Myocardial stress analysis method and device based on real-time movie imaging

Abstract

The present application discloses a method and device for myocardial stress analysis based on real-time movie imaging, wherein the method comprises acquiring real-time movie image data of cardiac magnetic resonance; when the real-time movie image data is real-time movie image data of heart rate regularity, extracting a movie image sequence of a single cardiac cycle, the sequence comprising N-frame cardiac cycle images of multiple cardiac cycles; acquiring a cardiac cycle registration image corresponding to the N-frame cardiac cycle image; determining the endocardial image boundary corresponding to each frame in the N-frame cardiac cycle registration image; performing myocardial segmentation on the corresponding endocardial image boundary in the first frame cardiac cycle registration image to obtain an initial length L0 of multiple initial myocardial segments; tracking the deformation length L of the matching myocardial segments corresponding to each initial myocardial segment in the remaining N‑1 frames cardiac cycle registration images according to the initial myocardial segmentation; calculating the myocardial stress strain parameters corresponding to the single cardiac cycle according to each initial length L0 and each deformation length L to perform myocardial stress analysis.

Description

Technical Field

[0001] The present application relates to the field of magnetic resonance technology, and in particular to a method and device for myocardial stress analysis based on real-time movie imaging. Background Art

[0002] Myocardial stress refers to the interaction force between myocardial cells during contraction and relaxation of the heart, which can be divided into radial stress, circumferential stress and longitudinal stress. By quantitatively evaluating the percentage of change in unit length of the myocardium from end-diastole to end-systole in a specified direction, the deformation capacity of myocardial fibers under tension can be obtained, thereby determining whether there is an abnormality in myocardial contraction or relaxation, and thus detecting whether the heart function is abnormal.

[0003] In the prior art, cardiac magnetic resonance segmented cine imaging is a commonly used imaging method for measuring cardiac function in clinical practice. Generally, partial K-space data is collected within one cardiac cycle. Since the complete K-space data is completed within multiple cardiac cycles, this method is more sensitive to respiratory movement. In order to eliminate the impact of respiratory movement on imaging quality, the subject needs to hold his breath repeatedly during the scanning process. If the subject does not hold his breath enough or has an irregular heart rate, the accuracy deviation of the collected data is large, which will affect the results of myocardial stress analysis. Therefore, the use of segmented cine imaging technology for myocardial stress analysis has great limitations.

[0004] Therefore, how to achieve more accurate analysis of myocardial stress is a technical problem that needs to be urgently solved by technical personnel in this field. Summary of the invention

[0005] The main purpose of this application is to provide a myocardial stress analysis method and device based on real-time movie imaging, aiming to achieve accurate analysis of myocardial stress.

[0006] To achieve the above objectives, the present application provides a myocardial stress analysis method based on real-time movie imaging, the method comprising:

[0007] Acquire real-time movie image data corresponding to cardiac magnetic resonance;

[0008] When the real-time movie image data is real-time movie image data of heart rate regularity, extracting a movie image sequence of a single cardiac cycle from the real-time movie image data, the movie image sequence includes N frames of cardiac cycle diagrams of multiple cardiac cycles, where N is a positive integer;

[0009] enhancing the contrast of the N-frame cardiac cycle images, and performing image rigid registration on the N-frame cardiac cycle images to obtain a cardiac cycle registration image corresponding to the N-frame cardiac cycle images;

[0010] Determine the endocardial image boundary corresponding to each frame in the N frames of cardiac cycle registration images;

[0011] Performing myocardial segmentation on the boundary of the endocardium image corresponding to the first frame of the cardiac cycle registration image to obtain multiple initial myocardial segments and recording an initial length L0 of each initial myocardial segment;

[0012] According to the initial myocardial segment, the matching myocardial segments corresponding to each of the initial myocardial segments in the remaining N-1 frames of the cardiac cycle registration map are tracked, and the deformation length L of each of the matching myocardial segments is recorded;

[0013] The myocardial stress strain parameters corresponding to the single cardiac cycle are calculated according to each of the initial lengths L0 and each of the deformation lengths L to perform myocardial stress analysis.

[0014] To achieve the above objectives, the present application provides a myocardial stress analysis method based on real-time movie imaging, the method comprising:

[0015] Acquire real-time movie image data corresponding to cardiac magnetic resonance;

[0016] When the real-time movie image data is real-time movie image data corresponding to arrhythmia, extracting a plurality of movie image sequences corresponding to a plurality of cardiac cycles from the real-time movie image data, each of the movie image sequences comprising N frames of cardiac cycle diagrams, where N is a positive integer;

[0017] enhancing the contrast of the N-frame cardiac cycle images, and performing image rigid registration on the N-frame cardiac cycle images to obtain a cardiac cycle registration image corresponding to the N-frame cardiac cycle images;

[0018] Determine the endocardial image boundary corresponding to each frame in the N frames of cardiac cycle registration images;

[0019] Performing myocardial segmentation on the boundary of the endocardium image corresponding to the first frame of the cardiac cycle registration image to obtain multiple initial myocardial segments and recording an initial length L0 of each initial myocardial segment;

[0020] According to the initial myocardial segment, the matching myocardial segments corresponding to each of the initial myocardial segments in the remaining N-1 frames of the cardiac cycle registration map are tracked, and the deformation length L of each of the matching myocardial segments is recorded;

[0021] Myocardial stress strain parameters corresponding to a plurality of cardiac cycles are calculated according to each of the initial lengths L0 and each of the deformation lengths L to perform myocardial stress analysis.

[0022] To achieve the above-mentioned purpose, the present application also provides a myocardial stress analysis device, which includes a processor, a memory, and a storage, wherein the memory stores a myocardial stress analysis program that can be executed by the processor, and when the computer program is executed by the processor, the aforementioned myocardial stress analysis method is implemented.

