Method and system for visualizing cardiac electrical activation signals using magnetic field inversion source reconstruction

Abstract

The application provides a magnetic field inverse source reconstruction cardiac electric excitation signal imaging visualization method and system, and relates to the data visualization field, and the method comprises the following steps: obtaining a three-dimensional cardiac anatomical model containing cardiac and four-chamber anatomical boundary information of the heart, and discrete cardiac magnetic signal data with spatial coordinates and signal intensity changing with time; and converting the discrete cardiac magnetic signal data into an electric signal representing cardiac electric excitation activity through a source reconstruction algorithm; initializing a three-dimensional body data space by using the spatial bounding box information of the three-dimensional cardiac anatomical model; calculating an initial electric signal intensity based on the converted electric signal to obtain three-dimensional body data containing the initial electric signal intensity; performing smoothing processing on the three-dimensional body data to generate a smoothed electric signal field, and performing clipping on the smoothed electric signal field by using the anatomical boundary to generate a final electric signal body data; and finally performing volume rendering imaging. The application can convert the source reconstruction electric signal into a four-dimensional dynamic image based on anatomical constraints.

Description

Technical Field

[0001] This invention relates to the field of data visualization technology, and in particular to a method and system for imaging and visualizing cardiac electrical excitation signals reconstructed from a magnetic field reverse source. Background Technology

[0002] Magnetocardiography (MCG), a highly sensitive detection method, can capture weak magnetic signals generated by the electrophysiological activity of the myocardium. Since the cardiac magnetic field is generated by ion currents during the depolarization and repolarization processes of myocardial cells, the spatiotemporal evolution of the magnetic signal directly reflects the state of the heart's internal electrophysiological activity. Therefore, reconstructing the electrical signal intensity from the magnetic field is of significant reference value for assessing the electrophysiological characteristics of the heart. With advancements in sensor technology, modern magnetocardiography (MCG) systems can achieve multi-channel synchronous acquisition, and algorithms can inversely reconstruct the dynamic trajectory of the cardiac magnetic field as a raw current, evolving over time in three-dimensional space.

[0003] With the maturation of four-dimensional magnetocardiography (MCC) acquisition technology, the processing of massive amounts of time-varying magnetic signal data has become a focal point. The extremely low intensity and complex spatial distribution of MCC signals require algorithms that not only possess efficient data processing capabilities but also deeply integrate the abstract magnetic field strength values ​​with the actual electrical signals and the patient's real cardiac anatomy. Although existing high-performance computers can process large-scale point cloud data, how to transform these discrete physical detection points into a biologically meaningful continuous field distribution remains a research hotspot in the field of biomedical engineering.

[0004] Existing visualization systems mostly display the original magnetic signals in two dimensions or simple three-dimensional magnetic poles, failing to effectively convert the detected magnetic signals into anatomically significant electrical signal fields. This prevents researchers from directly observing the three-dimensional conduction process of electrical excitation waves (excitation waves) within myocardial tissue, and makes them susceptible to interference from external anatomical noise. Summary of the Invention

[0005] This invention provides a method and system for imaging and visualizing cardiac electrical excitation signals reconstructed from a magnetic field reverse source. This method addresses the shortcomings of existing technologies, such as the lack of display dimension for magnetic-to-electric signals and the inability to intuitively and continuously reconstruct the cardiac electrical excitation conduction process in three-dimensional space. It enables four-dimensional visualization imaging of the dynamic propagation of cardiac electrical activity over time within three-dimensional anatomical structures.

