A method, system, device, medium, and product for assessing right ventricular function

Abstract

The application discloses a method, system, device, medium and product for evaluating right ventricular function, and relates to the technical field of medical image processing and ultrasonic diagnosis. The method comprises the following steps: acquiring a sequence of multiple two-dimensional ultrasonic images of a right ventricle of a heart and spatial pose information of the sequence, and generating point cloud data; identifying an end-diastolic frame, an end-systolic frame and an anatomical structure point, and acquiring sparse point cloud data of the anatomical structure point; constructing a three-dimensional grid model of the end-diastolic frame and the end-systolic frame; determining multiple tracking points of a tricuspid annulus and acquiring a three-dimensional coordinate trajectory; calculating longitudinal displacement, area change rate and peak time dispersion based on the trajectory, and completing function evaluation. The application overcomes the limitations of angle dependence and point substitution of traditional TAPSE technology, realizes comprehensive evaluation of right ventricular function from three dimensions of longitudinal displacement, area change rate and systolic synchrony, and improves the accuracy and clinical value of evaluation.

Description

Technical Field

[0001] This application relates to the fields of medical image processing and ultrasound diagnostic technology, and in particular to a method, system, device, medium and product for assessing right ventricular function. Background Technology

[0002] Accurate assessment of right ventricular function is of significant clinical value in the diagnosis, treatment, and prognosis of cardiovascular diseases. Due to the complex morphology of the right ventricle, which crescent-shaped and encloses the left ventricle, it is difficult to describe using simple geometric models; therefore, its functional assessment has always been a challenge in the field of echocardiography.

[0003] Currently, the most commonly used indicator for assessing right ventricular systolic function in clinical practice is the tricuspid annular systolic displacement (TAPSE). This method uses M-mode ultrasound to measure the displacement of the tricuspid valve lateral annulus along its long axis during systole, indirectly reflecting the longitudinal systolic function of the right ventricle. Due to its ease of operation and high repeatability, TAPSE has become a convenient indicator for assessing right ventricular function in routine examinations. However, TAPSE has the following technical limitations: (1) Angle dependence: the sampling line must be parallel to the direction of ventricular wall motion, and angle deviation will lead to an underestimation of the measurement value; (2) Operator dependence: the measurement results are affected by the operator's positioning experience; (3) Inherent defects of using a point to represent the surface: only measuring the displacement of a single point cannot reflect the overall movement of the tricuspid annulus, and it is easy to misjudge when there is local abnormal ventricular wall motion. In recent years, tissue Doppler or speckle tracking techniques have mostly been limited to two-dimensional planar analysis, which is difficult to fully capture the complex deformation of the tricuspid annulus in three-dimensional space.

[0004] Existing technologies include several methods for processing two-dimensional ultrasound images and creating three-dimensional models. For example, Chinese patent CN107527316B discloses "a method for constructing point cloud data from arbitrary points in a two-dimensional ultrasound image sequence," which obtains spatial information of points through probe pose binding; Chinese patent CN111166388B discloses "a method for constructing a three-dimensional model based on two-dimensional ultrasound image cognition," which constructs a model by identifying anatomical structure points and combining them with a knowledge base. However, these technical solutions mainly focus on point cloud generation and three-dimensional modeling, and have not been further applied to the functional assessment of specific organs. They lack multi-point tracking of the tricuspid valve annulus and the calculation of multi-dimensional functional indicators, thus failing to solve the aforementioned problems of traditional TAPSE technology. Summary of the Invention

[0005] The purpose of this application is to provide a method, system, device, medium, and product for assessing right ventricular function. By fusing two-dimensional ultrasound image sequences with spatial pose information, three-dimensional point cloud data and a three-dimensional mesh model can be constructed to achieve multi-point tracking of the tricuspid annulus and calculate assessment indicators in three dimensions: longitudinal displacement, annular area change, and contraction synchronicity.

[0006] To achieve the above objectives, this application provides the following solution:

[0007] Firstly, this application provides a method for assessing right ventricular function, including:

[0008] Acquire a multi-frame two-dimensional ultrasound image sequence of the right ventricle of the heart and its corresponding spatial pose information, and generate point cloud data containing spatial coordinates and pose information;

[0009] Identify end-diastolic frames, end-systolic frames, and anatomical structure points in the image sequence, and obtain sparse point cloud data of anatomical structure points on the end-diastolic and end-systolic frames;

[0010] Based on the sparse point cloud data and the cardiac knowledge base, a three-dimensional mesh model of the end-diastolic frame and the end-systolic frame is constructed.

[0011] Based on the sparse point cloud data of the tricuspid annulus, multiple tracking points of the tricuspid annulus are determined on the three-dimensional mesh model, and the three-dimensional coordinate trajectory of each tracking point in each frame of the cardiac cycle is obtained.

[0012] Based on the three-dimensional coordinate trajectory, multiple indices are calculated to assess tricuspid valve annular changes. These indices include a displacement index for assessing right ventricular longitudinal contraction, an area index for assessing right ventricular annular area changes, and a time dispersion index for assessing right ventricular contraction synchronicity.