[0023] The present application provides a method and device for myocardial stress analysis based on real-time movie imaging, wherein the method comprises acquiring real-time movie image data corresponding to cardiac magnetic resonance; when the real-time movie image data is real-time movie image data of heart rate regularity, extracting a movie image sequence of a single cardiac cycle from the real-time movie image data, wherein the movie image sequence comprises N frames of cardiac cycle diagrams of multiple cardiac cycles, where N is a positive integer; enhancing the contrast of the N frames of the cardiac cycle diagrams, and performing image rigid registration on the N frames of the cardiac cycle diagrams to obtain a cardiac cycle registration diagram corresponding to the N frames of the cardiac cycle diagrams; determining the N frames of the cardiac cycle diagrams; Endocardial image boundaries corresponding to each frame in the cycle registration map; myocardial segmentation corresponding to the endocardial image boundaries in the first frame of the cardiac cycle registration map to obtain multiple initial myocardial segments and record the initial length L0 of each initial myocardial segment; tracking the matching myocardial segments corresponding to each of the initial myocardial segments in the remaining N-1 frames of the cardiac cycle registration map according to the initial myocardial segmentation, and recording the deformation length L of each matching myocardial segment; calculating the myocardial stress strain parameters corresponding to the single cardiac cycle according to each of the initial lengths L0 and each of the deformation lengths L to perform myocardial stress analysis. The real-time movie image data acquisition used in the myocardial stress analysis method provided in the present application can realize data acquisition when the subject is in a free breathing state, which greatly improves the accuracy of myocardial stress analysis using real-time movie image data. By performing myocardial stress analysis on the real-time movie image data corresponding to cardiac magnetic resonance, the accuracy of myocardial stress analysis can be effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 A flowchart of the steps of a myocardial stress analysis method based on real-time movie imaging provided in this application;

[0026] Figure 2 A schematic diagram of rearranging multiple cardiac cycle images to obtain a single cardiac cycle image using the ECG delay time extracted from the K-space in the present application;

[0027] Figure 3 A flowchart of the steps of another myocardial stress analysis method based on real-time movie imaging provided in this application;

[0028] Figure 4 Schematic diagram of the rigid registration of the central motion periodogram in this application;

[0029] Figure 5 A flowchart of the steps of another myocardial stress analysis method based on real-time movie imaging provided in this application;

[0030] Fig. 6A This is a schematic diagram of the endocardial border outlined in the first frame of the cardiac cycle registration image;

[0031] Figure 6B This is a schematic diagram of the endocardial border traced in the registration image of the fourth frame of the cardiac cycle;

[0032] Figure 7 A flowchart of the steps of another myocardial stress analysis method based on real-time movie imaging provided in this application;

[0033] Figure 8 A schematic diagram of combining a pixel matching template in a first frame of cardiac cycle registration image with N-1 second discrete point pixel blocks corresponding to the remaining N-1 frames of cardiac cycle registration images;

[0034] Fig. 9 is a schematic diagram of the second endocardial boundary obtained by fitting the second discrete points;

[0035] Fig.10 A flowchart of the steps of another myocardial stress analysis method based on real-time movie imaging provided in this application;

[0036] Fig.11 A schematic diagram of the structure for segmenting the myocardium based on the boundary of the endocardial image;

[0037] Fig.12 A flowchart of the steps of another myocardial stress analysis method based on real-time movie imaging provided in this application;

[0038] Fig.13 A flowchart of the steps of another myocardial stress analysis method based on real-time movie imaging provided in this application;

[0039] Fig.14 A schematic diagram of feature tracking of a first myocardial discrete point of the initial myocardial segmentation in a second frame cardiac cycle registration image;

[0040] Fig.15 A flowchart of the steps of another myocardial stress analysis method based on real-time movie imaging provided in this application;

[0041] Fig.16 A schematic block diagram of a myocardial stress analysis device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0043] It should be noted that the descriptions of "first", "second", etc. in this application are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" or "second" may explicitly or implicitly include at least one of the features.

[0044] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0045] In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the fact that they can be implemented by ordinary technicians in this field. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0046] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0047] See also Figure 1 , Figure 1The present invention is a flow chart of a myocardial stress analysis method based on real-time movie imaging. The method is applied to a myocardial stress analysis device. The myocardial stress analysis device can be a server, or a mobile phone, a tablet computer, a wearable device, a vehicle-mounted device, an augmented reality (AR) / virtual reality (VR) device, a laptop computer, a personal computer (PC), a netbook, a personal digital assistant (PDA) and other terminal devices. The embodiment of the present application does not impose any restrictions on the specific type of the myocardial stress analysis device.

[0048] like Figure 1 As shown, the method includes steps S11 to S17.

[0049] Step S11: Acquire real-time movie image data corresponding to cardiac magnetic resonance.

[0050] Real-time cine cardiac magnetic resonance imaging uses technology to appropriately reduce spatial resolution, parallel imaging technology and special reconstruction algorithms, such as the sliding window method and echo sharing method, to complete the acquisition of all K-space data within one cardiac cycle. Therefore, the image quality is less affected by respiratory movement, and data acquisition can be achieved when the subject is breathing freely, greatly improving the accuracy of myocardial stress analysis using real-time cine image data.

[0051] When magnetic resonance is performed on the subject's heart, the acquired magnetic resonance real-time movie image data includes real-time movie image data of heart rate regularity and real-time movie image data of heart rate irregularity.

[0052] Step S12: When the real-time movie image data is real-time movie image data of heart rate regularity, extract a movie image sequence of a single cardiac cycle from the real-time movie image data, wherein the movie image sequence includes N frames of cardiac cycle diagrams of multiple cardiac cycles, where N is a positive integer.

[0053] Since the spatial resolution of the real-time movie image data is low, the real-time movie image data of multiple cardiac cycles are rearranged, and N frames of cardiac cycle images of multiple cardiac cycles are extracted and spliced ​​to obtain a single cardiac cycle movie image sequence with a spatial resolution that meets the preset value. Wherein N is a positive integer and N≥30.

[0054] In some embodiments, extracting a movie image sequence of a single cardiac cycle from the real-time movie image data includes:

[0055] Extract the ECG delay time recorded in the k-space data corresponding to each frame of the real-time movie image;

[0056] Sorting the ECG delay times and generating an index;

[0057] The image frame sequence in the real-time movie image is rearranged according to the index to obtain a movie image sequence of a single cardiac cycle with a time resolution that meets a preset value.

[0058] like Figure 2 As shown, specifically, the ECG delay time corresponding to the k-space data corresponding to each frame of the real-time movie image data is extracted, the ECG delay time is sorted in ascending or descending order to generate an index, and the order of the image frames in the real-time movie image data is rearranged according to the index to obtain a movie image sequence of a single cardiac cycle whose time resolution meets the preset value. The movie image sequence of a single cardiac cycle includes N frames of cardiac cycle diagrams of multiple cardiac cycles.

[0059] Step S13: enhancing the contrast of the N frames of cardiac cycle graphs, and performing image rigid registration on the N frames of cardiac cycle graphs to obtain cardiac cycle registration graphs corresponding to the N frames of cardiac cycle graphs.

[0060] The acquired N-frame cardiac cycle images are contrast enhanced by an image contrast enhancement method to improve the image contrast, thereby improving the image quality of the N-frame cardiac cycle images, and the N-frame cardiac cycle images after contrast enhancement are rigidly registered to obtain cardiac cycle registration images corresponding to the N-frame cardiac cycle images, thereby eliminating the inter-frame heart shift caused by respiratory movement. For example, the image enhancement method can be an image enhancement method based on a Laplace operator, or an image enhancement method based on histogram equalization, which is not limited here.