[0006] This invention provides an imaging and visualization method for reconstructing cardiac electrical excitation signals using a magnetic field inverse source, comprising: A three-dimensional cardiac anatomical model containing information on the anatomical boundaries of the heart and its four chambers is obtained, along with discrete cardiac magnetic signal data with spatial coordinates and signal intensity varying over time. The discrete cardiac magnetic signal data is then converted into electrical signals characterizing cardiac electrical excitation activity using a source reconstruction algorithm. The three-dimensional volume data space is initialized using the spatial bounding box information of the three-dimensional cardiac anatomy model, and the initial electrical signal intensity is calculated for the voxel points in the three-dimensional volume data space based on the converted electrical signal, so as to obtain three-dimensional volume data containing the initial electrical signal intensity. The three-dimensional volume data is smoothed to generate a smooth electrical signal field; Using the anatomical boundaries provided by the three-dimensional cardiac anatomy model, the smooth electrical signal field is clipped, and the signal intensity in the region outside the anatomical boundaries is set to zero to generate the final electrical signal volume data. The final electrical signal volume data is volumetrically plotted to generate a visualization image showing the dynamic propagation process of the electrical signal within the three-dimensional cardiac anatomical model.

[0007] According to the magnetic field reverse source reconstruction imaging visualization method for cardiac electrical excitation signals provided by the present invention, the step of calculating the initial electrical signal intensity for voxel points in the three-dimensional volume data space includes: For any voxel point to be calculated in the three-dimensional volume data space, the K nearest sampling points are retrieved in the converted electrical signal using the kd-tree algorithm; Calculate the electrical signal contribution intensity of each of the K sampling points to the voxel point. The electrical signal contribution intensity is positively correlated with the electrical signal intensity of the sampling point at the corresponding time and negatively correlated with the distance from the voxel point to the sampling point. The maximum value of the electrical signal contribution intensity among the K sampling points is taken as the initial electrical signal intensity of the voxel point.

[0008] According to the magnetic field reverse source reconstruction imaging visualization method for cardiac electrical excitation signals provided by the present invention, the formula for calculating the contribution intensity of the electrical signal is as follows: , Where I represents the electrical signal contribution intensity of any of the K sampling points corresponding to the voxel point to be calculated to that voxel point. This represents the electrical signal strength value at that sampling point. The distance from the voxel to the sampling point. The radius of influence of the sampling point.

[0009] The magnetic field reverse source reconstruction imaging visualization method for cardiac electrical excitation signals provided by the present invention further includes, before calculating the initial electrical signal intensity for voxel points in the three-dimensional volume data space: The surface vertices and spatial coordinates of the converted electrical signals of the three-dimensional cardiac anatomy model are obtained and mapped to foreground points in the three-dimensional volume data space. Calculate the distance field from each voxel point in the three-dimensional volume data space to the foreground point, and extract voxel points whose distance field values ​​are less than a preset threshold to form narrowband candidate regions; The initial signal intensity is calculated only for voxel points located within the narrowband candidate region, while the signal intensity of voxel points outside the narrowband candidate region is set to zero.

[0010] According to the magnetic field reverse source reconstruction imaging visualization method for cardiac electrical excitation signals provided by the present invention, the step of cropping the smoothed electrical signal field using the anatomical boundaries provided by the three-dimensional cardiac anatomical model includes: The internal space of the three-dimensional cardiac anatomy model is filled to generate a binary solid heart mask image; The solid heart mask image is multiplied on a voxel-by-voxel basis with the smoothed electrical signal field, preserving the electrical signal intensity inside the mask.

[0011] According to the magnetic field reverse source reconstruction imaging visualization method for cardiac electrical excitation signals provided by the present invention, the volume rendering of the final electrical signal volume data includes: A logarithmic transformation mapping is performed on the electrical signal strength values ​​in the final electrical signal volume data to enhance the display dynamics of weak electrical signals; Based on the preset transparency and color mapping tables, the electrical signal intensity values ​​after logarithmic transformation are assigned color and transparency, and the maximum density projection method is used for rendering to generate the visualization image.

[0012] According to the magnetic field reverse source reconstruction imaging visualization method for cardiac electrical excitation signals provided by the present invention, the smoothing process adopts a Gaussian smoothing algorithm, and the Gaussian smoothing parameters sigma, the influence radius when calculating the electrical signal, and the transparency settings when drawing the volume are dynamically adjusted according to the feedback of the visualization effect.

