Coronary artery phase analysis method and device and storage medium

Abstract

The invention relates to the technical field of medical images, and discloses a coronary artery phase analysis method and device and a storage medium. The heart comprises a plurality of coronary arteries, for at least one coronary artery, first a plurality of phase reconstructed image sequences of the heart are acquired. Then, coronary artery motion data of the coronary artery and heart motion data corresponding to the position of the coronary artery are acquired based on the plurality of phase reconstruction image sequences. Then, based on the heart motion data, motion differences of different coronary artery partitions are analyzed, and a target interval corresponding to a target period of the coronary artery is obtained; and finally, based on the target interval, obtaining a target phase corresponding to the coronary artery on the coronary artery motion data, improving the accuracy of the optimal phase in the heart reconstruction image, reducing motion artifact interference, and generating a clearer coronary artery image.

Description

Technical Field

[0001] This invention relates to the field of medical imaging technology, and in particular to a method, apparatus and storage medium for coronary artery phase analysis. Background Art

[0002] Cardiovascular diseases have a high incidence rate and are highly fatal in my country. Cardiac computed tomography (CT) imaging, with its non-invasive, efficient, and multi-dimensional imaging characteristics, has revolutionized the diagnosis and treatment of cardiovascular diseases, playing a significant role in disease prevention, precision intervention, and optimization of medical resources.

[0003] Among related technologies, the optimal phase method focuses on changes in the overall coronary arteries. However, due to the periodic motion of the heart, artifacts of varying degrees appear in the coronary arteries at different stages of the cardiac cycle, leading to blurred vessels and structural distortion, thus affecting the accuracy of diagnosis. Therefore, a new method for coronary artery phase analysis is needed. Summary of the Invention

[0004] The embodiments described in this specification aim to at least partially address one of the technical problems in the related art. To this end, the embodiments described in this specification provide a method, apparatus, and storage medium for coronary artery phase analysis.

[0005] This specification provides a method for coronary artery phase analysis. The heart includes multiple coronary arteries. For at least one coronary artery, the method includes:

[0006] Acquire multiple phase reconstructed image sequences of the heart;

[0007] Based on multiple phase reconstruction image sequences, coronary artery motion data and cardiac motion data corresponding to the location of the coronary artery are obtained;

[0008] Based on the cardiac motion data, the target interval corresponding to the target period of the coronary artery is obtained;

[0009] Based on the target interval, the target phase corresponding to the coronary artery is obtained from the coronary motion data.

[0010] In one implementation, acquiring coronary artery motion data and cardiac motion data corresponding to the location of the coronary arteries based on multiple phase-reconstructed image sequences includes:

[0011] Multiple phase reconstruction image sequences are segmented to obtain multiple phase mask image sequences, wherein the mask image at each position in the phase mask image sequence defines the pericardial fat region and the atrioventricular region;

[0012] Based on the overlap of the target pericardial fat region in the intermediate phase reconstruction image subsequences of adjacent phase reconstruction image sequences, the coronary artery motion data is obtained, wherein the intermediate phase reconstruction image subsequence includes a reconstruction image at at least one location, and the target pericardial fat region of the intermediate phase reconstruction image subsequence is a region determined by the intermediate phase mask image subsequence and connected to the coronary artery.

[0013] Based on the area changes of the atrial and ventricular regions in the mask images of the target location in multiple phase mask image sequences, cardiac motion data corresponding to the coronary artery location is obtained, wherein the mask image of the target location corresponds to a specific location on the coronary artery.

[0014] In one implementation, the target pericardial fat region of the phase-reconstructed image subsequence is determined by the following method:

[0015] Based on the segmentation center, the coronary artery region is divided into adjacent phase mask image subsequences to obtain the pericardial fat region corresponding to the coronary artery location;

[0016] Based on the overlap of pericardial fat regions in adjacent phase mask image subsequences, the total pericardial fat region is obtained;

[0017] Based on the mapping of the total pericardial fat region, the target pericardial fat region of the adjacent phase reconstruction image subsequence is determined.

[0018] In one implementation, the segmentation center is determined in the following manner:

[0019] Based on the area of ​​the fat region at the center of the mask image at each position in the phase mask image sequence, the key mask image is determined;

[0020] Based on the pericardial fat region in the key mask image, the intersection of the diagonals of the circumscribed polygon of the pericardial fat region is determined as the segmentation center.

[0021] In one implementation, obtaining the coronary artery motion data based on the overlap of the target pericardial fat region in intermediate phase reconstruction image subsequences of adjacent phase reconstruction image sequences includes:

[0022] Based on the number of overlapping images and the selected phase position, determine the first phase reconstruction image subsequence and the second phase reconstruction image subsequence in the adjacent phase reconstruction image sequence;

[0023] Based on the overlap between the target pericardial fat region of the first phase reconstruction image subsequence and the target pericardial fat region of the second phase reconstruction image subsequence, the coronary motion image corresponding to the coronary artery is determined;

[0024] Based on the coronary motion image corresponding to the coronary artery, the coronary motion number corresponding to the coronary artery is determined.

[0025] In one implementation, determining the coronary motion image corresponding to the coronary artery based on the overlap between the target pericardial fat region of the first phase reconstruction image subsequence and the target pericardial fat region of the second phase reconstruction image subsequence includes:

[0026] The position of the motion phase is determined based on the selected phase position;

[0027] Based on the overlap between the target pericardial fat region of the first phase reconstructed image subsequence and the target pericardial fat region of the second phase reconstructed image subsequence, the motion map value corresponding to the position of the coronary artery in the motion phase is determined;

[0028] Based on the motion map values ​​corresponding to each phase position of the coronary artery, a coronary motion image corresponding to the coronary artery is generated.

[0029] In one implementation, determining the coronary motion data corresponding to the coronary artery based on the coronary motion image corresponding to the coronary artery within the effective location range includes:

[0030] Within the effective location range of the coronary motion image corresponding to the coronary artery, motion data corresponding to the motion phase is determined based on the selected motion phase position;

[0031] Coronary artery motion data corresponding to the aforementioned motion phase is constructed based on the motion data.

[0032] In one implementation, obtaining cardiac motion data corresponding to the coronary artery location based on the area change of the atrial and ventricular regions in the mask images of the target location in multiple phase mask image sequences includes:

[0033] The target location of the coronary artery is determined based on the area of ​​the atrial and ventricular regions in the phase mask image sequence.

[0034] Based on the area changes of the atrial and ventricular regions of the mask images corresponding to the target location of the coronary artery in multiple phase mask image sequences, cardiac motion data corresponding to the location of the coronary artery are determined.

[0035] In one implementation, the target interval includes a target systolic interval and a target diastolic interval, and the step of obtaining the target interval corresponding to the target period of the coronary artery based on the cardiac motion data includes:

[0036] Based on the changing trends of the cardiac motion data, the end-systolic interval is determined;

[0037] Based on the mapping of the end-systolic interval, the target systolic interval corresponding to the target period of the coronary artery is determined; and / or,

[0038] Based on the changing trends of the cardiac motion data, the end-diastolic interval is determined;

[0039] Based on the mapping of the end-diastolic interval, the target diastolic interval corresponding to the target period of the coronary artery is determined.

