory amplitude under each respiratory gating are determined. This method can obtain the sample predicted respiratory motion signals under multiple respiratory gatings by dividing the sample predicted respiratory motion signal, and accurately determine the sample predicted respiratory amplitude and the sample true respiratory amplitude under each respiratory gating based on the multiple sample predicted respiratory motion signals.

[0100] Based on the above embodiments, this embodiment is to Figure 4 The relevant contents of step S403 in the above description of "determining the predicted respiratory amplitude of each sample under respiratory gating based on the predicted respiratory motion signal of each sample under respiratory gating" are introduced and explained. Figure 5 As shown, as a non-limiting example, the above step S403 may include the following content:

[0101] S501 , for any respiratory gating, obtaining an average value of the sample predicted respiratory motion signal under the respiratory gating.

[0102] In this embodiment, since a sample predicted respiratory motion signal under respiratory gating is also a respiratory motion signal within a short scanning time, it includes multiple respiratory motion quantization values. The computer device can calculate the average value of all respiratory motion quantization values ​​under respiratory gating, and use the average value to represent the respiratory motion quantization value under respiratory gating.

[0103] S502, obtaining a difference between the average value and a reference average value.

[0104] In this embodiment, after obtaining the average value under each respiratory gating, any average value can be used as a reference average value, for example, the smallest average value among all average values ​​can be used as the reference average value. The computer device can calculate the difference between any average value under respiratory gating and the reference average value.

[0105] S503, determining the difference as the sample predicted respiratory amplitude under respiratory gating.

[0106] In this embodiment, when the difference between the average value under respiratory gating and the reference average is obtained, the difference represents the respiratory amplitude difference between the respiratory motion quantization value under respiratory gating and the lowest respiratory motion quantization value, that is, the sample predicted respiratory amplitude under respiratory gating.

[0107] In the above image reconstruction method, for any respiratory gating, the average value of the sample predicted respiratory motion signal under the respiratory gating is obtained; the difference between the average value and the reference average value is obtained; and the difference is determined as the sample predicted respiratory amplitude under the respiratory gating. This method obtains the average value under each respiratory gating and accurately obtains the sample predicted respiratory amplitude under each respiratory gating based on the difference between the average value and the reference average value.

[0108] Based on the above embodiments, this embodiment is to Figure 4 The relevant contents of step S403 in the above description of "predicting the respiratory motion signal based on the samples under each respiratory gating, and determining the true respiratory amplitude of the samples under each respiratory gating" are introduced and explained. Figure 6 As shown, as a non-limiting example, the above step S403 may include the following content:

[0109] S601 , based on the predicted respiratory motion signal of each sample under respiratory gating, determine each sample PET raw scan data under respiratory gating from the sample PET raw scan data.

[0110] In this embodiment, after dividing the sample predicted respiratory motion signal into multiple sample predicted respiratory motion signals under respiratory gating, the computer device can divide the sample PET raw scan data according to the division rule to obtain the sample PET raw scan data under each respiratory gating.

[0111] S602, performing image reconstruction on each sample PET raw scan data under respiratory gating to obtain a plurality of sample reconstructed images.

[0112] In this embodiment, for any respiratory gating, the computer device can use an image reconstruction algorithm to reconstruct the sample PET raw scan data under the respiratory gating to obtain the sample reconstructed image under the respiratory gating. Based on this method, the sample PET raw scan data under each respiratory gating is image reconstructed to obtain multiple sample reconstructed images. That is to say, there is a one-to-one mapping relationship between the sample reconstructed image and the respiratory gating. Among them, the image reconstruction algorithm can be an iterative algorithm or an analytical algorithm. For example, the iterative algorithm can be a maximum likelihood expectation method (MLEM), an ordered subset maximum likelihood method (OSEM) or a mean average precision (MAP) of all classes, etc. The analytical algorithm can be an image reconstruction method based on the projection slice theorem (Filtered Back Projection, FBP) or a 3D rapid prototyping method (3D-Rapid Prototyping), etc. It should be noted that the PET image can be an image without attenuation correction, or an image after attenuation correction. Alternatively, the PET image may be an image reconstructed by a conventional image reconstruction method, or may be a direct back-projection image. This embodiment does not limit the type of PET image.

