Motion image processing method and apparatus, and cardiac image reconstruction processing method

Abstract

The present application relates to a motion image processing method and apparatus, and a cardiac image reconstruction processing method. The motion image processing method comprises: respectively acquiring a multi-temporal motion template image of a motion-controlled model in a target motion mode and an actual motion image of a motion-uncontrolled entity; and on the basis of a registration parameter obtained by means of the multi-temporal motion template image and the actual motion image, performing augmentation on the actual motion image, so as to obtain a multi-temporal motion image of the motion-uncontrolled entity in the target motion mode. The cardiac image reconstruction processing method comprises: acquiring target scanning data of a target heart; on the basis of a preset mode, determining a plurality of reconstruction center moments within an exposure duration; on the basis of the plurality of reconstruction center moments, reconstructing the target scanning data, so as to obtain cardiac images at the plurality of reconstruction center moments; and on the basis of the cardiac images at the plurality of reconstruction center moments, determining a cardiac image at a target reconstruction center moment.

Description

Method and device for processing moving image and method for reconstructing heart image

[0001] Related applications

[0002] The present application claims priority to the Chinese patent application No. 2024110749091, filed on August 6, 2024, and entitled "Image augmentation method and method for removing image motion artifact, device and equipment", the contents of which are hereby incorporated by reference in its entirety.

[0003] The present application claims priority to the Chinese patent application No. 2024113800757, filed on September 29, 2024, and entitled "Method for reconstructing heart image, method for determining phase of heart image and device", the contents of which are hereby incorporated by reference in its entirety. TECHNICAL FIELD

[0004] The present application relates to the technical field of image processing, in particular to a method and device for processing moving image, computer equipment, computer readable storage medium and computer program product, a method and device for reconstructing heart image, computer equipment, computer readable storage medium and computer program product, and a method and device for removing motion artifact of heart image, computer equipment, computer readable storage medium and computer program product. BACKGROUND

[0005] By means of CT (Computed Tomography) technology and MR (Magnetic Resonance) technology, a part (solid object, non-digital object) in a human or animal body can be scanned to obtain a corresponding image. In some scenarios (such as a model training scenario for removing motion artifacts), it may be necessary to obtain an image of the same part under a specific motion pattern, such as obtaining a gold standard in a model training set. However, since the human body is in motion at all times, it is difficult to accurately obtain the corresponding image of a part with non-autonomous motion, as the motion process of the part is not controlled.

[0006] For the imaging needs of the heart, a typical non-autonomous motion part, the scanning technology for heart imaging includes a post-gating scanning technology (retrospective ECG gating scanning technology) and a pre-gating scanning technology (prospective ECG triggering scanning technology). The two aforementioned scanning technologies for heart imaging have a high dependence on electrocardiogram waveforms in the processing process in order to reconstruct the heart image at the reconstruction phase required by the doctor. In the case of no electrocardiogram waveform or inaccurate electrocardiogram waveform, it is difficult to obtain the heart image at the reconstruction phase required by the doctor. SUMMARY

[0007] The application provides a moving image processing method, which comprises:

[0008] respectively acquiring a plurality of time-phase moving template images of the moving controlled model under a target moving mode, and a moving actual image of a moving uncontrolled entity;

[0009] augmenting the moving actual image based on registration parameters obtained from the plurality of time-phase moving template images and the moving actual image, to obtain a plurality of time-phase moving images of the moving uncontrolled entity under the target moving mode.

[0010] In one embodiment, augmenting the moving actual image based on registration parameters obtained from the plurality of time-phase moving template images and the moving actual image, to obtain a plurality of time-phase moving images of the moving uncontrolled entity under the target moving mode, comprises:

[0011] determining a target time-phase moving template image from the plurality of time-phase moving template images according to a part shape similarity between each time-phase moving template image in the plurality of time-phase moving template images and the moving actual image;

[0012] performing conversion processing on the plurality of time-phase moving template images based on registration parameters between the target time-phase moving template image and the moving actual image, to obtain the plurality of time-phase moving images.

[0013] In one embodiment, performing conversion processing on the plurality of time-phase moving template images based on registration parameters between the target time-phase moving template image and the moving actual image comprises:

[0014] performing conversion processing on each time-phase moving template image in the plurality of time-phase moving template images according to registration parameters between the target time-phase moving template image and the moving actual image.

[0015] In one embodiment, performing conversion processing on the plurality of time-phase moving template images based on registration parameters between the target time-phase moving template image and the moving actual image comprises:

[0016] performing conversion processing on the target time-phase moving template image according to registration parameters between the target time-phase moving template image and the moving actual image, to obtain a target time-phase moving image;

[0017] performing conversion processing on the target time-phase moving image according to registration parameters between the target time-phase moving template image and a non-target time-phase moving template image in the plurality of time-phase moving template images.

[0018] The application provides a method for removing image moving artifacts, which comprises:

[0019] inputting the image to be de-artifacted into the de-artifacting model to obtain a target image; the image to be de-artifacted includes at least one of the actual motion image and the multi-phase motion image;

[0020] The de-artifacting model is obtained in the following manner:

[0021] simulated acquisition and reconstruction based on the multi-phase motion image to obtain a multi-phase motion artifact image of the motion uncontrolled entity under the target motion mode;

[0022] model training based on the multi-phase motion image and the multi-phase motion artifact image to obtain the de-artifacting model.

[0023] In one embodiment, simulated acquisition and reconstruction based on the multi-phase motion image to obtain a multi-phase motion artifact image of the motion uncontrolled entity under the target motion mode, includes:

[0024] for each phase, simulated acquisition based on the motion image of the phase in the multi-phase motion image and a plurality of adjacent phase motion images to obtain mixed-phase simulated raw data associated with the phase;

[0025] reconstruction based on the mixed-phase simulated raw data associated with the phase to obtain the phase motion artifact image.

[0026] In one embodiment, simulated acquisition based on the motion image of the phase in the multi-phase motion image and a plurality of adjacent phase motion images to obtain mixed-phase simulated raw data associated with the phase, includes:

[0027] simulated acquisition of the motion image of the phase to obtain pure-phase simulated raw data of the motion image of the phase;

[0028] simulated acquisition of each adjacent phase motion image to obtain pure-phase simulated raw data of each adjacent phase motion image;

[0029] splicing of the pure-phase simulated raw data of the motion image of the phase and the pure-phase simulated raw data of each adjacent phase motion image to obtain mixed-phase simulated raw data associated with the phase.

[0030] The present application provides a motion image processing device, which comprises:

[0031] an image acquisition module configured to acquire a multi-phase motion template image of a motion controlled model under a target motion mode and an actual motion image of a motion uncontrolled entity, respectively;

[0032] An image augmentation module is configured to augment the actual motion image based on the registration parameters obtained from the multi-phase motion template image and the actual motion image, to obtain multi-phase motion images of the non-controlled entity under the target motion mode.

[0033] The application provides a computer device, comprising a memory and a processor, the memory stores a computer program, and the processor implements the image motion artifact removal method when executing the computer program.

[0034] The application provides a reconstruction processing method of a cardiac image, the method comprising:

[0035] Obtaining target scanning data of a target heart; the target scanning data is obtained by continuously exposing the target heart by a CT scanning device according to a set exposure time length;

[0036] According to a preset mode, a plurality of reconstruction center moments are determined in the exposure time length; the preset mode makes the number of the plurality of reconstruction center moments lower than a threshold value and the cardiac morphology difference degrees between cardiac images of adjacent reconstruction center moments all reach a difference degree threshold value;

[0037] According to the plurality of reconstruction center moments, the target scanning data is reconstructed to obtain cardiac images of the plurality of reconstruction center moments;

[0038] According to the cardiac images of the plurality of reconstruction center moments, a cardiac image of a target reconstruction center moment is determined.

[0039] In one embodiment, the plurality of reconstruction center moments are determined in the exposure time length according to the preset mode, comprising:

[0040] For any exposure time length proportion, sample scanning data matched with the exposure time length proportion is obtained, and whether the interval time length proportion is associated with the exposure time length proportion is determined according to whether the number of reconstruction center moments determined in the sample scanning data according to the interval time length proportion is lower than a number threshold value and whether the cardiac morphology difference degrees between cardiac images of adjacent reconstruction center moments all reach a difference degree threshold value, to obtain an association relationship between different exposure time length proportions and interval time length proportions;

[0041] According to the association relationship and a target exposure time length proportion, a target interval time length proportion is obtained; the target exposure time length proportion is a proportion of the exposure time length in a cardiac cycle of the target heart;

[0042] According to the target interval time length proportion and the cardiac cycle of the target heart, the plurality of reconstruction center moments are determined in the exposure time length.

[0043] In an embodiment, the determining the plurality of reconstruction center instants in the exposure time length according to the target interval time length ratio and the cardiac cycle of the target heart comprises:

[0044] According to the target interval time length ratio and the cardiac cycle of the target heart, a target interval time length is obtained.

[0045] According to the scanning data range required for reconstructing a single-instant cardiac image, a first reconstruction center instant is determined.

[0046] Taking the first reconstruction center instant as a sampling starting position, the exposure time length is sampled according to the target interval time length, and a plurality of reconstruction center instants are obtained.

[0047] In an embodiment, the exposure time length is determined in the following manner:

[0048] The heart rate of the target heart is obtained.

[0049] According to the heart rate and a set ratio, the exposure time length is determined.

[0050] In an embodiment, the cardiac image at the target reconstruction center instant is determined according to the cardiac images at the plurality of reconstruction center instants, comprising:

[0051] A preset evaluation algorithm is obtained.

[0052] According to the preset evaluation algorithm, the cardiac image at the target reconstruction center instant is determined from the cardiac images at the plurality of reconstruction center instants.

[0053] In an embodiment, the method further comprises:

[0054] If it is determined that the cardiac image at the target reconstruction center instant needs to be motion corrected, the cardiac image at the target reconstruction center instant is motion corrected to obtain a target image.

[0055] In an embodiment, the method further comprises:

[0056] According to the cardiac images at the plurality of reconstruction center instants, cardiac cavity volumes at the plurality of reconstruction center instants are determined.

[0057] According to the corresponding relationship between a virtual cardiac cavity volume change curve and a virtual electrocardiogram waveform, and the cardiac cavity volumes at the plurality of reconstruction center instants, positions corresponding to the reconstruction center instants on the virtual electrocardiogram waveform are determined.

[0058] According to the positions, reconstruction phases corresponding to the cardiac images at the reconstruction center instants are determined.

[0059] In one embodiment, the corresponding positions of the reconstruction center moments on the virtual electrocardiogram waveform are determined according to the correspondence between the virtual heart chamber volume change curve and the virtual electrocardiogram waveform, and the heart chamber volumes of the reconstruction center moments, including:

[0060] The first reconstruction center moment heart chamber volume is determined from the heart chamber volumes of the reconstruction center moments.

[0061] The corresponding position of the first reconstruction center moment on the virtual electrocardiogram waveform is determined according to the correspondence between the virtual heart chamber volume change curve and the virtual electrocardiogram waveform, and the first reconstruction center moment heart chamber volume.

[0062] The corresponding position of the intermediate reconstruction center moment on the virtual electrocardiogram waveform is determined according to the corresponding position of the first reconstruction center moment on the virtual electrocardiogram waveform, and the interval duration between the intermediate reconstruction center moment and the first reconstruction center moment.

[0063] In one embodiment, the corresponding positions of the reconstruction center moments on the virtual electrocardiogram waveform are determined according to the correspondence between the virtual heart chamber volume change curve and the virtual electrocardiogram waveform, and the heart chamber volumes of the reconstruction center moments, including:

[0064] The last reconstruction center moment heart chamber volume is determined from the heart chamber volumes of the reconstruction center moments.

[0065] The corresponding position of the last reconstruction center moment on the virtual electrocardiogram waveform is determined according to the correspondence between the virtual heart chamber volume change curve and the virtual electrocardiogram waveform, and the last reconstruction center moment heart chamber volume.

[0066] The corresponding position of the intermediate reconstruction center moment on the virtual electrocardiogram waveform is determined according to the corresponding position of the last reconstruction center moment on the virtual electrocardiogram waveform, and the interval duration between the intermediate reconstruction center moment and the last reconstruction center moment.

[0067] In one embodiment, if the reconstruction phase includes an absolute reconstruction phase, the virtual electrocardiogram waveform acquisition step includes:

[0068] Acquiring a standard electrocardiogram waveform;

[0069] According to the target heart cardiac cycle, the standard electrocardiogram waveform is compressed or stretched to obtain a virtual electrocardiogram waveform.

[0070] The application provides a computer device, including a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned heart image reconstruction processing method when executing the computer program.

[0071] The application provides a method for removing motion artifacts of a cardiac image, and the method comprises the following steps:

[0072] obtaining target scanning data of a target heart, wherein the target scanning data is obtained by continuously exposing the target heart to a CT scanning device for a preset exposure time;

[0073] determining a plurality of reconstruction center moments in the exposure time according to a preset mode, wherein the preset mode is configured to make the number of the plurality of reconstruction center moments lower than a threshold value and the cardiac morphology difference between cardiac images of adjacent reconstruction center moments reach a difference threshold value;

[0074] reconstructing the target scanning data according to the plurality of reconstruction center moments to obtain cardiac images of the plurality of reconstruction center moments;

[0075] determining a cardiac image of a target reconstruction center moment according to the cardiac images of the plurality of reconstruction center moments;

[0076] inputting the cardiac image of the target reconstruction center moment into a de-artifact model to obtain a target image, wherein the de-artifact model is configured to remove at least one of the actual motion image and the multi-phase motion image;

[0077] wherein the de-artifact model is obtained in the following manner:

[0078] obtaining a multi-phase motion image obtained by the motion image processing method;

[0079] performing simulation acquisition and reconstruction based on the multi-phase motion image to obtain a multi-phase motion artifact image of a motion uncontrolled entity under the target motion mode;

[0080] performing model training based on the multi-phase motion image and the multi-phase motion artifact image to obtain the de-artifact model.