[0040] In one implementation, obtaining the target phase corresponding to the coronary artery from the coronary motion data based on the target interval includes:

[0041] Based on the coronary motion data, the phases within the target systolic interval that meet the preset screening criteria are determined as the target systolic phases corresponding to the coronary artery;

[0042] Based on the coronary artery motion data, the phases within the target diastolic interval that meet the preset screening criteria are determined as the target diastolic phases corresponding to the coronary artery.

[0043] In one implementation, for at least one coronary artery, the coronary artery is divided into multiple regions according to its location on the heart, each region corresponding to a target phase.

[0044] This specification provides a coronary artery phase analysis device. The heart includes multiple coronary arteries. For at least one coronary artery, the device includes:

[0045] The reconstructed image acquisition module is used to acquire multiple phase reconstructed image sequences of the heart;

[0046] The motion data acquisition module is used to acquire coronary motion data of the coronary artery and cardiac motion data corresponding to the location of the coronary artery based on multiple phase reconstruction image sequences.

[0047] The target interval determination module is used to obtain the target interval corresponding to the target period of the coronary artery based on the cardiac motion data.

[0048] The target phase determination module is used to obtain the target phase corresponding to the coronary artery based on the target interval and the coronary artery motion data.

[0049] This specification provides a medical imaging device, which includes: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions executable by the one or more processors, which, when executed by the one or more processors, cause the one or more processors to perform the steps of the method described in any of the above embodiments.

[0050] This specification provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any of the above embodiments.

[0051] This specification provides a computer program product that includes instructions that, when executed by a processor of a computer device, enable the computer device to perform the steps of the method described in any of the above embodiments.

[0052] In the above-described embodiments, the heart includes multiple coronary arteries. For at least one coronary artery, firstly, multiple phase reconstruction image sequences of the heart are acquired. Next, based on the multiple phase reconstruction image sequences, coronary artery motion data and cardiac motion data corresponding to the location of the coronary artery are acquired. Then, based on the cardiac motion data, the motion differences between different coronary artery regions are analyzed to obtain the target interval corresponding to the target phase of the coronary artery. Finally, based on the target interval, the target phase corresponding to the coronary artery is obtained from the coronary artery motion data. This process can improve the accuracy of the optimal phase in the reconstructed cardiac images, reduce the interference of motion artifacts, and ultimately generate clearer coronary artery images, improving diagnostic accuracy. Attached Figure Description

[0053] Figure 1a A schematic flowchart of the coronary artery phase analysis method provided in the embodiments of this specification;

[0054] Figure 1b A schematic diagram of the reconstructed image provided for an embodiment of this specification;

[0055] Figure 2a A schematic diagram of the process for acquiring coronary artery motion data and cardiac motion data provided for the embodiments of this specification;

[0056] Figure 2b A schematic diagram of the mask image corresponding to the reconstructed image provided in the embodiments of this specification;

[0057] Figure 3a A schematic diagram of the process for obtaining cardiac motion data corresponding to the location of the coronary arteries, provided for the embodiments of this specification;

[0058] Figure 3bA schematic diagram illustrating the reference positions of the left and right crowns for embodiments of this specification;

[0059] Figure 4a A flowchart illustrating the process of determining the target pericardial fat region provided for embodiments of this specification;

[0060] Figure 4b A schematic diagram illustrating the determination of the mask image subsequence and reconstructed image subsequence corresponding to the selected phase position provided in the embodiments of this specification;

[0061] Figure 4c A schematic diagram illustrating the division of coronary artery regions in a phase mask image provided for embodiments of this specification;

[0062] Figure 5a A flowchart illustrating the determination of the segmentation center is provided for the implementation of this specification.

[0063] Figure 5b A schematic diagram for determining the segmentation center provided for the implementation of this specification;

[0064] Figure 6a A schematic diagram of the process for acquiring coronary artery motion data provided in the embodiments of this specification;

[0065] Figure 6b A schematic diagram illustrating the effective location range provided for embodiments of this specification;

[0066] Figure 6c A schematic diagram of the coronary motion curve corresponding to the coronary artery provided for the embodiments of this specification;

[0067] Figure 7a A flowchart illustrating the process of determining coronary motion data corresponding to a coronary artery, provided for the implementation of this specification;

[0068] Figure 7b A schematic diagram illustrating the acquisition of a set of target phases provided for the implementation of this specification;

[0069] Figure 8 A schematic diagram illustrating the determination of the target pericardial fat region provided for embodiments of this specification;

[0070] Figure 9 A flowchart illustrating the process of determining the coronary motion image corresponding to the coronary artery, provided for the embodiments of this specification;

[0071] Figure 10a A schematic flowchart illustrating the process of determining the target systolic phase corresponding to the coronary artery, provided for the implementation of this specification;

[0072] Figure 10b A schematic diagram illustrating the determination of the end-systolic and end-diastolic intervals for the purposes of this specification;

[0073] Figure 10c A schematic diagram illustrating the target systolic phase and target diastolic phase provided for embodiments of this specification;

[0074] Figure 11 A flowchart illustrating the process of determining the target diastolic phase corresponding to the coronary artery, provided for the implementation of this specification;

[0075] Figure 12 A schematic diagram of the coronary artery phase analysis device provided for embodiments of this specification;

[0076] Figure 13 An internal structural diagram of a computer device provided for embodiments of this specification. Detailed Implementation

[0077] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0078] Cardiovascular disease is a prevalent and potentially fatal condition. Cardiac CTA (computed tomography angiography), as a non-invasive, efficient, and multi-dimensional vascular imaging technique, plays a crucial role in the diagnosis and treatment of cardiovascular diseases, particularly in disease prevention, precision intervention, and optimization of medical resources. However, the heart is in a constant state of cyclical motion, which affects the image quality of the coronary arteries. Especially during the cardiac cycle, motion-induced artifacts can lead to blurred blood vessels, structural distortion, and even affect diagnostic accuracy.

[0079] The contraction and relaxation of the heart are not merely mechanical movements, but rather highly temporal processes involving coordination among different tissues. Furthermore, the different coronary arteries of the heart have different spatial distributions and motion characteristics, which makes the optimal imaging timing (i.e., optimal phase) for each coronary artery during the cardiac cycle different.

[0080] In related technologies, when searching for the optimal phase of the coronary artery, the performance of the entire coronary system is usually taken into account. However, since the motion states of different coronary arteries are different, the optimal phase of the overall coronary artery does not mean that the image quality of all coronary arteries is at its best. Motion artifacts will still occur in some coronary arteries, affecting doctors' assessment and diagnosis of coronary artery diseases.

[0081] Based on this, the embodiments of this specification provide a method for coronary artery phase analysis. The heart includes multiple coronary arteries. For at least one coronary artery, firstly, multiple phase reconstruction image sequences of the heart are acquired. Next, based on the multiple phase reconstruction image sequences, coronary artery motion data and cardiac motion data corresponding to the location of the coronary arteries are acquired. Then, based on the cardiac motion data, motion differences in different coronary artery regions are analyzed to obtain the target interval corresponding to the target phase of the coronary artery. Finally, based on the target interval, the target phase corresponding to the coronary artery is obtained from the coronary artery motion data. This process can improve the accuracy of the optimal phase in the reconstructed cardiac images, reduce the interference of motion artifacts, and ultimately generate clearer coronary artery images, improving diagnostic accuracy.

