Method, device, medium and product for determining a sequence of ct images

Abstract

Embodiments of the present application disclose a CT image sequence determination method, device, medium and product, and belong to the technical field of image processing. The method comprises: in response to a first trigger operation, determining at least one motion phase to be displayed, and obtaining target projection data for a predetermined part of a target object, the motion phase being a motion phase of a driving part, and the predetermined part being a region of interest which changes with the motion of the driving part; determining a first image sequence according to the target projection data, the first image sequence comprising a second image sequence corresponding to each motion phase in the at least one motion phase, and the second image sequence comprising a first image under each sub-phase in the corresponding motion phase, the first image comprising actual or expected contrast information of the region of interest under the corresponding sub-phase. The technical scheme provided by the embodiments of the present application can solve the problem that the reconstructed CT image in the prior art cannot accurately display the state information of a moving organ or tissue.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of image processing technology, and in particular to a method, device, medium and product for determining CT image sequences. Background Technology

[0002] Currently, cone-beam computed tomography (CBCT) is an essential tool for interventional diagnosis and treatment of critical illnesses such as cardiovascular and cerebrovascular diseases, structural heart disease, and oncological diseases.

[0003] Cone-beam CT has a very slow rotation speed, low temporal resolution, and low temporal sampling rate. For moving organs or tissues such as the heart, cardiovascular system, pulmonary vessels, and abdominal vessels, the images reconstructed by existing technologies cannot accurately display the state information of moving organs or tissues due to the influence of factors such as cardiac motion, respiratory motion, and abdominal organ motion. Summary of the Invention

[0004] This invention provides a method, device, medium, and product for determining CT image sequences, in order to solve the problem that the images reconstructed by the prior art cannot accurately display the state information of moving organs or moving tissues.

[0005] According to one aspect of the present invention, a method for determining a CT image sequence is provided, the method comprising:

[0006] In response to a first triggering operation, at least one motion phase to be displayed is determined, and target projection data for a predetermined part of the target object is acquired, wherein the motion phase is the motion phase of the driving part, and the region of interest of the predetermined part changes with the motion of the driving part.

[0007] A first image sequence is determined based on the target projection data. The first image sequence includes a second image sequence corresponding to each of the at least one motion phases. The second image sequence includes a first image under each sub-phase within the corresponding motion phase. The first image includes the actual or expected contrast information of the region of interest under the corresponding sub-phase. The motion phase includes at least one of the sub-phases.

[0008] The second image sequence corresponding to each of the at least one motion phase is displayed.

[0009] According to another aspect of the present invention, a CT image sequence determination apparatus is provided, the apparatus comprising:

[0010] The first triggering module is used to respond to a first triggering operation, determine at least one motion phase to be displayed, and acquire target projection data for a predetermined part of the target object, wherein the motion phase is the motion phase of the driving part, and the region of interest of the predetermined part changes with the motion of the driving part.

[0011] A first image sequence module is used to determine a first image sequence based on the target projection data. The first image sequence includes a second image sequence corresponding to each of the at least one motion phase. The second image sequence includes a first image under each sub-phase within the corresponding motion phase. The first image includes the actual or expected contrast information of the region of interest under the corresponding sub-phase. The motion phase includes at least one of the sub-phases.

[0012] The first display module is used to display the second image sequence corresponding to the at least one motion phase.

[0013] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0014] One or more processors;

[0015] Storage device for storing one or more programs.

[0016] When one or more programs are executed by one or more processors, the one or more processors implement the CT image sequence determination method as described in any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, which stores computer instructions for causing a processor to execute and implement the CT image sequence determination method of any embodiment of the present invention.

[0018] According to another aspect of the present invention, a computer program product is provided, which, when executed by a processor, implements the CT image sequence determination method as described in any embodiment of the present invention.

[0019] The technical solution of the CT image sequence determination method provided in this invention is that the motion phase is the motion phase of the driving part, and one motion phase includes at least one sub-phase; the region of interest of the predetermined part changes with the motion of the driving part, and the first image sequence includes a second image sequence corresponding to each motion phase in the at least one motion phase. Each first image in the second image sequence includes the actual contrast information or the expected contrast information of the region of interest under the corresponding sub-phase. Therefore, each first image is only used to reflect the state of the region of interest under the corresponding sub-phase. Compared with the prior art, which reflects the motion state of the region of interest under one or more motion cycles through one image, it has higher accuracy. In this way, by displaying the second image sequence corresponding to the at least one motion phase, the motion change and contrast change of the region of interest during the at least one motion phase of the driving part can be displayed intuitively and accurately.

[0020] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A flowchart of a CT image sequence determination method provided in an embodiment of the present invention;

[0023] Figure 2A This is a schematic diagram illustrating the change of the reconstruction time curve of the right coronary artery root over time, provided in an embodiment of the present invention.

[0024] Figure 2B This is a schematic diagram illustrating the change of reconstruction time curve over time for the right coronary artery branch, as provided in an embodiment of the present invention.

[0025] Figure 2C This is a schematic diagram illustrating the change of the reconstruction time curve of the left coronary artery root over time, provided in an embodiment of the present invention.

[0026] Figure 2D A schematic diagram illustrating the change of reconstruction time curve over time for the left coronary artery branch, provided in an embodiment of the present invention;

[0027] Figure 3The true image corresponding to the first image of the coronary artery under three predetermined motion phases in the first and fifth motion cycles provided in the embodiments of the present invention;

[0028] Figure 4A A contrast diagram of the coronary artery under different viewing angles at time steps 0 and 20 when the driving part is in diastole, as provided in an embodiment of the present invention:

[0029] Figure 4B A contrast diagram of the coronary artery under different viewing angles at time steps of 40 and 60, provided as an embodiment of the present invention, when the driving part is in diastole.

[0030] Figure 5A A contrast diagram of the coronary artery under different viewing angles at time steps 0 and 20 when the driving part is in the contraction phase, as provided in this embodiment of the invention:

[0031] Figure 5B A contrast diagram of the coronary artery under different viewing angles at time steps of 40 and 60, provided as an embodiment of the present invention, when the driving part is in diastole.

[0032] Figure 6 This is another flowchart of the CT image sequence determination method provided in the embodiments of the present invention;

[0033] Figure 7 A flowchart of a first image sequence determination method provided in an embodiment of the present invention;

[0034] Figure 8 Another flowchart of the first image sequence determination method provided in the embodiments of the present invention;

[0035] Figure 9 Another flowchart of the first image sequence determination method provided in the embodiments of the present invention;

[0036] Figure 10 This is another flowchart of the CT image sequence determination method provided in the embodiments of the present invention;

[0037] Figure 11 A flowchart of a third image sequence determination method provided in an embodiment of the present invention;

[0038] Figure 12 Another flowchart of the third image sequence determination method provided in the embodiments of the present invention;

[0039] Figure 13 A flowchart of a target projection data determination method provided in an embodiment of the present invention;

[0040] Figure 14A This is a schematic diagram of the CT image sequence determination device provided in an embodiment of the present invention;

[0041] Figure 14B This is another schematic diagram of the CT image sequence determination device provided in an embodiment of the present invention;

[0042] Figure 14C This is another schematic diagram of the CT image sequence determination device provided in an embodiment of the present invention;

[0043] Figure 14D This is another schematic diagram of the CT image sequence determination device provided in an embodiment of the present invention;

[0044] Figure 15 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0045] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0046] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0047] Figure 1 This is a flowchart illustrating the CT image sequence determination method provided in this embodiment of the invention. This embodiment is applicable to situations where, based on target projection data of a predetermined location, an image sequence is reconstructed that reflects the changes in the region of interest of the predetermined location as the driving location moves. This method can be executed by a CT image sequence determination device, which can be implemented in hardware and / or software and can be configured in electronic devices such as computers or servers. Figure 1 As shown, the method in this embodiment includes:

[0048] S110. In response to the first triggering operation, determine at least one motion phase to be displayed, and acquire target projection data for a predetermined part of the target object, wherein the motion phase is the motion phase of the driving part, and the region of interest of the predetermined part changes with the motion of the driving part.

[0049] Motion phase can be understood as the state of motion, such as the heart's contraction state and relaxation state.

[0050] The first triggering operation can be a determination operation based on the input content, which includes cycle identifiers, time period information, or start and end phase identifiers, etc. The start and end phase identifiers include the cycle identifier and the phase identifier of the motion cycle to which the start motion phase belongs; the end phase identifier includes the cycle identifier and the phase identifier of the motion cycle to which the end motion phase belongs.

[0051] For example, the input includes start and end phase identifiers. In this case, the motion phase corresponding to the start phase identifier, the motion phase corresponding to the end phase identifier, and all motion phases in between are taken as the at least one motion phase.

[0052] For example, the input includes time period information. In this case, all motion phases of the driving part within the time period corresponding to the time period information are taken as the at least one motion phase.

[0053] The target object is the CT scan subject. The driving site refers to a part that can move voluntarily, such as the heart or lungs. The predetermined site is the brain, heart, lungs, abdomen, or periphery, etc. The region of interest is the blood vessels or tissue within the predetermined site.

[0054] Target projection data can be acquired through various forms of CT equipment, such as traditional single-energy CT equipment, newer dual-energy CT or multi-energy CT equipment. These can be either slowly rotating cone-beam CT devices, rapidly rotating diagnostic CT devices, or other types of CT equipment. Slowly rotating cone-beam CT devices include C-arm cone-beam CT, O-arm cone-beam CT, interventional CT, digital subtraction angiography systems, and minimally invasive interventional surgery systems commonly used in interventional therapy rooms, as well as cone-beam CT commonly used in radiotherapy rooms, and high spatial resolution micro-CT, with common rotation speeds of 2-60 seconds per revolution. Rapidly rotating diagnostic CT devices include multi-slice diagnostic CT, dual-source multi-slice diagnostic CT commonly used in diagnostic radiology departments, or sliding-rail CT commonly used in hybrid operating rooms, with common rotation speeds of 0.1-2 seconds per revolution. Other types of CT equipment include linear orbit CT, reverse geometry CT, non-rotating static structural CT, breast tomography equipment, and breast CT equipment. In other words, the technical solution provided in this embodiment of the invention is not affected by the rotation speed of the CT equipment during its specific implementation. The projection data acquired by dual-energy CT and multi-energy CT equipment can be used to obtain target projection data through subtraction; specific implementation methods can be found in subsequent embodiments. Contrast projection data acquired under single-energy conditions can be directly used as target projection data.

[0055] S120. Determine a first image sequence based on the target projection data. The first image sequence includes a second image sequence corresponding to each motion phase in the at least one motion phase. The second image sequence includes a first image under each sub-phase within the corresponding motion phase. The first image includes the actual or expected contrast information of the region of interest under the corresponding sub-phase. The motion phase includes at least one sub-phase.

[0056] The first image corresponds to a sub-phase, therefore the first image can reflect the state of the region of interest under the sub-phase, such as contrast state, motion state, etc. Since the temporal resolution of the sub-phase is higher than that of the motion phase, the second image sequence in this embodiment has a higher temporal resolution.

[0057] When an X-ray beam passes through an object, the difference in the original line integral signal caused by the different attenuation characteristics and thickness of the tissue along the path results in different reconstructed pixel values ​​at different spatial locations, which is called contrast.

[0058] In this embodiment, the actual contrast information refers to the contrast information directly determined based on the projection data; the expected contrast information refers to the contrast information modeled based on the projection data at the corresponding projection viewpoint at a nearby time. Using the expected contrast information can improve the resolution of the second image sequence, that is, increase the number of sub-phases corresponding to each motion phase. This allows the second image sequence to display continuous and subtle motion and contrast changes in the region of interest as the driving part moves.

[0059] If all the first images in the second image sequence only include actual contrast information, then the maximum temporal resolution of the second image sequence is the temporal resolution corresponding to the projection viewpoint.

[0060] Since the contrast agent concentration changes continuously, the contrast also changes continuously. By introducing contrast, the resolution of the second image sequence can be made greater than the temporal resolution corresponding to a single projection viewpoint.

[0061] The first image in this embodiment is equipped with an image identifier.

[0062] In one embodiment, the image identifier includes only the phase identifier of the motion phase of the first image and the sub-phase identifier of its sub-phase, but not the period identifier of the period to which the first image belongs. This embodiment is suitable for cases where periodization is not distinguished. For example, the data acquisition duration corresponding to the target projection data is less than one motion cycle of the driving part; or, although the motion of the driving part is periodic, each motion cycle is not exactly the same, or in some cases, each motion cycle is not completely different, therefore, the motion cycle is not distinguished.