[0081] In one embodiment, the simulation acquisition and reconstruction based on the multi-phase motion image to obtain a multi-phase motion artifact image of a motion uncontrolled entity under the target motion mode comprises:

[0082] for each phase, performing simulation acquisition based on the motion image of the phase and a plurality of adjacent phase motion images in the multi-phase motion image to obtain mixed-phase simulation raw data associated with the phase;

[0083] performing reconstruction based on the mixed-phase simulation raw data associated with the phase to obtain the motion artifact image of the phase.

[0084] In one embodiment, the simulation acquisition based on the motion image of the phase and a plurality of adjacent phase motion images in the multi-phase motion image to obtain mixed-phase simulation raw data associated with the phase comprises:

[0085] The phase motion image is simulated to obtain pure-phase simulated raw data of the phase motion image;

[0086] Each adjacent phase motion image is simulated to obtain pure-phase simulated raw data of each adjacent phase motion image;

[0087] The pure-phase simulated raw data of the phase motion image and the pure-phase simulated raw data of each adjacent phase motion image are spliced to obtain mixed-phase simulated raw data associated with the phase.

[0088] In one embodiment, the plurality of reconstruction center moments in the exposure duration are determined according to a preset manner, comprising:

[0089] For any exposure duration proportion, sample scanning data matched with the exposure duration proportion is obtained, and whether the number of reconstruction center moments determined in the sample scanning data according to the interval duration proportion is lower than the number threshold and whether the cardiac morphology difference between adjacent reconstruction center moments reaches the difference threshold, to determine whether the interval duration proportion is associated with the exposure duration proportion, to obtain the association relationship between different exposure duration proportions and interval duration proportions;

[0090] According to the association relationship and a target exposure duration proportion, a target interval duration proportion is obtained; the target exposure duration proportion is the proportion of the exposure duration in the cardiac cycle of the target heart;

[0091] According to the target interval duration proportion and the cardiac cycle of the target heart, a plurality of reconstruction center moments in the exposure duration are determined.

[0092] In one embodiment, the plurality of reconstruction center moments in the exposure duration are determined according to the target interval duration proportion and the cardiac cycle of the target heart, comprising:

[0093] According to the target interval duration proportion and the cardiac cycle of the target heart, a target interval duration is obtained;

[0094] According to the scanning data range required for reconstructing a single-moment cardiac image, a first reconstruction center moment is determined;

[0095] Taking the first reconstruction center moment as a sampling starting position, the exposure duration is sampled according to the target interval duration to obtain a plurality of reconstruction center moments.

[0096] In one embodiment, the determination manner of the exposure duration comprises:

[0097] The heart rate of the target heart is obtained;

[0098] determine the exposure duration according to the heart rate and a set ratio.

[0099] In one embodiment, the method further comprises:

[0100] obtaining a preset evaluation algorithm;

[0101] determining the cardiac image at the target reconstruction center time from the cardiac images at the plurality of reconstruction center times according to the preset evaluation algorithm.

[0102] In one embodiment, the method further comprises:

[0103] determining the cardiac cavity volume at each reconstruction center time according to the cardiac images at the plurality of reconstruction center times;

[0104] determining the position corresponding to each reconstruction center time on the virtual electrocardiogram waveform according to the correspondence between the virtual cardiac cavity volume change curve and the virtual electrocardiogram waveform and the cardiac cavity volume at the plurality of reconstruction center times;

[0105] determining the reconstruction phase corresponding to the cardiac image at each reconstruction center time according to the position.

[0106] In one embodiment, the method further comprises:

[0107] determining the cardiac cavity volume at the first reconstruction center time from the cardiac cavity volumes at the plurality of reconstruction center times;

[0108] determining the position corresponding to the first reconstruction center time on the virtual electrocardiogram waveform according to the correspondence between the virtual cardiac cavity volume change curve and the virtual electrocardiogram waveform and the cardiac cavity volume at the first reconstruction center time;

[0109] determining the position corresponding to the intermediate reconstruction center time on the virtual electrocardiogram waveform according to the position corresponding to the first reconstruction center time on the virtual electrocardiogram waveform and the interval duration between the intermediate reconstruction center time and the first reconstruction center time.

[0110] In one embodiment, the method further comprises:

[0111] determining the cardiac cavity volume at the last reconstruction center time from the cardiac cavity volumes at the plurality of reconstruction center times;

[0112] determining a position corresponding to the last reconstruction center time on the virtual electrocardiogram waveform according to the corresponding relationship between the virtual heart chamber volume variation curve and the virtual electrocardiogram waveform and the chamber volume at the last reconstruction center time;

[0113] determining a position corresponding to the intermediate reconstruction center time on the virtual electrocardiogram waveform according to the position corresponding to the last reconstruction center time on the virtual electrocardiogram waveform and the interval length between the intermediate reconstruction center time and the last reconstruction center time.

[0114] In one embodiment, if the reconstruction time phase includes an absolute reconstruction time phase, the obtaining step of the virtual electrocardiogram waveform includes:

[0115] obtaining a standard electrocardiogram waveform;

[0116] compressing or stretching the standard electrocardiogram waveform to obtain the virtual electrocardiogram waveform according to the cardiac cycle of the target heart.

[0117] The present application provides a computer device, comprising a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned method for removing motion artifacts of a heart image when executing the computer program. BRIEF DESCRIPTION OF DRAWINGS

[0118] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor based on the disclosed drawings.

[0119] Fig. 1 is a flowchart of a motion image processing method in one embodiment;

[0120] Fig. 2 is a flowchart of a method for removing image motion artifacts in one embodiment;

[0121] Fig. 3 is a schematic diagram of the overall flow of a motion image processing method in one embodiment;

[0122] Fig. 4 is a structural block diagram of a motion image processing device in one embodiment;

[0123] Fig. 5 is an internal structure diagram of a computer device in one embodiment;

[0124] Fig. 6 is a flowchart of a reconstruction processing method for a heart image in one embodiment;

[0125] Fig. 7 is a flowchart of a reconstruction processing method for a heart image in another embodiment;

[0126] FIG. 8 is a schematic diagram of the correspondence between the volume change curve of the central chamber and the electrocardiogram waveform in an embodiment;

[0127] FIG. 9 is a schematic diagram of the flow of the method provided by the present application in an embodiment;

[0128] FIG. 10 is a schematic diagram of the cardiac images at multiple reconstruction center times in an embodiment. DETAILED DESCRIPTION

[0129] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by those skilled in the art without any creative work fall within the scope of protection of the present application.

[0130] The phrase "embodiment" mentioned in the present application means that the specific features, structures or characteristics described in connection with the embodiment can be included in at least one embodiment of the present application. The appearance of this phrase at various places in the specification does not necessarily mean the same embodiment, nor is it an independent or alternative embodiment to other embodiments. Those skilled in the art explicitly and implicitly understand that the embodiments described in the present application can be combined with other embodiments.

[0131] The motion image processing method provided by the present application can be executed by a computer device, including the steps shown in FIG. 1:

[0132] In step S101, the multi-time phase motion template images of the motion controlled model under the target motion mode and the motion actual images of the motion uncontrolled entity are acquired respectively.

[0133] The motion controlled model and the motion uncontrolled entity need to belong to the same part, for example, both belong to the heart; the motion controlled model and the motion uncontrolled entity do not need to come from the same individual. A motion controlled model can be set up, which can be used for different individuals to obtain the multi-time phase motion images of the corresponding motion uncontrolled entity under the target motion mode.

[0134] The motion non-controlled entity can refer to an entity in a natural state, whose motion process is spontaneously regulated by internal physiological mechanism, autonomous rhythm or environmental stimulation, etc., and is not controlled by external active intervention or preset program, for example, the human heart is a typical motion non-controlled entity, the actual motion process of the contraction and diastole of the human heart is not controlled by the outside, and the natural periodic phase change is presented. The motion controlled model can refer to a simulation system constructed by computer algorithm, physical device, etc., which can actively regulate the motion state according to the preset parameters, logical rules or external instructions. The motion controlled model can simulate the motion characteristics of the target entity, and the motion process of the motion controlled model has predictability and adjustability, and by intervening the variables of the motion controlled model, the specific motion effect of the motion non-controlled entity can be simulated. The motion controlled model can include a heart model simulating the multi-phase motion of a real heart. Taking the heart as an example, a user can adjust the periodic motion amplitude, motion frequency, etc. of the heart model (motion controlled model) through a computer device, so that the heart model is in a target motion mode, and in this state, the multi-phase motion image of the heart model is obtained. In order to distinguish, the multi-phase motion image of the heart model obtained at this time is called a multi-phase motion template image.

[0135] After CT or MR scanning of the heart (motion non-controlled entity) of a certain individual, the motion image of the heart of the individual in at least a certain phase can be obtained, which is called a motion actual image.

[0136] In step S102, the motion actual image is augmented based on the registration parameters obtained from the multi-phase motion template image and the motion actual image, and the multi-phase motion image of the motion non-controlled entity in the target motion mode is obtained.

[0137] The registration parameters can be obtained according to the multi-phase motion template image and the motion actual image. The augmented image obtained according to the registration parameters is used as the multi-phase motion image of the motion non-controlled entity. The motion actual image belongs to an image obtained by real scanning, and the augmented image belongs to an image obtained by non-real scanning. Therefore, the multi-phase motion image can also be called a multi-phase motion virtual image.

[0138] Since the multi-phase motion template image is an image of the motion controlled model in the target motion mode, the multi-phase motion image obtained by augmenting the motion actual image based on the multi-phase motion template image presents the target motion mode.

[0139] Correspondingly, when the multi-phase motion template image is a full-phase motion template image, the obtained multi-phase motion image can be a full-phase motion image.

[0140] In the motion image processing method, the motion controlled model is a digital object, a non-human or animal entity object, and its motion process is controllable. Therefore, the motion process of the motion controlled model is regulated to obtain a multi-phase motion template image of the motion controlled model under a target motion mode. The multi-phase motion template image and the motion actual image of the non-controlled entity are registered to obtain a registration parameter. The motion actual image is augmented based on the registration parameter to obtain a multi-phase motion image of the non-controlled entity under the target motion mode. According to the scheme provided in the application, an image of a part of a human or animal body affected by non-autonomic motion under a specific motion mode can be obtained.

[0141] In one embodiment, the multi-phase motion image of the non-controlled entity under the target motion mode is obtained by augmenting the motion actual image based on the registration parameter obtained from the multi-phase motion template image and the motion actual image. Specifically, the steps can include the following: determining a target phase motion template image from the multi-phase motion template image according to the part shape similarity between each phase motion template image in the multi-phase motion template image and the motion actual image; and converting the multi-phase motion template image based on the registration parameter between the target phase motion template image and the motion actual image to obtain the multi-phase motion image.

[0142] After obtaining the multi-phase motion template image, the part shape similarity between each phase motion template image and the motion actual image can be calculated. The motion template image with the highest part shape similarity in the multi-phase motion template image is determined, and the motion template image with the highest part shape similarity is used as the target phase motion template image. The images other than the target phase motion template image in the multi-phase motion template image are non-target phase motion template images.

[0143] Then, the target phase motion template image and the motion actual image can be registered using various registration methods (such as point cloud registration, image optical flow registration, deep learning-based registration, etc.) to obtain a registration parameter (the registration parameter can include a registration vector field, which can be denoted as Θ).

[0144] The multi-phase motion template image is converted based on the registration parameter to complete the augmentation of the motion actual image, and the multi-phase motion image of the non-controlled entity under the target motion mode is obtained.

[0145] In the above embodiment, the target phase motion template image is determined according to the part shape similarity between each phase motion template image and the motion actual image, and more accurate registration parameters are obtained to obtain more accurate multi-phase motion images.

[0146] In one embodiment, the conversion processing of the multi-time-phase motion template images based on the registration parameters between the target time-phase motion template image and the motion actual image comprises: performing conversion processing on each time-phase motion template image in the multi-time-phase motion template images according to the registration parameters between the target time-phase motion template image and the motion actual image.

[0147] For example, the multi-time-phase motion template images comprise {m1, m2, m3, m4, m5, m6, m7, m8, m9, m 10} and the motion actual image is denoted as n.

[0148] If the target time-phase motion template image is m3, the registration parameters between m3 and n can be calculated and applied to each time-phase motion template image in {m1, m2, m4, m5, m6, m7, m8, m9, m 10 , so that {m'1, m'2, m'3, m'4, m'5, m'6, m'7, m'8, m'9, m 10} can be obtained. {m'1, m'2, m'3, m'4, m'5, m'6, m'7, m'8, m'9, m 10} can be used as the multi-time-phase motion images of the motion uncontrolled entity under the target motion mode.

[0149] In this embodiment, the conversion processing of each time-phase motion template image in the multi-time-phase motion template images according to the registration parameters between the target time-phase motion template image and the motion actual image can quickly obtain the multi-time-phase motion images and reduce the processing complexity.

[0150] In one embodiment, the conversion processing of the multi-time-phase motion template images based on the registration parameters between the target time-phase motion template image and the motion actual image comprises: performing conversion processing on the target time-phase motion template image according to the registration parameters between the target time-phase motion template image and the motion actual image to obtain a target time-phase motion image; and performing conversion processing on the target time-phase motion image according to the registration parameters between the target time-phase motion template image and the non-target time-phase motion template images in the multi-time-phase motion template images.

[0151] For example, the multi-time-phase motion template images comprise {m1, m2, m3, m4, m5, m6, m7, m8, m9, m 10} and the motion actual image is denoted as n.

[0152] If the target time-phase motion template image is m3, the registration parameters between m3 and n can be calculated and applied to m3, so that m'3 can be obtained as the target time-phase motion image.