[0113] S603, reconstructing images of each sample to determine the actual respiratory amplitude of each sample under respiratory gating.

[0114] In this embodiment, for any sample reconstructed image, the computer device may input the sample reconstructed image into a preset respiratory amplitude determination model, and analyze the sample reconstructed image according to the respiratory amplitude determination model to obtain the sample real respiratory amplitude under the respiratory gating. Alternatively, the computer device may obtain the deformation between each sample reconstructed image, and determine the sample real respiratory amplitude under each respiratory gating based on the image deformation.

[0115] In the above-mentioned image reconstruction method, based on the sample prediction respiratory motion signal under each respiratory gating, the sample PET raw scan data under each respiratory gating is determined from the sample PET raw scan data; image reconstruction is performed on each sample PET raw scan data under respiratory gating to obtain multiple sample reconstructed images; and the real respiratory amplitude of the sample under each respiratory gating is determined based on each sample reconstructed image. This method can accurately obtain the real respiratory amplitude of each sample under respiratory gating by performing image reconstruction on the sample PET raw scan data under each respiratory gating and based on the image reconstruction result.

[0116] Based on the above embodiments, this embodiment is to Figure 6 The relevant contents of "reconstructing images according to each sample and determining the true respiratory amplitude of each sample under respiratory gating" in step S603 are introduced and explained.

[0117] like Figure 7 As shown, as a non-limiting example, the above step S603 may include the following content:

[0118] S701 , for any respiratory gating, registering the sample reconstructed image with the reference reconstructed image to obtain an image deformation matrix under the respiratory gating.

[0119] In this embodiment, the reference reconstructed image can be any one of all the sample reconstructed images of the respiratory gating, for example, the reference reconstructed image is the sample reconstructed image of the first respiratory gating. For any respiratory gating, the computer device can align the sample reconstructed image under the respiratory gating with the reference reconstructed image through an image registration method to obtain the image deformation matrix under the respiratory gating. Among them, the image registration method can be a registration method based on rigid change, affine change, B-spline or optical flow method, etc., and can also be an image registration model based on a deep learning network.

[0120] S702, determining the average value of the image deformation matrix as the actual breathing amplitude of the sample under respiratory gating.

[0121] In this embodiment, an image deformation matrix under respiratory gating represents multiple respiratory amplitude quantization values ​​under the respiratory gating. The computer device can calculate the average value of the image deformation matrix under respiratory gating, and use the average value as the real respiratory amplitude of the sample under the respiratory gating. It should be noted that the average value can be the overall mean of the image deformation matrix or the mean in a certain direction.

[0122] In the above image reconstruction method, for any respiratory gating, the sample reconstructed image is registered with the reference reconstructed image to obtain the image deformation matrix under the respiratory gating; the average value of the image deformation matrix is ​​determined as the sample true respiratory amplitude under the respiratory gating. This method can accurately obtain the image deformation matrix under each respiratory gating by image registration, so as to obtain a more accurate sample true respiratory amplitude, that is, the average value of the image deformation matrix.

[0123] Based on the above embodiments, this embodiment is to Figure 2 The relevant contents of "determining the user's predicted respiratory motion signal based on the PET raw scan data of the user's target part" in step S201 are introduced and explained. Figure 8 As shown, as a non-limiting example, the above step S201 may include the following content:

[0124] S801, obtaining a mask image of a target part in a preset mode.

[0125] In this embodiment, the computer device can use a pre-trained image segmentation network to perform image segmentation on the original image under a preset mode to obtain a mask image corresponding to the original image. When the scanning device is a PET device, the original image of the target part is a PET image, and the corresponding mask image is a PET mask image; when the scanning device is a PET-CT device, the original image of the target part is a CT image, and the corresponding mask image is a CT mask image; when the scanning device is a PET-MRI image, the original image of the target part is an MRI image, and the corresponding mask image is an MRI mask image. Alternatively, the original image can also be a multi-channel image integrated from PET, CT, MRI images, etc., or a PET direct back projection image, a CT attenuation coefficient image, or a derivative image of PET / CT / MRI. This embodiment does not limit the category of the mask image.

[0126] S802, determining the PET raw scan data of the target part from the PET raw scan data based on the mask image.