[0063] In one embodiment, the image identifier includes a period identifier of the motion cycle to which the first image belongs, a phase identifier of the motion phase it is in, and a sub-phase identifier of the sub-phase it is in. This embodiment is suitable for situations where the acquisition time of the target projection data is longer than one motion cycle, and where it is assumed that the motion of the driving part is the same in different motion cycles.

[0064] If the image identifier includes a period identifier, the input content corresponding to the first trigger operation in S110 can also be configured to include a period identifier. In this case, all motion phases of the driving part under the motion cycle corresponding to the period identifier during the target projection data acquisition period are taken as the at least one motion phase.

[0065] S130, Display the second image sequence corresponding to each of the at least one motion phase.

[0066] For example, in video format, the second image sequence corresponding to each of the at least one motion phase is sequentially displayed in a visualization interface.

[0067] For example, the second image sequence corresponding to each of the at least one motion phase is displayed in a visualization interface in the form of an image array.

[0068] By analyzing the changes in contrast of the region of interest over time using the second image sequence corresponding to at least one motion phase, the physiological state of the region of interest can be determined.

[0069] Figure 2A This diagram illustrates the change of the reconstruction time curve of the right coronary artery root over time, as provided in an embodiment of the present invention. In this diagram, the dashed line represents the change of the reconstruction time curve of the right coronary artery root over time, determined based on the second image sequences corresponding to at least one motion phase; the solid line represents the change of the reconstruction time curve of the coronary artery root over time, simulated based on human-like font models, cardiac motion characteristics, and dynamic changes in coronary blood flow. The reconstruction time curve characterizes the change of the CT value of a target point of interest (POI) within the region of interest in the first image sequence set over time, where the spatial position of the POI changes with the motion of the region of interest. For CT images, the CT value is the pixel intensity value.

[0070] And so on, Figures 2B-2D These are schematic diagrams showing the reconstruction time curves over time for the right coronary artery branch, the left coronary artery root, and the left coronary artery branch, respectively. Figures 2A-2D It can be seen that the dashed lines and solid lines in the attached figures have a high degree of overlap, except in... Figure 2C There is a slight deviation in the middle time segment (of the left coronary artery branch). This indicates that the second image sequence corresponding to each of the at least one motion phase provided in the embodiments of the present invention has high accuracy.

[0071] Figure 3 The six images on the left are the first images and ground truth images of the coronary artery under three predetermined motion phases within the first motion cycle provided by the embodiments of the present invention; Figure 3 The six images on the right are the first and ground truth images of the coronary artery under three predetermined motion phases within the fifth motion cycle provided in this embodiment of the invention. These three predetermined motion phases are the 0% motion phase, 20% motion phase, and 50% motion phase, respectively. The N% motion phase can be understood as the motion phase corresponding to N% of the motion cycle duration. For example, if the first motion cycle duration is 1 second, the 0% motion phase refers to the motion state of the driving part at the beginning of the first motion cycle; the 20% motion phase refers to the motion state of the driving part at 0.2 seconds of the first motion cycle.

[0072] from Figure 3 It can be seen that the first images under the three predetermined motion phases within the first motion cycle have a high similarity to the corresponding ground truth, indicating that the first images determined by the embodiments of the present invention have high accuracy; from Figure 3It can also be seen that, over time, the contrast of the first image corresponding to a later predetermined motion phase is higher than the contrast of the first image corresponding to an earlier predetermined motion phase. Specifically, the influence of contrast agent on non-coronary artery regions is negligible. Therefore, the CT values ​​of non-coronary artery regions in the first image corresponding to a later predetermined motion phase are the same as those in the first image corresponding to an earlier predetermined motion phase, while the CT values ​​of coronary artery regions are different. Furthermore, the coronary artery CT value in the first image corresponding to a later predetermined motion phase is higher than that in the first image corresponding to an earlier predetermined motion phase. This results in a visually significant difference between the coronary artery CT value and the non-coronary artery regions in the first image corresponding to a later predetermined motion phase compared to the difference between the coronary artery CT value and the non-coronary artery regions in the first image corresponding to an earlier predetermined motion phase.

[0073] Figure 4A A contrast diagram of the coronary artery under different viewing angles at time steps 0 and 20 when the driving part is in diastole, as provided in an embodiment of the present invention: Figure 4B This is a contrast diagram of the coronary artery under different viewing angles at time steps of 40 and 60, provided as an embodiment of the present invention, when the driving part is in diastole. Figure 4A and Figure 4B It can be seen that the contrast of the coronary arteries gradually increases with the increase of the time step. For any given time step, the projection image of the first image of the coronary artery at each viewing angle has a high similarity to the corresponding ground truth image, indicating that the first image determined by the embodiments of the present invention has high accuracy.

[0074] Figure 5A A contrast diagram of the coronary artery under different viewing angles at time steps 0 and 20 when the driving part is in the contraction phase, as provided in this embodiment of the invention: Figure 5B This is a contrast diagram of the coronary artery under different viewing angles at time steps of 40 and 60, provided as an embodiment of the present invention, when the driving part is in diastole. Figure 5A and Figure 5B It can be seen that the contrast of the coronary arteries gradually increases with the increase of the time step. For any given time step, the projection image of the first image of the coronary artery at each viewing angle has a high similarity to the corresponding ground truth image, indicating that the first image determined by the embodiments of the present invention has high accuracy.

[0075] Depend on Figure 4A , Figure 4B , Figure 5A and Figure 5B It can be seen that the first images under each motion phase provided in the embodiments of the present invention have high accuracy.

[0076] The technical solution of the CT image sequence determination method provided in this invention is that the motion phase is the motion phase of the driving part, and one motion phase includes at least one sub-phase; the region of interest of the predetermined part changes with the motion of the driving part, and the first image sequence includes a second image sequence corresponding to each motion phase in the at least one motion phase, and each first image in the second image sequence includes the actual contrast information or the expected contrast information of the region of interest under each sub-phase. Therefore, each first image is only used to reflect the state of the region of interest under the corresponding sub-phase, which has higher accuracy than the prior art which reflects the motion state of the region of interest under one or more motion cycles through one image; in this way, by displaying the second image sequence corresponding to the at least one motion phase, the motion change and contrast change of the region of interest during the at least one motion phase of the driving part can be displayed intuitively and accurately.

[0077] Based on the aforementioned embodiments, the time curves corresponding to each voxel in the first image sequence are smoothed to update the first image sequence.

[0078] Specifically, after the first image sequence is determined, the time curve corresponding to each voxel in the first image sequence is determined; for the time curve corresponding to each voxel, deconvolution or other methods are used to smooth it to improve its smoothness, thereby reducing the error caused by tissue overlap and improving the overall image quality.

[0079] Based on the aforementioned embodiments, target dynamic information for the region of interest is determined from CT image cinematography; the target dynamic information includes information on the change of contrast agent intensity over time and / or information on the change of motion state over time.

[0080] This embodiment may use a time-intensity curve to display the change in contrast agent intensity over time. The horizontal axis of this curve represents time, and the vertical axis represents contrast agent intensity. The curve showing the change in contrast agent intensity over time is crucial for transforming static anatomical images into dynamic functional information, revealing the blood flow state within the region of interest. The change in motion state over time can reflect the motion patterns of the region of interest. Taking blood as an example, the change in blood motion state over time can reveal the flow patterns, obstructions, and perfusion status of organs within blood vessels, making it an important indicator for diagnosing vascular diseases.

[0081] Figure 6 This is another flowchart of the CT image sequence determination method provided in this embodiment of the invention. This embodiment refines the determination step of "the second image sequence corresponding to each of the at least one motion phase" in the aforementioned embodiments. Figure 6 As shown, the method includes:

[0082] S210. In response to the first triggering operation, at least one motion phase to be displayed is determined, and target projection data for a predetermined part of the target object is obtained, wherein the motion phase is the motion phase of the driving part, and the region of interest of the predetermined part changes with the motion of the driving part.

[0083] S220. Determine a first image sequence based on the target projection data. The first image sequence includes a second image sequence under each motion phase in the motion phase set. The motion phase set includes all motion phases that occur at the driving part during the target projection data acquisition. The second image sequence includes a first image under each sub-phase within the corresponding motion phase. The first image includes the actual or expected contrast information of the region of interest under the corresponding sub-phase. The motion phase includes at least one sub-phase.

[0084] This step aims to reconstruct a second image sequence based on the target projection data, simultaneously reconstructing all motion phases of the driven part during the target projection data acquisition, rather than reconstructing only the second image sequence of each motion phase within that at least one motion phase.

[0085] S2301. Determine the phase identification information corresponding to the at least one motion phase. The phase identification information includes the period identifier and motion phase identifier of the motion cycle to which each motion phase belongs in the at least one motion phase.

[0086] In scenarios where motion periods are not distinguished, the period identifier of each motion phase in the at least one motion phase is a predetermined value.

[0087] It should be noted that this embodiment does not limit the execution order of S220 and S2301.

[0088] S2302. Based on the phase identification information, select a second image sequence from the first image sequence that corresponds to the at least one motion phase.

[0089] If the phase identification information includes a combination of identifiers for each motion phase in the at least one motion phase, and the identifier combination includes a period identifier and a phase identifier, then a first image corresponding to each identifier combination is selected from the first image sequence, and the selected first image combination is used as a second image sequence for the domestic identifier combination, thereby obtaining the second image sequence corresponding to the at least one motion phase respectively; if the phase identification information includes a combination of identifiers for the starting motion phase and a combination of identifiers for the ending motion phase in the at least one motion phase, then the starting identifier combination of the starting motion phase and the ending identifier combination of the ending motion phase, as well as the set of identifier combinations for all motion phases between the two, are determined, and a second image sequence corresponding to each identifier combination in the set is selected from the first image sequence, thereby obtaining the second image sequence corresponding to the at least one motion phase respectively.

[0090] S2303, Display the second image sequence corresponding to each of the at least one motion phase.

[0091] The technical solution provided by the embodiments of the present invention first determines a first image sequence that includes second image sequences under all motion phases, and then selects a second image sequence corresponding to the at least one motion phase from the first image sequence, thereby improving the flexibility of determining the second image sequence corresponding to the at least one motion phase.

[0092] Figure 7 This is a flowchart illustrating a first image sequence determination method provided in an embodiment of the present invention. This embodiment refines the first image sequence determination steps in the aforementioned embodiments. Figure 7 As shown, the method includes:

[0093] S3201. Based on the target projection data, the parameters of the joint model are optimized to obtain the target joint model. The joint model includes a first term and a second term. The first term is a function used to describe the dynamic change of contrast, with motion phase and sub-phase as independent variables. The second term is a function used to describe the change of contrast with space, with motion phase as independent variable. Both motion phase and sub-phase have time as independent variables.

[0094] The joint model is solved based on the target projection data, and the optimal parameters obtained are substituted into the joint model to obtain the target joint model.

[0095] Since the joint target model is based on the parameter optimization of the target projection data, and the target projection data includes the motion state information and contrast information of the predetermined part under multiple motion phases, the joint target model can determine the second image sequence under each motion phase based on the target projection data of the target object, and can ensure that the determined second image sequence has high accuracy.

[0096] In one embodiment, all target projection data is directly input into the joint model, allowing the joint model to perform parameter optimization based on all received target projection data. The joint model optimization process corresponding to this embodiment is relatively simple.

[0097] In one embodiment, the joint model is determined based on the principles of the Gaussian model. For each spatiotemporal location, the pixel intensity value at the current spatiotemporal location is determined by the sum of the pixel intensity values ​​of all Gaussian kernels at the current spatiotemporal location. The spatiotemporal location corresponds to the target combination result, which includes the target spatial location, the target motion phase, and the target sub-phase. For each Gaussian kernel, the pixel intensity value of the current Gaussian kernel at the current spatiotemporal location is determined based on the pixel intensity value of the current Gaussian kernel center at the target sub-phase and the spatial distribution attribute value of the current Gaussian kernel. The spatial distribution attribute value of the current Gaussian kernel is determined based on the difference between the covariance matrix of the current Gaussian kernel and the target value. The target difference is the difference between the target spatial location and the center spatial location of the current Gaussian kernel. The covariance matrix and the center spatial location both use the motion phase as independent variables, while the pixel intensity value uses the motion phase and sub-phase as independent variables. Specifically:

[0098] In the context of CBCT (cone-beam computed tomography) temporal-resolved imaging, addressing the core challenges of coronary artery involvement due to cardiac motion and the dynamic changes in contrast agent concentration over time during contrast agent injection, this paper proposes a joint model based on radiative Gaussian splatting technology to construct a model for the combined temporal evolution of motion and concentration. This model is trained using all projection data acquired during the contrast enhancement phase. Simultaneously, leveraging the synchronous triggering mechanism between the electrocardiogram (ECG) device and the CBCT scanning system, it accurately obtains a one-to-one correspondence table between projection data and motion phases within the scanning cycle, achieving spatiotemporal alignment between projection data and cardiac rhythm.