[0013] This invention also provides an imaging and visualization system for reconstructing cardiac electrical excitation signals using a magnetic field reverse source, comprising the following modules: The acquisition module is used to acquire a three-dimensional cardiac anatomical model containing anatomical boundary information of the heart and its four chambers, as well as discrete cardiac magnetic signal data with spatial coordinates and signal intensity varying over time; and to convert the discrete cardiac magnetic signal data into electrical signals characterizing cardiac electrical excitation activity through a source reconstruction algorithm. The initial signal intensity calculation module is used to initialize the three-dimensional volume data space using the spatial bounding box information of the three-dimensional cardiac anatomy model, and calculate the initial electrical signal intensity for the voxel points in the three-dimensional volume data space based on the converted electrical signal, so as to obtain three-dimensional volume data containing the initial electrical signal intensity. A smoothing module is used to smooth the three-dimensional volume data and generate a smooth electrical signal field. The trimming module is used to trim the smooth electrical signal field using the anatomical boundary provided by the three-dimensional cardiac anatomy model, setting the signal intensity in the region outside the anatomical boundary to zero, and generating the final electrical signal volume data. The visualization module is used to perform volume rendering on the final electrical signal volume data and generate a visualization image that shows the dynamic propagation process of the electrical signal inside the three-dimensional cardiac anatomical model.

[0014] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the magnetic field reverse source reconstruction cardiac electrical excitation signal imaging visualization method as described above.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the magnetic field reverse source reconstruction cardiac electrical excitation signal imaging visualization method as described above.

[0016] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the magnetic field reverse source reconstruction imaging visualization method for cardiac electrical excitation signals as described above.

[0017] The present invention provides a magnetic field reverse source reconstruction method and system for imaging and visualizing cardiac electrical excitation signals. This method acquires a three-dimensional cardiac anatomical model containing anatomical boundary information and discrete cardiac magnetic signal data with spatial coordinates and signal intensity varying over time. A source reconstruction algorithm converts the discrete cardiac magnetic signal data into electrical signals characterizing cardiac electrical excitation activity. The system initializes the volume data space using model bounding box information and calculates initial electrical signal intensities for voxel points in the three-dimensional volume data space based on the converted electrical signals, resulting in three-dimensional volume data containing these initial electrical signal intensities. By restoring discrete physical detection values ​​to a spatially continuous distribution with biomedical significance, this method facilitates the realization of… Panoramic observation in the spatial dimension lays the foundation for the data; combined with smoothing of the three-dimensional volume data and cropping based on the anatomical boundaries, signal jumps caused by discrete acquisition are eliminated and the signal display is strictly confined to the internal structure of the heart anatomy, effectively eliminating interference from invalid signals or artifacts outside the heart, and ensuring the anatomical accuracy and reference value of the imaging results; finally, by volume rendering of the electrical signal volume data, a visual image of dynamic propagation within the anatomical model is generated, allowing users to intuitively observe the complete conduction path and evolution of cardiac electrical signals, such as excitation waves, within the heart over time, thus providing rich imaging data references for the assessment of cardiac physiological characteristics. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is a schematic flowchart of the magnetic field reverse source reconstruction imaging visualization method for cardiac electrical excitation signals provided by the present invention.

[0020] Figure 2 This is a schematic diagram of the structure of the magnetic field reverse source reconstruction imaging visualization system for cardiac electrical excitation signals provided by the present invention.

[0021] Figure 3 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0023] In existing magnetocardiography (MCG) hardware architectures, sensor arrays are typically housed within a cryogenic container above the chest wall to acquire magnetic signals from different detection channels. The output data from these systems is usually time-domain waveforms corresponding to each independent channel. Traditionally, these signals are mapped onto a two-dimensional mesh and interpolated to generate a magnetic field intensity contour map. However, this display method ignores the thickness of the heart and the spatial anatomical relationships between the atria and ventricles, resulting in blurred depth perception and propagation paths for the signals. Even with the introduction of a three-dimensional mesh model of the heart, existing visualization logic often merely displays the signals as background textures or external light spots, failing to address the problem of continuous signal reconstruction within the spatial volume data.