[0153] Since m3 is a target time-phase motion template image, {m1, m2, m4, m5, m6, m7, m8, m9, m 10} are non-target time-phase motion template images. A registration parameter between each non-target time-phase motion template image in {m1, m2, m4, m5, m6, m7, m8, m9, m 10} and m3 can be calculated, and each registration parameter is applied to m’3 to obtain {m’1, m’2, m’4, m’5, m’6, m’7, m’8, m’9, m’ 10}; for example, a registration parameter between m1 and m3 can be calculated, and the registration parameter is applied to m’3 to obtain m’1.

[0154] According to {m’1, m’2, m’4, m’5, m’6, m’7, m’8, m’9, m’ 10} and m’3, {m’1, m’2, m’3, m’4, m’5, m’6, m’7, m’8, m’9, m’ 10} can be obtained, and {m’1, m’2, m’3, m’4, m’5, m’6, m’7, m’8, m’9, m’ 10} is a multi-time-phase motion image of a motion non-controlled entity under a target motion mode.

[0155] In this embodiment, the target time-phase motion template image is converted according to the registration parameter between the target time-phase motion template image and the motion actual image to obtain a target time-phase motion image; and the target time-phase motion image is converted according to the registration parameter between the target time-phase motion template image and the non-target time-phase motion template image in the multi-time-phase motion template image, so that the processing flexibility can be improved.

[0156] In one embodiment, the motion image processing method includes the steps shown in FIG. 2:

[0157] In step S201, a to-be-removed-artifact image is input into a de-artifact model to obtain a target image; the to-be-removed-artifact image includes at least one of the motion actual image and the multi-time-phase motion image;

[0158] The de-artifact model is obtained in the following manner:

[0159] The motion non-controlled entity under the target motion mode is simulated to obtain a multi-time-phase motion artifact image of the motion non-controlled entity under the target motion mode;

[0160] The model is trained according to the multi-time-phase motion image and the multi-time-phase motion artifact image to obtain the de-artifact model.

[0161] After determining that a certain image is to be de-artifacted, the image can be referred to as a to-be-de-artifacted image. The to-be-de-artifacted image generated by uncontrollable motion of a human body can be input into a de-artifacting model. For example, the motion actual image before augmentation can be input into the de-artifacting model, or the multi-time-phase motion image after augmentation can be input into the de-artifacting model. The image output by the de-artifacting model can be taken as a target image, and the target image can be further processed. For example, the motion actual image before augmentation can be input into the de-artifacting model, and the obtained target image can be augmented. Compared with the to-be-de-artifacted image, the target image has a lighter motion artifact. The artifact generated by non-autonomous motion includes, for example, breathing motion that is difficult for some patients to control when scanning a certain part, and for example, the beating of a heart when scanning the heart.

[0162] Taking the scenario of CT scanning of a heart as an example, because the beating of the heart causes the CT scanned image to have a relatively serious artifact, the CT scanned image, i.e., the motion actual image of the uncontrollable entity, can be taken as a to-be-de-artifacted image, and the to-be-de-artifacted image can be input into a de-artifacting model to obtain a target image with a lighter artifact.

[0163] Exemplarily, taking any time phase in the multi-time phase as an example, for the time phase x, the time phase x motion image can be determined in the multi-time-phase motion image, the time phase x motion image can be taken as a gold standard image, the time phase x motion artifact image can be determined in the multi-time-phase motion artifact image, and the time phase x motion artifact image can be taken as an image input into the model. The training data includes the time phase x motion image taken as the gold standard image and the time phase x motion artifact image taken as the image input into the model. According to the model training based on the training data, the de-artifacting model can be obtained. Through the above pairing training corresponding to the time phase, the model can accurately learn the mapping relationship between the artifact features unique to each time phase and the real structure, and the accuracy of the model in removing the artifact of a specific time phase is improved.

[0164] The image involved in the embodiment can belong to a CT modality, an MR modality, or a PET (Positron Emission Computed Tomography) modality.

[0165] In the embodiment, the multi-time-phase motion image of the uncontrollable entity in the target motion mode can be obtained by means of the foregoing motion image processing method. The multi-time-phase motion artifact image of the uncontrollable entity in the target motion mode can be obtained based on the simulated acquisition and reconstruction of the multi-time-phase motion image. The de-artifacting model with better de-artifacting performance can be obtained by taking the multi-time-phase motion artifact image as part of the training data, and the removal effect of the motion artifact is improved.

[0166] In one embodiment, based on the multi-phase motion image, simulated acquisition and reconstruction are performed to obtain a multi-phase motion artifact image of the motion uncontrolled entity under the target motion mode. Specifically, for each phase, based on the motion image of the phase and a plurality of adjacent phase motion images in the multi-phase motion image, simulated acquisition is performed to obtain mixed-phase simulated raw data associated with the phase; and based on the mixed-phase simulated raw data associated with the phase, reconstruction is performed to obtain the phase motion artifact image.

[0167] In medical scanning imaging technology, real scanning acquisition is performed to obtain data, for example, CT scanning, MR scanning or PET scanning, and the data obtained by real scanning acquisition is reconstructed to obtain a real image. The data obtained by real scanning can be referred to as actual acquired raw data. The process of simulating the foregoing real scanning acquisition is taken as an example of simulated acquisition, and the data obtained by simulated acquisition can be referred to as simulated raw data. In CT modality medical images, simulated acquisition can involve CT forward projection mode, and in MRI modality medical images, simulated acquisition can involve nuclear magnetic Fourier transform mode.

[0168] Taking phase y as an example, the motion images corresponding to the phases adjacent to phase y in the multi-phase motion image are referred to as a plurality of adjacent phase motion images of phase y. The motion image of phase y can be determined in the multi-phase motion image, and phases adjacent to y can be determined with phase y as the center. For example, phases y-5 to y-1 and phases y+1 to y+5 can be taken as phases adjacent to y.

[0169] In the motion image of phase y and the plurality of adjacent phase motion images of phase y, simulated acquisition is respectively performed on each phase motion image, and simulated raw data corresponding to each phase motion image can be obtained. Since such simulated raw data only corresponds to a single specific phase and does not contain information of other phases, it can be referred to as pure-phase simulated raw data.

[0170] For the motion image of phase y and the plurality of adjacent phase motion images of phase y, the pure-phase simulated raw data corresponding to each phase motion image is integrated to obtain simulated raw data. Since the simulated raw data contains information of a plurality of phases, it can be referred to as mixed-phase simulated raw data. Since the mixed-phase simulated raw data is obtained according to the motion image of phase y and its adjacent motion images, the mixed-phase simulated raw data can be associated with phase y.

[0171] After obtaining the mixed-phase simulated raw data associated with phase y, reconstruction can be performed. Since the mixed-phase simulated raw data contains information of a plurality of phases, the image obtained by reconstruction has serious motion artifacts, and the image obtained by reconstruction can be taken as a motion artifact image of phase y.

[0172] In the above embodiment, for each phase, based on the phase motion image and a plurality of adjacent phase motion images in the multi-phase motion image, simulated acquisition is performed to obtain mixed-phase simulated raw data associated with the phase; based on the mixed-phase simulated raw data associated with the phase, reconstruction is performed to obtain the phase motion artifact image with more serious motion artifacts, which is used as training data to train the de-artifact model.

[0173] In one embodiment, based on the phase motion image and a plurality of adjacent phase motion images in the multi-phase motion image, simulated acquisition is performed to obtain mixed-phase simulated raw data associated with the phase, which can specifically include: performing simulated acquisition on the phase motion image to obtain pure-phase simulated raw data of the phase motion image; performing simulated acquisition on each adjacent phase motion image to obtain pure-phase simulated raw data of each adjacent phase motion image; and splicing the pure-phase simulated raw data of the phase motion image and the pure-phase simulated raw data of each adjacent phase motion image to obtain the mixed-phase simulated raw data associated with the phase.

[0174] Taking the phase y and its adjacent phases y-5 to y-1 and y+1 to y+5 as an example:

[0175] For each phase in [y-5, y+5], simulated acquisition can be performed on each phase motion image to obtain pure-phase simulated raw data of each phase in [y-5, y+5]. Then, part of data is extracted from the pure-phase simulated raw data of each phase in [y-5, y+5], respectively. The extracted data can be referred to as sub pure-phase simulated raw data. For example, part of data is extracted from the pure-phase simulated raw data corresponding to y-5, which is referred to as sub pure-phase simulated raw data corresponding to y-5. Part of data is extracted from the pure-phase simulated raw data corresponding to y-4, which is referred to as sub pure-phase simulated raw data corresponding to y-4. After obtaining the sub pure-phase simulated raw data corresponding to each phase in [y-5, y+5] in this way, these sub pure-phase simulated raw data are spliced to obtain mixed-phase simulated raw data associated with the phase y.

[0176] In one embodiment, the pure-phase simulated raw data of the phase motion image and the pure-phase simulated raw data of each adjacent phase motion image are spliced to obtain mixed-phase simulated raw data associated with the phase, which can specifically include: sub pure-phase simulated raw data of a corresponding data acquisition period is extracted from the pure-phase simulated raw data of each phase motion image in the order of the phases; the earlier the phase, the earlier the corresponding data acquisition period; and the extracted sub pure-phase simulated raw data are spliced to obtain the mixed-phase simulated raw data associated with the phase.

[0177] In medical scanning imaging technology, a real scan is a scan lasting for a period of time. Since the simulation acquisition simulates the foregoing scan process, the simulation acquisition process also simulates a scan process lasting for a period of time first, and then acquires data formed in the scan process. The simulation acquisition obtained simulation raw data are different in acquisition time, for example, part of the data in the simulation raw data corresponds to a data acquisition time period of t0 to t1, and part of the data corresponds to a data acquisition time period of t1 to t2.

[0178] Exemplarily, after obtaining pure-phase simulation raw data of each phase in [y-5, y+5], the data acquisition time corresponding to the simulation acquisition can be divided according to the number of phases contained in [y-5, y+5] to obtain a plurality of data acquisition time periods, for example, [t0, t1], [t1, t2], [t2, t3], [t3, t4], [t4, t5], [t5, t6], [t6, t7], [t7, t8], [t8, t9], [t9, t10], [t10, t11], and the data acquisition time periods are increasingly late.

[0179] From y-5 to y+5, the phases gradually become later, and therefore, the data of the data acquisition time period [t0, t1] is extracted from the pure-phase simulation raw data corresponding to y-5, the part of data is referred to as the sub pure-phase simulation raw data corresponding to y-5, the data of the data acquisition time period [t1, t2] is extracted from the pure-phase simulation raw data corresponding to y-4, the part of data is referred to as the sub pure-phase simulation raw data corresponding to y-4, the data of the data acquisition time period [t2, t3] is extracted from the pure-phase simulation raw data corresponding to y-3, the part of data is referred to as the sub pure-phase simulation raw data corresponding to y-3, the data of the data acquisition time period [t3, t4] is extracted from the pure-phase simulation raw data corresponding to y-2, the part of data is referred to as the sub pure-phase simulation raw data corresponding to y-2, and in this way, the sub pure-phase simulation raw data corresponding to y-1 to y+5 can be obtained.

[0180] After obtaining the sub pure-phase simulation raw data corresponding to y-5 to y+5, the sub pure-phase simulation raw data can be spliced, and the splicing result is used as the mixed-phase simulation raw data associated with the phase y.

[0181] In order to better understand the foregoing method, an application example of the method of the present application is described in detail below. In the application example, a part of the type of heart is taken as an example, and accordingly, the motion controlled model is referred to as a heart model, and the running uncontrolled entity is the heart of a certain individual (individual A) in the application example.

[0182] Referring to FIG. 3, a full-time phase motion template image of the heart model under the target motion mode and a motion actual image of the heart of individual A at a certain time phase are acquired; a part shape similarity of each time phase motion template image in the full-time phase motion template image and the motion actual image is calculated, and a motion template image with the highest part shape similarity in the full-time phase motion template image is taken as a target time phase motion template image. The motion actual image and the target time phase motion template image are registered to obtain a registration vector field, and the registration vector field is applied to the full-time phase motion template image to obtain a full-time phase motion image of the heart of individual A under the target motion mode.

[0183] The process of acquiring the motion artifact image is introduced by taking time phase y and time phases adjacent to y as examples, i.e., y-5 to y-1 and y+1 to y+5:

[0184] In the full-time phase motion image, the motion images of each time phase in [y-5, y+5] are determined, and each time phase motion image is simulated and acquired to obtain the pure phase simulated projection data corresponding to each time phase. According to the order of time phases, the sub pure phase simulated projection data of the corresponding data acquisition period is extracted from the pure phase simulated projection data corresponding to each time phase. The earlier the time phase is, the earlier the corresponding data acquisition period is. For example, the data of the data acquisition period [t0, t1] is extracted from the pure phase simulated projection data corresponding to y-5, and this part of data is referred to as the sub pure phase simulated projection data corresponding to y-5. The data of the data acquisition period [t1, t2] is extracted from the pure phase simulated projection data corresponding to y-4, and this part of data is referred to as the sub pure phase simulated projection data corresponding to y-4. In this way, after the sub pure phase simulated projection data corresponding to each of [y-5, y+5] is obtained, these sub pure phase simulated projection data can be spliced, and the splicing result is taken as the mixed phase simulated projection data associated with time phase y. The mixed phase simulated projection data associated with time phase y is reconstructed to obtain the motion artifact image corresponding to time phase y.

[0185] The motion image of time phase y in the full-time phase motion image is taken as the gold standard image, and the motion artifact image corresponding to time phase y is taken as the input model image. In this way, the model is trained to obtain the de-artifact model.