[0099] In the joint modeling framework, the model achieves co-modeling of motion and concentration through decoupling design: a unified implicit neural representation (INR) serves as the core of concentration constraints, responsible for learning and outputting globally consistent spatiotemporal evolution laws of contrast agent concentration. The concentration information generated is mapped to independent three-dimensional point clouds corresponding to each heartbeat phase (i.e., the motion field carrier represented by the radiation Gaussian model). Specifically, it can be represented as:

[0100] ;

[0101] ;

[0102] in, For spatial location, For the motion phase, For sub-phase, In three-dimensional space, spatial position Motion phase With sub-phase The pixel intensity value at the corresponding spatiotemporal location; To be identified as in three-dimensional space Gaussian kernel in spatial position Motion phase With sub-phase The pixel intensity value at the corresponding spatiotemporal location; To be identified as Gaussian kernel in motion phase The central spatial location below, The first item refers to the identifier as Gaussian kernel in motion phase Sub-phase At that time, the pixel intensity value at its central spatial location; It indicates the first The covariance matrix of Gaussian kernels; This is the second item.

[0103] As can be seen from the joint model expression, it does not require a specific number of sub-phases included in each motion phase. Therefore, each motion phase can include any number of sub-phases, and correspondingly, the second image sequence can include any number of first images. It is understandable that, since each first image can reflect the state of the corresponding region of interest under the corresponding sub-phase, when the number of sub-phases included in the motion phase is sufficiently large, each sub-phase corresponds to only one instant. Each first image in the second image sequence only needs to reflect the state of the region of interest at the corresponding instant, improving the matching degree between the first image and the state of the region of interest. Therefore, the second image sequence can accurately display the state changes of the region of interest under the corresponding motion phase.

[0104] This joint model allows for the display of the motion state and contrast distribution of a region of interest (ROI) under defined motion phases and sub-phases by specifying the motion phase (motion state) and sub-phases; it also allows for the display of multiple different contrast distributions of the ROI under defined motion phases by specifying the motion phase; and further allows for the display of multiple different motion states of the ROI under defined contrast (sub-phases) by specifying the contrast (see subsequent embodiments for details). In other words, the contrast distribution and motion phase are decoupled in this embodiment. Compared to existing four-dimensional medical imaging technologies that strictly bind the contrast distribution and motion phase together, this invention provides users with more dimensional image information. It is easy to understand that by specifying the motion phase of the target joint model and ensuring that the motion phase includes only one sub-phase, a first image that strictly corresponds to the motion phase and the contrast distribution under that motion phase can be obtained.

[0105] During the joint model parameter optimization process, or rather, during the joint model training process, gradient backpropagation is performed based on the error loss of the projection domain (such as the projection reconstruction residual). While constraining the accuracy of the INR concentration representation, the spatial coordinates of each phase point cloud are simultaneously corrected (optimizing the geometric accuracy of the motion field) and the concentration parameters carried by the point cloud (strengthening the matching degree between concentration and local structure). Ultimately, collaborative modeling of motion and concentration under a decoupled architecture is achieved, which can be expressed as:

[0106] ;

[0107] in, This indicates the model's prediction at time [time]. and projection angle The projection below, Indicates at time and projection angle Label projection below, Indicates L1 loss, Represents structural similarity loss. The weights represent the structural similarity loss.

[0108] It should be noted that neural radiance fields (NeRF) or implicit neural representations (INR) can also be used to directly analyze... Perform modeling. The model can be trained using the loss function described above.

[0109] It should be noted that the second image sequence may include only the first type of first images, or it may include both the first type of first images and the second type of first images. The first type of first images includes the actual contrast information of the region of interest (ROI) under the corresponding sub-phase, while the second type of first images includes the expected contrast information of the ROI under the corresponding sub-phase. The target projection data includes the actual projection viewpoint under the sub-phase corresponding to the first type of first images, and the actual contrast information of the first type of first images is determined based on the projection data under the actual projection viewpoint corresponding to the first images. The target projection data does not include the actual projection viewpoint under the sub-phase corresponding to the second type of first images; therefore, the expected correspondence information of the ROI under the corresponding sub-phase in the second type of first images is calculated by the joint target model based on the target projection data, motion phase, and sub-phase.

[0110] S3202. Determine the first image sequence based on the joint target model.

[0111] In one embodiment, after substituting the optimal parameters into the target joint model, the target joint model can be run to obtain a first image sequence. The first image sequence includes a second image sequence for each motion phase in the motion phase set, wherein the motion phase set includes all motion phases that occur during the acquisition of target projection data by the driving part. This embodiment is adapted to reconstruct a first image sequence including all first images, and then select a second image sequence from the first image sequence corresponding to each of the at least one motion phase.

[0112] In one embodiment, the phase identifiers corresponding to the at least one motion phase are sequentially input into the target joint model to obtain a second image sequence for each motion phase among the at least one motion phase; the union of the second image sequences corresponding to the at least one motion phase is the first image sequence. Specifically, the formula above is used to... The phase identifiers are sequentially replaced with phase identifiers corresponding to each of the at least one motion phase. Executing this formula yields a second image sequence for each of the at least one motion phase. The phase identifier can be an absolute phase identifier, which can be a percentage of the total scan time. This embodiment is suitable for reconstructing only the second image sequences corresponding to the at least one motion phase, without needing to reconstruct second image sequences for other motion phases. This significantly reduces the amount of data computation during image reconstruction, thereby improving the determination speed of the first image sequence.

[0113] The technical solution provided by the embodiments of the present invention, since the target joint model is a joint model after parameter optimization, and the parameter optimization is completed based on the target projection data, can determine a first image sequence including at least one second image sequence based on the target joint model. The second image sequence corresponds to the motion phase and includes the first image under each sub-phase within the corresponding motion phase. In this way, the second image sequence can accurately show the motion state change and contrast change of the region of interest within the corresponding motion phase.

[0114] In the joint modeling phase, a decoupled design was used to guide motion field modeling with concentration modeling, effectively overcoming the core bottleneck of accurately modeling cardiac vessel shapes in complex dynamic scenarios, and successfully obtaining high-precision motion fields corresponding to each heartbeat phase. Considering that a unified INR might reduce concentration accuracy at a given moment due to the need to accommodate high temporal resolution concentration changes across phases, and to further improve the accuracy of contrast agent concentration changes and meet the requirements of coronary perfusion function assessment for detailed concentration distribution, this embodiment utilizes the optimized motion field to perform phase-by-phase concentration fine-tuning. Assuming a desired reconstruction accuracy of 80%, if the reconstruction accuracy of the joint modeling does not meet this expectation, phase-by-phase fine-tuning is required; otherwise, no fine-tuning is needed.

[0115] In the fine-tuning phase, the independent motion fields of each phase serve as spatial constraints, focusing the parameter optimization of the radiation Gaussian model on the concentration representation dimension. For each motion phase, based on the precise spatial coordinates of the point cloud at that phase (determined by the motion field), the core optimization objective is the concentration difference between the reconstruction result of the projection domain and the actual projection data. By backpropagating the gradient through error loss, the concentration representation of INR at that phase is constrained, ensuring a precise match between the concentration distribution and the vascular spatial structure (fixed by the motion field). Ultimately, this achieves refined optimization of the concentration at each phase, further improving the overall accuracy of the motion-concentration joint model.

[0116] Specifically, based on the aforementioned embodiments, the target identifier of the sub-phase also includes the period identifier of its corresponding period. After the first image sequence is determined, the following steps are performed:

[0117] Step a1: Based on the associated gating information corresponding to the target projection data, divide the target projection data into at least one first projection data group. The first projection data group includes projection data under the same motion phase in all motion cycles.

[0118] If at least one motion phase participates in the projection data acquisition within a motion cycle, or in other words, the CT device acquires projection data for that at least one motion phase within each motion cycle of the driving part, then the target projection data is divided into first projection data sets corresponding to each of the at least one motion phase. Each first projection data set includes projection data under the same motion phase across all motion cycles. For example, if each motion cycle has three motion phases, and only the first and second motion phases participate in the projection data acquisition, then the target projection data is divided into two first projection data sets. One first projection data set includes projection data under the first motion phase across all motion cycles, and the other first projection data set includes projection data under the second motion phase across all motion cycles. Since the first projection data sets include projection data under the same motion phase across all motion cycles, each first projection data set can be considered as including projection data of the region of interest under the same motion state, and the accuracy of its image reconstruction result is less affected by the motion of the driving part.

[0119] If the state changes of the region of interest in the predetermined site are mainly affected by the heart, the associated gating information is ECG gating information, which includes the original ECG waveform data and key time point markers parsed from the ECG waveform data, such as R wave markers; if the state changes of the region of interest in the predetermined site are mainly affected by respiration, the associated gating information is respiratory gating information, which includes respiratory waveform data and key time point markers parsed from the respiratory waveform data, such as peak markers and trough markers.

[0120] In a single-energy non-subtraction scenario, the associated gating information is the ECG gating information or respiratory gating information that is synchronously acquired during the target projection data acquisition process.

[0121] In the subtraction angiography scenario, the associated gating information is the ECG gating information or respiratory gating information synchronously acquired during the acquisition of the first or second projection data. The target projection data is determined based on the first and second projection data; the specific determination method is described in the following embodiments.

[0122] Step a2: For each first projection data group, determine the first image combination corresponding to the first projection data combination in the first image sequence, and the first forward projection data of the first image combination under the viewpoint combination corresponding to the first projection data group; determine the difference between the first forward projection data and the first projection data group; if the difference does not meet the predetermined accuracy condition, adjust the parameters of the target joint model according to the difference; update the first image combination for the first projection data group based on the target joint model after parameter adjustment.

[0123] For each first projection data group, the target motion phase corresponding to that first projection data group is determined, as well as the first image combination in the first image sequence corresponding to that target motion phase. The first image combination includes a second image sequence for the target motion phase within each motion cycle. The first image combination is forward-projected under the viewpoint combination corresponding to the first projection data group to obtain first forward-projection data. Then, the difference between the first forward-projection data and the first projection data group is determined. If the difference meets a predetermined accuracy condition, it indicates that the first image combination has high image quality; otherwise, it indicates that the image quality of the first image combination is low. The parameters of the target joint model are adjusted according to the difference, and the first image combination under the corresponding motion phase is re-determined based on the target joint model with adjusted parameters. This first image combination is the updated first image combination.

[0124] Step a3: Determine the updated first image sequence based on all the updated first image combinations.

[0125] The updated first image sequence is obtained by replacing the corresponding first image in the first image sequence with the updated first image combination.

[0126] This embodiment determines whether the first image combination under the motion phase corresponding to each first projection data group has high image quality by determining whether the difference between the first forward projection data and the first projection data group in the first image sequence meets a predetermined accuracy condition. If the difference does not meet the predetermined accuracy condition, the parameters of the target joint model are updated based on the corresponding first projection data group, and the first image combination under the corresponding motion phase is re-determined based on the updated target joint model, thereby achieving the purpose of updating the first image sequence. This embodiment ensures the accuracy of the first image sequence through fine-tuning of model parameters.

[0127] Based on the aforementioned embodiments, step S3201, which optimizes the parameters of the joint model based on the target projection data to obtain the target joint model, can be further refined as follows: obtaining the association gating information of the target projection data; dividing the target projection data into at least one first projection data group based on the association gating information, wherein the first projection data group includes projection data under the same motion phase in all motion cycles; and optimizing the parameters of the joint model based on each first projection data group to obtain the target joint model.