[0024] To address the aforementioned shortcomings, this application provides an imaging and visualization method for reconstructing cardiac electrical excitation signals using a magnetic field inverse source, such as... Figure 1 As shown, the method includes the following steps: Step S100: Obtain a three-dimensional cardiac anatomical model containing information on the anatomical boundaries of the heart and its four chambers, as well as discrete cardiac magnetic signal data with spatial coordinates and signal intensity varying over time; and convert the discrete cardiac magnetic signal data into electrical signals characterizing cardiac electrical excitation activity using a source reconstruction algorithm. Step S200: Initialize the three-dimensional volume data space using the spatial bounding box information of the three-dimensional cardiac anatomy model, and calculate the initial electrical signal intensity for the voxel points in the three-dimensional volume data space based on the converted electrical signal to obtain three-dimensional volume data containing the initial electrical signal intensity. Step S300: Smooth the three-dimensional volume data to generate a smooth electrical signal field; Step S400: Using the anatomical boundary provided by the three-dimensional cardiac anatomy model, the smooth electrical signal field is clipped, and the signal intensity located outside the anatomical boundary is set to zero to generate the final electrical signal volume data. Step S500: Perform volume rendering on the final electrical signal volume data to generate a visualization image showing the dynamic propagation process of the electrical signal within the three-dimensional cardiac anatomical model.

[0025] In existing magnetocardiography (MCG) imaging, magnetic signals are essentially a physical representation of cardiac electrophysiological activity. This embodiment utilizes acquired discrete cardiac magnetic signal data to reconstruct the corresponding electrical signals using a source reconstruction algorithm, and then performs field reconstruction and visualization in three-dimensional anatomical space.

[0026] Specifically, this invention introduces an anatomical model of the heart and its four chambers as a constraint framework, expanding discrete probe data into volumetric data covering the entire heart region through a specific spatial weighting algorithm. Subsequently, numerical abrupt changes are eliminated through smoothing, and the signal field is physically confined and clipped using anatomical boundaries. Finally, logarithmic enhancement and volumetric rendering techniques are combined to visually reproduce the dynamic propagation of electrocardiographic excitation waves within the three-dimensional heart structure. This method not only improves the intuitiveness of imaging but also eliminates invalid signals through anatomical constraints, significantly enhancing the reliability of auxiliary analysis. This invention utilizes discrete cardiac magnetic signal data acquired by a magnetocardiograph, and through spatial weighting and physical transformation algorithms, reconstructs and visualizes the corresponding electrical signal distribution in three-dimensional anatomical space. This processing allows the operator to directly observe the three-dimensional conduction process of cardiac electrical excitation activity (i.e., excitation waves) in the atria and ventricles, transcending abstract magnetic field lines.

[0027] The method described in this embodiment of the invention can be executed by a computer device with graphics processing capabilities, such as a medical imaging workstation connected to a magnetocardiograph.

[0028] The above steps will be explained in detail below through specific embodiments.

[0029] Step S100: Obtain a three-dimensional cardiac anatomical model containing information on the anatomical boundaries of the heart and its four chambers, as well as discrete cardiac magnetic signal data with spatial coordinates and signal intensity varying over time; and convert the discrete cardiac magnetic signal data into electrical signals characterizing cardiac electrical excitation activity using a source reconstruction algorithm. First, a three-dimensional cardiac anatomical model containing information about the heart and the anatomical boundaries of its four chambers was acquired, along with discrete cardiac magnetic signal data with spatial coordinates and signal intensity varying over time. This three-dimensional cardiac anatomical model defines the external contour and internal chamber interfaces of the heart in the form of a mesh, and can be referred to as a mesh model.

[0030] It should be noted that the discrete cardiac magnetic signal data is acquired by a sensor array arranged around the heart. During the sampling period, the spatial coordinates of each sampling point remain fixed, while the recorded magnetic induction intensity value fluctuates dynamically over time, thus forming a four-dimensional spatiotemporal dataset.