[0127] In this embodiment, since the mask image includes segmentation information of various parts of the user, the computer device can determine the PET raw scan data corresponding to the target part from the PET raw scan data based on the segmentation information of the target part in the mask image.

[0128] S803, determining the mass center movement information of the target part from the PET raw scan data of the target part, and determining the mass center movement information as the predicted respiratory motion signal of the user.

[0129] In this embodiment, the computer device can use a centroid-based analysis method to analyze the PET raw scan data of the target area to determine the movement information of the center of mass in the target area during the scanning time. The movement information of the center of mass represents the user's respiratory motion signal during the scanning time, and the center of mass movement information is determined as the user's predicted respiratory motion signal.

[0130] In the above-mentioned image reconstruction method, a mask image of the target part under a preset mode is obtained; based on the mask image, the PET raw scan data of the target part is determined from the PET raw scan data; the center of mass movement information of the target part is determined from the PET raw scan data of the target part, and the center of mass movement information is determined as the user's predicted respiratory motion signal. This method can accurately determine the required PET raw scan data of the target part from the PET raw scan data through the mask image. And the center of mass movement information of the target part can be accurately determined from the PET raw scan data of the target part, so that the user's predicted respiratory motion signal can be accurately obtained.

[0131] Based on the above embodiments, this embodiment is to Figure 2 The relevant contents of "correcting the predicted respiratory motion signal by the respiratory correction factor to obtain the user's real respiratory motion signal" in step S202 are introduced and explained. Fig. 9 As shown, as a non-limiting example, the above step S202 may include the following content:

[0132] S901, calculating the product of the breathing correction factor and the predicted breathing motion signal.

[0133] In this embodiment, the predicted respiratory motion signal is a plurality of respiratory motion signals within a scanning time period. Therefore, for any respiratory motion signal, the computer device can calculate the product of the respiratory motion signal and the respiratory correction factor to obtain the real respiratory motion signal corresponding to the respiratory motion signal.

[0134] S902: Determine the product as the user's real respiratory motion signal.

[0135] In this embodiment, after the product results of all respiratory motion signals are obtained, all product results are fused, and the fused respiratory motion signal is used as the user's real respiratory motion signal. It should be noted that, in this embodiment, the respiratory correction factor is the slope value K obtained by linear regression, and the product of the slope K and the predicted respiratory motion signal S can be expressed as: F = K*S. Wherein, F represents the user's real respiratory motion signal.

[0136] In the above-mentioned image reconstruction method, the product of the breathing correction factor and the predicted breathing motion signal is calculated; and the product is determined as the user's real breathing motion signal. In the process of correcting the predicted breathing motion signal, the method can calculate the product of the breathing correction factor and the predicted breathing motion signal, and the user's real breathing motion signal can be accurately obtained according to the product result.

[0137] Based on the above embodiments, this embodiment is to Figure 2The relevant contents of step S203 in “performing image reconstruction according to the real respiratory motion signal and the PET original scan data to obtain a reconstructed image of the target part” are introduced and explained. Fig.10 As shown, as a non-limiting example, the above step S203 may include the following content:

[0138] S1001, obtaining multiple respiratory motion amplitudes of a real respiratory motion signal.

[0139] In this embodiment, the computer device can divide the real respiratory motion signal to obtain multiple real respiratory motion signals under respiratory gating, calculate the average value of each real respiratory motion signal under respiratory gating, and calculate the difference between the average value under each respiratory gating and the reference average value, and determine the difference as the respiratory motion amplitude under multiple respiratory gating.

[0140] S1002, comparing the amplitude of each respiratory movement with a preset amplitude threshold.

[0141] In this embodiment, since an increase in the respiratory motion amplitude will make the artifacts in the image more obvious, the preset amplitude threshold refers to the maximum tolerable respiratory motion amplitude Mmax. After the computer device obtains the respiratory motion amplitudes under multiple respiratory gatings, each respiratory motion amplitude under respiratory gating can be compared with the preset amplitude threshold to obtain a comparison result. Among them, the comparison result includes: the respiratory motion amplitude is greater than the preset amplitude threshold and the respiratory motion amplitude is less than or equal to the preset amplitude threshold.