[0128] Specifically, regarding the determination of the first projection data group: In response to the motion phase number selection operation or input operation, a motion phase positioning algorithm corresponding to the number of motion phases is determined, where the number of motion phases is the total number of motion phases within one motion cycle; based on the motion phase positioning algorithm, the associated gating information is analyzed to determine the time period corresponding to each motion phase within each motion cycle, and the projection data corresponding to the time period; the projection data corresponding to the same motion phase participating in data acquisition within all motion cycles are taken as the first projection data group of the corresponding motion phase.

[0129] In this embodiment, the motion cycle type corresponding to the target projection data is configured as optional. If the cycle type is configured as the first type, the target projection data is considered as projection data under one motion cycle, and all motion phases of the driving part belong to a default motion cycle, such as a motion cycle identified as 1. If the cycle type is configured as the second type, the target projection data is considered as periodic projection data, and all motion phases of the driving part are distributed within one or more motion cycles; in this case, when two motion phases belong to different motion cycles, their cycle identifiers are different.

[0130] A corresponding motion phase localization algorithm is pre-matched for each number of motion phases. The phase localization algorithm is used to identify the start and end times of each motion phase in the associated gating information. For example, if the processor detects a selection operation or input operation, it determines the number of motion phases corresponding to that operation, then determines the motion phase localization algorithm corresponding to that number of motion phases, and then analyzes the associated gating information based on the motion phase algorithm to determine the time period corresponding to each motion phase within each motion cycle. For the Nth motion phase, based on the time period corresponding to each motion phase, the projection data of the Nth motion phase within all motion cycles is extracted from the target projection data to obtain the first projection data set for the Nth motion phase.

[0131] Since the first projection data set under each motion phase carries the phase identification information of the corresponding motion phase, the first projection data sets under each motion phase can be input into the joint model sequentially, or all the first projection data sets under each motion phase can be input into the joint model at once. The joint model performs parameter optimization based on the received first projection data sets to obtain the target joint model. In this embodiment, the joint model gradually completes parameter optimization through the first projection data sets under different motion phases. Verification shows that this parameter optimization method can significantly improve the accuracy of parameter optimization.

[0132] Figure 8 This is a flowchart of a first image sequence determination method provided in an embodiment of the present invention, used to refine the first image sequence determination process in the foregoing embodiments. For example... Figure 8 As shown, the method includes:

[0133] S4201. Based on the associated gating information corresponding to the target projection data, the target projection data is divided into at least one first projection data group, and the first projection data group includes the projection data under the same motion phase in all motion cycles.

[0134] S4202. Determine a first image combination for each first projection data group, wherein the first image combination includes the first image under the corresponding motion phase in all motion cycles.

[0135] In one embodiment, for each first projection data combination, it is input into a pre-trained machine learning model to obtain a first image combination for that first projection data group; then, first images in the first image combination whose corresponding motion phase is not within at least one motion phase are deleted to update the corresponding first image combination; the first image combination includes first images under associated motion phases in all motion cycles; the associated motion phase is the motion phase corresponding to the first projection data combination in each motion cycle. For example, for a first projection data group with a first motion phase, the corresponding first image combination includes first images under the first motion phase in each motion cycle.

[0136] In one embodiment, for each first projection data group, a target intermediate image corresponding to the first projection data group is determined; by mapping the contrast information in the projection data under the projection viewpoint corresponding to the target motion phase of each motion cycle to the target intermediate image, a first image under each target motion phase is determined; the summation result of the first images under all target motion phases is used as the first image combination for the first projection data group, where the target motion phase is the motion phase that has a corresponding relationship with the first projection data group.

[0137] For example, a motion cycle includes a first motion phase, a second motion phase, and a third motion phase, wherein the at least one motion phase includes the first motion phase and the second motion phase within the first motion cycle. Taking a first projection data set under the first motion phase as an example, a target intermediate image is obtained by reconstructing its three-dimensional image; the contrast information corresponding to the projection viewpoint under the first motion phase within the first motion cycle is mapped to the target intermediate image to obtain a first image for the first motion phase within the first motion cycle; this first image is the first image combination for the first projection data set.

[0138] For example, a motion cycle includes a first motion phase, a second motion phase, and a third motion phase. The at least one motion phase includes the first motion phase, the second motion phase, and the third motion phase within the first motion cycle, as well as the first motion phase within the second motion cycle. Taking a first projection data set under the first motion phase as an example: Three-dimensional image reconstruction is performed on the first projection data set to obtain a target intermediate image; the contrast information corresponding to the projection viewpoint under the first motion phase within the first motion cycle is mapped to the target intermediate image to obtain a first image for the first motion phase within the first motion cycle; the contrast information corresponding to the projection viewpoint under the first motion phase within the target motion cycle is mapped to the target intermediate image to obtain a first image for the first motion phase within the second motion cycle; the combination of the first image under the first motion phase within the first motion cycle and the first image under the first motion phase within the second motion cycle is used as a first image combination for the first projection data set.

[0139] This embodiment may, but is not limited to, determining the target intermediate image corresponding to the first projection data group based on a machine learning model; or, determining the target intermediate image corresponding to the first projection data group based on an ultra-sparse viewpoint 3D image reconstruction method.

[0140] Among them, the ultra-sparse viewpoint 3D reconstruction method attempts to recover the 3D shape and structure of a predetermined part using projection data from a small number of viewpoints.

[0141] Ultrasparse viewpoint 3D image reconstruction methods can be one of the following reconstruction methods: analytical methods, model-based iterative methods, machine learning methods, deep learning methods, artificial intelligence methods, model-based iterative methods and deep learning methods, or a combination of at least two of these reconstruction methods, or improvements or variations of the following reconstruction methods: analytical methods, model-based iterative methods, machine learning methods, deep learning methods, artificial intelligence methods, model-based iterative methods or deep learning methods.

[0142] The projection method can be one of the projection methods such as ray-driven, voxel-driven, and distance-driven projection methods, or a combination of two or more of them.

[0143] S4203. Determine the first image sequence based on all combinations of first images.

[0144] Arrange the first images within all the first image combinations in chronological order to obtain the first image sequence.

[0145] In this embodiment, a motion phase includes only one sub-phase. The temporal resolution of the second image sequence is the same as the temporal resolution of the motion phase. Therefore, the temporal resolution of the second image sequence can be improved by increasing the temporal resolution of the motion phase, that is, dividing a motion cycle into the required number of motion phases.

[0146] The technical solution provided by the embodiments of the present invention has a small change in the state of the region of interest when the driving part is in the same motion phase within different motion cycles. Therefore, the first projection data group corresponding to each motion phase in the phase deduplication result is determined, and the first image combination corresponding to each first projection data group is determined, so that the first image combination includes the first image under the corresponding target motion phase; then, the first image sequence is determined according to all the first image combinations, which can ensure the accuracy of the first image sequence.

[0147] Figure 9 This is a flowchart of a first image sequence determination method provided in an embodiment of the present invention, used to refine the first image sequence determination process in the foregoing embodiments. For example... Figure 9 As shown, the method includes:

[0148] S5201. Determine the four-dimensional auxiliary image based on the target projection data.

[0149] Based on the associated gating information of the target projection data, the target projection data is divided into at least one first projection data group. Each first projection data group includes projection data under the same motion phase within all motion cycles. The 3D image reconstruction result of each first projection data group is determined, resulting in an auxiliary intermediate image corresponding to each first projection data group. Since there is a correspondence between the first projection data group, the motion phase, and the auxiliary intermediate image, the motion phase and the auxiliary intermediate image are also corresponding. Therefore, the four-dimensional auxiliary image corresponding to all auxiliary intermediate images is determined based on the order of motion phase occurrence. Because this four-dimensional auxiliary image has temporal information, it is dynamic, and the 3D motion field is different under each motion phase.

[0150] Understandably, other methods can also be used to determine the four-dimensional auxiliary image corresponding to the target projection data, such as machine learning models.

[0151] This embodiment may use a method based on ultra-sparse viewpoint 3D reconstruction to determine the 3D image reconstruction results of each first projection data group.

[0152] S5202. Determine at least one motion cycle to which the at least one motion phase belongs, and extract projection data corresponding to each motion cycle in the at least one motion cycle from the target projection data based on the association gating information of the target projection data, to obtain a second projection data group for each motion cycle.

[0153] If the at least one motion phase belongs to a motion cycle, then the projection data corresponding to the motion cycle is extracted from the target projection data to obtain a second projection data set for the motion cycle; if the at least one motion phase is distributed across at least two motion cycles, then the projection data corresponding to each of the at least two motion cycles is extracted from the target projection data to obtain a second projection data set for each motion cycle.

[0154] S5203. For any motion cycle, based on the four-dimensional auxiliary image, determine the three-dimensional deformation motion field of all motion phases relative to each motion phase, and obtain the first deformation motion field combination for each motion phase. The number of three-dimensional deformation motion fields in the first deformation motion field combination is equal to the number of motion phases included in one motion cycle.

[0155] A three-dimensional deformation motion field is a vector field that defines the motion vector of every point in three-dimensional space from one time frame to the next. Simply put, it describes the motion throughout the entire three-dimensional space. It allows motion to be more than just simple rigid transformations (such as translation and rotation), but can include complex, non-rigid changes such as stretching, bending, and compression. It assigns a three-dimensional motion vector to each point in space, which indicates the displacement of that point from time t to time t+1.

[0156] The method for determining the three-dimensional deformation (elastic) motion field can be one of the following: analytical method, model-based iterative method, machine learning method, deep learning method, artificial intelligence method, model-based iterative method and deep learning method, or an improvement of one of them, or a combination of two or more of them.

[0157] S5204. For each motion phase in the at least one motion phase, based on the first deformation motion field combination corresponding to all motion phases within the motion phase and the second projection data group under the motion period of the motion phase, determine the second image sequence for the motion phase.

[0158] If a motion phase belongs to the Nth motion cycle, then based on the first deformation motion field combination corresponding to all motion phases within that motion phase and the second projection data group under the Nth motion cycle, a second image sequence for that motion phase is determined, wherein N is greater than or equal to 1 and less than or equal to the total number of motion cycles that occur at the driving part during the target projection data acquisition period, and the total number of motion cycles is the rounded-up result of the actual number of motion cycles.

[0159] For each motion phase under a given motion phase, the current motion phase is taken as the reference motion phase. The second projection data group corresponding to the motion cycle of the reference motion phase is determined, as well as the reference projection data corresponding to the reference motion phase in the second projection data group, and the projection data under each other motion phase. Based on the projection data under each other motion phase and the three-dimensional deformation motion field of each other motion phase compared to the reference motion phase, the projection data under each other motion phase is updated. Based on the reference projection data corresponding to the reference motion phase and the updated projection data under each other motion phase, the first image for the reference motion phase is determined. This process is repeated to obtain the first images under all motion phases within the motion phase, thereby obtaining the second image sequence under the motion phase.

[0160] The process involves determining the first image for the reference motion phase based on the reference projection data corresponding to the reference motion phase and the updated projection data for each of the other motion phases.

[0161] In one embodiment, based on the second set of deformed motion fields and the second projection data group under the motion period of the motion phase, a second intermediate image combination for the motion phase is determined. The number of intermediate images in the second intermediate image combination is equal to the number of projection view combination combinations included in the motion period. According to the contrast information corresponding to the projection view under each motion phase in the motion period, the corresponding intermediate images in the second intermediate image combination are updated to obtain the second image sequence for the motion phase.

[0162] This embodiment can be based on an ultra-limited viewpoint, a motion-compensated 3D reconstruction method, and a first deformation motion field combination to determine a second intermediate image combination corresponding to the second projection data group under the motion cycle, so as to improve the image quality of each intermediate image in the second intermediate image combination.

[0163] Specifically, after determining the reference projection data corresponding to the reference motion phase and the updated projection data under each other motion phase, the union of the reference projection data and the updated projection data under each other motion phase is determined; the union is then used to reconstruct an image based on the ultra-limited viewpoint and motion compensation 3D reconstruction method to obtain the first image for the reference motion phase; and so on, a second intermediate image combination corresponding to the second projection data group under the motion period can be obtained.

[0164] After the second intermediate image combination is determined, for each motion phase, the contrast information corresponding to the projection viewpoint under that motion phase within the corresponding motion cycle is mapped to the corresponding intermediate image in the second intermediate image combination. During the mapping process, the first deformation motion field combination for that motion phase is considered to update the intermediate image. In this way, the updated intermediate image includes both the contrast information under the corresponding motion phase and conforms to the motion state corresponding to the corresponding motion phase.