[0031] Then, the input raw magnetic field data is processed using a source reconstruction algorithm. It should be noted that the source reconstruction algorithm falls under the category of data preprocessing; it reconstructs the original current dipoles or electrophysiological activity information inside the heart by inversely solving the magnetic field, thereby obtaining discrete electrical signal data. Each signal point carries precise three-dimensional spatial coordinates. and the electrical signal strength value at that moment electrical signal strength value Changes over time.

[0032] In this invention, the input data for the visualization process is the cardiac electrical excitation signal, i.e., the electrical signal, converted by the source reconstruction algorithm. This step uses inverse magnetic field calculation to convert the physically detected magnetic induction intensity into electrophysiological signal data characterizing the distribution of current or potential changes within the heart.

[0033] Step S200: Convert the discrete cardiac magnetic signal data into an electrical signal characterizing cardiac electrical excitation activity using a source reconstruction algorithm; initialize the three-dimensional volume data space using the spatial bounding box information of the three-dimensional cardiac anatomy model; and calculate the initial electrical signal intensity for voxel points in the three-dimensional volume data space based on the converted electrical signal to obtain three-dimensional volume data containing the initial electrical signal intensity. After acquiring the electrical signal data, a physical field reconstruction is performed in a three-dimensional anatomical space. Specifically, after obtaining the converted electrical signal, the system extracts the spatial features of the cardiac anatomical model, establishes a volume data space that can encompass the complete cardiac structure, and then uses neighborhood retrieval and distance decay functions based on the kd-tree (k-dimensional tree) algorithm to map the discrete electrical signal intensity onto voxels in the volume space, thereby completing the conversion from physical detection data to anatomical field data.

[0034] Specifically, during the initialization of 3D volumetric data, the system determines the imaging region based on the spatial bounding box information of the heart mesh model. A mesh can be constructed based on the manually set sampling interval (spacing) and origin information. Specifically, the vertex with the smallest coordinate value among all vertices of the mesh is extracted as the origin, and a mesh is divided within a cuboid bounding box according to the preset sampling interval, thus initializing a 3D volumetric data space with all zero values. Here, origin is set to the coordinates of the vertex with the smallest coordinate value within the bounding cuboid corresponding to the input heart and its four chambers mesh. This setting ensures that the volumetric data space perfectly encompasses the cardiac anatomy with minimal redundancy.

[0035] To improve computational efficiency, narrowband extraction is performed before intensity calculation. The system considers not only signal points but also the mesh vertices of the heart and four chambers. Based on the nearest neighbor rule, the nearest points in the 3D volume data are found and set as foreground points, and the distance field of the entire space is calculated. Points with distance field values ​​less than a preset narrowband threshold (thres) are extracted to form narrowbands. Subsequent signal intensity calculations are only performed on points within these narrowbands, thus reducing the number of computation points and improving speed.

[0036] Specifically, in this embodiment, the spatial coordinates corresponding to the vertices of the model surface and the converted electrical signals are mapped to foreground points in the volume data space, and the distance field from each voxel point in the space to the foreground point is calculated. The system extracts only voxel points with distance field values ​​less than a preset narrowband threshold (thres) to form narrowband candidate regions. Subsequent signal intensity calculations are performed only on voxel points within this narrowband region; the signal intensity of voxel points outside the narrowband is directly set to zero. This method avoids traversing the massive number of invalid voxel points in the entire space, thereby reducing computational load and improving the generation speed of four-dimensional images.

[0037] When calculating the initial electrical signal intensity of a voxel point within a narrowband, the system uses the kd-tree algorithm to calculate its K nearest neighbors with the converted electrical signal.