[0142] S1003, determining PET target scan data corresponding to a respiratory motion amplitude that is less than or equal to a preset amplitude threshold from the PET original scan data.

[0143] In this embodiment, since the respiratory motion signal corresponding to the respiratory motion amplitude less than or equal to the preset amplitude threshold is relatively stable, the computer device can use the PET scan data corresponding to the respiratory motion amplitude less than or equal to the preset amplitude threshold as the original data for image reconstruction.

[0144] S1004, reconstructing an image of the target part based on the PET target scanning data to obtain a reconstructed image of the target part.

[0145] In this embodiment, when the scanning device is a PET device, the computer device can use an image reconstruction algorithm to perform image reconstruction on the PET target scanning data to obtain a reconstructed image of the target part. The image reconstruction algorithm can be an iterative algorithm or an analytical algorithm.

[0146] In the above-mentioned image reconstruction method, the respiratory motion amplitude of the real respiratory motion signal is obtained; the respiratory motion amplitude is compared with a preset amplitude threshold; the PET target scan data corresponding to the respiratory motion amplitude less than the preset amplitude threshold is determined from the PET original scan data; the image of the target part is reconstructed based on the PET target scan data to obtain a reconstructed image of the target part. By comparing the respiratory motion amplitude with the preset amplitude threshold, the method can accurately screen out a relatively stable area from the real respiratory motion signal, and reconstruct the image of the target part based on the PET target scan data corresponding to the relatively stable area, so that the artifacts in the obtained reconstructed image are less affected and the quality of the reconstructed image is higher.

[0147] As a specific embodiment of the present application, Fig.11 As shown, the image reconstruction method includes:

[0148] S1101, obtaining a mask image of a target part in a preset mode;

[0149] S1102, determining PET raw scan data of a target part from the PET raw scan data based on the mask image;

[0150] S1103, determining the centroid movement information of the target part from the PET raw scan data of the target part, and determining the centroid movement information as the predicted respiratory motion signal of the user;

[0151] S1104, calculating the product of the breathing correction factor and the predicted breathing motion signal;

[0152] S1105, determining the product as the user's real respiratory motion signal;

[0153] S1106, acquiring multiple respiratory motion amplitudes of the real respiratory motion signal;

[0154] S1107, comparing each respiratory movement amplitude with a preset amplitude threshold;

[0155] S1108, determining PET target scan data corresponding to a respiratory motion amplitude that is less than or equal to a preset amplitude threshold from the PET raw scan data;

[0156] S1109, reconstructing the image of the target part based on the PET image corresponding to the PET target scanning data to obtain a reconstructed image of the target part.

[0157] It should be understood that, although the steps in the flowcharts involved in the above 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 is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may 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 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.

[0158] Based on the same inventive concept, the embodiment of the present application also provides an image reconstruction device for implementing the above-mentioned image reconstruction method. The implementation solution provided by the device to solve the problem is similar to the implementation solution recorded in the above-mentioned method, so the specific limitations in one or more image reconstruction device embodiments provided below can refer to the limitations on the image reconstruction method above, and will not be repeated here.

[0159] In one embodiment, Fig.12 As shown, an image reconstruction device is provided, comprising: a determination module 11, a correction module 12 and a reconstruction module 13, wherein:

[0160] A determination module 11, used to determine the user's predicted respiratory motion signal based on the PET raw scan data of the user's target part;

[0161] A correction module 12, used to correct the predicted respiratory motion signal by a respiratory correction factor to obtain a real respiratory motion signal of the user; the respiratory correction factor represents a mapping relationship between the real respiratory amplitude of the user and the predicted respiratory amplitude;

[0162] The reconstruction module 13 is used to perform image reconstruction according to the real respiratory motion signal and the PET original scanning data to obtain a reconstructed image of the target part.

[0163] In one embodiment, the above-mentioned image reconstruction device further includes: an acquisition module, an amplitude determination module and a processing module, wherein:

[0164] An acquisition module, used for acquiring sample PET raw scan data;

[0165] An amplitude determination module, used for determining a plurality of sample predicted breathing amplitudes and a plurality of sample actual breathing amplitudes according to the sample PET raw scan data;

[0166] The processing module is used to perform regression processing on the predicted breathing amplitude of each sample and the actual breathing amplitude of each sample to obtain a breathing correction factor.