[0165] Of course, once the combination of each first deformation motion field and the second projection data group under the motion period of the motion phase are determined, other image processing methods can also be used to determine the second image sequence corresponding to the at least one motion phase, such as machine learning models.

[0166] S5205, take the union of all second image sequences as the first image sequence.

[0167] The union of the second image sequences corresponding to each motion phase in at least one motion phase is the first image sequence.

[0168] The technical solution provided by the embodiments of the present invention determines the second image sequence under each motion phase through motion compensation, thereby obtaining a first image sequence including the second image sequence under each motion phase; and this embodiment can reconstruct only the second image sequence corresponding to each motion phase in the at least one motion phase, without having to reconstruct the second image sequence under each motion phase in all motion cycles, thereby significantly reducing the amount of data processing and improving the determination speed of the first image sequence.

[0169] Based on the aforementioned embodiments, after determining the first deformation motion field combination corresponding to each motion phase, the union of the first deformation motion field combinations corresponding to all motion phases within one motion cycle is taken as the first deformation motion field set. Based on a predetermined deformation motion field interpolation method, a second deformation motion field set corresponding to the first deformation motion field set is determined. The first deformation motion field set includes the first deformation motion field combination under each motion phase, and the second deformation motion field set includes the second deformation motion field combination under each projection viewpoint combination. All projection views corresponding to each motion phase are divided into at least two projection viewpoint combinations, each including one projection viewpoint or multiple adjacent projection viewpoints. For each motion phase within the at least one motion phase, based on the first deformation motion field combinations corresponding to all motion phases within the motion phase combination and the second projection data group within the motion cycle of the motion phase, a second image sequence for the motion phase is determined. The motion phase combination includes all motion phases within one motion cycle.

[0170] Specifically, for each motion phase, all corresponding projection viewpoints are divided into at least two projection viewpoint combinations. Each projection viewpoint combination includes at least one projection viewpoint, and when a projection viewpoint combination includes at least two projection viewpoints, these at least two projection viewpoints are adjacent. The motion field of this motion phase can be regarded as the three-dimensional motion field under its intermediate projection viewpoint combination. The three-dimensional deformation motion field between this motion phase and adjacent motion phases is the superposition result of the three-dimensional deformation motion fields of each projection viewpoint combination in the projection viewpoint set relative to the intermediate projection viewpoint combination. The projection viewpoint set includes all projection viewpoint combinations between the intermediate projection viewpoint combination of this motion phase and the intermediate projection viewpoint combination of adjacent motion phases. Therefore, by using deformation field interpolation methods, such as thin plate spline interpolation and other existing interpolation methods, the three-dimensional deformation motion field of each projection viewpoint combination in the projection viewpoint set relative to the intermediate projection viewpoint combination can be obtained. By analogy, the three-dimensional deformation motion field of all projection viewpoint combinations relative to the intermediate projection viewpoint combination can be obtained. Adding the three-dimensional deformation motion field of each projection viewpoint combination relative to itself, a second deformation motion field combination for this projection viewpoint combination can be obtained. At this point, for each motion phase, based on the second set of deformed motion fields and the second set of projection data within the motion period of that motion phase, a second image sequence for that motion phase can be determined. Each first image in this second image sequence corresponds to a combination of projection viewpoints. It can be understood that when the combination of projection viewpoints includes only one projection viewpoint, each first image in the second image sequence corresponds to one projection viewpoint. Each first image reflects the motion state and contrast of the region of interest at the moment corresponding to that first image's projection viewpoint.

[0171] Figure 10This is a flowchart of a first image sequence determination method provided in an embodiment of the present invention, used to refine the first image sequence determination process in the foregoing embodiments. For example... Figure 10 As shown, the method includes:

[0172] S610. In response to the first triggering operation, determine at least one motion phase to be displayed, and acquire target projection data for a predetermined part of the target object, wherein the motion phase is the motion phase of the driving part, and the region of interest of the predetermined part changes with the motion of the driving part.

[0173] S6201. Determine the four-dimensional auxiliary image based on the target projection data.

[0174] S6202. Based on the associated gating information of the target projection data, the target projection data is divided into projection data corresponding to each motion cycle, to obtain a second projection data group for each motion cycle.

[0175] Based directly on the associated gating information, the target projection data is divided into a second projection data group corresponding to each motion cycle according to the motion cycle.

[0176] S6203. For any motion cycle, based on the four-dimensional auxiliary image, determine the three-dimensional deformation motion field of all motion phases relative to each motion phase, and obtain the first deformation motion field combination for each motion phase. The number of three-dimensional deformation motion fields in the first deformation motion field combination is equal to the number of motion phases included in one motion cycle. Each motion phase corresponds to one or more projection viewpoints.

[0177] S6204. For each motion cycle, based on each first deformation motion field combination and the second projection data group under the motion cycle in which the motion phase is located, determine a third image combination for each motion cycle. The third image combination includes the second image sequence under each motion phase in the corresponding motion cycle.

[0178] For each motion cycle, for each motion phase, it is taken as a reference motion phase. The second projection data set corresponding to the motion cycle is determined, as well as the reference projection data corresponding to the reference motion phase in the second projection data set, and the projection data under each other motion phase. Based on the projection data under each other motion phase and the three-dimensional deformation motion field of each other motion phase compared to the reference motion phase, the projection data under each other motion phase is updated. Based on the reference projection data corresponding to the reference motion phase and the updated projection data under each other motion phase, a first image for the reference motion phase is determined. By analogy, a second image sequence under that motion phase is obtained. The union of the second image sequences under all motion phases within the motion cycle is taken as a third image combination for the motion cycle.

[0179] S6205, take the union of all third images as the first image sequence.

[0180] S630. Select a second image sequence from the first image sequence that corresponds to the at least one motion phase, and display the second image sequence that corresponds to the at least one motion phase.

[0181] The summation of all third images is taken as a first image sequence, which includes a first image for each motion phase of each motion cycle. Based on the phase identifier of each motion phase in the at least one motion phase, the first image corresponding to each motion phase is selected from the first image sequence to obtain a second image sequence for each motion phase, and the second image sequence corresponding to the at least one motion phase is displayed.

[0182] The technical solution provided by this invention determines the third image combination under each motion cycle through motion compensation, thereby obtaining a first image sequence including the third image combination under each motion cycle, specifically including the first image under each motion cycle and each motion phase; then, according to the identifier combination of each motion phase in the at least one motion phase, the first image corresponding to each motion phase is selected from the first image sequence to obtain the second image sequence corresponding to each motion phase, and the second image sequence corresponding to the at least one motion phase is displayed, thus achieving the purpose of determining the second image sequence corresponding to the at least one motion phase by combining full reconstruction and selection.

[0183] Figure 11 This is a flowchart of a third image sequence determination method provided in an embodiment of the present invention. This embodiment adds a third image sequence determination step based on the aforementioned embodiments. Figure 11 As shown, the method includes:

[0184] S701, in response to a second trigger operation for contrast, determines the target contrast.

[0185] For example, the interactive interface displays a contrast input box. After the user enters the target contrast in the contrast input box, a second trigger operation is generated by clicking or touching the confirmation option; upon detecting the second trigger operation, the processor determines the target contrast corresponding to the second trigger operation.

[0186] For example, the interactive interface displays candidate contrast ratios. The user triggers a second action by clicking or touching the desired contrast ratio; upon detecting this second trigger action, the processor uses the contrast ratio clicked or touched by the user as the target contrast ratio.

[0187] For example, the interactive interface displays a second image sequence corresponding to each of the at least one motion phase. The user selects a target first image from the second image sequence corresponding to each of the at least one motion phase by clicking or touching. In response to the selection of the target first image, the contrast of the target first image is used as the target contrast. This embodiment allows the user to intuitively select the desired target contrast.

[0188] For example, a sub-phase input option or selection option is displayed on the interactive interface; in response to a sub-phase input operation or a sub-phase selection operation, the contrast corresponding to the input sub-phase or the selected sub-phase is used as the target contrast.

[0189] S702. Determine a third image sequence corresponding to the target contrast based on the target projection data. The third image sequence is used to display the motion state changes of the region of interest within at least one motion cycle under the target contrast.

[0190] In one embodiment, the sub-phase corresponding to the target contrast is input into the target joint model, and the target joint model lets... It is the pixel intensity value corresponding to the sub-phase, and determines the third image sequence corresponding to the sub-phase. This third image sequence is used to show the motion state changes of the region of interest within at least one motion cycle under the target contrast, and the at least one motion cycle can be selected as all motion cycles experienced by a predetermined part during the target projection data acquisition.

[0191] Specifically, a sub-phase with target contrast is determined based on the first image sequence, and then the sub-phase with the target contrast is input into the target joint model so that the target joint model outputs a third image sequence corresponding to the target contrast.

[0192] S703, Display the third image sequence corresponding to the target contrast.

[0193] The third image sequence is displayed in the visualization interface as a video or image array.

[0194] Compared to existing technologies, where a single contrast level typically corresponds to only one image, users cannot observe the motion state of different motion phases at the same contrast level. In this embodiment, however, users can intuitively observe the motion state of the region of interest at the target contrast level across all motion phases within at least one motion cycle through the third image sequence. This provides users with more diagnostic assistance information and helps improve the accuracy of their clinical diagnoses.

[0195] In one embodiment, the third image sequence specifically includes second images of the region of interest at each sub-phase within at least one motion cycle, under target contrast. Therefore, the number of second images in the third image sequence is at least equal to the number of sub-phases within each motion cycle.

[0196] It should be noted that the first trigger operation and the second trigger operation can occur simultaneously or one after the other; and the second trigger operation can occur before or after the first trigger operation.

[0197] It should be noted that S701-S703 in this embodiment of the invention and S110-S130 in the previous embodiment have no execution order restrictions. Their specific execution order depends on the triggering order of the user's first trigger operation and the second trigger operation.

[0198] The technical solution provided by the embodiments of the present invention displays the changes in the motion state of the region of interest within at least one motion cycle under the target contrast through a third image sequence, and shows the user the state of different motion phases of the region of interest under the target contrast.

[0199] Based on the foregoing embodiments, S701, in response to the second trigger operation for contrast, determines the target contrast, which can be refined to determine the target contrast and a default motion period, such as 1 or all motion periods, in response to the second trigger operation for contrast. Accordingly, the third image sequence determined in S702 includes the second image under each sub-phase within the default motion period, or includes the second image under each sub-phase within all motion periods. In one embodiment, the default third image sequence corresponds to only one default motion period.

[0200] Based on the foregoing embodiment, S701, in response to the second trigger operation for contrast, determining the target contrast before, after, or simultaneously, further includes determining at least one motion cycle in response to the trigger operation for the cycle identifier; correspondingly, the third image sequence determined in S702 includes the second image at each motion phase within the at least one motion cycle. This embodiment allows the user to determine at least one motion cycle corresponding to the third image sequence as needed.

[0201] Figure 12 This flowchart illustrates a third image sequence determination method provided in an embodiment of the present invention, used to refine the third image sequence determination steps in the aforementioned embodiments. For example... Figure 12 As shown, the method includes:

[0202] S8501. Determine the four-dimensional auxiliary image based on the target projection data.

[0203] S8502. Based on the associated gating information of the target projection data, the target projection data is divided into second projection data groups corresponding to each motion cycle.

[0204] S8503. For any motion cycle, determine the three-dimensional deformation motion field of all motion phases relative to each motion phase based on the four-dimensional auxiliary image, and obtain the first deformation motion field combination for each motion phase.

[0205] S8504. For each motion cycle, based on the second projection data group and each first deformation motion field combination under the motion cycle, determine the first intermediate image combination for the motion cycle, and map the target contrast onto each intermediate image in the first intermediate image combination to obtain the second image combination corresponding to the target contrast.

[0206] The number of intermediate images in the first intermediate image combination is equal to the number of three-dimensional deformation motion fields in the first deformation motion field combination, and the number of three-dimensional deformation motion fields in the first deformation motion field combination is equal to the number of motion phases included in one motion cycle.