[0013] Secondly, this application provides a system for assessing right ventricular function, comprising:

[0014] The data acquisition module is used to acquire a multi-frame two-dimensional ultrasound image sequence of the right ventricle of the heart and its corresponding spatial pose information, and generate point cloud data containing spatial coordinates and pose information;

[0015] A point cloud construction module is used to identify end-diastolic frames, end-systolic frames, and anatomical structure points in the image sequence, and to acquire sparse point cloud data of anatomical structure points on the end-diastolic frames and end-systolic frames.

[0016] A three-dimensional mesh model construction module is used to construct three-dimensional mesh models of the end-diastolic frames and end-systolic frames based on the sparse point cloud data and the cardiac knowledge base.

[0017] The trajectory acquisition module is used to determine multiple tracking points of the tricuspid annulus on the three-dimensional mesh model based on the sparse point cloud data of the tricuspid annulus, and to acquire the three-dimensional coordinate trajectory of each tracking point in each frame of the cardiac cycle.

[0018] The functional evaluation module is used to calculate multiple indicators for evaluating tricuspid valve annular changes based on the three-dimensional coordinate trajectory. The indicators include a displacement indicator for evaluating right ventricular longitudinal contraction, an area indicator for evaluating right ventricular annular area changes, and a time dispersion indicator for evaluating right ventricular contraction synchronicity.

[0019] The data acquisition module, point cloud construction module, 3D mesh model construction module, trajectory acquisition module, and functional evaluation module are connected in sequence.

[0020] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0021] This application provides a method, system, device, medium, and product for assessing right ventricular function. It acquires a multi-frame two-dimensional ultrasound image sequence of the right ventricle and its corresponding spatial pose information, generating point cloud data containing spatial coordinates and pose information. It identifies end-diastolic frames, end-systolic frames, and anatomical structure points in the image sequence, obtaining sparse point cloud data of anatomical structure points on the end-diastolic and end-systolic frames. This solves the problem of existing two-dimensional ultrasound's inability to accurately reconstruct the right ventricular morphology, achieving three-dimensional spatial localization and key anatomical structure point extraction based on conventional two-dimensional ultrasound. Based on the sparse point cloud data and a pre-set cardiac knowledge base, a three-dimensional mesh model of the end-diastolic and end-systolic frames is constructed. Based on the sparse point cloud data of the tricuspid valve annulus on the three-dimensional mesh model, the anatomical structure is determined... This method utilizes multiple tracking points on the tricuspid annulus and acquires the three-dimensional coordinate trajectories of each tracking point in each frame of the cardiac cycle. This solves the problems of traditional TAPSE technology, which relies on single-point measurements and is highly angle-dependent, enabling three-dimensional dynamic tracking of the overall movement of the tricuspid annulus. Furthermore, it calculates multiple indicators for assessing tricuspid annular changes based on the three-dimensional coordinate trajectories. These indicators include displacement indicators for assessing right ventricular longitudinal contraction, area indicators for assessing right ventricular annular area changes, and time dispersion indicators for assessing right ventricular contractile synchronicity. This addresses the limitation of existing technologies in comprehensively assessing overall right ventricular function, achieving a comprehensive assessment of right ventricular function from three dimensions: longitudinal contractile capacity, radial contractile force, and contractile synchronicity. This significantly improves the accuracy and clinical value of the assessment. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is an application environment diagram of a method for assessing right ventricular function according to an embodiment of this application;

[0024] Figure 2 This is a flowchart illustrating a method for assessing right ventricular function according to an embodiment of this application;

[0025] Figure 3 This is a schematic diagram of the structure of a dynamic three-dimensional mesh model of the right ventricle in the end-diastolic frame and the end-systolic frame, representing a method for evaluating right ventricular function according to an embodiment of this application.

[0026] Figure 4 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0027] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0028] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0029] This application provides a method for assessing right ventricular function, which can be applied to, for example... Figure 1In the application environment shown, the ultrasound acquisition device 102 communicates with the server 104 via a network. The data storage system can store multiple frames of two-dimensional ultrasound image sequences and spatial pose information that the server 104 needs to process. The data storage system can be set up separately, integrated into the server 104, or placed in the cloud or on another server. The ultrasound acquisition device 102 includes a two-dimensional ultrasound probe, a pose sensor, and a connected control host. The two-dimensional ultrasound probe is used to acquire multiple frames of two-dimensional ultrasound image sequences of the right ventricle of the heart, and the pose sensor is used to synchronously record the probe's spatial pose information corresponding to each frame of image. The ultrasound acquisition device 102 can send the acquired two-dimensional image sequences and their corresponding spatial pose information to the server 104. After receiving the two-dimensional image sequence and spatial pose information, server 104 executes the method provided in this application embodiment: based on the image sequence and spatial pose information, it generates point cloud data containing spatial coordinates and pose information; it identifies end-diastolic frames, end-systolic frames, and anatomical structure points in the image sequence, and acquires sparse point cloud data of anatomical structure points on the end-diastolic and end-systolic frames; based on the sparse point cloud data and a cardiac knowledge base, it constructs a three-dimensional mesh model of the end-diastolic and end-systolic frames; based on the sparse point cloud data of the tricuspid annulus on the three-dimensional mesh model, it determines multiple tracking points of the tricuspid annulus and acquires the three-dimensional coordinate trajectory of each tracking point in each frame of the cardiac cycle; based on the three-dimensional coordinate trajectory, it calculates multiple indicators for evaluating tricuspid annular changes, including a displacement indicator reflecting right ventricular longitudinal contraction, an area indicator reflecting right ventricular annular area changes, and a time dispersion indicator reflecting right ventricular contraction synchronicity, thereby generating an evaluation result. Server 104 can feed back the evaluation result to ultrasound acquisition device 102 or doctor's workstation. The doctor's workstation can be, but is not limited to, various desktop computers, laptops, tablets, etc., used to display assessment results, including a visualization of the dynamic three-dimensional mesh model of the right ventricle, displacement-time curves of each tracking point, dynamic change curves of the tricuspid annulus area over time, and various calculated functional indicators.