[0082] This specification provides a method for coronary artery phase analysis. The heart includes multiple coronary arteries. For at least one coronary artery, please refer to [link to documentation]. Figure 1a The coronary artery phase analysis method may include the following steps:

[0083] S110. Obtain multiple phase reconstruction image sequences of the heart.

[0084] The coronary artery can be at least one of the left anterior descending artery, the left circumflex artery, or the right coronary artery.

[0085] Specifically, the range of cardiac imaging is determined based on clinical needs, and the parameters of the medical imaging equipment are preset accordingly. The medical imaging equipment scans the patient under these parameter settings and reads the corresponding scan data. The cardiac cycle refers to the time required for the heart to complete one full contraction and relaxation process, usually defined by the RR interval. By analyzing the electrocardiogram signals in the scan data, the start and end points of the cardiac cycle are determined, thereby calculating the phase range used for reconstruction. The phase range is divided according to the preset phase intervals to determine multiple phases. Then, by reconstructing images at each location within each phase, a reconstructed image corresponding to each location of each phase is generated. All the reconstructed images can constitute a sequence of multiple phase reconstructed images. For an example, please refer to [link to example]. Figure 1b , Figure 1b To reconstruct the image.

[0086] S120. Based on multiple phase reconstruction image sequences, obtain coronary artery motion data and cardiac motion data corresponding to the location of the coronary arteries.

[0087] S130. Based on cardiac motion data, obtain the target interval corresponding to the target period of the coronary artery.

[0088] S140. Based on the target interval, obtain the target phase corresponding to the coronary artery from the coronary motion data.

[0089] Specifically, because different coronary arteries in the heart have different spatial distributions and motion characteristics, the optimal imaging timing (i.e., optimal phase) during the cardiac cycle may vary for each coronary artery or different regions within the same coronary artery. Therefore, each coronary artery or a specific region within the same coronary artery needs to be analyzed independently to determine the corresponding target phase. Coronary arteries are not fixed structures but move continuously with the contraction and relaxation of the heart during its pulsation. Due to differences in the cardiac physiological characteristics of each patient, adjustments are necessary to determine the appropriate phase range. Considering the differences in the distribution of different coronary arteries within the heart, it is necessary to calculate the motion characteristics corresponding to different regions separately and use analysis methods based on local motion characteristics to analyze different regions individually. This not only improves the accuracy of motion data but also allows for personalized phase optimization for specific regions. Therefore, to comprehensively capture the dynamic changes of coronary arteries throughout the cardiac cycle, analysis based on image sequences reconstructed from multiple phases can obtain information such as the location and morphological changes of coronary arteries at different time points, thereby obtaining coronary artery motion data. Analyzing multiple phase-reconstructed image sequences allows for the acquisition of cardiac motion data corresponding to the location of the coronary arteries. Then, by analyzing this cardiac motion data, a target interval corresponding to the target phase of the coronary arteries can be defined. Finally, by analyzing coronary artery motion data within the target interval, the target phase of the coronary arteries can be determined more accurately, providing a precise basis for subsequent coronary artery assessment and disease diagnosis.

[0090] In the above embodiments, the heart includes multiple coronary arteries. For at least one coronary artery, firstly, multiple phase reconstruction image sequences of the heart are acquired. Next, based on the multiple phase reconstruction image sequences, coronary artery motion data and cardiac motion data corresponding to the location of the coronary artery are acquired. Then, based on the cardiac motion data, the motion differences between different coronary artery regions are analyzed to obtain the target interval corresponding to the target phase of the coronary artery. Finally, based on the target interval, the target phase corresponding to the coronary artery is obtained from the coronary artery motion data. This process can improve the accuracy of the optimal phase in the reconstructed cardiac images, reduce the interference of motion artifacts, and ultimately generate clearer coronary artery images, improving diagnostic accuracy.

[0091] In some implementations, please refer to Figure 2a Based on multiple phase-reconstructed image sequences, coronary artery motion data and cardiac motion data corresponding to the location of the coronary arteries can be obtained, which may include the following steps:

[0092] S210. Segment the multiple phase reconstruction image sequences to obtain multiple phase mask image sequences.

[0093] Specifically, after image reconstruction is completed, the trained model is used to segment multiple phase-reconstructed image sequences, distinguishing cardiac-related structures in the mask image at each location in the phase-mask image sequence, resulting in multiple phase-mask image sequences. The mask image at each location in the phase-mask image sequence defines the pericardial fat region and the atrioventricular region. For example, for... Figure 1b The reconstructed image is segmented to obtain a mask image. Please refer to [link / reference]. Figure 2b , Figure 2b for Figure 1b The mask image corresponding to the reconstructed image in the image. Figure 2b The masked regions include the pericardial fat region a and the atrioventricular region b.

[0094] S220. Based on the overlap of the target pericardial fat region in the mid-phase reconstruction image subsequence of adjacent phase reconstruction image sequences, obtain coronary motion data.

[0095] The phase reconstruction image subsequence includes a reconstructed image at at least one location, and the target pericardial fat region of the phase reconstruction image subsequence is the region determined by the phase mask image subsequence and connected to the coronary artery.

[0096] In some cases, during the physiological activity of the heart, the coronary arteries move continuously along with various complex patterns of motion, including the heart's contraction and relaxation. Each heartbeat, with its contraction and relaxation, causes corresponding displacement of surrounding tissues, especially the pericardial fat region.

[0097] Specifically, firstly, corresponding phase-reconstructed image subsequences are selected from adjacent phase-reconstructed image sequences. Since adjacent phase-reconstructed image subsequences correspond to adjacent phase-mask image subsequences, the target pericardial fat region corresponding to each adjacent phase can be determined based on the pericardial fat region defined in the mask image at each position in the adjacent phase-mask image subsequence. Next, since the movement of the pericardial fat region is synchronous and correlated with the movement of the coronary arteries, coronary artery motion data can be obtained by analyzing the overlap of the target pericardial fat regions corresponding to adjacent phases. The overlap is determined by calculating the pixel difference. When the coronary arteries are at rest, the overlap of the target pericardial fat regions of adjacent phases is high, and the pixel difference is small; while when the coronary arteries are in motion, the number of overlapping pixels in the target pericardial fat regions of adjacent phases decreases, and the difference increases. Therefore, the magnitude of the difference can represent the magnitude of the motion amplitude.

[0098] S230. Based on the area changes of the atrial and ventricular regions of the mask images at the target location in multiple phase mask image sequences, obtain cardiac motion data corresponding to the coronary artery location.

[0099] The mask image of the target location corresponds to a specific location on the coronary artery.

[0100] Specifically, the heart's movement inevitably exerts traction and compression on the coronary arteries, thus affecting blood flow within the arteries and their mechanical environment; a close interrelationship exists between the two. By analyzing mask images of target locations in multiple phase mask image sequences, the area changes of the atrial and ventricular regions during the cardiac cycle can be precisely tracked, obtaining cardiac motion data corresponding to the coronary artery locations. The area changes of the atrial and ventricular regions during the cardiac cycle reflect the heart's systolic and diastolic movement patterns, thus providing crucial data support for studying the dynamic response of the coronary arteries during the cardiac cycle.