[0186] Since the motion-controlled model is a digitalized object, a non-human or animal entity object, and its motion process is controllable, the application instance regulates the motion process of the motion-controlled model to obtain a multi-phase motion template image of the motion-controlled model under a target motion mode, performs augmentation on the motion actual image according to the registration parameters obtained from the multi-phase motion template image and the motion actual image of the motion-uncontrolled entity, and obtains a multi-phase motion image of the motion-uncontrolled entity under the target motion mode. According to the scheme provided in the application, an image of a part of a human or animal body affected by non-autonomic motion under a specific motion mode can be obtained. In addition, simulation acquisition and reconstruction are performed with the aid of the multi-phase motion image of the motion-uncontrolled entity under the target motion mode to obtain a multi-phase motion artifact image of the motion-uncontrolled entity under the target motion mode. In this way, model training is performed to obtain a better de-artifact model, and the removal effect of motion artifacts is improved.

[0187] It should be understood that, although each step in the flowchart involved in each embodiment as described above is displayed in sequence according to the indication of the arrow, these steps are not necessarily executed in sequence according to the indication of the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other sequences. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or steps or stages in other steps.

[0188] Based on the same inventive concept, the embodiments of the present application also provide a motion image processing device for implementing the motion image processing method described above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more motion image processing device embodiments provided below can refer to the limitations of the motion image processing method in the above text, which will not be repeated here.

[0189] In one embodiment, as shown in FIG. 4, a motion image processing device is provided, which includes:

[0190] The image acquisition module 401 is configured to acquire a multi-phase motion template image of a motion-controlled model under a target motion mode and a motion actual image of a motion-uncontrolled entity, respectively.

[0191] The image augmentation module 402 is configured to augment the actual motion image based on the registration parameters obtained from the multi-phase motion template image and the actual motion image, to obtain a multi-phase motion image of the motion non-controlled entity under the target motion mode.

[0192] In an embodiment, the image augmentation module 402 is further configured to: determine a target phase motion template image from the multi-phase motion template image according to a part shape similarity between each phase motion template image in the multi-phase motion template image and the actual motion image; and convert the multi-phase motion template image based on the registration parameters between the target phase motion template image and the actual motion image, to obtain the multi-phase motion image.

[0193] In an embodiment, the image augmentation module 402 is further configured to: convert each phase motion template image in the multi-phase motion template image according to the registration parameters between the target phase motion template image and the actual motion image.

[0194] In an embodiment, the image augmentation module 402 is further configured to: convert the target phase motion template image according to the registration parameters between the target phase motion template image and the actual motion image, to obtain a target phase motion image; and convert the target phase motion image according to the registration parameters between the target phase motion template image and a non-target phase motion template image in the multi-phase motion template image.

[0195] In an embodiment, the motion image processing apparatus further comprises:

[0196] The artifact removal module 403 is configured to input a to-be-artifact-removed image into an artifact removal model to obtain a target image, wherein the to-be-artifact-removed image comprises at least one of the actual motion image and the multi-phase motion image.

[0197] The model training module 404 is configured to: simulate acquisition and reconstruction based on the multi-phase motion image to obtain a multi-phase motion artifact image of the motion non-controlled entity under the target motion mode; and train the model based on the multi-phase motion image and the multi-phase motion artifact image, to obtain the artifact removal model.

[0198] In an embodiment, the model training module 404 is further configured to, for each phase, simulate acquisition based on a phase motion image in the multi-phase motion image and a plurality of adjacent phase motion images associated with the phase to obtain mixed-phase simulation raw data associated with the phase; and reconstruct based on the mixed-phase simulation raw data associated with the phase to obtain the phase motion artifact image.

[0199] In an embodiment, the model training module 404 is further configured to simulate acquisition of the time-phase motion image to obtain pure-phase simulated raw data of the time-phase motion image; simulate acquisition of each adjacent time-phase motion image to obtain pure-phase simulated raw data of each adjacent time-phase motion image; and splice the pure-phase simulated raw data of the time-phase motion image and the pure-phase simulated raw data of each adjacent time-phase motion image to obtain mixed-phase simulated raw data associated with the time phase.

[0200] The modules in the motion image processing apparatus described above can be implemented in whole or in part by software, hardware, and combinations thereof. The modules described above can be embedded in a processor in the computer device in hardware form or independent of the processor in the computer device, or can be stored in a memory in the computer device in software form to be invoked by the processor to perform the operations corresponding to the modules.

[0201] In an exemplary embodiment, a computer device is provided, and an internal structure diagram of the computer device can be as shown in FIG. 5. The computer device includes a processor, a memory, an input / output interface (I / O), and a communication interface. The processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is configured 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 running the operating system and the computer program in the non-volatile storage medium. The database of the computer device is configured to store data related to the methods described above. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to communicate with external terminals through a network connection. The computer program is executed by the processor to implement a motion image processing method or a method for removing image motion artifacts.

[0202] Those skilled in the art can understand that the structure shown in FIG. 5 is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. A specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0203] In an embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the method embodiments described above.

[0204] In an embodiment, a computer readable storage medium is provided, and a computer program is stored on the computer readable storage medium, and the computer program is executed by a processor to implement the steps in each of the above method embodiments.

[0205] In an embodiment, a computer program product is provided, and a computer program is stored on the computer program product, and the computer program is executed by a processor to implement the steps in each of the above method embodiments.

[0206] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0207] 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 relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. In the embodiments provided in the present application, any reference to memory, database or other medium can include at least one of non-volatile memory and volatile memory. The non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, 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. The volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0208] It should be noted that the backdoor scanning technology mainly uses the synchronous acquisition technology of electrocardiograph (ECG) and CT scanning. In the process of exposing the heart, the electrocardiograph waveform of the heart is recorded synchronously. Then, according to the synchronously recorded electrocardiograph waveform, the scanning data corresponding to the reconstruction phase required by the doctor can be selected from the entire scanning data obtained by exposure, so as to reconstruct the heart image of the reconstruction phase required by the doctor.

[0209] In the former gating scan technology, a doctor sets a delay time according to a required reconstruction phase, then continuously monitors an electrocardiogram waveform of the heart, and exposes the heart according to the set delay time when the R wave is monitored to obtain scan data corresponding to the required reconstruction phase of the doctor to reconstruct a heart image of the required reconstruction phase of the doctor.

[0210] Both of the above-mentioned scan schemes for heart imaging need to rely on the acquired electrocardiogram waveform. If the electrocardiogram waveform cannot be acquired or the acquired electrocardiogram waveform is inaccurate, it is difficult to obtain a heart image of a required reconstruction phase.

[0211] The present application provides a reconstruction processing method of a heart image, which can be used on different collimation CT scanning devices without acquiring an electrocardiogram waveform, for example, can be used on a large collimation (greater than or equal to 16 cm) CT scanning device, and can also be used on a small collimation (less than 16 cm) CT scanning device. The method provided by the present application can also be applied to a single-source CT scanning device or a multi-source CT scanning device.

[0212] The reconstruction processing method of the heart image provided by the present application can be applied to coronary calcification score scanning and coronary CTA scanning (Coronary Artery CT Angiography, CCTA), and can also be widely applied to other heart-related CT scanning scenes involving heart motion affecting image quality, especially scenes requiring clear structures of a specific heart motion phase (such as end diastole and end systole).

[0213] Taking the coronary calcification score scanning scene or the coronary CTA scanning scene as an example, if the method provided by the present application is used after the subject is positioned, the electrocardiogram waveform of the subject can not be collected, and the doctor can trigger the CT scanning device to expose the heart (which can be referred to as a target heart) of the subject at any time point. The CT scanning device continuously exposes the target heart for a set exposure time to obtain target scan data of the target heart. The target scan data can be reconstructed according to multiple reconstruction center time points to obtain heart images of the multiple reconstruction center time points, which can cover most of the reconstruction phases as much as possible.

[0214] In one embodiment, the present application provides a reconstruction processing method of a heart image, which includes the steps shown in FIG. 6:

[0215] Step S601, obtaining target scan data of a target heart; the target scan data is obtained by continuously exposing the target heart by a CT scanning device for a set exposure time.

[0216] Specifically, the exposure starting moment of the target heart can be determined without the electrocardiogram waveform of the target heart, and the CT scanning device can expose the target heart at any time point and continuously expose the target heart for a set time length, so that the data sensed by the detector is obtained as the target scanning data of the target heart. The set time length in the set time length of the continuous exposure can be referred to as an exposure time length. During the exposure of the CT scanning device, the detector and the ray source of the CT scanning device can work in an axial scan mode, in which the detector and the ray source of the CT scanning device rotate as a whole around the target object, and the target object is fixed on the examination bed to remain stationary. The CT scanning device can also work in other scanning modes, such as a spiral scan mode, a dynamic volume scan mode, etc., as long as the mode can continuously expose the target heart for the set exposure time length to obtain the target scanning data. The target scanning data is obtained by the CT scanning device continuously exposing the target heart for the set exposure time length, which eliminates the dependence on the ECG signal, thereby avoiding the limitations of the ECG signal itself, such as not obtaining a high-quality ECG signal, not obtaining an ECG signal, or the limitations of the applicable population of the ECG signal, etc. In addition, the scanning operation is simplified without the ECG-related operation.

[0217] In the exposure of the target heart, the exposure time length used can be obtained according to the heart rate of the target heart, and the exposure time length used can also be obtained without considering the heart rate of the target heart.

[0218] When the exposure time length used is obtained according to the heart rate of the target heart, the exposure time length is self-adaptive to the heart rate of the target heart. At this time, the determination of the exposure time length includes the following steps: obtaining the heart rate of the target heart; and determining the exposure time length according to the heart rate and a set proportion.

[0219] Specifically, the heart rate of the target heart before the exposure starts can be obtained, the heart cycle of the target heart can be determined according to the heart rate, and then the heart cycle is multiplied by the set proportion to obtain the exposure time length, or the product is adjusted to obtain the exposure time length. The value of the set proportion can be set as needed. When the set proportion is greater than or equal to 1, the exposure time length includes at least the time length corresponding to one heart cycle. When the set proportion is greater than or equal to 0.5 and less than 1, the exposure time length includes at least the time length corresponding to half a heart cycle.

[0220] For example, if the heart rate of the target heart is 60 beats per minute and the heart cycle of the target heart is 1000 ms, and the set proportion is 1, the product of the heart cycle and the set proportion is 1000 ms. 1000 ms can be directly used as the exposure time length, or 1000 ms can be adjusted, and the adjustment result is used as the exposure time length.

[0221] The manner of obtaining the heart rate of the target heart before the start of exposure can include but is not limited to the following manners:

[0222] Manner one: in the case that the target object wears a wearable device, the heart rate of the target heart of the target object before the start of exposure can be obtained according to the wearable device;

[0223] Manner two: in the case that the CT scanning device has a heart rate detection function, the heart rate detection unit of the CT scanning device can be called to obtain the heart rate of the target heart before the start of exposure.

[0224] When the exposure duration used can be determined without considering the heart rate of the target heart, the determination of the exposure duration includes the following manners:

[0225] Manner one: a time value input by a user, such as 1000 ms, can be obtained and used as the exposure duration. The time value input by the user can not need to consider the individual heart rate of the target heart, and the user can include a doctor.

[0226] Manner two: a fixed heart rate with low correlation to the individual target heart can be obtained, and a fixed cardiac cycle is obtained according to the fixed heart rate. The fixed cardiac cycle is multiplied by a set ratio to obtain the exposure duration. The set ratio can be pre-set or input by a user. The value of the set ratio can be determined in combination with the imaging requirements of the heart, the characteristics of the scanning device, and the clinical application scenarios.

[0227] In step S602, a plurality of reconstruction center instants are determined in the exposure duration according to a preset manner. The preset manner makes the number of the plurality of reconstruction center instants lower than a threshold value, and the heart shape difference degrees between the cardiac images of adjacent reconstruction center instants reach a difference degree threshold value.

[0228] According to the preset manner, a target interval duration can be determined, and sampling is performed in the exposure duration according to the target interval duration to obtain N reconstruction center instants (N is an integer greater than 1). The number of reconstruction center instants will not be too large, and the heart shape difference degrees between the cardiac images of the target heart at adjacent reconstruction center instants can all reach the difference degree threshold value. Therefore, the reconstruction phases of the cardiac images of the N reconstruction center instants are mostly different, and can cover most of the reconstruction phases. The present application adopts a continuous exposure manner, and the heart shape difference degrees between adjacent reconstruction center instants in the actual scanning data are used to capture the heart shape changes, and then a plurality of reconstruction center instants are obtained, i.e., motion phasing. Compared with the electrical signal phasing scheme which depends on the ECG signal, this manner can more accurately capture the instants when the heart is actually stationary or motionally stable, and thus reduces the artifacts caused by phase mismatch from the root.

[0229] Step S603, reconstruct the target scan data according to the plurality of reconstruction center instants to obtain the plurality of reconstruction center instant heart images.

[0230] After determining the N reconstruction center instants, for each reconstruction center instant, the following steps can be performed:

[0231] The scan data segment centered at the reconstruction center instant can be determined in the scan data according to the scan data range required by the reconstruction of the single-instant heart image, for example, if the reconstruction center instant is the 500th ms, and the axial scan mode is taken as an example, the scan data range required by the reconstruction of the single-instant heart image is the scan data with the axial scan angle from 0 to 360 degrees, and the rotation time corresponding to the scan data range is T, then the scan data segment of [500-T / 2, 500+T / 2] can be determined in the scan data, and the scan data segment is centered at the reconstruction center instant (i.e. the 500th ms). After obtaining the scan data segment centered at the reconstruction center instant, the scan data segment can be reconstructed to obtain the reconstruction center instant heart image.

[0232] In this way, the N reconstruction center instant heart images can be obtained.

[0233] Step S604, determine the target reconstruction center instant heart image according to the plurality of reconstruction center instant heart images.

[0234] After obtaining the N reconstruction center instant heart images, these heart images can be retained in the background without being fed back to the user. The N reconstruction center instant heart images can be automatically evaluated from the quality of each heart image or the motion degree corresponding to each heart image, and the best reconstruction center instant heart image is determined therefrom, and the best reconstruction center instant heart image is taken as the target reconstruction center instant heart image. The user can independently select the best image from the N reconstruction center instant heart images as the target reconstruction center instant heart image, for example, according to the observed image quality; or the best phase selection algorithm or model can be used to automatically complete the screening of the best reconstruction center instant heart image in the background as the target reconstruction center instant heart image. Subsequently, the target reconstruction center instant heart image is fed back to the user for analysis.