[0030] Furthermore, in some embodiments, the method can also be implemented by either server 104 or ultrasonic acquisition device 102. For example, ultrasonic acquisition device 102 can directly process and evaluate the acquired two-dimensional image sequence and spatial pose information, or server 104 can obtain the two-dimensional image sequence and spatial pose information from the data storage system for processing. Server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers, or it can be a cloud server.

[0031] In one exemplary embodiment, such as Figure 2As shown, a method for assessing right ventricular function is provided. This method is executed by a computer device, specifically a terminal or server, or both. In this embodiment, the method is described using server 104 as an example, and includes the following steps 1 to 5. Wherein:

[0032] Step 1: Acquire a multi-frame two-dimensional ultrasound image sequence of the right ventricle of the heart and its corresponding spatial pose information, and generate point cloud data containing spatial coordinates and pose information;

[0033] Step 2: Identify the end-diastolic frames, end-systolic frames, and anatomical structure points in the image sequence, and obtain sparse point cloud data of the anatomical structure points on the end-diastolic frames and end-systolic frames;

[0034] Step 3: Construct a three-dimensional mesh model of the end-diastolic frame and end-systolic frame based on the sparse point cloud data and the cardiac knowledge base;

[0035] Step 4: Based on the sparse point cloud data of the tricuspid annulus, determine multiple tracking points of the tricuspid annulus on the three-dimensional mesh model, and obtain the three-dimensional coordinate trajectory of each tracking point in each frame of the cardiac cycle.

[0036] Step 5: Based on the three-dimensional coordinate trajectory, calculate multiple indicators for evaluating tricuspid valve annular changes, including displacement indicators for evaluating right ventricular longitudinal contraction, area indicators for evaluating right ventricular annular area changes, and time dispersion indicators for evaluating right ventricular contraction synchronicity.

[0037] Steps 1 to 5 of this application involve acquiring a multi-frame two-dimensional ultrasound image sequence of the right ventricle and its corresponding spatial pose information, generating point cloud data containing spatial coordinates and pose information, and identifying end-diastolic frames, end-systolic frames, and anatomical structure points in the image sequence. This process obtains sparse point cloud data of anatomical structure points on the end-diastolic and end-systolic frames, solving the problem of existing two-dimensional ultrasound methods that struggle to accurately reconstruct the right ventricular morphology. This achieves three-dimensional spatial localization and extraction of key anatomical structure points based on conventional two-dimensional ultrasound. Furthermore, a three-dimensional mesh model of the end-diastolic and end-systolic frames is constructed based on the point cloud data and a pre-set cardiac knowledge base. Multiple tracking points of the tricuspid annulus are then determined on this three-dimensional mesh model based on the sparse point cloud data of the tricuspid annulus. This invention acquires the three-dimensional coordinate trajectories of each tracking point in each frame of the cardiac cycle, solving the problems of traditional TAPSE technology which relies only on single-point measurements and has strong angle dependence. It achieves three-dimensional dynamic tracking of the overall movement of the tricuspid annulus. Based on the three-dimensional coordinate trajectory, multiple indicators for evaluating tricuspid annular changes are calculated. These indicators include displacement indicators for assessing right ventricular longitudinal contraction, area indicators for assessing right ventricular annular area changes, and time dispersion indicators for assessing right ventricular contraction synchronicity. This addresses the problem that existing technologies cannot comprehensively assess overall right ventricular function, achieving a comprehensive assessment of right ventricular function from three dimensions: longitudinal contractile capacity, radial contractile force, and contraction synchronicity, significantly improving the accuracy and clinical value of the assessment. Furthermore, this application can generate a comprehensive assessment report based on the calculated displacement, area, and time dispersion indicators. The comprehensive assessment report includes a visualization of the right ventricular dynamic three-dimensional mesh model, displacement-time curves for each tracking point, and dynamic change curves of the tricuspid annular area over time, facilitating intuitive diagnosis and efficacy evaluation by clinicians.