[0101] In the above embodiments, multiple phase reconstruction image sequences are segmented to obtain multiple phase mask image sequences. Based on the overlap of the target pericardial fat region in the phase reconstruction image subsequences of adjacent phase reconstruction image sequences, coronary artery motion data is obtained. Based on the area change of the atrial and ventricular regions in the mask images of the target location in the multiple phase mask image sequences, cardiac motion data corresponding to the coronary artery location is obtained, providing a data basis for subsequent determination of the target phase.

[0102] In some implementations, please refer to Figure 3a Obtaining cardiac motion data corresponding to the coronary artery location based on the area changes of the atrial and ventricular regions in mask images of the target location in multiple phase mask image sequences may include the following steps:

[0103] S310. Determine the target location of the coronary artery based on the area of ​​the atrial and ventricular regions in the phase mask image sequence.

[0104] S320. Based on the area changes of the atrial and ventricular regions of the mask images corresponding to the target location of the coronary artery in multiple phase mask image sequences, determine the cardiac motion data corresponding to the location of the coronary artery.

[0105] Specifically, in the phase mask image sequence, any phase corresponding to the mask image sequence is selected as the target phase mask image sequence. For each position in the target phase mask image sequence, the area of ​​the atrial and ventricular regions in the target phase mask image corresponding to that position is calculated. Next, the target location of the coronary artery is determined based on the position corresponding to the image with the largest atrial and ventricular mask area among all images. Then, the area of ​​the atrial and ventricular regions in the mask image corresponding to the target location in the phase mask image sequence is calculated. Finally, the obtained atrial and ventricular region areas corresponding to the target locations in multiple phases are arranged according to the phases of the phase mask images to constitute the cardiac motion data corresponding to the coronary artery locations.

[0106] In some implementations, the coronary artery can be the left anterior descending artery or the left circumflex artery, in which case the target location can be the left coronary reference location; if the coronary artery is the right coronary artery, then the target location can be the right coronary reference location. The right coronary reference location is determined by calculating the difference between the left coronary reference location and a location calculation threshold. For example, please refer to [link to relevant documentation]. Figure 3b , Figure 3b PosLCA in the figure is the reference position for the left crown. Figure 3b PosRCA in the diagram represents the reference position for the right crown.

[0107] In the above embodiments, the target location of the coronary artery is determined based on the area of ​​the atrial and ventricular regions in the phase mask image sequence. Based on the area changes of the atrial and ventricular regions in the mask images corresponding to the target location of the coronary artery in multiple phase mask image sequences, the cardiac motion data corresponding to the coronary artery location is determined, so as to improve the accuracy of determining the target phase by combining the coronary motion data.

[0108] In some implementations, please refer to Figure 4a The target pericardial fat region of the phase reconstruction image subsequence was determined by the following method:

[0109] S410. Based on the segmentation center, the coronary artery region is divided into subsequences of adjacent phase mask images to obtain the pericardial fat region corresponding to the location of the coronary artery.

[0110] Specifically, the phase mask image sequence is analyzed to select the required mask images. Then, data processing is performed using the selected mask images to determine the segmentation center corresponding to that phase. Within the phase mask image sequence, a specific location within a phase is selected as the chosen phase position. Then, starting from the chosen phase position, the selection is incremented along the position direction until the number of selected mask images reaches the number of overlapping images. Finally, the selected mask images are called phase mask image subsequences. The segmentation centers corresponding to adjacent phase mask image subsequences are mapped onto each mask image of the corresponding phase mask image subsequence to determine the corresponding position of the segmentation center in each mask image. Then, based on the corresponding position of the segmentation center determined in each mask image, the mask image is divided into coronary artery regions to obtain the pericardial fat region corresponding to the coronary artery location.

[0111] For example, please refer to Figure 4b For example, the number of overlapping images is PosThickness. See also Figure 4b Select k-phase phase kThe q position in the sequence is used as the selected phase position. Then, in the phase mask image sequence, starting from the selected phase position, the positions are incremented sequentially along the vertical axis until the number of selected mask images reaches the PosThickness (the number of overlapping images). Finally, the selected mask images are Masked. k,q Mask k,q+1 Mask k,q+PosThickness-1 This is called the k-phase mask image subsequence. The k-phase phase... k adjacent k+1 phase k+1 Starting from position q, the values ​​are incremented sequentially along the vertical axis until the number of selected mask images reaches the PosThickness value for overlapping images. Finally, the selected mask images are... k+1,q Mask k+1,q+1 Mask k+1,q+PosThickness-1 This is called the phase k+1 mask image subsequence.

[0112] Please see Figure 4c The coronary artery can be the right coronary artery (RCA). The mask image is divided into coronary artery regions by segmentation center. The area to the left of the segmentation center is defined as the right coronary region, i.e., the RCA region (red area). The pericardial fat region within the RCA region is called the pericardial fat region corresponding to the location of the right coronary artery.

[0113] The coronary artery can be the left circumflex branch (LCX). The coronary artery region is divided into parts of the mask image by segmentation center. The lower right half of the segmentation center is defined as the left circumflex branch region, i.e., the LCX region (blue area). The pericardial fat region within the LCX region is called the pericardial fat region corresponding to the location of the left circumflex branch.

[0114] The coronary arteries are composed of the left anterior descending artery (LAD). The mask image is divided into coronary artery regions using a segmentation center. The upper right half of the segmentation center is defined as the LAD region (green area), and the pericardial fat region within the LAD region is referred to as the pericardial fat region corresponding to the location of the left circumflex artery.

[0115] S420. Based on the overlap of pericardial fat regions in adjacent phase mask image subsequences, obtain the total pericardial fat region.

[0116] S430. Based on the total pericardial fat region, the target pericardial fat region of the adjacent phase reconstruction image subsequence is determined by mapping.

[0117] Specifically, the total pericardial fat region is determined by statistically analyzing the overlap of pericardial fat regions in adjacent phase mask image subsequences. The total number of mask pixels is determined by calculating the number of pixels within this total pericardial fat region. Mapping of the total pericardial fat region is performed within adjacent phase reconstructed image subsequences to determine the corresponding regions. The pixel values ​​of each pixel location within the region corresponding to the determined total pericardial fat region in any phase reconstructed image subsequence are added together, and the results are integrated to determine the target pericardial fat region for that phase reconstructed image subsequence. Repeating this process yields the target pericardial fat region for another phase reconstructed image subsequence within adjacent phase reconstructed image subsequences.

[0118] In the above implementation, the coronary artery region is divided into adjacent phase mask image subsequences based on the segmentation center to obtain the pericardial fat region corresponding to the coronary artery location. Based on the overlap of the pericardial fat regions of adjacent phase mask image subsequences, the total pericardial fat region is obtained. Based on the total pericardial fat region, the target pericardial fat region of adjacent phase reconstructed image subsequences is determined by mapping, providing a data basis for subsequent determination of motion map values.