[0061] See also Figure 3 In some embodiments, the performing of image rigid registration on the N frames of cardiac cycle images to obtain the corresponding N frames of cardiac cycle registration images includes:

[0062] Step S131: selecting a first cardiac region from the cardiac cycle image of the first frame, and recording first position information of the first cardiac region;

[0063] Step S132: performing feature matching between the first cardiac region and the remaining N-1 frames of the cardiac cycle graph;

[0064] Step S133: acquiring a second cardiac region corresponding to and matching the first cardiac region in each frame of the remaining N-1 frames of the cardiac cycle graph, and recording second position information corresponding to the second cardiac region;

[0065] Step S134: using the cardiac cycle diagram of the first frame as a template, and moving the remaining N-1 frames of the cardiac cycle diagram according to the first position information and the second position information to obtain corresponding N frames of the cardiac cycle registration diagram.

[0066] Specifically, the user finds the area where the heart is located from the first frame of the cardiac cycle diagram, and sends a selection command to the computer device through an input device such as a touch screen, a mouse, or a keyboard. The computer device selects the first heart area from the first frame of the cardiac cycle diagram according to the selection command of the user, and records the first position information corresponding to the first heart area. Preferably, the geometric center coordinates of the first heart area are selected as the first position information.

[0067] The first heart region selected in the first cardiac cycle image is used as a matching template, and image feature matching is performed with each frame of the remaining N-1 cardiac cycle images to obtain the corresponding heart image region in each frame of the remaining N-1 cardiac cycle images.

[0068] For example, in this embodiment, the normalized cross-correlation coefficient method is used to perform image feature matching as an example for description.

[0069] Each of the remaining N-1 cardiac cycle images is the image to be matched I. The pixel size of the image to be matched I is known. Assume that the pixel size of the image to be matched I is M×N, and the pixel size of the matching template T is m×n. A sub-image Ix,y with a pixel size of m×n is randomly selected from the image I. The coordinates of its upper left corner in the image to be matched I are (x, y). It can be seen that the coordinate range is 0≤x≤Mm, 0≤y≤Nn, where M and N are the number of rows and columns of pixels in the image to be matched, respectively, and m and n are the number of rows and columns of pixels in the template, respectively.

[0070] The normalized cross-correlation value R(x, y) between the sub-image I(x, y) and the matching template T is defined as follows:

[0071]

[0072] In the formula, (i, j) is the coordinate of the pixel in the matching template T, is the average pixel value of sub-image Ix,y, is the average pixel value of the matching template T.

[0073] By calculating the normalized cross-correlation value R, the highest point of the normalized cross-correlation value R in each frame of the cardiac cycle diagram to be matched is used as the best matching point, and the position information of the point is used as the second position information.

[0074] After obtaining the first position information of the first frame of the cardiac cycle graph and the second position information of each frame of the remaining N-1 frames of the cardiac cycle graph, the first frame of the cardiac cycle graph is used as a template to move the remaining N-1 frames of the cardiac cycle graph, thereby realizing the calibration of the N frames of the cardiac cycle graph to obtain the corresponding N frames of the cardiac cycle registration graph.

[0075] In some embodiments, performing feature matching on the first cardiac region and the remaining N-1 frames of the cardiac cycle graph includes:

[0076] Selecting a region to be matched in each of the remaining N-1 frames of the cardiac cycle graph, wherein the area of ​​the region to be matched is larger than the area of ​​the first heart region;

[0077] Perform feature matching on the first heart region and the region to be matched.

[0078] Specifically, a matching region is selected in the N-1 frame cardiac cycle image to be matched, and the area of ​​the matching region is smaller than the area of ​​the current frame cardiac cycle image and larger than the area of ​​the matching template. The matching template is matched with the matching region, thereby effectively reducing the time used for matching.

[0079] like Figure 4 As shown, the fourth frame of the cardiac cycle diagram to be matched is taken as an example for explanation, part a is the first frame of the cardiac cycle diagram, b is the fourth frame of the cardiac cycle diagram to be matched, the framed area is the area to be matched corresponding to the fourth frame, c is the fourth frame of the cardiac cycle diagram after matching, and the horizontal line shows the change in heart position before and after alignment.

[0080] Step S14: determining the endocardial image boundary corresponding to each frame in the N frames of the cardiac cycle registration image.

[0081] Each frame of the cardiac cycle registration image includes an endocardial area image. After obtaining N frames of cardiac cycle registration images, it is necessary to obtain the endocardial boundary corresponding to the endocardial area image in the center of each frame of the cardiac cycle registration image.

[0082] See also Figure 5 In some embodiments, determining the endocardial image boundary corresponding to each frame in the N frames of the cardiac cycle registration map includes:

[0083] Step S141: outlining a first endocardial boundary corresponding to the endocardial region image in the first frame of the cardiac cycle registration image, and discretizing the first endocardial boundary into a preset number of first discrete points.

[0084] Specifically, the user finds the area where the endocardial region image is located from the first frame cardiac cycle registration image, and gives a delineation instruction to the computer device through an input device such as a touch screen, a mouse, or a keyboard. The computer device delineates the first endocardial boundary B0 corresponding to the endocardial region image from the first frame cardiac cycle registration image according to the user's delineation instruction, such as Fig. 6A The first endocardial boundary is discretized into a preset number of first discrete points by using an interpolation method, for example, into 300-500 first discrete points. The number of the first discrete points can be set as required and is not limited here.

[0085] Step S142: Determine a corresponding second discrete tracking point in each of the remaining N-1 frames of the cardiac cycle registration image according to the first discrete point.

[0086] The first discrete points discretized by the first endocardial boundary are matched with the corresponding second discrete tracking points in each of the remaining N-1 frames of the cardiac cycle registration image.

[0087] See also Figure 7 In some embodiments, determining a corresponding second discrete tracking point in each frame of the remaining N-1 frames of the cardiac cycle registration map according to the first discrete point includes:

[0088] Step S1421: selecting a pixel strip of a preset thickness in the radial direction of each of the first discrete points as a pixel matching template;

[0089] Step S1422: performing feature matching between each of the pixel matching templates and each of the remaining N-1 frames of the cardiac cycle registration images;

[0090] Step S1423: acquiring a second discrete point pixel block corresponding to and matching the pixel matching template in each of the remaining N-1 frames of the cardiac cycle registration image;

[0091] Step S1424: selecting a matching template area in each of the pixel matching templates, and obtaining a matching area with the maximum matching degree with the matching template area in the second discrete point pixel block, and using the points in the matching area as the second discrete tracking points.