[0235] The way of obtaining the motion degree corresponding to each heart image can specifically include: obtaining the motion degree corresponding to the heart image according to the heart shape difference between the heart image and the heart image of the previous reconstruction center instant. The greater the heart shape difference, the greater the motion degree.

[0236] While feeding the heart image of the target reconstruction center moment to the doctor, the reconstruction phase corresponding to the heart image can also be fed to the doctor, and the determination manner of the reconstruction phase is introduced in subsequent embodiments.

[0237] In the reconstruction processing method of the heart image, the CT scanning device can continuously expose the target heart according to the set exposure time without relying on the ECG waveform of the target heart to trigger exposure, and obtain target scanning data. After obtaining the target scanning data, a plurality of reconstruction center moments are determined in the exposure time according to a preset manner. The preset manner can make the number of the plurality of reconstruction center moments less than a threshold value, and the heart morphology difference between the heart images of adjacent reconstruction center moments reaches a difference threshold value. According to the preset manner, the number of heart images required for reconstruction is small when the plurality of reconstruction center moments are reconstructed from the target scanning data, which ensures the reconstruction processing efficiency. In addition, the plurality of reconstruction center moments of the heart image can cover most of the reconstruction phases as much as possible, so that the heart image of the required reconstruction phase of the doctor can be hit with a high probability. After obtaining the plurality of reconstruction center moments of the heart image, the target reconstruction center moment of the heart image can be determined to feed the doctor, thereby improving the processing efficiency of the doctor. The embodiments of the present application consider the number of reconstruction center moments and the heart morphology difference, select a plurality of reconstruction center moments, and effectively reduce the image blur or artifacts caused by the periodic motion of the heart (such as heartbeat), so as to obtain the target reconstruction center moment of the heart image that meets the requirements in the continuous exposure CT scanning.

[0238] In one embodiment, according to the plurality of reconstruction center moments of the heart image, the target reconstruction center moment of the heart image is determined, comprising: obtaining a preset evaluation algorithm; and determining the target reconstruction center moment of the heart image from the plurality of reconstruction center moments of the heart image according to the preset evaluation algorithm.

[0239] The corresponding evaluation algorithm can be constructed according to the corresponding evaluation dimension, so as to obtain the preset evaluation algorithm. For example, the evaluation dimension includes the motion degree of each heart image. When the target reconstruction center moment of the heart image is determined from the plurality of reconstruction center moments of the heart image according to the preset evaluation algorithm corresponding to the evaluation dimension, the following steps can be performed:

[0240] For the non-first reconstruction center moment, the heart morphology difference between the heart image of the reconstruction center moment and the heart image of the previous reconstruction center moment can be obtained. The greater the heart morphology difference, the greater the motion degree corresponding to the heart image of the reconstruction center moment. In this way, the motion degree corresponding to the heart image of each reconstruction center moment except the first reconstruction center moment can be obtained. The heart image with a motion degree less than a threshold value or the smallest motion degree is selected as the target reconstruction center moment of the heart image.

[0241] In one embodiment, the method for reconstructing a cardiac image provided by the present application further comprises: if it is determined that the cardiac image at the target reconstruction center time needs motion correction, performing motion correction on the cardiac image at the target reconstruction center time to obtain a target image.

[0242] After determining the cardiac image at the target reconstruction center time, before feeding back the cardiac image to the user, it can be automatically determined whether the cardiac image needs motion correction, and if so, the motion correction algorithm or motion correction model can be used to perform motion correction on the cardiac image to obtain a target image and feed it back to the user for analysis.

[0243] In one embodiment, according to a preset manner, a plurality of reconstruction center times are determined in the exposure duration, comprising: obtaining a target interval duration ratio according to a preset correlation and a target exposure duration ratio; the target exposure duration ratio is the ratio of the exposure duration in the cardiac cycle of the target heart; the preset correlation includes the correlation between each exposure duration ratio and each interval duration ratio; according to the target interval duration ratio and the cardiac cycle of the target heart, a plurality of reconstruction center times are determined in the exposure duration.

[0244] Specifically, the ratio of the exposure duration in the cardiac cycle is referred to as the exposure duration ratio.

[0245] According to the ratio of the exposure duration used to obtain the target scanning data in the cardiac cycle of the target heart, the target exposure duration ratio is obtained. According to the preset correlation, the interval duration ratio matching the target exposure duration ratio is determined as the target interval duration ratio. According to the target interval duration ratio and the cardiac cycle of the target heart, a plurality of reconstruction center times can be determined in the exposure duration used to obtain the target scanning data.

[0246] In one embodiment, the determination manner of the preset correlation comprises: for any exposure duration ratio, sample scanning data matching the exposure duration ratio is obtained; in the sample scanning data, if the number of reconstruction center times determined according to the interval duration ratio is lower than the number threshold, and the cardiac morphology difference degree between adjacent reconstruction center times reaches the difference degree threshold, it is determined that the interval duration ratio and the exposure duration ratio have a correlation.

[0247] Specifically, for a certain exposure duration ratio, for example 80%, there are multiple combinations of exposure duration and cardiac cycle that meet the exposure duration ratio of 80%; for example, combination ①, the exposure duration is 800 ms and the cardiac cycle is 1000 ms; combination ②: the exposure duration is 720 ms and the cardiac cycle is 900 ms.

[0248] If a sample scanning data is obtained by continuously exposing a heart for 800 ms, and a heart cycle of the heart is 1000 ms, the sample scanning data matches the exposure time length ratio of 80%.

[0249] If a sample scanning data is obtained by continuously exposing a heart for 720 ms, and a heart cycle of the heart (which can be referred to as a sample heart cycle) is 900 ms, the sample scanning data matches the exposure time length ratio of 80%.

[0250] The exposure time length used to obtain the sample scanning data can be referred to as a sample exposure time length, and the heart cycle of the heart corresponding to the sample scanning data can be referred to as a sample heart cycle.

[0251] After obtaining the sample scanning data matching the exposure time length ratio of 80%, the sample exposure time length and the sample heart cycle corresponding to the sample scanning data can be obtained, and then the interval time length ratio can be adjusted multiple times. After each adjustment of the interval time length ratio, the following steps are performed:

[0252] According to the product of the interval time length ratio and the sample heart cycle, an interval time length is obtained, which is referred to as a sample interval time length. A plurality of reconstruction center instants are determined in the sample exposure time length according to the sample interval time length, and a plurality of heart images of the reconstruction center instants are reconstructed in combination with the sample scanning data. It is determined whether the heart morphology difference degrees between the heart images of adjacent reconstruction center instants all reach a difference degree threshold value and whether the number of determined reconstruction center instants is lower than a number threshold value.

[0253] If any heart morphology difference degree does not reach the difference degree threshold value or the number of determined reconstruction center instants is not lower than the number threshold value, the next adjustment is continued.

[0254] If the heart morphology difference degrees all reach the difference degree threshold value and the number of determined reconstruction center instants is lower than the number threshold value, the next adjustment is stopped, and the interval time length ratio is associated with the exposure time length ratio of 80%.

[0255] The heart morphology difference degrees between the heart images of adjacent reconstruction center instants can be predicted by a neural network. Specifically, the heart images of adjacent reconstruction center instants can be input into the neural network to enable the neural network to predict the heart morphology difference degrees between the heart images of adjacent reconstruction center instants.

[0256] In the above manner, the interval time length ratio associated with each exposure time length ratio can be determined, and a preset association relationship between different exposure time length ratios and interval time length ratios is obtained.

[0257] The number of the plurality of reconstruction center instants determined in this way is not excessive, and the difference in the cardiac morphology between the cardiac images of adjacent reconstruction center instants meets the requirement, indicating that the reconstruction phases of the cardiac images corresponding to the plurality of reconstruction center instants are mostly different, so as to cover most of the reconstruction phases.

[0258] In one embodiment, according to the target interval length proportion and the cardiac cycle of the target heart, the plurality of reconstruction center instants are determined in the exposure length, including: according to the target interval length proportion and the cardiac cycle of the target heart, obtaining the target interval length; determining the first reconstruction center instant according to the scanning data range required by the single-time-point cardiac image; taking the first reconstruction center instant as the sampling starting position, sampling the exposure length according to the target interval length, to obtain the plurality of reconstruction center instants.

[0259] After obtaining the target interval length proportion, the target interval length proportion and the cardiac cycle of the target heart can be multiplied, and the product is taken as the target interval length.

[0260] Exemplarily, the scanning data range required by the single-time-point cardiac image is the scanning data of the CT scanning device axial scanning angle from 0 to 360 degrees, and the scanning data range corresponds to a rotation time of T, and then the first reconstruction center instant t1 can be T / 2 ms.

[0261] After determining the first reconstruction center instant, the exposure length used to obtain the target scanning data is sampled according to the target interval length, taking the first reconstruction center instant as the sampling starting position, to obtain the plurality of reconstruction center instants. For example, the second reconstruction center instant t2=t1+target interval length, and the third reconstruction center instant t3=t1+2×target interval length.

[0262] In this embodiment, the first reconstruction center instant is determined according to the scanning data range required by the single-time-point cardiac image, which can avoid that there is not enough scanning data for reconstruction at the first reconstruction center instant, and the obtained first reconstruction center instant is more reasonable.

[0263] Reconstruction phase: a phase in a single cardiac cycle is reconstructed into a cardiac image, which can be referred to as a reconstruction phase. The reconstruction phase includes a relative reconstruction phase and an absolute reconstruction phase. A single cardiac cycle can be represented by the interval time of two adjacent R waves (denoted as R-R). For example, if the reconstruction is performed at 80% of the R-R interval time, the reconstruction phase at this time can be referred to as a relative reconstruction phase; for another example, if the reconstruction is performed 300 ms after the R wave delay, the reconstruction phase at this time can be referred to as an absolute reconstruction phase.

[0264] The reconstruction processing method of the cardiac image can further include the steps shown in FIG. 7 to determine the reconstruction phase corresponding to the cardiac image, and the determined reconstruction phase can be used in cardiac function analysis.

[0265] In step S701, a plurality of cardiac images at reconstruction center moments are acquired.

[0266] The plurality of cardiac images at reconstruction center moments can be obtained according to the reconstruction processing method of the cardiac image, and the specific introduction can refer to the above embodiments.

[0267] The plurality of cardiac images at reconstruction center moments can also be obtained by other methods. For example, after obtaining the target scanning data of the target heart, the user can randomly select a plurality of reconstruction center moments in the exposure time length used to obtain the target scanning data, so as to obtain a plurality of cardiac images at the reconstruction center moments randomly selected by the user.

[0268] In step S702, according to the plurality of cardiac images at reconstruction center moments, a plurality of heart cavity volumes at reconstruction center moments are determined.

[0269] For example, if N reconstruction center moments are determined, after obtaining the cardiac images at N reconstruction center moments, the cardiac images at each reconstruction center moment can be analyzed to determine the heart cavity volume at each reconstruction center moment. The heart cavity volume can be the ventricular volume or atrial volume of the heart.

[0270] In step S703, according to the corresponding relationship between the virtual heart cavity volume change curve and the virtual electrocardiogram waveform, and the plurality of heart cavity volumes at reconstruction center moments, the positions corresponding to each reconstruction center moment on the virtual electrocardiogram waveform are determined.

[0271] During the movement of the heart, the cardiac electrical signal changes with time, thereby forming an electrocardiogram waveform, and during the movement of the heart, the size of the heart cavity volume changes with time, thereby forming a heart cavity volume change curve. The electrocardiogram waveform and the heart cavity volume change curve have a corresponding relationship. Referring to FIG. 8, FIG. 8 shows the corresponding relationship between the ventricular volume change curve and the electrocardiogram waveform.

[0272] The standard ventricular volume change curve, the standard electrocardiogram waveform, and the corresponding relationship between them can be obtained from textbooks and the like.

[0273] When the reconstruction phase includes a relative reconstruction phase, the standard ventricular volume change curve can be directly used as the virtual ventricular volume change curve, the standard electrocardiogram waveform can be directly used as the virtual electrocardiogram waveform, and the corresponding relationship between the standard ventricular volume change curve and the standard electrocardiogram waveform can be directly used as the corresponding relationship between the virtual heart cavity volume change curve and the virtual electrocardiogram waveform.

[0274] When the reconstruction phase comprises an absolute reconstruction phase, the step of obtaining the virtual electrocardiogram waveform comprises: obtaining a standard electrocardiogram waveform; and compressing or stretching the standard electrocardiogram waveform to obtain the virtual electrocardiogram waveform, according to the cardiac cycle of the target heart.

[0275] Exemplarily, the cardiac cycle corresponding to the standard electrocardiogram waveform can be 900 ms, and the cardiac cycle of the target heart can be 1000 ms. In this case, the standard electrocardiogram waveform can be stretched so that the cardiac cycle corresponding to the standard electrocardiogram waveform becomes the cardiac cycle of the target heart, and the stretched standard electrocardiogram waveform is taken as the virtual electrocardiogram waveform.

[0276] Exemplarily, the cardiac cycle corresponding to the standard electrocardiogram waveform can be 900 ms, and the cardiac cycle of the target heart can be 800 ms. In this case, the standard electrocardiogram waveform can be compressed so that the cardiac cycle corresponding to the standard electrocardiogram waveform becomes the cardiac cycle of the target heart, and the compressed standard electrocardiogram waveform is taken as the virtual electrocardiogram waveform.

[0277] In addition to compressing or stretching the standard electrocardiogram waveform, the standard ventricular volume change curve can also be compressed or stretched. Specifically, the standard ventricular volume change curve can be compressed according to the compression degree of the standard electrocardiogram waveform, or the standard ventricular volume change curve can be stretched according to the stretching degree of the standard electrocardiogram waveform. The standard ventricular volume change curve obtained after compression or stretching is taken as the virtual ventricular volume change curve.