[0207] For each motion phase within a motion cycle, the current motion phase is taken as the reference motion phase. A second projection data set corresponding to the motion cycle in which the reference motion phase is located is determined, along with reference projection data corresponding to the reference motion phase within the second projection data set, and projection data for each other motion phase. Based on the projection data for each other motion phase and the three-dimensional deformation motion field of each other motion phase compared to the reference motion phase, the projection data for each other motion phase is updated. The union of the reference projection data and the updated projection data for each other motion phase is determined. Image reconstruction is performed on this union to obtain an intermediate image for the reference motion phase. By analogy, intermediate images for all motion phases can be obtained, thus obtaining a first intermediate image combination for the motion cycle.

[0208] After the first intermediate image combination is determined, the target contrast is mapped onto each intermediate image in the first intermediate image combination to obtain the second image combination corresponding to the target contrast.

[0209] S8505, take the union of all the second images as the third image sequence.

[0210] The union of the second images from all motion cycles is taken as the third image sequence. Here, "all motion cycles" refers to all motion cycles occurring at the driving part during the target projection data acquisition period.

[0211] The technical solution provided by the embodiments of the present invention first determines a first intermediate image combination for a motion cycle by means of motion compensation, then maps the target contrast onto each intermediate image of the first intermediate image combination to obtain a second image combination for a motion cycle; then summarizes all the second image combinations together to obtain a third image sequence.

[0212] Based on the aforementioned embodiments, before S8502, which divides the target projection data into second projection data groups corresponding to each motion cycle based on the association gating information of the target projection data, the method further includes: determining a cycle identifier in response to a third trigger operation. At this point, S8502 can be refined to extract projection data corresponding to the motion cycle of the cycle identifier from the target projection data based on the association gating information of the target projection data, thus obtaining a second projection data group for the cycle corresponding to the cycle identifier. During subsequent image reconstruction, only the second image combination corresponding to the motion cycle of that cycle identifier needs to be reconstructed. This embodiment only reconstructs the second image under the motion cycle of interest to the user; therefore, the third image sequence also only includes the second image under the motion cycle of interest to the user. This improves the user experience and significantly reduces the amount of data computation during the determination of the third image sequence, thereby increasing the determination speed of the third image sequence.

[0213] Figure 13 This is a flowchart of a target projection data determination method provided in an embodiment of the present invention. This embodiment adds a step of the target projection data determination method before determining the first image sequence based on the target projection data in the aforementioned embodiments. For example... Figure 13 As shown, the method for determining the target projection data includes:

[0214] S901. Obtain first projection data and second projection data for a predetermined part of the target object, wherein the imaging conditions of the first projection data and the second projection data are different.

[0215] In one embodiment, the first projection data is acquired at a first energy level, and the second projection data is acquired at a second energy level. Furthermore, the target object is injected with contrast agent during both the acquisition of the first and second projection data. In other words, the first and second projection data are projection data acquired based on dual-energy imaging, and the imaging condition variable for both is energy.

[0216] In one embodiment, the first projection data is acquired with contrast agent injected, and the second projection data is acquired without contrast agent injection. That is, the imaging condition variable for both is whether or not contrast agent is injected.

[0217] S902. Determine target projection data based on the first projection data and the second projection data. The target projection data includes the change in target tissue information corresponding to the first projection data compared to the target tissue information corresponding to the second projection data, caused by imaging condition variables, under the condition that the reference tissue component values ​​are the same.

[0218] Reference tissue components refer to tissue components that are insensitive to imaging condition variables and need to be removed, such as bone tissue and diaphragm. Specifically, when the predetermined site only includes bones, such as the brain, the reference tissue is bones; when the predetermined site includes both bones and diaphragm, such as the abdomen, the reference tissue is both bones and diaphragm.

[0219] The embodiments of the present invention use bone and diaphragm as reference tissues to illustrate the technical solutions.

[0220] For scenarios where dual-energy imaging is used to acquire first and second projection data, the integral values ​​of each line in the first projection data are modified to obtain updated first projection data. The reference tissue component values ​​corresponding to the updated first projection data are consistent with the bone component values ​​and diaphragm component values ​​corresponding to the second projection data, respectively. The difference between the updated first and second projection data is used as the target projection data.

[0221] By modifying the line integral values ​​in the first projection data, the bone and diaphragm component values ​​corresponding to the modified first projection data are made consistent with those corresponding to the second projection data. Thus, the difference between the two—that is, the bone and diaphragm component values ​​corresponding to the target projection data—is zero. In this case, the target projection data can be considered to include only the change in target tissue information corresponding to the first projection data compared to the target tissue information corresponding to the second projection data.

[0222] For scenarios where the first projection data is acquired with contrast agent injection and the second projection data is acquired without contrast agent injection, the following steps are taken: First, determine if the single scan range of the CT equipment cannot completely cover the predetermined area. If so, the difference between the first and second projection data is used as the target projection data. If not, determine the first image reconstruction result corresponding to the first projection data and the second image reconstruction result corresponding to the second projection data. Using a reference tissue as a reference, register the second image reconstruction result to the first image reconstruction result to obtain the registration result. Based on the viewpoint combination corresponding to the first projection data, perform the forward projection of the second image reconstruction result in the registration result to obtain the second forward projection data. Based on the principle of subtraction within the same viewpoint, the difference between the first and second forward projection data is used as the target projection data.

[0223] Specifically, on the one hand, bone structures and the diaphragm are generally unaffected by contrast agents in CT imaging; on the other hand, bone structures and the diaphragm are relatively stable anatomical structures. Therefore, using bone structures and the diaphragm as reference benchmarks, after aligning the space of the second image reconstruction result to the space of the first image reconstruction result to obtain the registration result, the CT values ​​of the bone structures in the two images in this registration result are the same, and the CT values ​​of the diaphragm are also the same. Based on the viewpoint combination corresponding to the first projection data, the forward projection of the second image reconstruction result in the registration result is completed to obtain the second forward projection data. In this way, the viewpoint combination corresponding to the first projection data is the same as the viewpoint combination corresponding to the second forward projection data, and the bone component values ​​and diaphragm component values ​​in the first projection data are the same as those in the second forward projection data. Therefore, based on the principle of subtraction of the same viewpoint, the difference between the first projection data and the second forward projection data is determined, and this difference is used as the target projection data, which makes the bone component values ​​and diaphragm component values ​​in the target projection data all zero. Thus, the target projection data can be regarded as only including the change in the target tissue information corresponding to the first projection data compared to the target tissue information corresponding to the second projection data.

[0224] The technical solution provided by the embodiments of the present invention expands the application scenarios of the CT image sequence determination method described in the embodiments of the present invention by including the change in target tissue information corresponding to the first projection data compared to the target tissue information corresponding to the second projection data under the condition that the reference tissue component values ​​are the same, caused by imaging condition variables.

[0225] Based on the aforementioned example, in the single-energy subtraction scenario, S902 may further include: extracting non-interest tissue from the image reconstruction result corresponding to the target projection data to obtain a non-interest tissue segmentation result; determining the third forward projection data of the non-interest tissue segmentation result based on a predetermined viewpoint combination; using the difference between the target projection data and the third forward projection data as the updated target projection data, and the predetermined viewpoint combination as the viewpoint combination corresponding to the first projection data.

[0226] If the tissue of interest is blood vessels, then all tissues in the image reconstruction result other than blood vessels and reference tissue are considered non-tissues of interest.

[0227] Specifically, after the image reconstruction result is determined, the segmentation result of the non-interest-bearing tissue (NIT) corresponding to the image reconstruction result can be determined based on a pre-trained machine learning model; alternatively, the NIT can be extracted from the image reconstruction result first to obtain the NIT segmentation result, and then the difference between the image reconstruction result and the NIT segmentation result can be used as the NIT segmentation result. After the NIT segmentation result is determined, the viewpoint combination corresponding to the first projection data is used as a predetermined viewpoint combination, and the third forward projection data of the NIT segmentation result is determined based on the predetermined viewpoint combination. It can be understood that the third forward projection data only includes the line integral values ​​of the NIT. In this case, calculating the difference between the target projection data and the third forward projection data can be understood as setting the line integral values ​​of the NIT in the target projection data to zero, so that the difference only includes the line integral values ​​of the NIT, that is, the updated target projection data only includes the line integral values ​​of the NIT.

[0228] Figure 14A This is a schematic diagram of the CT image sequence determination device provided in an embodiment of the present invention. Figure 14A As shown, the CT image sequence determination device includes:

[0229] The first trigger module 110 is used to respond to the first trigger operation, determine at least one motion phase to be displayed, and acquire target projection data for a predetermined part of the target object, wherein the motion phase is the motion phase of the driving part, and the region of interest of the predetermined part changes with the motion of the driving part.

[0230] The first image sequence module 120 is used to determine a first image sequence based on the target projection data. The first image sequence includes a second image sequence corresponding to each of the at least one motion phases. The second image sequence includes a first image under each sub-phase within the corresponding motion phase. The first image includes the actual or expected contrast information of the region of interest under the corresponding sub-phase. The motion phase includes at least one of the sub-phases.

[0231] The first display module 130 is used to display the second image sequence corresponding to the at least one motion phase.

[0232] In one embodiment, the image identifier of the first image includes a phase identifier of the motion phase in which the first image is located and a sub-phase identifier of the sub-phase in which it is located, or includes a period identifier of the motion cycle to which the first image belongs, a phase identifier of the motion phase in which it is located, and a sub-phase identifier of the sub-phase in which it is located.

[0233] In one embodiment, the first image sequence includes the second image sequence under each of the motion phases in the motion phase set, the motion phase set including all the motion phases that occur during the target projection data acquisition by the driving part;

[0234] like Figure 14B As shown, the device also includes a selection module 140 disposed in front of the first display device, the selection module 140 being used for:

[0235] Determine phase identification information corresponding to the at least one motion phase, wherein the phase identification information includes the period identifier and motion phase identifier of the motion cycle to which each of the at least one motion phase belongs;

[0236] Based on the phase identification information, a second image sequence corresponding to each of the at least one motion phase is selected from the first image sequence.

[0237] In one embodiment, the first image sequence module 120 includes:

[0238] The parameter optimization unit is used to optimize the parameters of the joint model based on the target projection data to obtain the target joint model. The joint model includes a first term and a second term. The first term is a function used to describe the dynamic change of contrast. The first term uses the motion phase and the sub-phase as independent variables. The second term is a function used to describe the change of contrast with space. The second term uses the motion phase as an independent variable. Both the motion phase and the sub-phase use time as an independent variable.

[0239] The first image sequence unit is used to determine the first image sequence based on the target joint model.

[0240] In one embodiment, the joint model is determined based on the principles of the Gaussian model;

[0241] For each spatiotemporal location, the pixel intensity value at the current spatiotemporal location is determined by the sum of the pixel intensity values ​​of all Gaussian kernels at the current spatiotemporal location;

[0242] Wherein, the spatiotemporal position corresponds to the target combination result, and the target combination result includes the target spatial position, the target motion phase, and the target sub-phase; for each Gaussian kernel, the pixel intensity value of the current Gaussian kernel at the current spatiotemporal position is determined based on the pixel intensity value of the current Gaussian kernel center at the target sub-phase within the target motion phase and the spatial distribution attribute value of the current Gaussian kernel. The spatial distribution attribute value of the current Gaussian kernel is determined based on the covariance matrix of the current Gaussian kernel and the target difference value. The target difference value is the difference between the target spatial position and the center spatial position of the current Gaussian kernel at the target motion phase. The covariance matrix and the center spatial position both use the motion phase as independent variables.

[0243] In one embodiment, such as Figure 14C As shown, the first display module is preceded by a parameter optimization module 150, which is used for:

[0244] Based on the associated gating information corresponding to the target projection data, the target projection data is divided into at least one first projection data group, and the first projection data group includes projection data under the same motion phase in all motion cycles.

[0245] For each of the first projection data groups, determine the first image combination corresponding to the first projection data combination in the first image sequence, and the first forward projection data of the first image combination under the viewpoint combination corresponding to the first projection data group; determine the difference between the first forward projection data and the first projection data group; if the difference does not meet the predetermined accuracy condition, adjust the parameters of the target joint model according to the difference; update the first image combination for the first projection data group based on the parameter-adjusted target joint model;

[0246] Based on all the updated combinations of the first images, an updated sequence of the first images is determined.

[0247] In one embodiment, the parameter optimization unit is used for:

[0248] Obtain the associated gating information of the target projection data;

[0249] Based on the associated gating information, the target projection data is divided into at least one first projection data group, and the first projection data group includes projection data under the same motion phase in all motion cycles.