[0092] Exemplarily, a pixel strip b0 of preset thickness is selected in the radial direction of each first discrete point as a pixel matching template, and feature matching is performed between each pixel matching template and the corresponding preset area in each frame of the remaining N-1 frames of cardiac cycle registration images, so as to obtain a second discrete point pixel block b1 corresponding to the pixel matching template in each frame, wherein the feature matching method between the pixel template in the first frame of the cardiac cycle registration image and the remaining N-1 frames of the cardiac cycle registration image can adopt the normalized cross-correlation coefficient method, which will not be elaborated here.

[0093] Then, a matching template area is selected from each pixel matching template of the first frame of the cardiac cycle registration image, and the matching area with the maximum matching degree with the matching template area is obtained in the second discrete point pixel block in each frame of the remaining N-1 frames of the cardiac cycle registration image, and any point in the matching area is used as the second discrete tracking point.

[0094] like Figure 8 As shown, Figure 8 The method includes a pixel matching template obtained from the first frame of the cardiac cycle registration image and a second discrete point pixel block corresponding to the pixel matching template obtained from each frame of the remaining N-1 frames of the cardiac cycle registration image.

[0095] Step S143: modifying the second discrete tracking point to obtain a corresponding second discrete point.

[0096] Each first discrete point corresponding to the first frame of the cardiac cycle registration image can find a corresponding second discrete tracking point in each frame of the remaining N-1 frames of the cardiac cycle registration image. The second discrete tracking point may deviate from the preset range due to the introduction of matching errors during feature matching, and the second discrete tracking point needs to be corrected.

[0097] For example, the matching of a first discrete point is taken as an example for explanation.

[0098] A first discrete point in the first frame of the cardiac cycle registration map can obtain a corresponding second discrete tracking point in each frame of the remaining N-1 frames of the cardiac cycle registration map, that is, there are N-1 second discrete tracking points in total. The first discrete point and the N-1 second discrete tracking points can be fitted to form a curve, which matches the preset curve of myocardial contraction and relaxation when the heart beats. If a second discrete tracking point deviates significantly, that is, the curvature significantly exceeds the preset curvature value, it indicates that the second discrete tracking point is incorrectly matched, and it is necessary to abandon the second discrete tracking point and reset a second discrete tracking point according to the corresponding position of the preset curve, so that the curvature corresponding to the curve formed by the first discrete point and the N-1 second discrete tracking points meets the preset curvature value.

[0099] Step S144: determining a second endocardial boundary corresponding to the endocardial region image in each of the remaining N-1 frames of the cardiac cycle registration image according to the second discrete points.

[0100] In each frame of the remaining N-1 cardiac cycle registration images, a second discrete point corresponding to the first discrete point is obtained, and the second discrete point is fitted to obtain a second endocardial boundary corresponding to the endocardial region image of each frame of the remaining N-1 cardiac cycle registration images, such as Fig. 9 As shown, Fig. 9is the second endocardial boundary obtained by fitting the second discrete points.

[0101] See also Fig.10 In some embodiments, determining the second endocardial boundary corresponding to the endocardial region image in each frame of the remaining N-1 frames of the cardiac cycle registration image according to the second discrete point includes:

[0102] Step S1441: determining an initial second endocardial boundary corresponding to the endocardial region image in each frame of the remaining N-1 frames of the cardiac cycle registration image according to the second discrete points;

[0103] Step S1442: Detect whether there is an abnormal curve area in the initial second endocardial boundary, wherein the abnormal curve area is a second endocardial boundary area whose curvature exceeds a preset curvature value;

[0104] Step S1443: When an abnormal curve area exists in the initial second endocardial boundary, the abnormal curve area is corrected to obtain the corresponding second endocardial boundary.

[0105] For example, during the second endocardial boundary fitting process, when confirming multiple second discrete points in each frame of the remaining N-1 cardiac cycle registration images, there may be some second discrete point offsets, resulting in insufficient accuracy of the fitted second endocardial boundary.

[0106] Therefore, in this embodiment, the second discrete points are used to determine the initial second endocardial boundary corresponding to the endocardial region image in each frame of the remaining N-1 frames of the cardiac cycle registration map, and detect whether the initial second endocardial boundary has an abnormal curve area, wherein the abnormal curve area is the second endocardial boundary area corresponding to the curvature of the initial second endocardial boundary exceeding the preset curvature value. When the initial second endocardial boundary has an abnormal curve area, it indicates that there is a second discrete point matching error, and the second discrete point with the error is removed, and a new second discrete point is set to make the fitted second endocardial boundary smooth and continuous, that is, the curvature is within the preset range.

[0107] Step S15: performing myocardial segmentation on the boundary of the endocardium image corresponding to the first frame of the cardiac cycle registration image to obtain multiple initial myocardial segments and record the initial length L0 of each initial myocardial segment.

[0108] The myocardial 16-segment segmentation method is used to segment the endocardial image boundary of the first frame of the cardiac cycle registration image to obtain multiple initial myocardial segments, and the initial length L0 corresponding to each initial myocardial segment is recorded.

[0109] like Fig.11As shown, the myocardium 16-segment segmentation method is used to segment the endocardial image boundary of the first frame of the cardiac cycle registration image, and 6 initial myocardial segments D0 are obtained, and the initial length L0 corresponding to each initial myocardial segment D0 is recorded.

[0110] Step S16: According to the initial myocardial segment, the matching myocardial segments corresponding to each of the initial myocardial segments in the remaining N-1 frames of the cardiac cycle registration map are tracked, and the deformation length L of each of the matching myocardial segments is recorded.

[0111] In a cardiac cycle, since the heart is performing contraction or relaxation, the length of the myocardial segment corresponding to each frame image of the cardiac cycle is changing.

[0112] The initial myocardial segment is used as a matching template to track the corresponding matching myocardial segment in the next adjacent frame. After the matching myocardial segment matching the initial myocardial segment in the next adjacent frame is tracked, the current matching myocardial segment is used as a tracking template to track the matching myocardial segment matching the current matching template in the next adjacent frame until the corresponding matching myocardial segment is tracked in the last frame. That is, each initial myocardial segment obtained in the first frame cardiac cycle registration map is used to track the matching myocardial segment corresponding to each initial myocardial segment in the second frame cardiac cycle registration map. Then, the matching myocardial segment corresponding to each initial myocardial segment in the second frame cardiac cycle registration map is used as a matching template to track the corresponding matching myocardial segment in the next adjacent frame cardiac cycle registration map, and so on, until the corresponding matching myocardial segment is tracked in the last frame.

[0113] Alternatively, each initial myocardial segment obtained in the first frame of the cardiac cycle registration image is used to track the matching myocardial segments corresponding to each initial myocardial segment in each frame of the remaining N-1 frames of the cardiac cycle registration image, and the deformation length L of each matching myocardial segment in each frame is recorded.

[0114] See also Fig.12 In some embodiments, tracking the matching myocardial segments corresponding to each of the initial myocardial segments in the remaining N-1 frames of the cardiac cycle registration map according to the initial myocardial segments includes:

[0115] Step S161: discretizing the initial myocardial segment into a preset number of first myocardial discrete points.