[0038] In an exemplary embodiment, step 1 acquires a multi-frame two-dimensional ultrasound image sequence of the right ventricle of the heart and its corresponding spatial pose information, generating point cloud data containing spatial coordinates and pose information; specifically including steps S1.1 to S1.3, wherein:

[0039] S1.1. Use a two-dimensional ultrasound probe to acquire image sequences of multiple sections of the right ventricle of the heart to obtain a multi-frame two-dimensional ultrasound image sequence;

[0040] S1.2. Using a pose sensor fixed on a two-dimensional ultrasound probe and a reference sensor fixed on the patient's chest, the probe pose information corresponding to each frame of the two-dimensional image in the image sequence is recorded synchronously. The original measurement data of the sensor is converted into three-dimensional pose information in a spatial rectangular coordinate system through a coordinate transformation formula.

[0041] S1.3. The two-dimensional image information and the three-dimensional pose information are time-stamp aligned and fused to generate point cloud data containing spatial coordinates and pose information, and the timestamp information corresponding to each frame of the image is associated with it.

[0042] The two-dimensional ultrasound probe used here can be a conventional two-dimensional ultrasound probe. The point cloud data itself contains spatial coordinates and the corresponding pose information. Each pixel in each frame of the image is associated with the pose information at the time of acquisition, providing basic data for subsequent steps such as labeling anatomical structures and constructing a three-dimensional mesh model. The aforementioned synchronous recording is achieved by aligning the timestamps of the image frames with the pose data through a data cable connection between the pose sensor and the host computer.

[0043] The probe pose information adopts the probe's six degrees of freedom in space, which includes the probe's three-dimensional coordinates (x, y, z) in a spatial rectangular coordinate system and its rotation angles around each coordinate axis: pitch angle (rotation around the X-axis, i.e., up and down swing), yaw angle (rotation around the Y-axis, i.e., left and right swing), and roll angle (rotation around the Z-axis, i.e., rotation around the probe's own axis).

[0044] In an exemplary embodiment, step 2: identify end-diastolic frames, end-systolic frames, and anatomical structure points in the image sequence, and obtain sparse point cloud data of the anatomical structure points on the end-diastolic frames and end-systolic frames;

[0045] Specifically, this includes steps S2.1 to S2.2, wherein:

[0046] S2.1 Identify end-diastolic frames, end-systolic frames, and anatomical structure points in the image sequence;

[0047] S2.2 On the two-dimensional image information of the end-diastolic frame and the end-systolic frame, mark the anatomical structure points for subsequent three-dimensional reconstruction and dynamic tracking; combine the pose information of the point cloud data, and use the coordinate transformation formula to map the coordinates of the anatomical structure points in the two-dimensional image to three-dimensional spatial coordinates to generate sparse point cloud data of the anatomical structure points.

[0048] It should be noted that the aforementioned anatomical structures include at least the apex of the heart, the tricuspid annulus, and the ventricular septum. The marking of these anatomical structures can be automatically completed using artificial intelligence algorithms. The point cloud data in step S1.3 of this application refers to the full point cloud generated based on all image frames and pixels; the sparse point cloud data here refers to the point cloud extracted from the full point cloud, containing only the anatomical structures at key temporal phases.

[0049] In an exemplary embodiment, step 3: Based on the sparse point cloud data and the cardiac knowledge base, construct a three-dimensional mesh model of the end-diastolic frame and the end-systolic frame; specifically including steps S3.1 to S3.2, wherein:

[0050] S3.1 Using the sparse point cloud data and the cardiac knowledge base, generate an initial three-dimensional mesh model of the end-diastolic frame and the end-systolic frame;

[0051] S3.2. Using a non-rigid registration algorithm, the initial three-dimensional mesh model is fitted to each frame between the end-diastolic frame and the end-systolic frame to generate a dynamic three-dimensional mesh model of the right ventricle with the same topology.

[0052] Figure 3 The dynamic three-dimensional mesh model of the right ventricle is shown in the end-diastolic and end-systolic frames. In the figure, ring A represents the tricuspid annulus at end-diastole, and ring B represents the tricuspid annulus at end-systole.

[0053] It should be noted that the three-dimensional mesh models of the end-diastolic frames and end-systolic frames generated by the above method have the same number of vertices and connection relationships (i.e., the same topology), which provides a basis for inter-frame locking of the tricuspid loop tracking points in subsequent steps.

[0054] The aforementioned cardiac knowledge base includes cardiac structural models and prior knowledge of anatomical structures.

[0055] The aforementioned non-rigid registration algorithm is a registration technique that allows the template to undergo local deformation to match the target shape. In this application, a non-rigid registration algorithm is used to fit the initial 3D mesh model to each frame between the end-diastolic and end-systolic frames. This process allows the 3D mesh model to undergo local deformation based on the anatomical points of the current frame to adapt to the morphological changes of the heart at different time phases, while maintaining the topological structure of the 3D mesh model unchanged. This algorithm ensures that the 3D mesh models of the end-diastolic and end-systolic frames have the same topological structure.