[0119] In some implementations, please refer to Figure 5a The segmentation center is determined in the following way:

[0120] S510. Based on the area of ​​the fat region at the center of the mask image at each position in the phase mask image sequence, determine the key mask image.

[0121] S520. Based on the pericardial fat region in the key mask image, the intersection of the diagonals of the circumscribed polygon of the pericardial fat region is determined as the segmentation center.

[0122] Specifically, for any given phase mask image sequence, the area of ​​the pericardial fat region in the mask image corresponding to each position is calculated at each position in the sequence. Then, the image with the largest pericardial fat region area is selected as the key mask image. Next, based on the pericardial fat region in the key mask image, the circumscribed polygon of the pericardial fat region is determined; that is, the contour of the pericardial fat region is extracted, and the diagonal of the corresponding circumscribed polygon is calculated. Finally, the intersection of these two diagonals is found, and this intersection is used as the segmentation center.

[0123] For example, please refer to Figure 5b For phase k kThe process iterates through all locations and calculates the pericardial fat region in the mask image. The mask image with the largest pericardial fat region is determined as the key mask image. A circumscribed rectangle is taken for the pericardial fat region in the key mask image, and the intersection of the diagonals of the circumscribed rectangle is determined as the segmentation center.

[0124] In the above embodiments, a key mask image is determined based on the area of ​​the pericardial fat region at the center of the mask image at each position in the phase mask image sequence. Based on the pericardial fat region in the key mask image, the intersection of the diagonals of the circumscribed polygons of the pericardial fat region is determined as the segmentation center, providing a data basis for subsequent coronary artery region segmentation.

[0125] In some implementations, please refer to Figure 6a Obtaining coronary artery motion data based on the overlap of the target pericardial fat region in intermediate phase reconstructed image subsequences of adjacent phase reconstructed image sequences may include the following steps:

[0126] S610. Based on the number of overlapping images and the selected phase position, determine the first phase reconstruction image subsequence and the second phase reconstruction image subsequence in the adjacent phase reconstruction image sequence.

[0127] Specifically, since the choice of the number of overlapping images directly affects the accuracy and stability of the calculation results, the appropriate number of overlapping images must be determined based on actual needs and application scenarios. Too many overlapping images may lead to information loss, while too few may make the calculation results overly sensitive to small changes. In the phase reconstruction image sequence, a specific location within a phase is selected as the selected phase location. Then, starting from the selected phase location, the number of images is increased sequentially along the positional direction until the number of selected reconstructed images reaches the number of overlapping images. Finally, the selected reconstructed images are called the first phase reconstruction image subsequence. After determining the selected phase location, the same location in the next adjacent phase is used as the starting point, and the number of images is increased sequentially along the positional direction until the number of selected reconstructed images reaches the number of overlapping images. Finally, the selected masked partition images are called the second phase reconstruction image subsequence.

[0128] For example, the number of overlapping images is PosThickness. See also Figure 4b Select k-phase phase k The q position in the image is used as the selected phase position. Then, in the phase reconstruction image sequence, starting from the selected phase position, the positions are incremented sequentially along the position direction (vertical axis) until the number of selected reconstructed images reaches the PosThickness (number of overlapping images). Finally, the selected reconstructed images are... k,q Image k,q+1 Image k,q+PosThickness-1This is called the first phase reconstruction image subsequence. The k-phase phase... k adjacent k+1 phase k+1 Starting from position q, the number of images is incremented sequentially along the vertical axis until the number of overlapping images reaches the PosThickness. Finally, the selected reconstructed images are... k+1,q Image k+1,q+1 Image k+1,q+PosThickness-1 This is called the second-phase reconstructed image subsequence.

[0129] S620. Based on the overlap between the target pericardial fat region of the first phase reconstruction image subsequence and the target pericardial fat region of the second phase reconstruction image subsequence, determine the coronary motion image corresponding to the coronary artery.

[0130] S630. Based on the coronary motion image corresponding to the coronary artery, determine the coronary motion data corresponding to the coronary artery.

[0131] Specifically, the target pericardial fat region is determined in the first-phase reconstructed image subsequence corresponding to each selected phase position, and the target pericardial fat region is determined in the second-phase reconstructed image subsequence corresponding to each selected phase position. The overlap between the target pericardial fat regions of the first-phase and second-phase reconstructed image subsequences is processed and calculated to determine the coronary artery motion image corresponding to the coronary artery. Since the scan range needs to be slightly larger than the heart itself when scanning the heart to ensure complete cardiac image information is obtained, and since the coronary arteries do not cover the entire heart, the reconstructed cardiac image sequence may not have regions of interest at the beginning and end. By cropping the reconstructed image sequence to include the start and end positions of the coronary arteries, the effective location range can be determined. Next, the region corresponding to the effective location range is cropped from the coronary artery motion image corresponding to the coronary artery. Then, within the cropped region, the motion map values ​​corresponding to a certain phase are summed along the positional direction to obtain the motion data corresponding to that phase. The above operation is repeated for each phase within the extracted region to determine the corresponding motion data for each phase. Finally, the motion data corresponding to each phase are combined according to phase to obtain the coronary artery motion data.

[0132] For example, coronary artery motion data can be coronary artery motion curves. See also... Figure 6bIn the coronary artery motion image corresponding to the coronary artery, a region corresponding to the valid position range ValidPos is extracted. Within the extracted region, the corresponding motion map values ​​within a certain phase are added along the position direction, i.e., the vertical axis, to obtain the motion data for that phase. Finally, the motion data corresponding to each phase are combined according to phase. Please refer to [link to relevant documentation]. Figure 6c This yields the coronary motion curves corresponding to the coronary arteries. It's important to note that the color intensity in the coronary motion images reflects the degree of difference between adjacent images. Darker colors indicate smaller differences between adjacent images and a more stable motion state; lighter colors indicate larger differences between adjacent images and a more vigorous motion state. This translates to smaller curve values ​​in the coronary motion data, indicating more stable motion in the corresponding phase.

[0133] In the above implementation, the first phase reconstruction image subsequence and the second phase reconstruction image subsequence in the adjacent phase reconstruction image sequence are determined based on the number of overlapping images and the selected phase position. Based on the overlap between the target pericardial fat region of the first phase reconstruction image subsequence and the target pericardial fat region of the second phase reconstruction image subsequence, the coronary motion image corresponding to the coronary artery is determined. Based on the coronary motion image corresponding to the coronary artery, the coronary motion data corresponding to the coronary artery is determined, providing a data basis for subsequent determination of the optimal phase.

[0134] In some implementations, please refer to Figure 7a Determining coronary motion data based on coronary artery motion images within the effective location range can include the following steps:

[0135] S710. Within the effective location range of the coronary motion image corresponding to the coronary artery, determine the motion data corresponding to the motion phase based on the selected motion phase position.

[0136] S720. Construct coronary artery motion data corresponding to the coronary artery based on the motion data corresponding to the motion period.