[0116] A preset number of first myocardial discrete points is generated by interpolation for each initial myocardial segment D0, and the number of the first myocardial discrete points can be set as required.

[0117] Step S162: sequentially acquiring, based on each of the first myocardial discrete points, a second myocardial discrete point that matches the first myocardial discrete point in each of the remaining N-1 frames of the cardiac cycle registration image.

[0118] A preset size matching template is set with each first myocardial discrete point as the center, such as a size of 10×10 pixels. In the area corresponding to the endocardial boundary point closest to the corresponding first myocardial discrete point in the cardiac cycle registration map of the next frame, such as a size of 20×20 pixels, the normalized mutual correlation coefficient method described above is used to obtain the point with the largest mutual correlation coefficient with the matching template as the best matching point, which is used as the second myocardial discrete point corresponding to the current first myocardial discrete point, thereby obtaining the second myocardial discrete point matching the first myocardial discrete point in each frame.

[0119] See also Fig.13 In some embodiments, the step of sequentially acquiring, based on each of the first myocardial discrete points, a second myocardial discrete point that matches the first myocardial discrete point in each of the remaining N-1 frames of the cardiac cycle registration image comprises:

[0120] Step S1621: forwardly tracking the first tracking discrete point matching the first myocardial discrete point in each frame of the remaining N-1 frames of the cardiac cycle registration image according to each of the first myocardial discrete points.

[0121] In some embodiments, forward tracking the first tracking discrete point matching the first myocardial discrete point in each frame of the remaining N-1 frames of the cardiac cycle registration image according to each of the first myocardial discrete points includes:

[0122] Taking each of the first myocardial discrete points as the center, acquiring the corresponding first myocardial discrete area;

[0123] Setting a first myocardial mapping area corresponding to the first myocardial discrete area in the adjacent cardiac cycle registration image, wherein the first myocardial mapping area is an area on the boundary of the endocardial image where the point closest to the corresponding first myocardial discrete point is located, and the area of ​​the first myocardial mapping area is larger than the area of ​​the first myocardial discrete area;

[0124] Perform feature matching on the first myocardial discrete area and the first myocardial mapping area to obtain a point with the highest matching degree as a discrete mapping point;

[0125] The discrete mapping points are projected onto corresponding endocardial boundaries to obtain corresponding discrete projection points, and the discrete projection points are used as the first tracking discrete points.

[0126] See also Fig.14 , an example is given for describing the tracking of a first myocardial discrete point of an initial myocardial segment of a first frame cardiac cycle registration image in a second frame cardiac cycle registration image.

[0127] Specifically, a first myocardial discrete area S0 of a preset size is obtained with each of the first myocardial discrete points k0 as the center, such as a size of 10×10 pixels. A first myocardial mapping area S1 corresponding to the first myocardial discrete area S0 is set in the cardiac cycle registration image of the next adjacent frame, wherein the first myocardial mapping area S1 is an area where the point closest to the corresponding first myocardial discrete point k0 on the boundary of the endocardial image is located, and the area of ​​the first myocardial mapping area S1 is larger than the area of ​​the first myocardial discrete area S0, such as a size of 20×20 pixels.

[0128] The first myocardial discrete area S0 is used as a pixel matching template to perform feature matching with the first myocardial mapping area S1 to obtain the center point of the area with the highest matching degree as the discrete mapping point k1, wherein the feature matching method can adopt the normalized cross-correlation coefficient method mentioned above, which will not be repeated here.

[0129] The discrete mapping point k1 is projected onto the corresponding endocardial boundary to obtain a corresponding discrete projection point k2, and the discrete projection point k2 is used as the first tracking discrete point, thereby ensuring that the first tracking scattered point falls on the endocardial image boundary.

[0130] Step S1622: reversely trace the second tracking discrete point matching the first myocardial discrete point in each frame of the remaining N-1 frames of the cardiac cycle registration image according to each of the first myocardial discrete points.

[0131] Similarly, when tracking the second discrete point, a reverse tracking discrete area of ​​a preset size is obtained based on the first myocardial discrete point in the first frame cardiac cycle registration image as the center, for example, the size of the reverse tracking discrete area is 10×10 pixels.

[0132] Using the reverse tracking discrete area as the matching template, the matching is started from the last cardiac cycle registration map in the remaining N-1 frames of cardiac cycle registration maps until the second cardiac cycle registration map is matched, and the reverse tracking mapping area corresponding to the reverse tracking discrete area is searched in each frame from the last cardiac cycle registration map to the second cardiac cycle registration map. Among them, the reverse tracking mapping area is the area on the boundary of the endocardial image where the point closest to the corresponding reverse tracking discrete area is located, and the area of ​​the reverse tracking mapping area is larger than the area of ​​the reverse tracking discrete area, such as the size of the reverse tracking mapping area is 20×20 pixels.

[0133] The center point of the area in the reverse tracking mapping area with the highest matching degree with the reverse tracking discrete area is used as the reverse discrete mapping point, wherein the feature matching method can adopt the normalized mutual correlation coefficient method mentioned above, which will not be repeated here.

[0134] The reverse discrete mapping points are projected onto the corresponding endocardial boundary to obtain corresponding reverse discrete projection points, and the reverse discrete projection points are used as the second tracking discrete points, thereby ensuring that the second tracking scattered points fall on the endocardial image boundary.

[0135] Step S1623: Calculate the average value of the position coordinates of the first tracking discrete point and the second tracking discrete point in the cardiac cycle registration map to determine the corresponding second myocardial discrete point.

[0136] In this embodiment, in order to ensure the accuracy of the acquired second myocardial discrete point data, each first myocardial discrete point is used to forward track the first tracking discrete point matching the first myocardial discrete point in each frame of the remaining N-1 cardiac cycle registration images.

[0137] At the same time, each first myocardial discrete point is used to reversely track the second tracking discrete point matching the first myocardial discrete point in each frame of the cardiac cycle registration image of the remaining N-1 frames, and the average value of the position coordinates of the first tracking discrete point and the corresponding second tracking discrete point in each frame of the corresponding cardiac cycle registration image is calculated to determine and eliminate the data error that exists when the parameters of the second myocardial discrete point are obtained unidirectionally.

[0138] Step S163: acquiring the corresponding matching myocardial segment according to the second myocardial discrete points.

[0139] A pixel block in a preset area is selected with the second discrete point as the center, and the pixel block is spliced ​​using image splicing technology to obtain a corresponding matching myocardial segment.

[0140] Step S17: calculating the myocardial stress strain parameters corresponding to the single cardiac cycle according to each of the initial lengths L0 and each of the deformation lengths L, so as to perform myocardial stress analysis.