[0056] In a preferred embodiment, the same method can be used to fit the three-dimensional mesh model frame by frame onto the anatomical structure of each frame of the cardiac cycle, forming a set of dynamic three-dimensional mesh models of the right ventricle with the same topology. This set completely records the motion trajectory of the right ventricular wall in three-dimensional space and can be used for more comprehensive dynamic analysis.

[0057] Furthermore, the surface fitting algorithm may employ, but is not limited to, thin plate spline interpolation, radial basis functions, etc.; the non-rigid registration algorithm may employ, but is not limited to, free deformation models based on B-splines, Demons algorithm, etc.

[0058] In an exemplary embodiment, step 4: Based on the sparse point cloud data of the tricuspid annulus, multiple tracking points of the tricuspid annulus are determined on the three-dimensional mesh model, and the three-dimensional coordinate trajectory of each tracking point in each frame of the cardiac cycle is obtained; specifically including steps S4.1 to S4.2, wherein:

[0059] S4.1 Based on the generated right ventricular dynamic three-dimensional mesh model, taking the three-dimensional mesh model corresponding to the end-diastolic frame as the benchmark, and according to the sparse point cloud data of the tricuspid annulus, extract the vertex ring of the region where the tricuspid annulus is located, and uniformly select K points along the circumferential direction from the vertex ring as the tracking point set, where K≥3;

[0060] S4.2 Obtain the three-dimensional spatial coordinates of each tracking point in the tracking point set in all frames within the cardiac cycle, forming the three-dimensional coordinate trajectory of each tracking point.

[0061] It should be noted that the selection of the 3D mesh model corresponding to the end-diastolic frame as the reference frame in this step is based on a comprehensive consideration of cardiac physiological characteristics and technical repeatability. The end-diastolic frame is the moment when ventricular filling ends and contraction is about to begin. At this time, the ventricles are maximally filled with blood, the ventricular walls are in a relatively relaxed state with low tension, and the tricuspid annulus is at its most regular and easily identifiable. Furthermore, the end-diastolic frame is a widely used reference starting point in cardiac function assessment; commonly used clinical indicators such as ejection fraction are measured based on this phase.

[0062] The method of uniformly selecting K points along the circumference of the petal ring can be either uniform selection based on angle or uniform selection based on arc length.

[0063] Furthermore, since all the generated 3D mesh models have the same topology (the same number of vertices and connection relationships), the same vertex has the same index number in different frames. Based on this characteristic, after determining the tracking point in S4.1, the point corresponding to the tracking point can be directly locked in the 3D mesh model of each subsequent frame through the vertex index, without the need for repeated marking, which significantly improves the efficiency and accuracy of dynamic tracking.

[0064] The three-dimensional coordinate trajectory obtained by the above method fully records the dynamic deformation process of each point on the tricuspid annulus during the cardiac cycle, providing a data basis for calculating longitudinal displacement, area change rate and contraction synchronicity index in subsequent steps.

[0065] In an exemplary embodiment, step 5: Based on the three-dimensional coordinate trajectory, calculate multiple indices for evaluating tricuspid valve annular changes, including a displacement index for evaluating right ventricular longitudinal contraction, an area index for evaluating right ventricular annular area changes, and a time dispersion index for evaluating right ventricular contraction synchronicity; specifically including steps S5.1 to S5.3, wherein:

[0066] S5.1, Longitudinal Functional Assessment of the Right Ventricle

[0067] A local coordinate system is established with the apex of the heart as the origin and the normal vector of the valve annulus plane as the Z-axis. The longitudinal displacement (i.e., displacement index) of each tracking point from the end-diastolic frame to the end-systolic frame in the Z-axis direction is calculated to reflect the overall longitudinal contractile capacity of the right ventricle.

[0068] S5.2 Assessment of right ventricular basal systolic function

[0069] In each frame, the projected area of ​​the tricuspid annulus is calculated using the projected coordinates of K tracking points on the annulus plane; the projected area of ​​the end-diastolic frame is extracted as the maximum area. Extract the projected area of ​​the final contraction frame as the minimum area. Calculate the rate of change of area from the end of diastole to the end of contraction. (i.e., area index), used to reflect the radial contractile force at the base of the right ventricle.

[0070] S5.3 Assessment of right ventricular systolic synchrony

[0071] Based on the three-dimensional coordinate trajectory, a longitudinal displacement-time curve is generated for each tracking point. The peak time of each tracking point reaching its maximum displacement is recorded. The peak time dispersion of all tracking points is calculated. The peak time dispersion is the standard deviation or maximum time difference of the peak time (i.e., the time dispersion index), which is used to evaluate the synchronicity of contraction of each segment of the right ventricle.

[0072] It should be noted that:

[0073] The establishment of the aforementioned local coordinate system: In this step, a local coordinate system is established with the apex of the heart as the origin and the normal vector of the valve annulus plane as the Z-axis. The purpose is to eliminate the coordinate differences caused by different patient positions and probe scanning angles, so that the calculation of longitudinal displacement has a unified reference benchmark, ensuring the comparability and repeatability of the evaluation results.