[0167] In one embodiment, the amplitude determination module includes: a first determination unit, a division unit, and a second determination unit, wherein:

[0168] A first determination unit, used for determining a sample predicted respiratory motion signal according to the sample PET raw scan data;

[0169] A division unit, used for dividing the sample predicted respiratory motion signal to obtain the sample predicted respiratory motion signal under each respiratory gating;

[0170] The second determination unit is used to determine the sample predicted breathing amplitude and the sample actual breathing amplitude under each respiratory gating based on the sample predicted respiratory motion signal under each respiratory gating.

[0171] In one embodiment, the second determination unit is further used to obtain, for any respiratory gating, an average value of the sample predicted respiratory motion signal under respiratory gating; obtain the difference between the average value and the reference average value; and determine the difference as the sample predicted respiratory amplitude under respiratory gating.

[0172] In one embodiment, the second determination unit is also used to predict respiratory motion signals based on samples under each respiratory gating, determine each sample PET raw scan data under respiratory gating from the sample PET raw scan data; perform image reconstruction on the PET image corresponding to each sample PET raw scan data under respiratory gating to obtain multiple sample reconstructed images; and determine the actual respiratory amplitude of the sample under each respiratory gating based on each sample reconstructed image.

[0173] In one embodiment, the second determination unit is further used to align the sample reconstructed image with the reference reconstructed image for any respiratory gating to obtain an image deformation matrix under respiratory gating; and determine the average value of the image deformation matrix as the sample true respiratory amplitude under respiratory gating.

[0174] In one embodiment, the above-mentioned determination module includes: a first acquisition unit, a third determination unit and a fourth determination unit, wherein:

[0175] A first acquisition unit, used to acquire a mask image of a target part in a preset mode;

[0176] A third determining unit is used to determine the PET raw scan data of the target part from the PET raw scan data based on the mask image;

[0177] The fourth determination unit is used to determine the center of mass movement information of the target part from the PET raw scan data of the target part, and determine the center of mass movement information as the predicted respiratory motion signal of the user.

[0178] In one embodiment, the correction module comprises: a calculation unit and a fifth determination unit, wherein:

[0179] The calculation unit is used to calculate the product of the respiratory correction factor and the predicted respiratory motion signal.

[0180] The fifth determining unit is used to determine the product as the real respiratory motion signal of the user.

[0181] In one embodiment, the reconstruction module includes: a second acquisition unit, a comparison unit, a sixth determination unit and a reconstruction unit, wherein:

[0182] A second acquisition unit, used for acquiring multiple respiratory motion amplitudes of the real respiratory motion signal;

[0183] A comparison unit, used for comparing each respiratory movement amplitude with a preset amplitude threshold;

[0184] A sixth determining unit, configured to determine PET target scan data corresponding to a respiratory motion amplitude that is less than or equal to a preset amplitude threshold from the PET raw scan data;

[0185] The reconstruction unit is used to reconstruct the image of the target part based on the PET image corresponding to the PET target scanning data to obtain a reconstructed image of the target part.

[0186] In one embodiment, the reconstruction unit is also used to obtain an original medical image of the target part under a preset modality; and to reconstruct the image of the target part based on the PET image corresponding to the PET target scanning data and the original medical image to obtain a reconstructed image of the target part.

[0187] Each module in the above-mentioned 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 the operations corresponding to each module above.

[0188] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the content of any one embodiment of the above-mentioned image reconstruction method is implemented.

[0189] 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 content of any one embodiment of the above-mentioned image reconstruction method is implemented.

[0190] In one embodiment, a computer program product is provided, comprising a computer program, which implements the content of any one embodiment of the above-mentioned image reconstruction method when executed by a processor.

[0191] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0192] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods 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 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), magnetoresistive 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, etc., but are not limited to this.

[0193] 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 specification.

[0194] The above 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. An image reconstruction method, characterized in that: The method comprises: Determining a predicted respiratory motion signal of the user based on the PET raw scan data of the target part of the user; Correcting the predicted respiratory motion signal by a respiratory correction factor to obtain a real respiratory motion signal of the user; the respiratory correction factor represents a mapping relationship between the real respiratory amplitude and the predicted respiratory amplitude of the user; Image reconstruction is performed according to the real respiratory motion signal and the PET original scan data to obtain a reconstructed image of the target part.