[0278] After the standard ventricular volume change curve and the standard electrocardiogram waveform are compressed or stretched, the corresponding relationship still exists, and thus the corresponding relationship between the virtual ventricular volume change curve and the virtual electrocardiogram waveform can be obtained.

[0279] In the embodiment, when the reconstruction phase corresponding to the heart image is determined, the real electrocardiogram waveform of the corresponding heart during the exposure process does not need to be collected, and the virtual electrocardiogram waveform can be used instead.

[0280] In step S704, the reconstruction phase corresponding to the heart image at each reconstruction center time is determined according to the position.

[0281] Specifically, the corresponding relationship between the virtual electrocardiogram waveform and the virtual ventricular volume change curve can be obtained by the foregoing method, and the position corresponding to each reconstruction center time on the virtual electrocardiogram waveform can be determined according to the corresponding relationship and the ventricular volume at the N reconstruction center times.

[0282] For each reconstruction center moment, according to the position corresponding to the reconstruction center moment on the virtual electrocardiogram waveform and the positions of two adjacent R waves on the virtual electrocardiogram waveform, the percentage of the reconstruction center moment belonging to the R-R interval time can be determined, thereby determining the relative reconstruction phase corresponding to the cardiac image of the reconstruction center moment.

[0283] For each reconstruction center moment, according to the position corresponding to the reconstruction center moment on the virtual electrocardiogram waveform and the position of the R wave on the virtual electrocardiogram waveform, the time of the reconstruction center moment relative to the R wave delay can be determined, thereby determining the absolute reconstruction phase corresponding to the cardiac image of the reconstruction center moment.

[0284] In the embodiment, in the case of no electrocardiogram waveform of the target heart, the reconstruction phases corresponding to the cardiac images of the plurality of reconstruction center moments can be determined by means of the heart cavity volume embodied by the cardiac images of the plurality of reconstruction center moments, the virtual heart cavity volume change curve and the virtual electrocardiogram waveform, thereby reducing the dependence on the electrocardiogram waveform.

[0285] In one embodiment, according to the corresponding relationship between the virtual heart cavity volume change curve and the virtual electrocardiogram waveform, and the heart cavity volumes of the plurality of reconstruction center moments, the positions corresponding to the reconstruction center moments on the virtual electrocardiogram waveform are determined, comprising: determining the heart cavity volume of the first reconstruction center moment among the heart cavity volumes of the plurality of reconstruction center moments; determining the position corresponding to the first reconstruction center moment on the virtual electrocardiogram waveform according to the corresponding relationship between the virtual heart cavity volume change curve and the virtual electrocardiogram waveform, and the heart cavity volume of the first reconstruction center moment; and determining the position corresponding to the intermediate reconstruction center moment on the virtual electrocardiogram waveform according to the position corresponding to the first reconstruction center moment on the virtual electrocardiogram waveform and the interval length between the intermediate reconstruction center moment and the first reconstruction center moment.

[0286] In one embodiment, according to the corresponding relationship between the virtual heart cavity volume change curve and the virtual electrocardiogram waveform, and the heart cavity volumes of the plurality of reconstruction center moments, the positions corresponding to the reconstruction center moments on the virtual electrocardiogram waveform are determined, comprising: determining the heart cavity volume of the last reconstruction center moment among the heart cavity volumes of the plurality of reconstruction center moments; determining the position corresponding to the last reconstruction center moment on the virtual electrocardiogram waveform according to the corresponding relationship between the virtual heart cavity volume change curve and the virtual electrocardiogram waveform, and the heart cavity volume of the last reconstruction center moment; and determining the position corresponding to the intermediate reconstruction center moment on the virtual electrocardiogram waveform according to the position corresponding to the last reconstruction center moment on the virtual electrocardiogram waveform and the interval length between the intermediate reconstruction center moment and the last reconstruction center moment.

[0287] Taking N reconstruction center instants as an example, the heart cavity volume of the first reconstruction center instant t1 and the heart cavity volume of the last reconstruction center instant t N , according to the corresponding relationship between the virtual heart cavity volume change curve and the virtual electrocardiogram waveform, the heart cavity volume of the first reconstruction center instant t1 and the heart cavity volume of the last reconstruction center instant t N , the heart cavity volume of the first reconstruction center instant t1 and the heart cavity volume of the last reconstruction center instant t N on the virtual electrocardiogram waveform.

[0288] For the reconstruction center instants between the first reconstruction center instant t1 and the last reconstruction center instant t N , these reconstruction center instants can belong to the intermediate reconstruction center instants, taking the second reconstruction center instant t2 as an example:

[0289] The position of the second reconstruction center instant t2 on the virtual electrocardiogram waveform can be determined according to the position of the first reconstruction center instant t1 on the virtual electrocardiogram waveform, and the interval length between the second reconstruction center instant t2 and the first reconstruction center instant t1.

[0290] Alternatively, the position of the second reconstruction center instant t2 on the virtual electrocardiogram waveform can also be determined according to the position of the last reconstruction center instant t N on the virtual electrocardiogram waveform, and the interval length between the second reconstruction center instant t2 and the last reconstruction center instant t N .

[0291] In this embodiment, the positions of the first reconstruction center instant t1 and the last reconstruction center instant t N on the virtual electrocardiogram waveform can be more accurately determined in combination with the heart cavity volume of the first reconstruction center instant t1 and the heart cavity volume of the last reconstruction center instant t N , so as to avoid errors and obtain more accurate positions of the intermediate reconstruction center instants on the virtual electrocardiogram waveform.

[0292] In order to better understand the above method, an application example of the above method provided by the present application is described in detail below. The application example includes the steps shown in FIG. 9.

[0293] Step S901, setting the scanning parameters of the CT scanning device;

[0294] The scanning parameters can include a dose upper limit, a heart rate of the exposed heart, a set ratio, and an exposure time length; an exposure range (i.e., a ratio of the exposure time length to a cardiac cycle) of a conventional cardiac CTA axial scanning protocol is generally 30% to 80%, and the exposure range involved in the embodiment is longer, so that, in terms of dose, scanning can be performed using a low dose, and a denoising technology of resolution or deep learning can be used in reconstruction to achieve an image effect at a level equivalent to a conventional dose.

[0295] The CT scanning device can provide a dose modulation function, which provides a user to input a dose upper limit, and after a positioning image scan is completed, the CT scanning device automatically recommends a scanning voltage and a milliampere, and a corresponding denoising algorithm level according to a patient positioning image, the user input dose upper limit, a heart rate of the exposed heart, and a selected exposure time length, so as to further ensure image quality on the basis of reducing a scanning dose.

[0296] Step S902, exposing the target heart according to the scanning parameters to obtain target scanning data;

[0297] Step S903, determining whether a user selects a reconstruction center moment;

[0298] If yes, steps S904 and S905 are entered, and if no, steps S906 to S908 are entered;

[0299] Step S904, determining Y reconstruction center moments selected by the user and reconstructing heart images of the Y reconstruction center moments;

[0300] Step S905, selecting a certain reconstruction center moment selected by the user at the Y reconstruction center moments as a target reconstruction center moment;

[0301] Step S906, determining N reconstruction center moments in the exposure time length;

[0302] Step S907, determining an optimal reconstruction center moment in the N reconstruction center moments;

[0303] Step S908, selecting the optimal reconstruction center moment as the target reconstruction center moment;

[0304] Step S909, determining whether a heart image of the target reconstruction center moment needs to be motion corrected;

[0305] If yes, step S910 is entered, and if no, step S911 is entered;

[0306] Step S910, performing motion correction on the heart image of the target reconstruction center moment;

[0307] Step S911, outputting the heart image of the target reconstruction center moment for user analysis;

[0308] Step S912: output the reconstruction phase of the target reconstruction center moment.

[0309] In the process of automatically identifying the reconstruction phase of the target reconstruction center moment, if the user selects one reconstruction center moment in step S904, and the reconstruction center moment is single, the reconstruction center moment can be taken as the target reconstruction center moment. After obtaining the cardiac image of the target reconstruction center moment, the neural network obtained by means of deep learning can be used for image analysis to determine the reconstruction phase of the target reconstruction center moment and output to the user for analysis.

[0310] In the case where the reconstruction center moment is multiple, the reconstruction phase of the target reconstruction center moment can be determined by means of the above method.

[0311] FIG. 10 is a schematic diagram of the reconstruction result obtained by applying the method provided by the present application to CCTA. When the multiple reconstruction center moments are 10, the cardiac images of the 10 reconstruction center moments shown in FIG. 10 can be obtained. As can be seen from FIG. 10, the cardiac images of the 10 reconstruction center moments show different cardiac morphologies, showing the cardiac morphology from the systolic phase to the diastolic phase. The reconstruction phases of these cardiac images are different, basically covering the entire cardiac cycle. Further, according to the preset evaluation algorithm corresponding to the motion degree corresponding to each cardiac image, it can be determined that the cardiac image of the 4th reconstruction center moment belongs to the cardiac image of the target reconstruction center moment.

[0312] It should be understood that, although each step in the flowchart involved in each embodiment described above is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other sequences. Moreover, at least part of the steps in the flowchart involved in each embodiment described above can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or steps or stages in other steps.

[0313] Based on the same inventive concept, the present embodiment also provides a cardiac image reconstruction processing device for implementing the cardiac image reconstruction processing method described above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more cardiac image reconstruction processing device embodiments provided below can refer to the limitations of the cardiac image reconstruction processing method described above, which will not be repeated here.

[0314] In one embodiment, a reconstruction processing apparatus of a cardiac image is provided, comprising:

[0315] a scan data acquisition module configured to acquire target scan data of a target heart; the target scan data is obtained by a CT scanning device continuously exposing the target heart for a set exposure duration;

[0316] a reconstruction center time determination module configured to determine a plurality of reconstruction center times in the exposure duration according to a preset manner; the preset manner makes the number of the plurality of reconstruction center times lower than a threshold value and the cardiac morphology difference degree between cardiac images of adjacent reconstruction center times reach a difference degree threshold value;

[0317] a reconstruction processing module configured to reconstruct the target scan data according to the plurality of reconstruction center times to obtain cardiac images of the plurality of reconstruction center times;

[0318] The reconstruction processing module is further configured to determine a cardiac image of a target reconstruction center time according to the cardiac images of the plurality of reconstruction center times.

[0319] In one embodiment, the reconstruction center time determination module is further configured to: obtain a target interval duration ratio according to the preset correlation and a target exposure duration ratio; the target exposure duration ratio is a ratio of the exposure duration in a cardiac cycle of the target heart; the preset correlation includes a correlation between each exposure duration ratio and each interval duration ratio; and determine the plurality of reconstruction center times in the exposure duration according to the target interval duration ratio and the cardiac cycle of the target heart.

[0320] In one embodiment, the reconstruction center time determination module is further configured to: for any exposure duration ratio, acquire sample scan data matched with the exposure duration ratio; and if the number of reconstruction center times determined according to the interval duration ratio and the cardiac morphology difference degree between cardiac images of adjacent reconstruction center times both reach the difference degree threshold value in the sample scan data, determine that the interval duration ratio and the exposure duration ratio have a correlation.

[0321] In one embodiment, the reconstruction center time determination module is further configured to: obtain a target interval duration according to the target interval duration ratio and the cardiac cycle of the target heart; determine a first reconstruction center time according to a scan data range required for reconstructing a single-time cardiac image; and sample the exposure duration according to the target interval duration to obtain the plurality of reconstruction center times, taking the first reconstruction center time as a sampling starting position.

[0322] In one embodiment, the apparatus further comprises an exposure duration determination module configured to: acquire a heart rate of the target heart; and determine the exposure duration according to the heart rate and a preset ratio.

[0323] In one embodiment, the reconstruction processing module is further configured to: acquire a preset evaluation algorithm; and determine the target cardiac image at the target reconstruction center time from the cardiac images at the plurality of reconstruction center times according to the preset evaluation algorithm.

[0324] In one embodiment, the apparatus further comprises a motion correction module configured to: if it is determined that the target cardiac image at the target reconstruction center time needs to be corrected, correct the target cardiac image at the target reconstruction center time to obtain a target image.

[0325] In one embodiment, a reconstruction processing apparatus for cardiac images is provided, and the apparatus further comprises:

[0326] a volume determination module configured to determine a plurality of cardiac cavity volumes at a plurality of reconstruction center times according to the cardiac images at the plurality of reconstruction center times;

[0327] a position determination module configured to determine a position corresponding to each reconstruction center time on a virtual electrocardiogram waveform according to a correspondence between a virtual cardiac cavity volume change curve and the virtual electrocardiogram waveform and the plurality of cardiac cavity volumes at the plurality of reconstruction center times;

[0328] a phase determination module configured to determine a reconstruction phase corresponding to each cardiac image at the plurality of reconstruction center times according to the position.

[0329] In one embodiment, the position determination module is further configured to: determine a first cardiac cavity volume at a first reconstruction center time from the plurality of cardiac cavity volumes at the plurality of reconstruction center times; determine a position corresponding to the first reconstruction center time on the virtual electrocardiogram waveform according to the correspondence between the virtual cardiac cavity volume change curve and the virtual electrocardiogram waveform and the first cardiac cavity volume at the first reconstruction center time; and determine a position corresponding to an intermediate reconstruction center time on the virtual electrocardiogram waveform according to the position corresponding to the first reconstruction center time on the virtual electrocardiogram waveform and an interval between the intermediate reconstruction center time and the first reconstruction center time.