[0074] The above calculation of the area change rate (FAC) reflects the radial contractile force at the base of the right ventricle. In the end-diastolic frame, the tricuspid annulus dilates to its maximum area. In the final contraction frame, the valve ring shrinks to its minimum area. The area change rate (FAC) is an important supplementary indicator for assessing right ventricular systolic function, especially suitable for patients with regional wall motion abnormalities.

[0075] The aforementioned peak time dispersion reflects the synchronicity of contraction across different segments of the right ventricle. In a healthy heart, myocardial segments reach their maximum contractile displacement almost simultaneously, resulting in a small standard deviation of peak time. However, in cases of right bundle branch block, cardiomyopathy, or post-cardiac resynchronization therapy, segmental contraction becomes asynchronous, significantly increasing the peak time dispersion. Therefore, this indicator has significant clinical value in assessing myocardial electromechanical synchronicity.

[0076] The aforementioned longitudinal displacement, area change rate (FAC), and peak time dispersion assess right ventricular function from three dimensions: longitudinal contractile capacity, radial contractile force, and contractile synchronicity. These three factors complement each other, overcoming the limitations of the traditional TAPSE which relies solely on single-point longitudinal displacement, and providing a more comprehensive right ventricular function assessment system for clinical practice.

[0077] In another exemplary embodiment of this application, after step 5 above, the method may further include the following steps:

[0078] Step 6: Integrated Visualization and Report Generation

[0079] Based on the calculated longitudinal displacement, area change rate, and peak time dispersion (i.e., synchronicity index), combined with the visualization results of the right ventricular dynamic three-dimensional grid model, a comprehensive evaluation report is generated. The comprehensive evaluation report includes at least: a visualization of the right ventricular dynamic three-dimensional grid model, longitudinal displacement-time curves for each tracking point, dynamic change curves of the tricuspid annulus area over time, and a comparative analysis of the calculated longitudinal displacement, area change rate, and synchronicity index with the traditional tricuspid annulus systolic displacement (TAPSE) measurement results.

[0080] It should be noted that step 6 combines the quantitative calculation results with the visualization of the dynamic three-dimensional mesh model to generate an intuitive comprehensive assessment report, enabling clinicians to simultaneously obtain quantitative indicators of right ventricular function and intuitive morphological change information.

[0081] The comprehensive evaluation report includes a comparative analysis with traditional TAPSE, helping physicians quickly understand the advantages of this method compared to existing technologies. It overcomes the limitations of TAPSE's point-to-surface approach, achieving a multi-directional assessment of right ventricular longitudinal function, basal radial contractile force, and contractile synchronicity. The results of traditional TAPSE measurements can be automatically calculated by the system based on the same image sequence or manually entered by the operator. This report can be directly used for clinical diagnosis, surgical planning, and efficacy evaluation, enhancing the clinical application value of this method.

[0082] Furthermore, the synchronicity index described in this application refers to an indicator used to quantify the contractile synchronicity of different segments of the right ventricle, which is composed of the standard deviation of the peak time or the maximum time difference calculated in step 5. This index, together with longitudinal displacement and area change rate, constitutes a three-dimensional comprehensive assessment system for right ventricular function.

[0083] To further clarify the differences between the technical concept of this application and the prior art, especially the differences between patent document CN107527316B (hereinafter referred to as prior art 1) and patent document CN111166388B (hereinafter referred to as prior art 2), the following detailed description is provided:

[0084] The core contribution of prior art 1 (CN107527316B) lies in providing a technique for mapping arbitrary points on two-dimensional ultrasound images to three-dimensional space. This technique records the pose information of each frame of the image using a probe pose sensor, thereby generating point cloud data containing spatial coordinates. However, the technical solution of prior art 1 is general and not optimized for any specific organ or clinical task. Those skilled in the art, after reading prior art 1, can only learn how to generate point clouds, but cannot directly learn how to use the point clouds to assess right ventricular function.

[0085] This application borrows the point cloud generation method of prior art 1 in step 1, but further introduces a series of technical means for right ventricular function assessment in steps 2 to 5: First, in step 2, end-diastolic and end-systolic frames are identified, and sparse point cloud data of anatomical structure points are obtained; second, in step 3, a three-dimensional mesh model of the right ventricle with the same topology is constructed using a preset cardiac knowledge base through surface fitting and non-rigid registration algorithms; third, in step 4, multiple tracking points are uniformly selected based on the tricuspid annulus of the three-dimensional mesh model, and its three-dimensional coordinate trajectory is obtained; finally, in step 5, functional indicators in three dimensions—longitudinal displacement, area change rate, and peak time dispersion—are calculated based on the trajectory. The technical content of steps 2 to 5 is neither disclosed nor implied in prior art 1. In particular, the technical details in step 5, such as establishing a local coordinate system with the apex as the origin, calculating the area change rate, and calculating the peak time dispersion, are specifically designed by this application to overcome the defects of TAPSE technology, and its technical concept and implementation method exceed the scope of disclosure of prior art 1.