[0137] Specifically, the effective location range needs to be determined first within the coronary motion images corresponding to the coronary arteries. Then, within this limited range, different motion phases are selected at different locations based on the characteristics of the coronary motion images to determine the locations of the selected motion phases. Next, the motion image values ​​corresponding to the selected motion phase locations are summed to obtain the motion data corresponding to each motion phase. Finally, all the motion data corresponding to each motion phase are integrated to construct the coronary motion data corresponding to the coronary arteries.

[0138] For example, please refer to Figure 7bIn the coronary artery motion image corresponding to the coronary artery, the effective location range ValidPos is extracted. Within the extracted region, the motion data corresponding to a certain phase is obtained by summing the motion values ​​corresponding to the red curves. It should be noted that in this method, when the target phase is finally determined, a set of phases represented by the red curves is obtained, rather than a single phase. In other words, for at least one coronary artery, it can be divided into multiple regions according to its location on the heart, and each region corresponds to a target phase. The target phases of different regions may be the same or different. For example, a coronary artery can be divided into the ventricular coronary artery region adjacent to the ventricle and the atrial coronary artery region adjacent to the atrium (more detailed divisions are also possible, but not limited in this implementation). Due to the differences in motion between the ventricle and the atrium, the target phases corresponding to the ventricular coronary artery region and the atrial coronary artery region are usually different. Therefore, different locations on the same coronary artery may correspond to different target phases, that is, each coronary artery can correspond to a set of target phases.

[0139] In the above embodiments, within the effective location range of the coronary motion image corresponding to the coronary artery, motion data corresponding to the selected motion phase is determined based on the selected motion phase position, and coronary motion data corresponding to the coronary artery is constructed based on the motion data corresponding to the motion phase, providing a data basis for subsequent determination of the optimal phase.

[0140] In some implementations, please refer to Figure 8 For the first reconstructed partition image sequence {Image k,q Image k,q+1 Image k,q+PosThickness-1 The red area in the image represents the right coronal region. The pixel values ​​of each pixel position in the right coronal region of these first reconstructed partition images are added together, and the results are integrated to generate the first reconstructed partition image. kSumq .

[0141] For the first mask partition image sequence {Mask k,q Mask k,q+1 Mask k,q+PosThickness-1 The red area in the image represents the right crown partition. The mask regions within these right crown partitions are then combined using a union operation to generate the first combined mask partition image (Mask). kSumq .

[0142] For the second reconstructed partition image sequence {Image k+1,q Image k+1,q+1 Image k+1,q+PosThickness-1The red area in the image represents the right coronal region. The pixel values ​​of each pixel position in the right coronal region of these second reconstructed partition images are added together, and the results are integrated to generate the second reconstructed partition image. k+1Sumq .

[0143] For the second mask partition image sequence {Mask k+1,q Mask k+1,q+1 Mask k+1,q+PosThickness-1 The red area in the image represents the right crown partition. A union operation is performed on the mask regions within these right crown partitions to generate the second combined mask partition image (Mask). k+1Sumq .

[0144] Mask of the first composite mask partition image kSumq Second-and-after mask partition image like Mask k+1Sumq The red masked regions are subjected to a union operation, which combines the portions covered by all these masked regions to determine the total pericardial fat region. Based on the total masked partitioned image, the first reconstructed partitioned image is then constructed. kSumq Second, reconstruct the partitioned image. k+1Sumq Mapping is performed to determine the target pericardial fat region.

[0145] It should be noted that the above operation is repeated for the green area, which represents the left anterior descending artery region, to determine the coronary artery motion image corresponding to the left anterior descending artery region. The above operation is repeated for the blue area, which represents the left circumflex artery region, to determine the coronary artery motion image corresponding to the left circumflex artery region.

[0146] In some implementations, please refer to Figure 9 Based on the overlap of the target pericardial fat region in the first phase reconstruction image subsequence and the overlap of the target pericardial fat region in the second phase reconstruction image subsequence, the coronary motion image corresponding to the coronary artery is determined, which may include the following steps:

[0147] S910. Determine the phase position of the motion period based on the selected phase position.

[0148] Specifically, based on the selected phase position, the currently selected phase and the selected position within that phase can be determined. The moving phase is determined by performing calculations based on the currently selected phase and the next adjacent phase. Then, the selected position within the moving phase is taken as the moving phase position. For example, the currently selected phase and the next adjacent phase can be added together, and the result divided by 2 to obtain the moving phase. For instance, if the currently selected phase is Phase... k Then the phase of the movement = (Phase) k +Phase k+1 ) / 2.

[0149] S920. Based on the overlap between the target pericardial fat region of the first phase reconstructed image subsequence and the target pericardial fat region of the second phase reconstructed image subsequence, determine the motion map value corresponding to the position of the coronary artery in the motion phase.

[0150] S930. Based on the motion map values ​​corresponding to the phase position of each motion phase of the coronary artery, generate the coronary motion image corresponding to the coronary artery.

[0151] Specifically, for pixels at the same position in the target pericardial fat region of the first-phase reconstructed image subsequence and the target pericardial fat region of the second-phase reconstructed image subsequence, the difference between corresponding pixels is calculated to determine the overlap between the two target pericardial fat regions. Then, by traversing all pixels in the target pericardial fat regions of the first-phase and second-phase reconstructed image subsequences, the difference between each pair of corresponding pixels is accumulated to obtain the total reconstructed pixels. The total reconstructed pixels are divided by the total mask pixels, and the result is used as the motion map value corresponding to the coronary artery at that motion phase position. Then, the above process is repeated for all positions and all phases to determine the motion map value corresponding to each motion phase position of the coronary artery. Finally, the motion map values ​​corresponding to each motion phase position of the coronary artery are combined to generate the coronary artery motion image.

[0152] In the above embodiments, the motion phase position is determined based on the selected phase position, and the motion map value corresponding to the motion phase position of the coronary artery is determined based on the overlap between the target pericardial fat region of the first phase reconstructed image subsequence and the target pericardial fat region of the second phase reconstructed image subsequence. Based on the motion map value corresponding to each motion phase position of the coronary artery, a coronary motion image corresponding to the coronary artery is generated, which better reflects the degree of difference between adjacent images.

[0153] In some implementations, please refer to Figure 10a The target interval includes the target systolic interval and the target diastolic interval. Based on cardiac motion data, the target interval corresponding to the target period of the coronary artery is obtained, including:

[0154] S1010. Determine the end-systolic interval based on the changing trend of cardiac motion data.

[0155] Specifically, by analyzing the trends in cardiac motion data, the phase corresponding to the minimum value can be determined. Next, based on a pre-set empirical systolic threshold, the end-systolic interval can be determined with the phase corresponding to the minimum value as the center.

[0156] For example, the coronary artery could be the right coronary artery, and the cardiac motion data could be a cardiac motion curve. See also... Figure 10b , Figure 10b The HeartCurveRCA curve corresponding to the location of the right coronary artery in the figure represents the contraction and relaxation process of the heart; therefore, the curve trend is first decreasing and then increasing. Analyzing this HeartCurveRCA curve, the phase corresponding to the minimum HeartCurveMin is determined. Then, based on the empirical threshold RangeThresh and the phase corresponding to the minimum HeartCurveMin, the end-systolic interval SysRange is determined.

[0157] S1020. Based on the end-systolic interval, map to determine the target systolic interval corresponding to the target period of the coronary artery.