[0038] Specifically, for any voxel point to be calculated within a narrow band region extracted in the 3D volumetric data space, the k-dtree algorithm is used to quickly retrieve the K nearest neighbor points (K is a positive integer) in the converted electrical signal. For each selected sampling point, a distance attenuation algorithm is used to calculate its contribution intensity to the electrical signal of the current voxel. The influence radius of the sampling point is set as... If the distance between the voxel point and the sampling point Exceeding the radius of influence If the influence radius is within 0, then its contribution is 0; The contribution intensity of the electrical signal is calculated according to the following formula: (1), Where I represents the electrical signal contribution intensity of any of the K sampling points corresponding to the voxel point to be calculated to that voxel point. This represents the electrical signal strength value at that sampling point. The distance from the voxel to the sampling point. The radius of influence of the sampling point.

[0039] Finally, the system takes the maximum value among the electrical signal contribution intensities calculated from these K nearest neighbor points as the initial electrical signal intensity of the voxel point.

[0040] If a voxel is influenced by multiple neighboring points, the maximum value among them is taken as the initial electrical signal intensity of that voxel. After traversing all valid voxels, three-dimensional volume data containing the initial electrical signal intensity is obtained.

[0041] Formula (1) reflects the physical characteristic that the electrical signal attenuates linearly with distance. When K is greater than 1, the processor takes the maximum value among these K influencing intensities as the initial electrical signal intensity of that voxel point. After traversing all voxels within the narrow band, three-dimensional volume data containing the initial electrical signal intensity is obtained.

[0042] Step S300: Smooth the three-dimensional volume data to generate a smooth electrical signal field; This step primarily involves spatial smoothing of the field distribution. Since interpolation calculations may introduce unnatural variations at the signal source edges, the processor performs Gaussian smoothing on the obtained 3D volume data. By setting the Gaussian smoothing parameter sigma, a weighted average of the voxel values ​​is applied, thereby smoothing the initial electrical signal intensity distribution. This step simulates the physical characteristics of continuous conduction of cardiac electrical excitation in the myocardial syncytium, eliminating computational noise and generating a smooth electrical signal field with a natural transition.

[0043] Step S400: Using the anatomical boundary provided by the three-dimensional cardiac anatomy model, the smooth electrical signal field is clipped, and the signal intensity located outside the anatomical boundary is set to zero to generate the final electrical signal volume data. To ensure the anatomical accuracy of the visualization results, the smoothed electrical signal field was cropped using the boundaries provided by a 3D cardiac anatomical model. In the cropping step, the mesh interiors of the heart and its four chambers were filled into an image. Based on this filled image, the smoothed volume data was filtered. Only voxels whose pixel coordinates fell within the foreground of the filled image (i.e., the heart interior) had their signal intensity preserved; otherwise, they were set to 0. This process ensured a high degree of consistency between the spatial distribution of the signal and the anatomical structure.

[0044] More specifically, firstly, the signal intensity in the region outside the anatomical boundaries is set to zero; then, the space inside the heart and four chambers mesh is voxelized and filled to generate a binary solid heart mask image (mask) the same size as the volume data space. Subsequently, a voxel-by-voxel product operation (logical AND operation) is performed between the smoothed electrical signal field and this mask: the electrical signal intensity located outside the anatomical boundaries (i.e., outside the heart tissue) is forcibly reset to 0. In this way, the electrical signal is strictly locked within the myocardial tissue and heart chambers, avoiding observational errors caused by signals floating outside the heart, and ensuring the absolute anatomical accuracy of the visualization results.

[0045] Step S500: Perform volume rendering on the final electrical signal volume data to generate a visualization image showing the dynamic propagation process of the electrical signal within the three-dimensional cardiac anatomical model.

[0046] Because the initial electrical signal strength is usually extremely weak (typically on the order of magnitude of 1), To achieve observable visual contrast, a logarithmic transformation is performed before assigning colors. A color bar is then set, and volume rendering is performed using maximum density projection (MIP) to make the signal peaks clearly visible in three-dimensional space.

[0047] Specifically, the final electrical signal volume data is first subjected to a logarithmic transformation mapping to stretch the dynamic range, making even the slightest signal potential difference visible to the naked eye. Subsequently, the intensity values ​​are colored according to a preset color and transparency mapping table, and rendered using the maximum density projection (MIP) method. By continuously processing the electrical signal data at each moment along the time axis, the system ultimately generates a four-dimensional image representing the dynamic propagation of cardiac electrical excitation activity within the three-dimensional anatomical model.