[0250] Based on each of the first projection data groups, the parameters of the joint model are optimized to obtain the target joint model.

[0251] In one embodiment, such as Figure 14D As shown, the device also includes:

[0252] The second trigger module 160 is used to determine the target contrast in response to a second trigger operation for contrast.

[0253] The third image sequence module 170 is used to determine a third image sequence corresponding to the target contrast based on the target projection data. The third image sequence is used to display the motion state changes of the region of interest within at least one motion cycle under the target contrast.

[0254] The second display module 180 is used to display the third image sequence corresponding to the target contrast.

[0255] In one embodiment, the third image sequence module 170 is used for:

[0256] The sub-phase corresponding to the target contrast is input into the target joint model to obtain a third image sequence corresponding to the target contrast.

[0257] In one embodiment, the third image sequence module 170 is used for:

[0258] A four-dimensional auxiliary image is determined based on the target projection data;

[0259] Based on the associated gating information of the target projection data, the target projection data is divided into a second projection data group corresponding to each of the motion cycles.

[0260] For any of the motion cycles, the three-dimensional deformation motion field of all the motion phases relative to each of the motion phases is determined based on the four-dimensional auxiliary image, to obtain the first deformation motion field combination for each of the motion phases;

[0261] For each of the motion cycles, based on the second projection data group under the motion cycle and each of the first deformation motion field combinations, a first intermediate image combination for the motion cycle is determined, and the target contrast is mapped onto each of the intermediate images in the first intermediate image combination to obtain a second image combination corresponding to the target contrast.

[0262] The union of all the second images is taken as the third image sequence;

[0263] Wherein, the number of intermediate images in the first intermediate image combination is equal to the number of three-dimensional deformation motion fields in the first deformation motion field combination, and the number of three-dimensional deformation motion fields in the first deformation motion field combination is equal to the number of motion phases included in one motion cycle.

[0264] In one embodiment, the first image sequence module 120 includes:

[0265] The projection data partitioning unit is used to partition the target projection data into at least one first projection data group based on the associated gating information corresponding to the target projection data. The first projection data group includes projection data under the same motion phase in all motion cycles.

[0266] An image determination unit is configured to determine a first image combination for each of the first projection data groups, wherein the first image combination includes the first image under the corresponding motion phase within all motion cycles;

[0267] An image combination unit is configured to determine the first image sequence based on all the first image combinations.

[0268] In one embodiment, the image determination unit is used for:

[0269] For each of the first projection data groups, a target intermediate image corresponding to the first projection data group is determined; by mapping the contrast information in the projection data under the projection viewpoint corresponding to the target motion phase of each motion cycle to the target intermediate image, the first image under the target motion phase in each motion cycle is determined; the summary result of all the first images under the target motion phase is used as the first image combination for the first projection data group, where the target motion phase is the motion phase that has a corresponding relationship with the first projection data group.

[0270] In one embodiment, the image determination unit determines the target intermediate image corresponding to the first projection data group based on an ultra-sparse viewpoint 3D image reconstruction method.

[0271] In one embodiment, the first image sequence unit includes:

[0272] An auxiliary unit is used to determine a four-dimensional auxiliary image based on the target projection data;

[0273] The projection data segmentation unit is used to determine at least one motion cycle to which the at least one motion phase belongs, and extracts projection data corresponding to each of the at least one motion cycle from the target projection data based on the association gating information of the target projection data, so as to obtain a second projection data group for each of the motion cycles.

[0274] The deformation motion field unit is used to determine, based on the four-dimensional auxiliary image, the three-dimensional deformation motion field of all motion phases relative to each motion phase for any motion cycle, to obtain a first deformation motion field combination for each motion phase, wherein the number of three-dimensional deformation motion fields in the first deformation motion field combination is equal to the number of motion phases included in one motion cycle.

[0275] An image sequence unit is configured to determine a second image sequence for each of the at least one motion phases, based on a first deformation motion field combination corresponding to all the motion phases within a motion phase combination and a second projection data group under the motion cycle in which the motion phase is located. The motion phase combination includes all the motion phases within one motion cycle.

[0276] An image summarization unit is used to take the union of all the second image sequences as the first image sequence.

[0277] In one embodiment, the image sequence unit is used to: determine a second set of deformation motion fields corresponding to a first set of deformation motion fields based on a predetermined deformation motion field interpolation method. The first set of deformation motion fields includes combinations of first deformation motion fields under each motion phase, and the second set of deformation motion fields includes combinations of second deformation motion fields under each combination of projection viewpoints. All projection viewpoints corresponding to each motion phase are divided into at least two combinations of projection viewpoints, and the combination of projection viewpoints includes one projection viewpoint or multiple adjacent projection viewpoints.

[0278] For each of the at least one motion phases, a second image sequence is determined based on the second set of deformable motion fields and the second projection data set under the motion period of the motion phase.

[0279] In one embodiment, the image sequence unit includes:

[0280] The first subunit is used to determine a second intermediate image combination for the motion phase based on the second set of deformed motion fields and the second projection data group under the motion period of the motion phase. The number of intermediate images in the second intermediate image combination is equal to the number of projection view combination combinations included in the motion period.

[0281] The second subunit is used to update the corresponding intermediate image in the second intermediate image combination according to the contrast information corresponding to the projection viewpoint under each of the second deformation motion field combinations and each motion phase within the motion period, so as to obtain the second image sequence for the motion phase.

[0282] In one embodiment, the first subunit determines a second intermediate image combination for the motion phase based on an ultra-limited viewpoint, a motion-compensated 3D reconstruction method, and the second projection data set under the motion period of the motion phase.

[0283] In one embodiment, if the associated gating information is ECG gating information, then the movement cycle is the cardiac cycle;

[0284] If the associated gating information is respiratory gating information, then the movement cycle is the respiratory movement cycle.

[0285] In one embodiment, the first triggering unit includes:

[0286] The data acquisition unit is used to acquire first projection data and second projection data for a predetermined part of the target object, wherein the imaging conditions of the first projection data and the second projection data are different.

[0287] The projection data determination unit is used to determine target projection data based on the first projection data and the second projection data. The target projection data includes the amount of change in target tissue information corresponding to the first projection data compared to target tissue information corresponding to the second projection data, caused by imaging condition variables and under the condition that the reference tissue component values ​​are the same.

[0288] In one embodiment, the reference tissue is bone, or bone and diaphragm; the first projection data is acquired at a first energy level, and the second projection data is acquired at a second energy level; and when the first projection data and the second projection data are acquired, the projection data determination unit is used to:

[0289] The line integral values ​​in the first projection data are modified to obtain the updated first projection data, and the reference tissue component values ​​corresponding to the updated first projection data are consistent with the bone component values ​​and the diaphragm component values ​​corresponding to the second projection data, respectively.

[0290] The difference between the updated first projection data and the second projection data is used as the target projection data.

[0291] In one embodiment, the first projection data is acquired with the contrast agent injected, and the second projection data is acquired without the contrast agent injected; the projection data determining unit is used for:

[0292] If a single scan of the CT equipment cannot completely cover the predetermined area, the difference between the first projection data and the second projection data shall be used as the target projection data.

[0293] If the single scan range of the CT equipment completely covers the predetermined area, the target projection data is determined through the following steps:

[0294] Determine the first image reconstruction result corresponding to the first projection data, and determine the second image reconstruction result corresponding to the second projection data;

[0295] Using the reference tissue as a reference benchmark, the second image reconstruction result is registered to the first image reconstruction result to obtain a registration result;

[0296] Based on the viewpoint combination corresponding to the first projection data, the forward projection of the second image reconstruction result in the registration result is completed to obtain the second forward projection data;

[0297] Based on the principle of subtraction from the same viewpoint, the difference between the first projection data and the second forward projection data is taken as the target projection data.

[0298] In one embodiment, the triggering unit is further configured to:

[0299] Extract the non-interest tissue from the image reconstruction result corresponding to the target projection data to obtain the non-interest tissue segmentation result;

[0300] The third forward projection data of the segmentation result of the non-interest tissue is determined based on a predetermined viewpoint combination, wherein the predetermined viewpoint combination is the viewpoint combination corresponding to the first projection data.

[0301] The difference between the target projection data and the third forward projection data is used as the updated target projection data.

[0302] In one embodiment, the region of interest is a blood vessel or tissue in a predetermined location, such as the brain, heart, lungs, abdomen, or periphery.

[0303] In one embodiment, the first image sequence module is further configured to:

[0304] The time curves corresponding to each voxel in the first image sequence are smoothed to update the first image sequence.

[0305] In one embodiment, the second image sequence is also used for:

[0306] Based on the CT image cinematography, target dynamic information for the region of interest is determined;

[0307] The target dynamic information includes information on the change of contrast agent intensity over time and / or information on the change of motion state over time.

[0308] In one embodiment, the target projection data is acquired using a cone-beam CT device.

[0309] The CT image sequence determination device provided in this embodiment of the invention, since the motion phase is the motion phase of the driving part, one motion phase includes at least one sub-phase; the region of interest of the predetermined part changes with the motion of the driving part, the first image sequence includes a second image sequence corresponding to each motion phase in the at least one motion phase, and each first image in the second image sequence includes the actual contrast information or the expected contrast information of the region of interest under the corresponding sub-phase. Therefore, each first image is only used to reflect the state of the region of interest under the corresponding sub-phase, which has higher accuracy than the prior art which reflects the motion state of the region of interest under one or more motion cycles through one image; in this way, by displaying the second image sequence corresponding to the at least one motion phase, the motion change and contrast change of the region of interest during the at least one motion phase of the driving part can be displayed intuitively and accurately.

[0310] The CT image sequence determination device provided in the embodiments of the present invention can execute the CT image sequence determination method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0311] It is worth noting that the various units and modules included in the above-mentioned CT image sequence determination device are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the protection scope of the embodiments of the present invention.

[0312] Figure 15 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0313] like Figure 15As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0314] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0315] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as CT image sequence determination methods.

[0316] In some embodiments, the CT image sequence determination method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via read-only memory (ROM) 12 and / or communication unit 19. When the computer program is loaded into random access memory (RAM) 13 and executed by processor 11, one or more steps of the CT image sequence determination method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the CT image sequence determination method by any other suitable means (e.g., by means of firmware).

[0317] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0318] Computer programs for implementing the CT image sequence determination method of this disclosure can be written in any combination of one or more programming languages. These computer programs can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer programs can be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0319] This invention provides a computer-readable storage medium storing computer instructions for causing a processor to execute a CT image sequence determination method, including:

[0320] In response to a first triggering operation, at least one motion phase to be displayed is determined, and target projection data for a predetermined part of the target object is acquired, wherein the motion phase is the motion phase of the driving part, and the region of interest of the predetermined part changes with the motion of the driving part.

[0321] A first image sequence is determined based on the target projection data. The first image sequence includes a second image sequence corresponding to each of the at least one motion phases. The second image sequence includes a first image under each sub-phase within the corresponding motion phase. The first image includes the actual or expected contrast information of the region of interest under the corresponding sub-phase. The motion phase includes at least one of the sub-phases.

[0322] The second image sequence corresponding to each of the at least one motion phase is displayed.

[0323] In the context of embodiments of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0324] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0325] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0326] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0327] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication unit 19, or installed from storage unit 18, or installed from ROM 12. When the computer program is executed by processor 11, it performs the functions defined in the methods of the embodiments of the present invention.

[0328] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements a CT image sequence determination method according to any embodiment of the invention.

[0329] In implementing a computer program product, computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0330] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this disclosure can be achieved, and this is not limited herein.

[0331] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for determining a CT image sequence, characterized in that, include: In response to a first triggering operation, at least one motion phase to be displayed is determined, and target projection data for a predetermined part of the target object is acquired, wherein the motion phase is the motion phase of the driving part, and the region of interest of the predetermined part changes with the motion of the driving part. A first image sequence is determined based on the target projection data. The first image sequence includes a second image sequence corresponding to each of the at least one motion phases. The second image sequence includes a first image under each sub-phase within the corresponding motion phase. The first image includes the actual or expected contrast information of the region of interest under the corresponding sub-phase. The motion phase includes at least one of the sub-phases. The second image sequence corresponding to each of the at least one motion phase is displayed.