[0141] After obtaining the initial lengths L0 of the initial myocardial segments corresponding to the endocardial boundary in the center of the N-frame cardiac cycle registration image and the deformation lengths L of the matching myocardial segments, the strain coefficient S and strain rate SR of the endocardial boundary are fitted using a preset formula, thereby performing myocardial stress analysis based on the strain coefficient S and strain rate SR of the endocardial boundary.

[0142] For example, the strain factor S and strain rate SR can be calculated using the following formula:

[0143] Strain coefficient S: S = (L-L0) / L0

[0144] Strain rate SR:

[0145] See also Fig.15The present application also provides a myocardial stress analysis method based on real-time movie imaging, which includes steps S21 to S27.

[0146] Step S21: Acquire real-time movie image data corresponding to cardiac magnetic resonance.

[0147] Step S21 and Figure 1 The process is the same as step S11 and will not be described in detail here.

[0148] Step S22: When the real-time movie image data is real-time movie image data corresponding to arrhythmia, a plurality of movie image sequences corresponding to a plurality of cardiac cycles are extracted from the real-time movie image data, each of the movie image sequences comprising N frames of cardiac cycle diagrams, where N is a positive integer.

[0149] Step S22 and Figure 1 Different from step S11, in step S22, there is no need to rearrange the real-time movie image data of multiple cardiac cycles, and the real-time movie image data of multiple cardiac cycles are directly acquired, and image analysis is performed on the N frames of cardiac cycle diagrams of each of the multiple cardiac cycles.

[0150] Step S23: enhancing the contrast of the N frames of cardiac cycle images, and performing image rigid registration on the N frames of cardiac cycle images to obtain cardiac cycle registration images corresponding to the N frames of cardiac cycle images.

[0151] Step S23 and Figure 1 The process is the same as step S13 and will not be described in detail here.

[0152] Step S24: Determine the endocardial image boundary corresponding to each frame in the N frames of the cardiac cycle registration image.

[0153] Step S24 and Figure 1 The process is the same as step S14 and will not be described in detail here.

[0154] Step S25: performing myocardial segmentation on the boundary of the endocardium image corresponding to the first frame of the cardiac cycle registration image to obtain multiple initial myocardial segments and record the initial length L0 of each initial myocardial segment.

[0155] Step S25 and Figure 1 The process is the same as step S15 and will not be described in detail here.

[0156] Step S26: According to the initial myocardial segment, the matching myocardial segments corresponding to each of the initial myocardial segments in the remaining N-1 frames of the cardiac cycle registration map are tracked, and the deformation length L of each of the matching myocardial segments is recorded.

[0157] Step S26 and Figure 1 The process is the same as step S16 and will not be described in detail here.

[0158] Step S27: calculating myocardial stress strain parameters corresponding to a plurality of cardiac cycles according to each of the initial lengths L0 and each of the deformation lengths L, so as to perform myocardial stress analysis.

[0159] After calculating the initial lengths L0 of the initial myocardial segments corresponding to the central endocardial boundary of the cardiac cycle registration image of each N frames in multiple cardiac cycles and the deformation lengths L of each matching myocardial segment, the myocardial stress-strain parameters of the corresponding cardiac cycle can be calculated, and myocardial stress analysis is performed by using the myocardial stress-strain parameters corresponding to multiple cardiac cycles.

[0160] See also Fig.16 , Fig.16 1 is a schematic block diagram of a myocardial stress analysis device provided in an embodiment of the present application. The myocardial stress analysis device may include a server or a terminal device. The myocardial stress analysis device 30 includes a processor 301 and a memory 302, and the processor 301 and the memory 302 are connected via a bus, such as an I2C (Inter-integrated Circuit) bus.

[0161] Specifically, the processor 301 may be a micro-controller unit (MCU), a central processing unit (CPU) or a digital signal processor (DSP).

[0162] The memory 302 may be a Flash chip, a read-only memory (ROM) disk, an optical disk, a USB flash drive, or a mobile hard disk.

[0163] The processor 301 is used to run the computer program stored in the memory 302, and implement any one of the myocardial stress analysis methods provided in the embodiments of the present application when executing the computer program.

[0164] Exemplarily, the processor 301 is used to run a computer program stored in the memory, and implement the following steps when executing the computer program:

[0165] Acquire real-time movie image data corresponding to cardiac magnetic resonance;

[0166] When the real-time movie image data is real-time movie image data of heart rate regularity, extracting a movie image sequence of a single cardiac cycle from the real-time movie image data, the movie image sequence includes N frames of cardiac cycle diagrams of multiple cardiac cycles, where N is a positive integer;

[0167] enhancing the contrast of the N-frame cardiac cycle images, and performing image rigid registration on the N-frame cardiac cycle images to obtain a cardiac cycle registration image corresponding to the N-frame cardiac cycle images;

[0168] Determine the endocardial image boundary corresponding to each frame in the N frames of cardiac cycle registration images;

[0169] Performing myocardial segmentation on the boundary of the endocardium image corresponding to the first frame of the cardiac cycle registration image to obtain multiple initial myocardial segments and recording an initial length L0 of each initial myocardial segment;

[0170] According to the initial myocardial segment, the matching myocardial segments corresponding to each of the initial myocardial segments in the remaining N-1 frames of the cardiac cycle registration map are tracked, and the deformation length L of each of the matching myocardial segments is recorded;

[0171] The myocardial stress strain parameters corresponding to the single cardiac cycle are calculated according to each of the initial lengths L0 and each of the deformation lengths L to perform myocardial stress analysis.

[0172] In some embodiments, extracting a movie image sequence of a single cardiac cycle from the real-time movie image data includes:

[0173] Extract the ECG delay time recorded in the k-space data corresponding to each frame of the real-time movie image;

[0174] Sorting the ECG delay times and generating an index;

[0175] The image frame sequence in the real-time movie image is rearranged according to the index to obtain a movie image sequence of a single cardiac cycle with a time resolution that meets a preset value.

[0176] In some embodiments, determining the endocardial image boundary corresponding to each frame in the N frames of the cardiac cycle registration map includes:

[0177] Outline a first endocardial boundary corresponding to the endocardial region image in the first frame of the cardiac cycle registration image, and discretize the first endocardial boundary into a preset number of first discrete points;

[0178] Determine a corresponding second discrete tracking point in each frame of the remaining N-1 frames of the cardiac cycle registration image according to the first discrete point;

[0179] modifying the second discrete tracking point to obtain a corresponding second discrete point;

[0180] A second endocardial boundary corresponding to the endocardial region image is determined in each of the remaining N-1 frames of the cardiac cycle registration image according to the second discrete points.