[0048] In practice, users can dynamically adjust the Gaussian smoothing parameters through the interactive interface based on the real-time rendered visual feedback. Radius of influence when calculating electrical signal strength Including transparency gradients during volume rendering, etc., to obtain the clearest and most layered four-dimensional dynamic images or videos. By continuously processing the electrical signals at each moment on the time axis, the system ultimately generates a four-dimensional dynamic image of the generation, conduction, and dissipation of cardiac electrical signals within the three-dimensional anatomical model, allowing the operator to observe the electrophysiological activities inside the heart as if observing fluid propagation.

[0049] The method of this invention involves several core parameters, including the narrowband threshold. Radius of influence The neighborhood number is the number of sampling points whose spatial coordinates are closest to the retrieved point. Gaussian smoothing parameters Sampling interval And the transparency settings during volume rendering. In practice, these parameters need to be selected through repeated experiments to determine the optimal parameter combination. Among them, , The transparency setting has the most significant impact on the final rendering effect and is the focus of debugging.

[0050] This invention provides a method for visualizing cardiac electrical excitation signals reconstructed from a magnetic field inverse source. Through a distance- and intensity-based weighted interpolation algorithm combined with Gaussian smoothing, it smoothly extends isolated signal intensity points onto the entire three-dimensional volumetric data, solving the problem of existing technologies being unable to reconstruct continuous electrical excitation signals in three-dimensional space, greatly facilitating visualization. A narrowband computation strategy is introduced, restricting the calculation and updating of electrical signal intensity to a narrow region around the heart model and signal points. Compared to full-space voxel computation, this design effectively reduces unnecessary computation, significantly improving the generation speed of four-dimensional images and meeting the needs of real-time observation. By precisely cropping the smoothed three-dimensional volumetric data according to the input cardiac mesh, the electrical signal intensity is strictly limited to within the heart tissue. This design effectively eliminates visual artifacts caused by signal overflow outside the anatomical structure, improving imaging resolution. Furthermore, this invention addresses the characteristic of extremely weak electrical signals by using logarithmic transformation enhancement processing, enabling subtle changes in electrical signals to be clearly mapped to the visualization color range, enhancing the dynamic display of weak electrical signals.

[0051] In some specific embodiments of the present invention, such as Figure 2 As shown, this solution provides an imaging and visualization system for reconstructing cardiac electrical excitation signals using a magnetic field inverse source, comprising: The acquisition module 10 is used to acquire a three-dimensional cardiac anatomical model containing anatomical boundary information of the heart and the four chambers of the heart, as well as discrete cardiac magnetic signal data with spatial coordinates and signal intensity varying over time; and to convert the discrete cardiac magnetic signal data into electrical signals characterizing cardiac electrical excitation activity through a source reconstruction algorithm. The initial signal intensity calculation module 20 is used to initialize the three-dimensional volume data space using the spatial bounding box information of the three-dimensional cardiac anatomy model, and calculate the initial electrical signal intensity for the voxel points in the three-dimensional volume data space based on the converted electrical signal, so as to obtain three-dimensional volume data containing the initial electrical signal intensity. Smoothing module 30 is used to smooth the three-dimensional volume data and generate a smooth electrical signal field; The trimming module 40 is used to trim the smooth electrical signal field using the anatomical boundary provided by the three-dimensional cardiac anatomy model, setting the signal intensity in the region outside the anatomical boundary to zero, and generating the final electrical signal volume data. The visualization module 50 is used to perform volume rendering on the final electrical signal volume data and generate a visualization image showing the dynamic propagation process of the electrical signal inside the three-dimensional cardiac anatomical model.