[0330] In one embodiment, the position determination module is further configured to: determine the cardiac cavity volume at the last reconstruction center time from the plurality of cardiac cavity volumes at the reconstruction center times; determine a position corresponding to the cardiac cavity volume at the last reconstruction center time on the virtual electrocardiogram waveform according to a correspondence between the virtual cardiac cavity volume variation curve and the virtual electrocardiogram waveform and the cardiac cavity volume at the last reconstruction center time; and determine a position corresponding to the intermediate reconstruction center time on the virtual electrocardiogram waveform according to the position corresponding to the last reconstruction center time on the virtual electrocardiogram waveform and an interval between the intermediate reconstruction center time and the last reconstruction center time.

[0331] In one embodiment, if the reconstruction time phase comprises an absolute reconstruction time phase, the apparatus further comprises a virtual electrocardiogram waveform acquisition module configured to: acquire a standard electrocardiogram waveform; and compress or stretch the standard electrocardiogram waveform to obtain the virtual electrocardiogram waveform according to a cardiac cycle of the target heart.

[0332] The modules in the above-mentioned cardiac image reconstruction processing apparatus can be implemented in whole or in part by software, hardware, or a combination thereof. The modules can be embedded in or independent of a processor in a computer device in hardware form, or stored in a memory in a computer device in software form, so as to be called and executed by a processor to perform operations corresponding to the modules.

[0333] In one exemplary embodiment, a computer device is provided, and an internal structure diagram of the computer device can be as shown in FIG. 5. The computer device comprises a processor, a memory, an input / output interface (I / O), and a communication interface. The processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises 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 running the operating system and the computer program in the non-volatile storage medium. The database of the computer device is configured to store data related to the above-mentioned method. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to communicate with external terminals through a network connection. The computer program is executed by the processor to implement a cardiac image reconstruction processing method or a cardiac image time phase determination method.

[0334] Those skilled in the art can understand that the structure shown in FIG. 5 is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. A specific computer device can comprise more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0335] In an embodiment, a computer device is provided, comprising a memory and a processor, the memory storing a computer program, and the processor implementing the steps in the above-mentioned various method embodiments when executing the computer program.

[0336] In an embodiment, a computer readable storage medium is provided, storing a computer program, and the computer program implementing the steps in the above-mentioned various method embodiments when executed by a processor.

[0337] In an embodiment, a computer program product is provided, storing a computer program, and the computer program implementing the steps in the above-mentioned various method embodiments when executed by a processor.

[0338] The present application provides a method for removing motion artifacts of a cardiac image, and the specific implementation schemes of the method are the same as those described in the above-mentioned motion image processing method and cardiac image reconstruction processing method. The specific limitations in the following embodiments of the method for removing motion artifacts of a cardiac image can be referred to the limitations of the motion image processing method and cardiac image reconstruction processing method described above, and will not be repeated here.

[0339] In an embodiment, the present application provides a method for removing motion artifacts of a cardiac image, comprising:

[0340] obtaining target scanning data of a target heart; the target scanning data is obtained by continuously exposing the target heart by a CT scanning device according to a set exposure time;

[0341] determining a plurality of reconstruction center moments in the exposure time according to a preset manner; the preset manner makes the number of the plurality of reconstruction center moments lower than a threshold value, and the cardiac morphology difference degree between cardiac images of adjacent reconstruction center moments reaches a difference degree threshold value;

[0342] reconstructing the target scanning data according to the plurality of reconstruction center moments to obtain cardiac images of the plurality of reconstruction center moments;

[0343] determining a cardiac image of a target reconstruction center moment according to the cardiac images of the plurality of reconstruction center moments;

[0344] inputting the cardiac image of the target reconstruction center moment into a motion artifact removal model to obtain a target image;

[0345] wherein, the motion artifact removal model is obtained in the following manner:

[0346] obtaining a plurality of time-phase motion images obtained according to the above-mentioned motion image processing method;

[0347] simulate acquisition and reconstruction based on the multi-phase motion image to obtain a multi-phase motion artifact image of the motion non-controlled entity under the target motion mode;

[0348] train a model based on the multi-phase motion image and the multi-phase motion artifact image to obtain the de-artifact model.

[0349] The application adopts a continuous exposure mode, captures the heart shape change through the heart shape difference degree of adjacent reconstruction center time in actual scanning data, that is, obtains multiple reconstruction center time based on a motion phasing mode, provides a high-quality basic image for subsequent de-artifact processing, and reduces the artifact caused by time phase selection ambiguity. The de-artifact model obtained by training can accurately match the artifact characteristics (such as local myocardial motion abnormality, blurred artifact in rapid contraction period, etc.) generated under the multi-phase motion of the heart, and is more targeted for removing complex motion artifacts.

[0350] In one embodiment, simulate acquisition and reconstruction based on the multi-phase motion image to obtain a multi-phase motion artifact image of the motion non-controlled entity under the target motion mode, comprising:

[0351] For each phase, simulate acquisition based on the motion image of the phase in the multi-phase motion image and multiple adjacent phase motion images to obtain mixed-phase simulated raw data associated with the phase;

[0352] Reconstruct based on the mixed-phase simulated raw data associated with the phase to obtain the motion artifact image of the phase.

[0353] In one embodiment, simulate acquisition based on the motion image of the phase in the multi-phase motion image and multiple adjacent phase motion images to obtain mixed-phase simulated raw data associated with the phase, comprising:

[0354] Simulate acquisition of the motion image of the phase to obtain pure-phase simulated raw data of the motion image of the phase;

[0355] Simulate acquisition of each adjacent phase motion image to obtain pure-phase simulated raw data of each adjacent phase motion image;

[0356] Splice the pure-phase simulated raw data of the motion image of the phase and the pure-phase simulated raw data of each adjacent phase motion image to obtain the mixed-phase simulated raw data associated with the phase.

[0357] In one embodiment, the plurality of reconstruction center time is determined in the exposure time length according to a preset mode, comprising:

[0358] For any exposure time proportion, sample scanning data matching the exposure time proportion is obtained, and whether the interval time proportion is associated with the exposure time proportion is determined according to whether the number of reconstruction center time points determined in the sample scanning data according to the interval time proportion is lower than the number threshold and whether the cardiac morphology difference between adjacent reconstruction center time points reaches the difference threshold, to obtain the association relationship between different exposure time proportions and interval time proportions;

[0359] According to the association relationship and a target exposure time proportion, a target interval time proportion is obtained; the target exposure time proportion is the proportion of the exposure time in the cardiac cycle of the target heart;

[0360] According to the target interval time proportion and the cardiac cycle of the target heart, a plurality of reconstruction center time points in the exposure time are determined.

[0361] In one embodiment, the plurality of reconstruction center time points in the exposure time are determined according to the target interval time proportion and the cardiac cycle of the target heart, comprising:

[0362] According to the target interval time proportion and the cardiac cycle of the target heart, a target interval time is obtained;

[0363] According to the scanning data range required for reconstructing a single-time cardiac image, a first reconstruction center time point is determined;

[0364] Taking the first reconstruction center time point as a sampling starting position, the exposure time is sampled according to the target interval time to obtain a plurality of reconstruction center time points.

[0365] In one embodiment, the determination of the exposure time comprises:

[0366] The heart rate of the target heart is obtained;

[0367] According to the heart rate and a set proportion, the exposure time is determined.

[0368] In one embodiment, according to the cardiac images of the plurality of reconstruction center time points, the cardiac image of a target reconstruction center time point is determined, comprising:

[0369] A preset evaluation algorithm is obtained;

[0370] According to the preset evaluation algorithm, the cardiac image of the target reconstruction center time point is determined from the cardiac images of the plurality of reconstruction center time points.

[0371] In one embodiment, the method further comprises:

[0372] determining a plurality of center time instants of the heart cavity volume according to the plurality of center time instants of the heart images;

[0373] determining a position corresponding to each of the center time instants on the virtual electrocardiogram waveform according to the correspondence between the virtual heart cavity volume variation curve and the virtual electrocardiogram waveform and the plurality of center time instants of the heart cavity volume;

[0374] determining the reconstruction phase corresponding to each of the center time instants of the heart images according to the position.

[0375] In one embodiment, determining a position corresponding to each of the center time instants on the virtual electrocardiogram waveform according to the correspondence between the virtual heart cavity volume variation curve and the virtual electrocardiogram waveform and the plurality of center time instants of the heart cavity volume comprises:

[0376] determining the heart cavity volume of the first center time instant among the plurality of center time instants of the heart cavity volume;

[0377] determining a position corresponding to the first center time instant on the virtual electrocardiogram waveform according to the correspondence between the virtual heart cavity volume variation curve and the virtual electrocardiogram waveform and the heart cavity volume of the first center time instant;

[0378] determining a position corresponding to the intermediate center time instant on the virtual electrocardiogram waveform according to the position corresponding to the first center time instant on the virtual electrocardiogram waveform and the interval between the intermediate center time instant and the first center time instant.

[0379] In one embodiment, determining a position corresponding to each of the center time instants on the virtual electrocardiogram waveform according to the correspondence between the virtual heart cavity volume variation curve and the virtual electrocardiogram waveform and the plurality of center time instants of the heart cavity volume comprises:

[0380] determining the heart cavity volume of the last center time instant among the plurality of center time instants of the heart cavity volume;

[0381] determining a position corresponding to the last center time instant on the virtual electrocardiogram waveform according to the correspondence between the virtual heart cavity volume variation curve and the virtual electrocardiogram waveform and the heart cavity volume of the last center time instant;

[0382] determining a position corresponding to the intermediate center time instant on the virtual electrocardiogram waveform according to the position corresponding to the last center time instant on the virtual electrocardiogram waveform and the interval between the intermediate center time instant and the last center time instant.

[0383] In one embodiment, if the reconstruction phase comprises an absolute reconstruction phase, the step of obtaining the virtual electrocardiogram waveform comprises:

[0384] acquire a standard electrocardiogram waveform;

[0385] compress or stretch the standard electrocardiogram waveform according to a cardiac cycle of the target heart to obtain a virtual electrocardiogram waveform.

[0386] The application provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the above-mentioned method for removing motion artifacts of a cardiac image when executing the computer program. It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0387] 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. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to 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. The non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, 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. The volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0388] Each technical feature of the above-mentioned embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of each technical feature in the above-mentioned embodiments are not described, however, as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0389] The above embodiments only express several implementation ways of the present application, and the description is relatively specific and detailed, but it should not be understood as a limitation to the patent scope of the application. It should be pointed out that for ordinary skilled persons in the art, several modifications and improvements can be made without departing from the concept of the present application, which all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A method for processing a moving image, the method comprising: obtaining a plurality of time-phase motion template images of a motion-controlled model under a target motion mode, and a motion actual image of a motion-uncontrolled entity; and performing augmentation on the motion actual image based on registration parameters obtained from the plurality of time-phase motion template images and the motion actual image, to obtain a plurality of time-phase motion images of the motion-uncontrolled entity under the target motion mode. The augmentation on the motion actual image based on the registration parameters obtained from the plurality of time-phase motion template images and the motion actual image, to obtain a plurality of time-phase motion images of the motion-uncontrolled entity under the target motion mode, comprises: determining a target time-phase motion template image from the plurality of time-phase motion template images according to a part shape similarity between each time-phase motion template image in the plurality of time-phase motion template images and the motion actual image; and performing conversion processing on the plurality of time-phase motion template images based on registration parameters between the target time-phase motion template image and the motion actual image, to obtain the plurality of time-phase motion images. The conversion processing on the plurality of time-phase motion template images based on the registration parameters between the target time-phase motion template image and the motion actual image, comprises: performing conversion processing on each time-phase motion template image in the plurality of time-phase motion template images according to the registration parameters between the target time-phase motion template image and the motion actual image.

2. The method of claim 1, wherein, The conversion processing on the plurality of time-phase motion template images based on the registration parameters between the target time-phase motion template image and the motion actual image, comprises: performing conversion processing on the target time-phase motion template image according to the registration parameters between the target time-phase motion template image and the motion actual image, to obtain a target time-phase motion image; and performing conversion processing on the target time-phase motion image according to the registration parameters between the target time-phase motion template image and a non-target time-phase motion template image in the plurality of time-phase motion template images. 5.The method according to any one of claims 1 to 4, further comprising: inputting a to-be-artifact-removed image into an artifact-removing model to obtain a target image, the to-be-artifact-removed image comprising at least one of the motion actual image and the plurality of time-phase motion images; wherein the artifact-removing model is obtained in the following manner: performing simulated acquisition and reconstruction based on the plurality of time-phase motion images to obtain a plurality of time-phase motion artifact images of the motion-uncontrolled entity under the target motion mode; and performing model training based on the plurality of time-phase motion images and the plurality of time-phase motion artifact images to obtain the artifact-removing model. The performing simulated acquisition and reconstruction based on the plurality of time-phase motion images to obtain a plurality of time-phase motion artifact images of the motion-uncontrolled entity under the target motion mode, comprises: for each time phase, performing simulated acquisition based on a time-phase motion image in the plurality of time-phase motion images and a plurality of adjacent time-phase motion images associated with the time phase, to obtain mixed-phase simulated raw data associated with the time phase; and performing reconstruction based on the mixed-phase simulated raw data associated with the time phase, to obtain a time-phase motion artifact image.

3. The method of claim 2, wherein, The performing simulated acquisition and reconstruction based on the plurality of time-phase motion images to obtain a plurality of time-phase motion artifact images of the motion-uncontrolled entity under the target motion mode, comprises: for each time phase, performing simulated acquisition based on a time-phase motion image in the plurality of time-phase motion images and a plurality of adjacent time-phase motion images associated with the time phase, to obtain mixed-phase simulated raw data associated with the time phase; and performing reconstruction based on the mixed-phase simulated raw data associated with the time phase, to obtain a time-phase motion artifact image. ​ 4. The method of claim 2, wherein, ​ ​ ​ ​ ​ ​ ​ ​ 6. The method of claim 5, wherein, ​ ​ ​ 7. The method of claim 6, wherein, The mixed-phase simulation seismic data associated with the time phase is obtained based on the time phase motion image and a plurality of adjacent time phase motion images in the plurality of time phase motion images, and the method comprises the following steps of: simulating acquisition of the time phase motion image to obtain pure-phase simulation seismic data of the time phase motion image; simulating acquisition of each adjacent time phase motion image to obtain pure-phase simulation seismic data of each adjacent time phase motion image; splicing the pure-phase simulation seismic data of the time phase motion image and the pure-phase simulation seismic data of each adjacent time phase motion image to obtain the mixed-phase simulation seismic data associated with the time phase.