[0181] In some embodiments, determining a corresponding second discrete tracking point in each frame of the remaining N-1 frames of the cardiac cycle registration map according to the first discrete point includes:

[0182] Selecting a pixel strip of preset thickness in the radial direction of each of the first discrete points as a pixel matching template;

[0183] Perform feature matching on each of the pixel matching templates and each of the remaining N-1 frames of the cardiac cycle registration images;

[0184] Acquire a second discrete point pixel block corresponding to and matching the pixel matching template in each of the remaining N-1 frames of the cardiac cycle registration image;

[0185] A matching template area is selected in each of the pixel matching templates and a matching area with the maximum matching degree with the matching template area is obtained in the second discrete point pixel block, and points in the matching area are used as the second discrete tracking points.

[0186] In some embodiments, determining the second endocardial boundary corresponding to the endocardial region image in each frame of the remaining N-1 frames of the cardiac cycle registration image according to the second discrete point comprises:

[0187] Determine the initial second endocardial boundary corresponding to the endocardial region image in each frame of the remaining N-1 frames of the cardiac cycle registration image according to the second discrete point; detect whether there is an abnormal curve area in the initial second endocardial boundary, wherein the abnormal curve area is a second endocardial boundary area corresponding to a curvature of the initial second endocardial boundary exceeding a preset curvature value;

[0188] When an abnormal curve area exists in the initial second endocardial boundary, the abnormal curve area is corrected to obtain the corresponding second endocardial boundary.

[0189] In some embodiments, tracking the matching myocardial segments corresponding to each of the initial myocardial segments in the remaining N-1 frames of the cardiac cycle registration map according to the initial myocardial segments includes:

[0190] discretizing the initial myocardial segment into a preset number of first myocardial discrete points;

[0191] According to each of the first myocardial discrete points, sequentially acquire a second myocardial discrete point that matches the first myocardial discrete point in each frame of the remaining N-1 frames of the cardiac cycle registration image;

[0192] The corresponding matching myocardial segment is acquired according to the second myocardial discrete point.

[0193] In some embodiments, sequentially acquiring, based on each of the first myocardial discrete points, a second myocardial discrete point that matches the first myocardial discrete point in each of the remaining N-1 frames of the cardiac cycle registration image, comprises:

[0194] forwardly tracking, according to each of the first myocardial discrete points, a first tracking discrete point matching the first myocardial discrete point in each frame of the remaining N-1 frames of the cardiac cycle registration images;

[0195] According to each of the first myocardial discrete points, reverse tracking of a second tracking discrete point in each frame of the remaining N-1 frames of the cardiac cycle registration image that matches the first myocardial discrete point is performed;

[0196] The average position coordinates of the first tracking discrete point and the second tracking discrete point in the cardiac cycle registration image are calculated to determine the corresponding second myocardial discrete point.

[0197] In some embodiments, forward tracking the first tracking discrete point matching the first myocardial discrete point in each frame of the remaining N-1 frames of the cardiac cycle registration image according to each of the first myocardial discrete points includes:

[0198] Taking each of the first myocardial discrete points as the center, acquiring the corresponding first myocardial discrete area;

[0199] Setting a first myocardial mapping area corresponding to the first myocardial discrete area in the adjacent cardiac cycle registration image, wherein the first myocardial mapping area is an area on the boundary of the endocardial image where the point closest to the corresponding first myocardial discrete point is located, and the area of ​​the first myocardial mapping area is larger than the area of ​​the first myocardial discrete area;

[0200] Perform feature matching on the first myocardial discrete region and the first myocardial mapping region to obtain a second myocardial discrete region and determine a discrete center point of the second myocardial discrete region, wherein the second myocardial discrete region is the region in the first myocardial mapping region with the highest matching degree with the first myocardial discrete region;

[0201] The discrete center point is projected onto the corresponding endocardial boundary to obtain a corresponding discrete projection point, and the discrete projection point is used as the first tracking discrete point.

[0202] In some embodiments, the processor 301 is further configured to implement the following method steps:

[0203] Acquire real-time movie image data corresponding to cardiac magnetic resonance;

[0204] When the real-time movie image data is real-time movie image data corresponding to arrhythmia, extracting a plurality of movie image sequences corresponding to a plurality of cardiac cycles from the real-time movie image data, each of the movie image sequences comprising N frames of cardiac cycle diagrams, where N is a positive integer;

[0205] enhancing the contrast of the N-frame cardiac cycle images, and performing image rigid registration on the N-frame cardiac cycle images to obtain a cardiac cycle registration image corresponding to the N-frame cardiac cycle images;

[0206] Determine the endocardial image boundary corresponding to each frame in the N frames of cardiac cycle registration images;

[0207] Performing myocardial segmentation on the boundary of the endocardium image corresponding to the first frame of the cardiac cycle registration image to obtain multiple initial myocardial segments and recording an initial length L0 of each initial myocardial segment;

[0208] According to the initial myocardial segment, the matching myocardial segments corresponding to each of the initial myocardial segments in the remaining N-1 frames of the cardiac cycle registration map are tracked, and the deformation length L of each of the matching myocardial segments is recorded;

[0209] Myocardial stress strain parameters corresponding to a plurality of cardiac cycles are calculated according to each of the initial lengths L0 and each of the deformation lengths L to perform myocardial stress analysis.

[0210] A storage medium is also provided in an embodiment of the present application. The storage medium stores a computer program. The computer program includes myocardial stress analysis program instructions. The processor executes the myocardial stress analysis program instructions to implement the steps of any one of the myocardial stress analysis methods provided in the above embodiments.

[0211] The storage medium may be an internal storage unit of the myocardial stress analysis device described in any of the above embodiments, such as a memory or internal memory of the myocardial stress analysis device. The storage medium may also be an external storage device of the myocardial stress analysis device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., equipped on the myocardial stress analysis device.

[0212] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A myocardial stress analysis method based on real-time movie imaging, characterized in that: The method comprises: Acquire real-time movie image data corresponding to cardiac magnetic resonance; When the real-time movie image data is real-time movie image data of heart rate regularity, extracting a movie image sequence of a single cardiac cycle from the real-time movie image data, the movie image sequence includes N frames of cardiac cycle diagrams of multiple cardiac cycles, where N is a positive integer; enhancing the contrast of the N-frame cardiac cycle images, and performing image rigid registration on the N-frame cardiac cycle images to obtain a cardiac cycle registration image corresponding to the N-frame cardiac cycle images; Determine the endocardial image boundary corresponding to each frame in the N frames of cardiac cycle registration images; Performing myocardial segmentation on the boundary of the endocardium image corresponding to the first frame of the cardiac cycle registration image to obtain multiple initial myocardial segments and recording an initial length L0 of each initial myocardial segment; According to the initial myocardial segment, the matching myocardial segments corresponding to each of the initial myocardial segments in the remaining N-1 frames of the cardiac cycle registration map are tracked, and the deformation length L of each of the matching myocardial segments is recorded; The myocardial stress strain parameters corresponding to the single cardiac cycle are calculated according to each of the initial lengths L0 and each of the deformation lengths L to perform myocardial stress analysis.