[0086] The core contribution of prior art 2 (CN111166388B) lies in providing a method for constructing a three-dimensional model based on the recognition of anatomical structural points. This method identifies key anatomical structural points (such as the apex of the heart and the tricuspid valve) on two-dimensional images and combines this with a knowledge base for surface fitting and non-rigid registration, thereby constructing an accurate three-dimensional model. However, the technical solution of prior art 2 stops at the construction of the three-dimensional model; its ultimate goal is to obtain a more accurate model, rather than using the model for functional assessment. Those skilled in the art, after reading prior art 2, can only learn how to construct a three-dimensional model, but cannot directly learn how to use the constructed model to assess organ function.

[0087] This application borrows the modeling concept of prior art 2 in step 3, but in steps 4 and 5, it further utilizes the three-dimensional model as a tool for functional assessment, proposing an innovative scheme for multi-point tracking and multi-dimensional index calculation. Specifically: In step 4, based on a three-dimensional mesh model with the same topological structure, multiple tracking points are uniformly selected along the circumference of the tricuspid valve annulus, and the inter-frame automatic locking of tracking points is achieved by utilizing the model topological consistency. This technique fully utilizes the model characteristics (same topological structure) created by the modeling method of prior art 2, but prior art 2 itself does not disclose or imply how to use this characteristic for dynamic tracking; In step 5, based on the three-dimensional coordinate trajectory of the tracking points, three functional indices—longitudinal displacement, area change rate, and peak time dispersion—are calculated. The calculation methods for these indices (such as establishing a local coordinate system, calculating the projected area, and statistically analyzing the peak time) are not involved in prior art 2. In particular, the peak time dispersion index is specifically designed in this application for assessing right ventricular systolic synchronicity, and its clinical value and technical implementation exceed the scope of inspiration from prior art 2.

[0088] In summary, this application is not a simple superposition or combination of prior art 1 and prior art 2, but rather a creative improvement and expansion based on absorbing the technical ideas of both, specifically addressing the clinical need for right ventricular function assessment. Specifically, this application applies the point cloud generation technology of existing technology 1 and the anatomical modeling technology of existing technology 2 to right ventricular function assessment for the first time, opening up new directions for technological applications. Based on the three-dimensional model constructed by existing technology 2, a method is proposed to uniformly select multiple tracking points along the circumference of the tricuspid annulus, and to automatically lock the tracking points between frames using the topological consistency of the model, thereby achieving three-dimensional dynamic tracking of the overall movement of the tricuspid annulus and overcoming the limitations of single-point measurement in traditional TAPSE technology. At the same time, based on the three-dimensional coordinate trajectory of the tracking points, three functional indicators—longitudinal displacement, area change rate, and peak time dispersion—are creatively proposed, corresponding to the longitudinal contractile capacity, radial contractile force of the basal part, and contractile synchronicity of the right ventricle, respectively, forming a comprehensive assessment system for right ventricular function. The calculation methods and combined use of these indicators are not disclosed or implied in existing technologies. Through the above technical means, this application achieves a leap from single-point displacement measurement to three-dimensional annular overall functional assessment, effectively solving the technical defects of traditional TAPSE technology that rely on angles and substitute points for surfaces, and significantly improving the accuracy and clinical value of right ventricular function assessment.

[0089] This application also provides an application scenario that utilizes the aforementioned method, system, device, medium, and product for assessing right ventricular function. In clinical ultrasound examinations, physicians use a conventional two-dimensional ultrasound probe to acquire multi-sectional dynamic images of the patient's right ventricle. The system simultaneously records the probe's pose information and automatically performs three-dimensional reconstruction and functional assessment, generating a dynamic model of the right ventricle and quantitative indicators such as longitudinal displacement, area change rate, and synchronicity index. It also outputs a comprehensive assessment report including a comparison with traditional tricuspid annular systolic displacement (TAPSE). This scenario can be widely used in the daily diagnosis and treatment, surgical planning, and postoperative efficacy evaluation of departments such as cardiology, ultrasound, and cardiac surgery, providing clinicians with an accurate and repeatable quantitative analysis tool for right ventricular function based on the methods, systems, devices, media, and products of this application.

[0090] Based on the same inventive concept, this application also provides a system for assessing right ventricular function to implement the method for assessing right ventricular function described above. The solution provided by this system is similar to the implementation described in the above method; therefore, the specific limitations of one or more system embodiments for assessing right ventricular function provided below can be found in the limitations of the method for assessing right ventricular function described above, and will not be repeated here.

[0091] In one exemplary embodiment, a system for assessing right ventricular function is provided, comprising:

[0092] The data acquisition module is used to acquire a multi-frame two-dimensional ultrasound image sequence of the right ventricle of the heart and its corresponding spatial pose information, and generate point cloud data containing spatial coordinates and pose information;

[0093] A point cloud construction module is used to identify end-diastolic frames, end-systolic frames, and anatomical structure points in the image sequence, and to acquire sparse point cloud data of anatomical structure points on the end-diastolic frames and end-systolic frames.

[0094] A three-dimensional mesh model construction module is used to construct three-dimensional mesh models of the end-diastolic frames and end-systolic frames based on the sparse point cloud data and the cardiac knowledge base.