2. The method according to claim 1, characterized in that The process of obtaining the breathing correction factor includes: Obtain sample PET raw scan data; Determining a plurality of sample predicted respiratory amplitudes and a plurality of sample actual respiratory amplitudes according to the sample PET raw scan data; Regression processing is performed on the predicted breathing amplitude of each sample and the actual breathing amplitude of each sample to obtain the breathing correction factor.

3. The method according to claim 2, characterized in that Determining a plurality of sample predicted respiratory amplitudes and a plurality of sample actual respiratory amplitudes according to the sample PET raw scan data includes: Determining a sample predicted respiratory motion signal according to the sample PET raw scan data; Dividing the sample predicted respiratory motion signal to obtain each sample predicted respiratory motion signal under respiratory gating; Based on the sample predicted respiratory motion signal under each respiratory gating, the sample predicted respiratory amplitude and the sample actual respiratory amplitude under each respiratory gating are determined.

4. The method according to claim 3, characterized in that Determining the predicted respiratory amplitude of each sample under the respiratory gating based on the predicted respiratory motion signal of each sample under the respiratory gating includes: For any respiratory gating, obtaining the average value of the sample predicted respiratory motion signal under the respiratory gating; Obtaining a difference between the average value and a reference average value; The difference is determined as the sample predicted respiratory amplitude under the respiratory gating.

5. The method according to claim 3 or 4, characterized in that: Predicting the respiratory motion signal based on the samples under the respiratory gating, determining the actual respiratory amplitude of the samples under the respiratory gating, including: Based on the predicted respiratory motion signal of each sample under respiratory gating, determine each sample PET raw scan data under respiratory gating from the sample PET raw scan data; Performing image reconstruction on each sample PET raw scan data under respiratory gating to obtain multiple sample reconstructed images; The real respiratory amplitude of each sample under respiratory gating is determined according to the reconstructed image of each sample.

6. The method according to claim 5, characterized in that The reconstructing the image according to each sample and determining the real respiratory amplitude of each sample under the respiratory gating includes: For any respiratory gating, registering the sample reconstructed image with the reference reconstructed image to obtain an image deformation matrix under the respiratory gating; The average value of the image deformation matrix is ​​determined as the true respiratory amplitude of the sample under the respiratory gating.

7. The method according to any one of claims 1 to 4, characterized in that: Determining the predicted respiratory motion signal of the user according to the PET raw scan data of the target part of the user includes: Acquire a mask image of the target part in a preset mode; Based on the mask image, determining the PET raw scan data of the target part from the PET raw scan data; The center of mass movement information of the target part is determined from the PET raw scan data of the target part, and the center of mass movement information is determined as the predicted respiratory motion signal of the user.

8. The method according to any one of claims 1 to 4, characterized in that: The step of correcting the predicted respiratory motion signal by using a respiratory correction factor to obtain a real respiratory motion signal of the user includes: Calculating the product of the breathing correction factor and the predicted respiratory motion signal; The product is determined as the actual respiratory motion signal of the user.

9. The method according to any one of claims 1 to 4, characterized in that: The step of performing image reconstruction according to the real respiratory motion signal and the PET raw scan data to obtain a reconstructed image of the target part includes: Acquiring multiple respiratory motion amplitudes of the real respiratory motion signal; comparing each of the respiratory movement amplitudes with a preset amplitude threshold; Determine PET target scan data corresponding to a respiratory motion amplitude that is less than or equal to the preset amplitude threshold from the PET raw scan data; An image of the target part is reconstructed based on the PET target scanning data to obtain a reconstructed image of the target part.

10. An image reconstruction device, characterized in that: The device comprises: A determination module, used to determine a predicted respiratory motion signal of the user based on the PET raw scan data of the target part of the user; A correction module, used to correct the predicted respiratory motion signal by a respiratory correction factor to obtain a real respiratory motion signal of the user; the respiratory correction factor represents a mapping relationship between the real respiratory amplitude and the predicted respiratory amplitude of the user; The reconstruction module is used to perform image reconstruction according to the real respiratory motion signal and the PET original scanning data to obtain a reconstructed image of the target part.

Images