8. A motion image processing device, the device comprising: an image acquisition module configured to acquire a plurality of time phase motion template images of a motion controlled model under a target motion mode and a motion actual image of a motion uncontrolled entity, respectively; an image augmentation module configured to augment the motion actual image based on registration parameters obtained from the plurality of time phase motion template images and the motion actual image to obtain a plurality of time phase motion images of the motion uncontrolled entity under the target motion mode.

9. A computer device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the method of any one of claims 1 to 7 when executing the computer program.

10. A reconstruction processing method of a cardiac image, the method comprising: acquiring target scan data of a target heart; the target scan data being obtained by a CT scanning device continuously exposing the target heart for a set exposure duration; determining a plurality of reconstruction center time points in the exposure duration according to a preset manner; the preset manner is such that the number of the plurality of reconstruction center time points is lower than a threshold value and the cardiac morphology difference degrees between cardiac images of adjacent reconstruction center time points both reach a difference degree threshold value; reconstructing the target scan data according to the plurality of reconstruction center time points to obtain cardiac images of the plurality of reconstruction center time points; determining a cardiac image of a target reconstruction center time point according to the cardiac images of the plurality of reconstruction center time points.

11. The method of claim 10, wherein, the determining the plurality of reconstruction center time points in the exposure duration according to the preset manner comprises: obtaining a target interval duration ratio according to a preset correlation and a target exposure duration ratio; the target exposure duration ratio is a ratio of the exposure duration in a cardiac cycle of the target heart; the preset correlation comprises a correlation between each exposure duration ratio and each interval duration ratio; determining the plurality of reconstruction center time points in the exposure duration according to the target interval duration ratio and the cardiac cycle of the target heart.

12. The method of claim 11, wherein, the determination manner of the preset correlation comprises: for any exposure duration ratio, acquiring sample scan data matched with the exposure duration ratio; in the sample scan data, if the number of reconstruction center time points determined according to the interval duration ratio is lower than a number threshold value and the cardiac morphology difference degrees between cardiac images of adjacent reconstruction center time points both reach a difference degree threshold value, it is determined that the interval duration ratio and the exposure duration ratio have a correlation.

13. The method of claim 11, wherein, The method comprises the following steps: According to the target interval duration ratio and the cardiac cycle of the target heart, a plurality of reconstruction center time points are determined in the exposure duration. According to the target interval duration ratio and the cardiac cycle of the target heart, a target interval duration is obtained. A first reconstruction center time point is determined according to a scanning data range required for reconstructing a single-time-point cardiac image.

14. The method of claim 10, wherein, The exposure duration is sampled according to the target interval duration, and a plurality of reconstruction center time points are obtained. The exposure duration is determined in the following manner: The heart rate of the target heart is obtained.

15. The method of claim 10, wherein, According to the heart rate and a set ratio, the exposure duration is determined. According to the plurality of cardiac images at the plurality of reconstruction center time points, a cardiac image at a target reconstruction center time point is determined, comprising: A preset evaluation algorithm is obtained.

16. The method of claim 10, wherein, According to the preset evaluation algorithm, a cardiac image at a target reconstruction center time point is determined from the plurality of cardiac images at the plurality of reconstruction center time points. The method further comprises:

17. The method of any one of claims 10 to 16, wherein, If it is determined that the cardiac image at the target reconstruction center time point needs to be corrected, the cardiac image at the target reconstruction center time point is corrected to obtain a target image. The method further comprises: According to the plurality of cardiac images at the plurality of reconstruction center time points, a plurality of cardiac cavity volumes at the plurality of reconstruction center time points are determined. According to the corresponding relationship between the virtual cardiac cavity volume change curve and the virtual electrocardiogram waveform, and the plurality of cardiac cavity volumes at the plurality of reconstruction center time points, positions corresponding to the plurality of reconstruction center time points on the virtual electrocardiogram waveform are determined.

18. The method of claim 17, wherein, According to the positions, reconstruction phases corresponding to the plurality of cardiac images at the plurality of reconstruction center time points are determined. According to the corresponding relationship between the virtual cardiac cavity volume change curve and the virtual electrocardiogram waveform, and the plurality of cardiac cavity volumes at the plurality of reconstruction center time points, positions corresponding to the plurality of reconstruction center time points on the virtual electrocardiogram waveform are determined, comprising: A cardiac cavity volume at a first reconstruction center time point is determined from the plurality of cardiac cavity volumes at the plurality of reconstruction center time points. According to the corresponding relationship between the virtual cardiac cavity volume change curve and the virtual electrocardiogram waveform, and the cardiac cavity volume at the first reconstruction center time point, a position corresponding to the first reconstruction center time point on the virtual electrocardiogram waveform is determined.

19. The method of claim 17, wherein, According to the position corresponding to the first reconstruction center time point on the virtual electrocardiogram waveform, and an interval duration between the intermediate reconstruction center time point and the first reconstruction center time point, a position corresponding to the intermediate reconstruction center time point on the virtual electrocardiogram waveform is determined. According to the corresponding relationship between the virtual cardiac cavity volume change curve and the virtual electrocardiogram waveform, and the plurality of cardiac cavity volumes at the plurality of reconstruction center time points, positions corresponding to the plurality of reconstruction center time points on the virtual electrocardiogram waveform are determined, comprising: A cardiac cavity volume at a last reconstruction center time point is determined from the plurality of cardiac cavity volumes at the plurality of reconstruction center time points. According to the corresponding relationship between the virtual cardiac cavity volume change curve and the virtual electrocardiogram waveform, and the cardiac cavity volume at the last reconstruction center time point, a position corresponding to the last reconstruction center time point on the virtual electrocardiogram waveform is determined. According to a position corresponding to the intermediate reconstruction center moment on the virtual electrocardiogram waveform and an interval length between the intermediate reconstruction center moment and the last reconstruction center moment, a position corresponding to the intermediate reconstruction center moment on the virtual electrocardiogram waveform is determined.

20. The method of claim 17, wherein, If the reconstruction phase includes an absolute reconstruction phase, the virtual electrocardiogram waveform is obtained in the following manner: a standard electrocardiogram waveform is obtained; the standard electrocardiogram waveform is compressed or stretched according to a cardiac cycle of the target heart to obtain the virtual electrocardiogram waveform.

21. A computer device, comprising a memory and a processor, the memory storing a computer program, and the processor implementing the method of any one of claims 10-20 when executing the computer program.

22. A method for removing motion artifacts of a cardiac image, the method comprising: obtaining target scanning data of a target heart; the target scanning data is obtained by continuously exposing the target heart to a CT scanning device for a set exposure time; determining a plurality of reconstruction center moments in the exposure time according to a preset manner; the preset manner makes the number of the plurality of reconstruction center moments lower than a threshold value and the cardiac morphology difference degree between cardiac images of adjacent reconstruction center moments reach a difference degree threshold value; reconstructing the target scanning data according to the plurality of reconstruction center moments to obtain cardiac images of the plurality of reconstruction center moments; determining a cardiac image of a target reconstruction center moment according to the cardiac images of the plurality of reconstruction center moments; inputting the cardiac image of the target reconstruction center moment into a de-artifact model to obtain a target image; wherein the de-artifact model is obtained in the following manner: obtaining a plurality of time-phase motion images by the motion image processing method according to any one of claims 1-4; performing simulated acquisition and reconstruction based on the plurality of time-phase motion images to obtain a plurality of time-phase motion artifact images of a motion uncontrolled entity under the target motion mode; performing model training according to the plurality of time-phase motion images and the plurality of time-phase motion artifact images to obtain the de-artifact model.

23. The method of claim 22, wherein, performing simulated acquisition and reconstruction based on the plurality of time-phase motion images to obtain a plurality of time-phase motion artifact images of a motion uncontrolled entity under the target motion mode, comprising: for each time phase, performing simulated acquisition based on the time-phase motion image and a plurality of adjacent time-phase motion images in the plurality of time-phase motion images to obtain mixed-phase simulated raw data associated with the time phase; performing reconstruction based on the mixed-phase simulated raw data associated with the time phase to obtain the time-phase motion artifact image.

24. The method of claim 23, wherein, performing simulated acquisition based on the time-phase motion image and a plurality of adjacent time-phase motion images in the plurality of time-phase motion images to obtain mixed-phase simulated raw data associated with the time phase, comprising: performing simulated acquisition on the time-phase motion image to obtain pure-phase simulated raw data of the time-phase motion image; performing simulated acquisition on each adjacent time-phase motion image to obtain pure-phase simulated raw data of each adjacent time-phase motion image; splicing the pure-phase simulated raw data of the time-phase motion image and the pure-phase simulated raw data of each adjacent time-phase motion image to obtain the mixed-phase simulated raw data associated with the time phase.

25. The method of claim 22, wherein, The determining a plurality of reconstruction center instants in the exposure time length according to a preset mode comprises: For any exposure time length proportion, sample scanning data matched with the exposure time length proportion is acquired, whether the number of reconstruction center instants determined in the sample scanning data according to interval time length proportions is lower than a number threshold, and whether the cardiac morphology difference degrees between adjacent reconstruction center instants are all up to a difference degree threshold, are determined to determine whether the interval time length proportion is associated with the exposure time length proportion, to obtain an association relationship between different exposure time length proportions and interval time length proportions; According to the association relationship and a target exposure time length proportion, a target interval time length proportion is obtained; the target exposure time length proportion is a proportion of the exposure time length in a cardiac cycle of the target heart; According to the target interval time length proportion and the cardiac cycle of the target heart, a plurality of reconstruction center instants in the exposure time length are determined.

26. The method of claim 25, wherein, The determining a plurality of reconstruction center instants in the exposure time length according to the target interval time length proportion and the cardiac cycle of the target heart comprises: According to the target interval time length proportion and the cardiac cycle of the target heart, a target interval time length is obtained; According to a scanning data range required by a single reconstruction instant cardiac image, a first reconstruction center instant is determined; Taking the first reconstruction center instant as a sampling starting position, the exposure time length is sampled according to the target interval time length, to obtain a plurality of reconstruction center instants.

27. The method of claim 22, wherein, The determination mode of the exposure time length comprises: A heart rate of the target heart is acquired; According to the heart rate and a set ratio, the exposure time length is determined.

28. The method of claim 22, wherein, According to the cardiac images of the plurality of reconstruction center instants, a cardiac image of a target reconstruction center instant is determined, comprising: A preset evaluation algorithm is acquired; According to the preset evaluation algorithm, a cardiac image of a target reconstruction center instant is determined from the cardiac images of the plurality of reconstruction center instants.

29. The method of any one of claims 22 to 28, wherein, The method further comprises: According to the cardiac images of the plurality of reconstruction center instants, cardiac cavity volumes of the plurality of reconstruction center instants are determined; According to a corresponding relationship between a virtual cardiac cavity volume change curve and a virtual electrocardiogram waveform, and the cardiac cavity volumes of the plurality of reconstruction center instants, positions corresponding to the plurality of reconstruction center instants on the virtual electrocardiogram waveform are determined; According to the positions, reconstruction phases corresponding to the cardiac images of the plurality of reconstruction center instants are determined.

30. The method of claim 29, wherein, According to the corresponding relationship between the virtual cardiac cavity volume change curve and the virtual electrocardiogram waveform, and the cardiac cavity volumes of the plurality of reconstruction center instants, the positions corresponding to the plurality of reconstruction center instants on the virtual electrocardiogram waveform are determined, comprising: From the cardiac cavity volumes of the plurality of reconstruction center instants, a cardiac cavity volume of a first reconstruction center instant is determined; According to the corresponding relationship between the virtual cardiac cavity volume change curve and the virtual electrocardiogram waveform, and the cardiac cavity volume of the first reconstruction center instant, a position corresponding to the first reconstruction center instant on the virtual electrocardiogram waveform is determined; According to the position corresponding to the first reconstruction center instant on the virtual electrocardiogram waveform, and an interval time length between an intermediate reconstruction center instant and the first reconstruction center instant, a position corresponding to the intermediate reconstruction center instant on the virtual electrocardiogram waveform is determined.

31. The method of claim 29, wherein, According to the corresponding relationship between the virtual heart chamber volume change curve and the virtual electrocardiogram waveform, and the heart chamber volumes at the plurality of reconstruction center moments, positions corresponding to the plurality of reconstruction center moments on the virtual electrocardiogram waveform are determined, including: Among the heart chamber volumes at the plurality of reconstruction center moments, the heart chamber volume at the last reconstruction center moment is determined; According to the corresponding relationship between the virtual heart chamber volume change curve and the virtual electrocardiogram waveform, and the heart chamber volume at the last reconstruction center moment, a position corresponding to the last reconstruction center moment on the virtual electrocardiogram waveform is determined; According to the position corresponding to the last reconstruction center moment on the virtual electrocardiogram waveform, and the interval length between the intermediate reconstruction center moment and the last reconstruction center moment, a position corresponding to the intermediate reconstruction center moment on the virtual electrocardiogram waveform is determined.

32. The method of claim 29, wherein, If the reconstruction phase includes an absolute reconstruction phase, the virtual electrocardiogram waveform acquisition step includes: Acquiring a standard electrocardiogram waveform; According to a cardiac cycle of the target heart, the standard electrocardiogram waveform is compressed or stretched to obtain the virtual electrocardiogram waveform. 33.A computer device, comprising a memory and a processor, the memory stores a computer program, and the processor implements the method in any one of claims 22 to 32 when executing the computer program.

Images