[0095] The trajectory acquisition module is used to determine multiple tracking points of the tricuspid annulus on the three-dimensional mesh model based on the sparse point cloud data of the tricuspid annulus, and to acquire the three-dimensional coordinate trajectory of each tracking point in each frame of the cardiac cycle.

[0096] The functional evaluation module is used to calculate multiple indicators for evaluating tricuspid valve annular changes based on the three-dimensional coordinate trajectory. The indicators include a displacement indicator for evaluating right ventricular longitudinal contraction, an area indicator for evaluating right ventricular annular area changes, and a time dispersion indicator for evaluating right ventricular contraction synchronicity.

[0097] The data acquisition module, point cloud construction module, 3D mesh model construction module, trajectory acquisition module, and functional evaluation module are connected in sequence.

[0098] The above modules can be implemented by software, hardware, or a combination of both.

[0099] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 4 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and databases. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When executed by the processor, the computer program implements a method for evaluating right ventricular function.

[0100] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0101] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0102] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0103] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0104] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0105] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0106] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0107] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0108] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only intended to help understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for assessing right ventricular function, characterized in that, Includes the following steps: Acquire a multi-frame two-dimensional ultrasound image sequence of the right ventricle of the heart and its corresponding spatial pose information, and generate point cloud data containing spatial coordinates and pose information; Identify end-diastolic frames, end-systolic frames, and anatomical structure points in the image sequence, and obtain sparse point cloud data of anatomical structure points on the end-diastolic and end-systolic frames; Based on the sparse point cloud data and the cardiac knowledge base, a three-dimensional mesh model of the end-diastolic frame and the end-systolic frame is constructed. Based on the generated right ventricular dynamic three-dimensional mesh model, and taking the three-dimensional mesh model corresponding to the end-diastolic frame as the benchmark, the vertex ring of the region where the tricuspid annulus is located is extracted according to the sparse point cloud data of the tricuspid annulus. K points are uniformly selected from the vertex ring along the circumferential direction as the tracking point set, K≥3; and the three-dimensional mesh model of all frames in the cardiac cycle has the same topological structure. The points corresponding to the tracking points in different frames are locked by vertex index to obtain the three-dimensional coordinate trajectory of each tracking point in each frame in the cardiac cycle. Based on the three-dimensional coordinate trajectory, multiple indices are calculated to assess tricuspid valve annular changes. These indices include a displacement index for assessing right ventricular longitudinal contraction, an area index for assessing right ventricular annular area changes, and a time dispersion index for assessing right ventricular contraction synchronicity. Calculating the displacement index involves establishing a local coordinate system with the apex as the origin and the annular plane normal vector as the Z-axis, and calculating the longitudinal displacement of each tracking point along the Z-axis during systole. Calculating the area index involves using the projected coordinates of multiple tracking points on the annular plane to calculate the projected area of ​​the tricuspid valve annulus at end-diastole and end-systole, obtaining the area change rate. Calculating the time dispersion index involves generating a displacement-time curve for each tracking point, recording the peak time for each tracking point, and calculating its standard deviation or maximum time difference.

2. The method for assessing right ventricular function according to claim 1, characterized in that, The cardiac knowledge base includes cardiac structural models and prior knowledge of anatomical structures.

3. The method for assessing right ventricular function according to claim 1, characterized in that, The method further includes generating a comprehensive evaluation report based on the calculated displacement index, area index, and time dispersion index; the comprehensive evaluation report includes at least a visualization of the dynamic three-dimensional grid model of the right ventricle, displacement-time curves of each tracking point, and dynamic change curves of the tricuspid annulus area over time.

4. A system for assessing right ventricular function using the method of claim 1, characterized in that, include: The data acquisition module is used to acquire a multi-frame two-dimensional ultrasound image sequence of the right ventricle of the heart and its corresponding spatial pose information, and generate point cloud data containing spatial coordinates and pose information; A point cloud construction module is used to identify end-diastolic frames, end-systolic frames, and anatomical structure points in the image sequence, and to acquire sparse point cloud data of anatomical structure points on the end-diastolic frames and end-systolic frames. A three-dimensional mesh model construction module is used to construct three-dimensional mesh models of the end-diastolic frames and end-systolic frames based on the sparse point cloud data and the cardiac knowledge base. The trajectory acquisition module is used to determine multiple tracking points of the tricuspid annulus on the three-dimensional mesh model based on the sparse point cloud data of the tricuspid annulus, and to acquire the three-dimensional coordinate trajectory of each tracking point in each frame of the cardiac cycle. The functional evaluation module is used to calculate multiple indicators for evaluating tricuspid valve annular changes based on the three-dimensional coordinate trajectory. The indicators include a displacement indicator for evaluating right ventricular longitudinal contraction, an area indicator for evaluating right ventricular annular area changes, and a time dispersion indicator for evaluating right ventricular contraction synchronicity. The data acquisition module, point cloud construction module, 3D mesh model construction module, trajectory acquisition module, and functional evaluation module are connected in sequence.

5. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the method for assessing right ventricular function as claimed in any one of claims 1-3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the method for assessing right ventricular function as described in any one of claims 1-3.

7. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the method for assessing right ventricular function as described in any one of claims 1-3.

Images