2. The method according to claim 1, characterized in that The step of extracting a movie image sequence of a single cardiac cycle from the real-time movie image data comprises: Extract the ECG delay time recorded in the k-space data corresponding to each frame of the real-time movie image; Sorting the ECG delay times and generating an index; The image frame sequence in the real-time movie image is rearranged according to the index to obtain a movie image sequence of a single cardiac cycle with a time resolution that meets a preset value.

3. The method according to claim 1, characterized in that: Determining the endocardial image boundary corresponding to each frame in the N frames of the cardiac cycle registration map includes: Outline a first endocardial boundary corresponding to the endocardial region image in the first frame of the cardiac cycle registration image, and discretize the first endocardial boundary into a preset number of first discrete points; Determine a corresponding second discrete tracking point in each frame of the remaining N-1 frames of the cardiac cycle registration image according to the first discrete point; modifying the second discrete tracking point to obtain a corresponding second discrete point; A second endocardial boundary corresponding to the endocardial region image is determined in each of the remaining N-1 frames of the cardiac cycle registration image according to the second discrete points.

4. The method according to claim 3, characterized in that The step of determining a corresponding second discrete tracking point in each frame of the remaining N-1 frames of the cardiac cycle registration image according to the first discrete point comprises: Selecting a pixel strip of preset thickness in the radial direction of each of the first discrete points as a pixel matching template; Perform feature matching on each of the pixel matching templates and each of the remaining N-1 frames of the cardiac cycle registration images; Acquire a second discrete point pixel block corresponding to and matching the pixel matching template in each of the remaining N-1 frames of the cardiac cycle registration image; A matching template area is selected in each of the pixel matching templates and a matching area with the maximum matching degree with the matching template area is obtained in the second discrete point pixel block, and points in the matching area are used as the second discrete tracking points.

5. The method according to claim 3, characterized in that: The step of determining the second endocardial boundary corresponding to the endocardial region image in each frame of the remaining N-1 frames of the cardiac cycle registration image according to the second discrete point comprises: Determine the initial second endocardial boundary corresponding to the endocardial region image in each frame of the remaining N-1 frames of the cardiac cycle registration image according to the second discrete point; detect whether there is an abnormal curve area in the initial second endocardial boundary, wherein the abnormal curve area is a second endocardial boundary area corresponding to a curvature of the initial second endocardial boundary exceeding a preset curvature value; When an abnormal curve area exists in the initial second endocardial boundary, the abnormal curve area is corrected to obtain the corresponding second endocardial boundary.

6. The method according to claim 1, characterized in that The step of tracking the matching myocardial segments corresponding to the initial myocardial segments in the remaining N-1 frames of the cardiac cycle registration map according to the initial myocardial segments includes: discretizing the initial myocardial segment into a preset number of first myocardial discrete points; According to each of the first myocardial discrete points, sequentially acquire a second myocardial discrete point that matches the first myocardial discrete point in each frame of the remaining N-1 frames of the cardiac cycle registration image; The corresponding matching myocardial segment is acquired according to the second myocardial discrete point.

7. The method according to claim 6, characterized in that The step of sequentially acquiring, based on each of the first myocardial discrete points, a second myocardial discrete point that matches the first myocardial discrete point in each of the remaining N-1 frames of the cardiac cycle registration image comprises: forwardly tracking, according to each of the first myocardial discrete points, a first tracking discrete point matching the first myocardial discrete point in each frame of the remaining N-1 frames of the cardiac cycle registration images; According to each of the first myocardial discrete points, reverse tracking of a second tracking discrete point in each frame of the remaining N-1 frames of the cardiac cycle registration image that matches the first myocardial discrete point is performed; The average position coordinates of the first tracking discrete point and the second tracking discrete point in the cardiac cycle registration image are calculated to determine the corresponding second myocardial discrete point.

8. The method according to claim 7, characterized in that The step of forward tracking the first tracking discrete point matching the first myocardial discrete point in each frame of the remaining N-1 frames of the cardiac cycle registration image according to each of the first myocardial discrete points comprises: Taking each of the first myocardial discrete points as the center, acquiring the corresponding first myocardial discrete area; Setting a first myocardial mapping area corresponding to the first myocardial discrete area in the adjacent cardiac cycle registration image, wherein the first myocardial mapping area is an area on the boundary of the endocardial image where the point closest to the corresponding first myocardial discrete point is located, and the area of ​​the first myocardial mapping area is larger than the area of ​​the first myocardial discrete area; Perform feature matching on the first myocardial discrete region and the first myocardial mapping region to obtain a second myocardial discrete region and determine a discrete center point of the second myocardial discrete region, wherein the second myocardial discrete region is the region in the first myocardial mapping region with the highest matching degree with the first myocardial discrete region; The discrete center point is projected onto the corresponding endocardial boundary to obtain a corresponding discrete projection point, and the discrete projection point is used as the first tracking discrete point.

9. A method for myocardial stress analysis based on real-time movie imaging, characterized in that: The method comprises: Acquire real-time movie image data corresponding to cardiac magnetic resonance; When the real-time movie image data is real-time movie image data corresponding to arrhythmia, extracting a plurality of movie image sequences corresponding to a plurality of cardiac cycles from the real-time movie image data, each of the movie image sequences comprising N frames of cardiac cycle diagrams, where N is a positive integer; enhancing the contrast of the N-frame cardiac cycle images, and performing image rigid registration on the N-frame cardiac cycle images to obtain a cardiac cycle registration image corresponding to the N-frame cardiac cycle images; Determine the endocardial image boundary corresponding to each frame in the N frames of cardiac cycle registration images; Performing myocardial segmentation on the boundary of the endocardium image corresponding to the first frame of the cardiac cycle registration image to obtain multiple initial myocardial segments and recording an initial length L0 of each initial myocardial segment; According to the initial myocardial segment, the matching myocardial segments corresponding to each of the initial myocardial segments in the remaining N-1 frames of the cardiac cycle registration map are tracked, and the deformation length L of each of the matching myocardial segments is recorded; Myocardial stress strain parameters corresponding to a plurality of cardiac cycles are calculated according to each of the initial lengths L0 and each of the deformation lengths L to perform myocardial stress analysis.

10. A myocardial stress analysis device, characterized in that: The myocardial stress analysis device includes a processor, a memory, and a storage, wherein the memory stores a myocardial stress analysis program executable by the processor, and when the computer program is executed by the processor, the myocardial stress analysis method according to any one of claims 1 to 9 is implemented.

Images