[0158] S1030. In coronary motion data, the phases that meet the preset screening conditions within the target systolic interval are determined as the target systolic phases corresponding to the coronary arteries.

[0159] Specifically, cardiac motion data and coronary artery motion data are aligned and corrected to ensure consistency on the phase axis. Then, based on the end-systolic interval corresponding to the cardiac motion data, a mapping is performed to determine the corresponding phase interval in the coronary artery motion data, thus obtaining the target systolic interval corresponding to the target phase of the coronary artery. The coronary artery motion data within the target systolic interval is traversed, and phases that meet preset screening criteria are selected and determined as the target systolic phase corresponding to the coronary artery.

[0160] For example, coronary artery motion data can be coronary artery motion curves. See also... Figure 10c Using the end-systolic interval (SysRange) as the search interval, mapping is performed on the coronary motion curve (MotionCurveRCA) corresponding to the right coronary artery to determine the target systolic interval (SysRangeM) corresponding to the target phase of the right coronary artery. The coronary motion curve (MotionCurveRCA) within the target systolic interval (SysRangeM) is retrieved, and the phase corresponding to the minimum value among the extreme points within the target systolic interval (SysRangeM) is taken as the target systolic phase (OptimalSys).

[0161] In the above embodiments, the end-systolic interval is determined based on the changing trend of cardiac motion data. The end-systolic interval is then mapped to determine the target systolic interval corresponding to the target phase of the coronary artery. In the coronary motion data, the phases within the target systolic interval that meet the preset screening conditions are determined as the target systolic phases corresponding to the coronary artery. This can improve the accuracy of the optimal phase in the cardiac reconstruction image and reduce the interference of motion artifacts.

[0162] In some implementations, please refer to Figure 11 The target interval includes the target systolic interval and the target diastolic interval. Based on cardiac motion data, the target interval corresponding to the target period of the coronary artery is obtained, including:

[0163] S1110. Determine the end-diastolic interval based on the changing trend of cardiac motion data.

[0164] Specifically, by analyzing the trends in cardiac motion data, the phase corresponding to the maximum value can be determined. Next, based on a pre-set empirical diastolic threshold, the end-diastolic interval can be determined with the phase corresponding to the maximum value as the center. It should be noted that the empirical systolic and diastolic thresholds can be set to the same value, or different values ​​can be set depending on the specific circumstances. The settings of these two thresholds directly affect the definition of cardiac systole and diastole, and thus affect the relevant data range of the coronary arteries.

[0165] For example, the coronary artery could be the right coronary artery, and the cardiac motion data could be a cardiac motion curve. See also... Figure 10b , Figure 10b The HeartCurveRCA curve corresponding to the location of the right coronary artery in the figure represents the contraction and relaxation process of the heart; therefore, the curve trend is first decreasing and then increasing. Analyzing this HeartCurveRCA curve, the phase corresponding to the maximum value HeartCurveMax is determined. Then, based on the empirical threshold RangeThresh and the phase corresponding to the maximum value HeartCurveMax, the end-diastolic interval DiasRange is determined.

[0166] S1120. Based on the end-diastolic interval, map to determine the target diastolic interval corresponding to the target period of the coronary artery.

[0167] S1130. In coronary motion data, the phases within the target diastolic interval that meet the preset screening conditions are determined as the target diastolic phases corresponding to the coronary arteries.

[0168] Specifically, cardiac motion data and coronary artery motion data are aligned and corrected to ensure consistency on the phase axis. Then, based on the end-diastolic interval corresponding to the cardiac motion data, a mapping is performed to determine the corresponding phase interval in the coronary artery motion data, thus obtaining the target diastolic interval corresponding to the target phase of the coronary artery. The coronary artery motion data within the target diastolic interval are traversed, and phases that meet preset screening criteria are selected and determined as the target diastolic phase corresponding to the coronary artery.

[0169] For example, coronary artery motion data can be coronary artery motion curves. See also... Figure 10cUsing the end-diastolic range DiasRange as the search interval, mapping is performed on the coronary motion curve MotionCurveRCA corresponding to the right coronary artery to determine the target diastolic interval DiasRangeM corresponding to the target phase of the right coronary artery. The coronary motion curve MotionCurveRCA within the target diastolic interval DiasRangeM is retrieved, and the phase corresponding to the minimum value among the extreme points within the target diastolic interval DiasRangeM is taken as the target diastolic phase OptimalDias.

[0170] In the above embodiments, the end-diastolic interval is determined based on the changing trend of cardiac motion data. The end-diastolic interval is then mapped to determine the target diastolic interval corresponding to the target phase of the coronary artery. In the coronary motion data, the phase that meets the preset screening conditions within the target diastolic interval is determined as the target diastolic phase corresponding to the coronary artery. This can improve the accuracy of the optimal phase in the cardiac reconstruction image and reduce the interference of motion artifacts.

[0171] It should be noted that the above implementation method is only one optional method of the embodiments described in this specification. In other embodiments, deep learning can be used to obtain coronary artery motion data and cardiac motion data corresponding to the location of the coronary arteries. Specifically, before obtaining the target phase of the coronary artery, a coronary artery motion field output model and a cardiac motion field output model can be obtained through training. Then, multiple phase reconstruction image sequences are input into the coronary artery motion field output model and the cardiac motion field output model to obtain the corresponding coronary artery motion data and cardiac motion data. For obtaining the target phase of different regions of the same coronary artery, multiple phase reconstruction image block sequences can be extracted from multiple phase reconstruction image sequences according to the regions divided by the coronary artery. Then, multiple phase reconstruction image block sequences are input into the coronary artery motion field output model and the cardiac motion field output model to obtain the coronary artery motion data and cardiac motion data corresponding to a certain region of the coronary artery.

[0172] In the above implementation, the coronary artery motion field output model and the cardiac motion field output model can employ convolutional neural networks (CNNs) or recurrent neural networks (RNNs), or even support vector products and random forests. The training samples for the coronary artery motion field output model can be the coronary artery segmented from multiple phase-reconstructed image sequence samples, while the training samples for the cardiac motion field output model can be the cardiac segmented from multiple phase-reconstructed image sequence samples.

[0173] This specification provides a coronary artery phase analysis device 1200. The heart includes multiple coronary arteries. For at least one coronary artery, please refer to [link to documentation]. Figure 12The coronary artery phase analysis device 1200 includes: a reconstructed image acquisition module 1210, a motion data acquisition module 1220, a target interval determination module 1230, and a target phase determination module 1240.

[0174] The reconstructed image acquisition module 1210 is used to acquire multiple phase reconstructed image sequences of the heart;

[0175] The motion data acquisition module 1220 is used to acquire coronary motion data of the coronary artery and cardiac motion data corresponding to the location of the coronary artery based on multiple phase reconstruction image sequences.

[0176] The target interval determination module 1230 is used to obtain the target interval corresponding to the target period of the coronary artery based on the cardiac motion data.

[0177] The target phase determination module 1240 is used to obtain the target phase corresponding to the coronary artery based on the target interval and the coronary motion data.

[0178] For a detailed description of the coronary artery phase analysis device, please refer to the description of the coronary artery phase analysis method above, which will not be repeated here.