2. The method according to claim 1, characterized in that, The image identifier of the first image includes a phase identifier of the motion phase in which the first image is located and a sub-phase identifier of the sub-phase in which it is located, or it includes a period identifier of the motion cycle to which the first image belongs, a phase identifier of the motion phase in which it is located, and a sub-phase identifier of the sub-phase in which it is located.

3. The method according to claim 1, characterized in that, The first image sequence includes the second image sequence under each of the motion phases in the motion phase set, the motion phase set including all the motion phases that the driving part appears during the target projection data acquisition; Before displaying the second image sequence corresponding to the at least one motion phase, the method further includes: Determine phase identification information corresponding to the at least one motion phase, wherein the phase identification information includes the period identifier and motion phase identifier of the motion cycle to which each of the at least one motion phase belongs; Based on the phase identification information, a second image sequence corresponding to each of the at least one motion phase is selected from the first image sequence.

4. The method according to claim 3, characterized in that, Determining the first image sequence based on the target projection data includes: The joint model is obtained by optimizing the parameters of the joint model based on the target projection data. The joint model includes a first term and a second term. The first term is a function used to describe the dynamic change of contrast. The first term uses the motion phase and the sub-phase as independent variables. The second term is a function used to describe the change of contrast with space. The second term uses the motion phase as an independent variable. Both the motion phase and the sub-phase use time as an independent variable. The first image sequence is determined based on the target joint model.

5. The method according to claim 4, characterized in that, The joint model is determined based on the principles of the Gaussian model. For each spatiotemporal location, the pixel intensity value at the current spatiotemporal location is determined by the sum of the pixel intensity values ​​of all Gaussian kernels at the current spatiotemporal location; Wherein, the spatiotemporal position corresponds to the target combination result, and the target combination result includes the target spatial position, the target motion phase, and the target sub-phase; for each Gaussian kernel, the pixel intensity value of the current Gaussian kernel at the current spatiotemporal position is determined based on the pixel intensity value of the current Gaussian kernel center under the target motion phase and the target sub-phase and the spatial distribution attribute value of the current Gaussian kernel. The spatial distribution attribute value of the current Gaussian kernel is determined based on the covariance matrix of the current Gaussian kernel and the target difference value. The target difference value is the difference between the target spatial position and the center spatial position of the current Gaussian kernel under the target motion phase. The covariance matrix and the center spatial position both use the motion phase as independent variables.

6. The method according to claim 4, characterized in that, Before determining the phase identification information corresponding to the at least one motion phase, the method further includes: Based on the associated gating information corresponding to the target projection data, the target projection data is divided into at least one first projection data group, and the first projection data group includes projection data under the same motion phase in all motion cycles. For each of the first projection data groups, determine the first image combination corresponding to the first projection data combination in the first image sequence, and the first forward projection data of the first image combination under the viewpoint combination corresponding to the first projection data group; determine the difference between the first forward projection data and the first projection data group; if the difference does not meet the predetermined accuracy condition, adjust the parameters of the target joint model according to the difference; update the first image combination for the first projection data group based on the parameter-adjusted target joint model; Based on all the updated combinations of the first images, an updated sequence of the first images is determined.

7. The method according to claim 4, characterized in that, The step of optimizing the parameters of the joint model based on the target projection data to obtain the target joint model includes: Obtain the associated gating information of the target projection data; Based on the associated gating information, the target projection data is divided into at least one first projection data group, and the first projection data group includes projection data under the same motion phase in all motion cycles. Based on each of the first projection data groups, the parameters of the joint model are optimized to obtain the target joint model.

8. The method according to claim 4, characterized in that, Also includes: In response to a second trigger operation targeting contrast, the target contrast is determined; A third image sequence corresponding to the target contrast is determined based on the target projection data. The third image sequence is used to show the motion state changes of the region of interest within at least one motion cycle under the target contrast. The third image sequence corresponding to the target contrast is displayed.

9. The method according to claim 8, characterized in that, Determining the third image sequence corresponding to the target contrast based on the target projection data includes: The sub-phase corresponding to the target contrast is input into the target joint model to obtain a third image sequence corresponding to the target contrast.

10. The method according to claim 8, characterized in that, Determining the third image sequence corresponding to the target contrast based on the target projection data includes: A four-dimensional auxiliary image is determined based on the target projection data; Based on the associated gating information of the target projection data, the target projection data is divided into a second projection data group corresponding to each of the motion cycles. For any of the motion cycles, the three-dimensional deformation motion field of all the motion phases relative to each of the motion phases is determined based on the four-dimensional auxiliary image, to obtain the first deformation motion field combination for each of the motion phases; For each of the motion cycles, based on the second projection data group under the motion cycle and each of the first deformation motion field combinations, a first intermediate image combination for the motion cycle is determined, and the target contrast is mapped onto each of the intermediate images in the first intermediate image combination to obtain a second image combination corresponding to the target contrast. The union of all the second images is taken as the third image sequence; Wherein, the number of intermediate images in the first intermediate image combination is equal to the number of three-dimensional deformation motion fields in the first deformation motion field combination, and the number of three-dimensional deformation motion fields in the first deformation motion field combination is equal to the number of motion phases included in one motion cycle.

11. The method according to claim 3, characterized in that, Determining the first image sequence based on the target projection data includes: Based on the associated gating information corresponding to the target projection data, the target projection data is divided into at least one first projection data group, and the first projection data group includes projection data under the same motion phase in all motion cycles. A first image combination is determined for each of the first projection data groups, wherein the first image combination includes the first image under the corresponding motion phase in all motion cycles; The first image sequence is determined based on all the first image combinations.

12. The method according to claim 11, characterized in that, Determining the first image combination for each of the first projection data groups includes: For each of the first projection data groups, a target intermediate image corresponding to the first projection data group is determined; by mapping the contrast information in the projection data under the projection viewpoint corresponding to the target motion phase in each motion cycle to the target intermediate image, the first image under the target motion phase in each motion cycle is determined; the summary result of all the first images under the target motion phase is used as the first image combination for the first projection data group, where the target motion phase is the motion phase that has a corresponding relationship with the first projection data group.

13. The method according to claim 12, characterized in that, Determining the target intermediate image corresponding to the first projection data group includes: Based on the ultra-sparse perspective 3D image reconstruction method, the target intermediate image corresponding to the first projection data group is determined.

14. The method according to claim 2, characterized in that, The target projection data includes projection data under at least one motion cycle, and the step of determining the first image sequence based on the target projection data includes: A four-dimensional auxiliary image is determined based on the target projection data; Determine at least one motion cycle to which the at least one motion phase belongs, and based on the association gating information of the target projection data, extract projection data corresponding to each of the at least one motion cycles from the target projection data to obtain a second projection data set for each of the motion cycles; For any of the motion cycles, based on the four-dimensional auxiliary image, the three-dimensional deformation motion field of all the motion phases relative to each of the motion phases is determined, resulting in a first deformation motion field combination for each of the motion phases. The number of three-dimensional deformation motion fields in the first deformation motion field combination is equal to the number of motion phases included in one motion cycle. For each of the at least one motion phases, based on the first deformation motion field combination corresponding to all the motion phases in the motion phase combination and the second projection data group under the motion cycle of the motion phase, a second image sequence for the motion phase is determined, wherein the motion phase combination includes all the motion phases within one motion cycle; The union of all the second image sequences is taken as the first image sequence.

15. The method according to claim 14, characterized in that, Determining the second image sequence for the motion phase includes: Based on a predetermined deformation motion field interpolation method, a second deformation motion field set corresponding to a first deformation motion field set is determined. The first deformation motion field set includes combinations of first deformation motion fields under each motion phase. The second deformation motion field set includes combinations of second deformation motion fields under each projection viewpoint combination. All projection views corresponding to each motion phase are divided into at least two projection viewpoint combinations. Each projection viewpoint combination includes one projection viewpoint or multiple adjacent projection viewpoints. For each of the at least one motion phases, a second image sequence is determined based on the second set of deformable motion fields and the second projection data set under the motion period of the motion phase.

16. The method according to claim 15, characterized in that, The step of determining the second image sequence for the motion phase based on the second set of deformed motion fields and the second projection data group under the motion period of the motion phase includes: Based on the second set of deformed motion fields and the second set of projection data under the motion period of the motion phase, a second intermediate image combination for the motion phase is determined. The number of intermediate images in the second intermediate image combination is equal to the number of projection view combination combinations included in the motion period. Based on the contrast information corresponding to the projection viewpoint under each of the second deformation motion field combinations and each motion phase within the motion cycle, the corresponding intermediate image in the second intermediate image combination is updated to obtain the second image sequence for the motion phase.

17. The method according to claim 16, characterized in that, The step of determining a second intermediate image combination for the motion phase based on the second set of deformed motion fields and the second projection data group under the motion period of the motion phase includes: Based on the ultra-limited viewpoint, the motion-compensated 3D reconstruction method, and the second projection data group under the motion period of the motion phase, a second intermediate image combination for the motion phase is determined.

18. The method according to claim 11 or 14, characterized in that, If the associated gating information is ECG gating information, the movement cycle is the cardiac cycle; If the associated gating information is respiratory gating information, then the movement cycle is the respiratory movement cycle.

19. The method according to claim 1, characterized in that, The acquisition of target projection data for a predetermined part of the target object includes: Acquire first projection data and second projection data for a predetermined part of the target object, wherein the imaging conditions of the first projection data and the second projection data are different; Target projection data is determined based on the first projection data and the second projection data. The target projection data includes the change in target tissue information corresponding to the first projection data compared to the target tissue information corresponding to the second projection data, caused by imaging condition variables, under the condition that the reference tissue component values ​​are the same.

20. The method according to claim 19, characterized in that, The reference tissue is bone, or bone and diaphragm. The first projection data is acquired at a first energy level, and the second projection data is acquired at a second energy level. When the first and second projection data are acquired, the target object is injected with contrast agent. Determining the target projection data based on the first and second projection data includes: The line integral values ​​in the first projection data are modified to obtain the updated first projection data, and the reference tissue component values ​​corresponding to the updated first projection data are consistent with the bone component values ​​and the diaphragm component values ​​corresponding to the second projection data, respectively. The difference between the updated first projection data and the second projection data is used as the target projection data.

21. The method according to claim 19, characterized in that, The first projection data was obtained with the contrast agent injected, and the second projection data was obtained without the contrast agent injected; determining the target projection data based on the first projection data and the second projection data includes: If a single scan of the CT equipment cannot completely cover the predetermined area, the difference between the first projection data and the second projection data shall be used as the target projection data. If the single scan range of the CT equipment completely covers the predetermined area, the target projection data is determined through the following steps: Determine the first image reconstruction result corresponding to the first projection data, and determine the second image reconstruction result corresponding to the second projection data; Using the reference tissue as a reference benchmark, the second image reconstruction result is registered to the first image reconstruction result to obtain a registration result; Based on the viewpoint combination corresponding to the first projection data, the forward projection of the second image reconstruction result in the registration result is completed to obtain the second forward projection data; Based on the principle of subtraction from the same viewpoint, the difference between the first projection data and the second forward projection data is taken as the target projection data.

22. The method according to claim 21, characterized in that, After the target projection data is determined, the following is also included: Extract the non-interest tissue from the image reconstruction result corresponding to the target projection data to obtain the non-interest tissue segmentation result; The third forward projection data of the segmentation result of the non-interest tissue is determined based on a predetermined viewpoint combination, wherein the predetermined viewpoint combination is the viewpoint combination corresponding to the first projection data. The difference between the target projection data and the third forward projection data is used as the updated target projection data.

23. The method according to claim 1, characterized in that, The region of interest is a blood vessel or tissue in a predetermined location, such as the brain, heart, lungs, abdomen, or periphery.

24. The method according to any one of claims 1-3, characterized in that, After determining the first image sequence based on the target projection data, the method further includes: The time curves corresponding to each voxel in the first image sequence are smoothed to update the first image sequence.

25. The method according to any one of claims 1-3, characterized in that, After the second image sequence is determined, the process also includes: Determine the target dynamic information for the region of interest based on the first image sequence; The target dynamic information includes information on the change of contrast agent intensity over time and / or information on the change of motion state over time.

26. The method according to claim 1, characterized in that, The target projection data is acquired using a cone-beam CT scanner.

27. An electronic device, characterized in that, The electronic device includes: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the CT image sequence determination method according to any one of claims 1-26.

28. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the CT image sequence determination method according to any one of claims 1-26.

29. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the CT image sequence determination method according to any one of claims 1-26.

Images