[0179] This specification provides a medical imaging device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method steps described above.

[0180] In some embodiments, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 13 As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for coronary artery phase analysis. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0181] Those skilled in the art will understand that Figure 13 The structures shown are merely block diagrams of some structures related to the solutions disclosed in this specification, and do not constitute a limitation on the computer device to which the solutions disclosed in this specification are applied. Specifically, the computer device may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements.

[0182] In some embodiments, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method steps described above.

[0183] This specification provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method in any of the above embodiments.

[0184] One embodiment of this specification provides a computer program product including instructions that, when executed by a processor of a computer device, enable the computer device to perform the steps of the method described in any of the above embodiments.

[0185] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

Claims

1. A method for coronary artery phase analysis, characterized in that, The heart includes multiple coronary arteries, and for at least one coronary artery, the method includes: Acquire multiple phase reconstructed image sequences of the heart; Based on multiple phase reconstruction image sequences, coronary artery motion data and cardiac motion data corresponding to the location of the coronary artery are obtained; Based on the cardiac motion data, the target interval corresponding to the target period of the coronary artery is obtained; Based on the target interval, the target phase corresponding to the coronary artery is obtained from the coronary motion data.

2. The method according to claim 1, characterized in that, The step of acquiring coronary artery motion data and cardiac motion data corresponding to the location of the coronary arteries based on multiple phase-reconstructed image sequences includes: Multiple phase reconstruction image sequences are segmented to obtain multiple phase mask image sequences, wherein the mask image at each position in the phase mask image sequence defines the pericardial fat region and the atrioventricular region; Based on the overlap of the target pericardial fat region in the intermediate phase reconstruction image subsequences of adjacent phase reconstruction image sequences, the coronary artery motion data is obtained, wherein the intermediate phase reconstruction image subsequence includes a reconstruction image at at least one location, and the target pericardial fat region of the intermediate phase reconstruction image subsequence is a region determined by the intermediate phase mask image subsequence and connected to the coronary artery. Based on the area changes of the atrial and ventricular regions in the mask images of the target location in multiple phase mask image sequences, cardiac motion data corresponding to the coronary artery location is obtained, wherein the mask image of the target location corresponds to a specific location on the coronary artery.

3. The method according to claim 2, characterized in that, The target pericardial fat region of the phase-reconstructed image subsequence was determined using the following method: Based on the segmentation center, the coronary artery region is divided into adjacent phase mask image subsequences to obtain the pericardial fat region corresponding to the coronary artery location; Based on the overlap of pericardial fat regions in adjacent phase mask image subsequences, the total pericardial fat region is obtained; Based on the mapping of the total pericardial fat region, the target pericardial fat region of the adjacent phase reconstruction image subsequence is determined.

4. The method according to claim 3, characterized in that, The segmentation center is determined in the following manner: Based on the area of ​​the fat region at the center of the mask image at each position in the phase mask image sequence, the key mask image is determined; Based on the pericardial fat region in the key mask image, the intersection of the diagonals of the circumscribed polygon of the pericardial fat region is determined as the segmentation center.

5. The method according to claim 2, characterized in that, The method of obtaining coronary artery motion data based on the overlap of target pericardial fat regions in intermediate phase reconstruction image subsequences of adjacent phase reconstruction image sequences includes: Based on the number of overlapping images and the selected phase position, determine the first phase reconstruction image subsequence and the second phase reconstruction image subsequence in the adjacent phase reconstruction image sequence; Based on the overlap between the target pericardial fat region of the first phase reconstruction image subsequence and the target pericardial fat region of the second phase reconstruction image subsequence, the coronary motion image corresponding to the coronary artery is determined; Based on the coronary motion image corresponding to the coronary artery, the coronary motion data corresponding to the coronary artery is determined.

6. The method according to claim 5, characterized in that, The determination of the coronary motion image corresponding to the coronary artery based on the overlap between the target pericardial fat region of the first phase reconstruction image subsequence and the target pericardial fat region of the second phase reconstruction image subsequence includes: The position of the motion phase is determined based on the selected phase position; Based on the overlap between the target pericardial fat region of the first phase reconstructed image subsequence and the target pericardial fat region of the second phase reconstructed image subsequence, the motion map value corresponding to the position of the coronary artery in the motion phase is determined; Based on the motion map values ​​corresponding to each phase position of the coronary artery, a coronary motion image corresponding to the coronary artery is generated.

7. The method according to claim 5, characterized in that, The determination of coronary motion data corresponding to the coronary artery based on the coronary motion image corresponding to the coronary artery within the effective location range includes: Within the effective location range of the coronary motion image corresponding to the coronary artery, motion data corresponding to the motion phase is determined based on the selected motion phase position; Coronary artery motion data corresponding to the aforementioned motion phase is constructed based on the motion data.

8. The method according to claim 2, characterized in that, The method of obtaining cardiac motion data corresponding to the coronary artery location by measuring the area changes of the atrial and ventricular regions in the mask images of the target location in multiple phase mask image sequences includes: The target location of the coronary artery is determined based on the area of ​​the atrial and ventricular regions in the phase mask image sequence. Based on the area changes of the atrial and ventricular regions of the mask images corresponding to the target location of the coronary artery in multiple phase mask image sequences, cardiac motion data corresponding to the location of the coronary artery are determined.

9. The method according to claim 1, characterized in that, The target interval includes a target systolic interval and a target diastolic interval. Obtaining the target interval corresponding to the target period of the coronary artery based on the cardiac motion data includes: Based on the changing trends of the cardiac motion data, the end-systolic interval is determined; Based on the mapping of the end-systolic interval, the target systolic interval corresponding to the target phase of the coronary artery is determined; and / or, Based on the changing trends of the cardiac motion data, the end-diastolic interval is determined; Based on the mapping of the end-diastolic interval, the target diastolic interval corresponding to the target period of the coronary artery is determined.

10. The method according to claim 9, characterized in that, The step of obtaining the target phase corresponding to the coronary artery from the coronary motion data based on the target interval includes: Based on the coronary motion data, the phases within the target systolic interval that meet the preset screening criteria are determined as the target systolic phases corresponding to the coronary artery; Based on the coronary artery motion data, the phases within the target diastolic interval that meet the preset screening criteria are determined as the target diastolic phases corresponding to the coronary artery.

11. The method according to any one of claims 1-10, characterized in that, For at least one of the coronary arteries, the coronary artery is divided into multiple regions according to its location on the heart, and each region corresponds to a target phase.

12. A coronary artery phase analysis device, characterized in that, The heart includes multiple coronary arteries, and for at least one coronary artery, the device includes: The reconstructed image acquisition module is used to acquire multiple phase reconstructed image sequences of the heart; The motion data acquisition module is used to acquire coronary motion data of the coronary artery and cardiac motion data corresponding to the location of the coronary artery based on multiple phase reconstruction image sequences. The target interval determination module is used to obtain the target interval corresponding to the target period of the coronary artery based on the cardiac motion data. The target phase determination module is used to obtain the target phase corresponding to the coronary artery based on the target interval and the coronary artery motion data.

13. A medical imaging device, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 11.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 11.

Images