[0052] Possibly, calculating the initial electrical signal intensity for voxel points in the three-dimensional volume data space includes: For any voxel point to be calculated in the three-dimensional volume data space, the K nearest sampling points are retrieved in the converted electrical signal using the kd tree algorithm; Calculate the electrical signal contribution intensity of each of the K sampling points to the voxel point. The electrical signal contribution intensity is positively correlated with the electrical signal intensity of the sampling point at the corresponding time and negatively correlated with the distance from the voxel point to the sampling point. The maximum value of the electrical signal contribution intensity among the K sampling points is taken as the initial electrical signal intensity of the voxel point.

[0053] Specifically, this embodiment provides an implementation method for calculating the initial electrical signal strength.

[0054] Possibly, before calculating the initial electrical signal intensity for voxel points in the three-dimensional volume data space, the following steps are also included: The surface vertices and spatial coordinates of the converted electrical signals of the three-dimensional cardiac anatomy model are obtained and mapped to foreground points in the three-dimensional volume data space. Calculate the distance field from each voxel point in the three-dimensional volume data space to the foreground point, and extract voxel points whose distance field values ​​are less than a preset threshold to form narrowband candidate regions; The initial signal intensity is calculated only for voxel points located within the narrowband candidate region, while the signal intensity of voxel points outside the narrowband candidate region is set to zero.

[0055] Specifically, this embodiment provides an implementation method for forming narrowband candidate regions.

[0056] Figure 3 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3As shown, the electronic device may include: a processor 310, a communications interface 320, a memory 330, and a communications bus 340, wherein the processor 310, the communications interface 320, and the memory 330 communicate with each other through the communications bus 340. The processor 310 can call logic instructions in the memory 330 to execute a magnetic field reverse source reconstruction method for imaging and visualization of cardiac electrical excitation signals. This method includes: acquiring a three-dimensional cardiac anatomical model containing anatomical boundary information of the heart and its four chambers, and discrete cardiac magnetic signal data with spatial coordinates and signal intensity varying over time; converting the discrete cardiac magnetic signal data into electrical signals characterizing cardiac electrical excitation activity using a source reconstruction algorithm; initializing a three-dimensional volume data space using the spatial bounding box information of the three-dimensional cardiac anatomical model, and calculating initial electrical signal intensities for voxel points in the three-dimensional volume data space based on the converted electrical signals, obtaining three-dimensional volume data containing the initial electrical signal intensities; smoothing the three-dimensional volume data to generate a smoothed electrical signal field; cropping the smoothed electrical signal field using the anatomical boundaries provided by the three-dimensional cardiac anatomical model, setting the signal intensity in the region outside the anatomical boundaries to zero, generating final electrical signal volume data; and performing volume rendering on the final electrical signal volume data to generate a visualization image showing the dynamic propagation process of the electrical signals within the three-dimensional cardiac anatomical model.

[0057] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0058] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the magnetic field reverse source reconstruction cardiac electrical excitation signal imaging visualization method provided by the above methods. This method includes: acquiring a three-dimensional cardiac anatomical model containing anatomical boundary information of the heart and its four chambers, and discrete cardiac magnetic signal data with spatial coordinates and signal intensity varying over time; converting the discrete cardiac magnetic signal data into an electrical signal characterizing cardiac electrical excitation activity using a source reconstruction algorithm; and utilizing the three-dimensional cardiac... The spatial bounding box information of the cardiac anatomy model is used to initialize the three-dimensional volume data space. Based on the converted electrical signal, the initial electrical signal intensity is calculated for the voxel points in the three-dimensional volume data space to obtain three-dimensional volume data containing the initial electrical signal intensity. The three-dimensional volume data is then smoothed to generate a smooth electrical signal field. Using the anatomical boundary provided by the three-dimensional cardiac anatomy model, the smooth electrical signal field is clipped, and the signal intensity in the region outside the anatomical boundary is set to zero to generate the final electrical signal volume data. The final electrical signal volume data is then volumetrically rendered to generate a visualization image showing the dynamic propagation process of the electrical signal within the three-dimensional cardiac anatomy model.

[0059] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements a magnetic field reverse source reconstruction method for imaging and visualization of cardiac electrical excitation signals provided by the methods described above. This method includes: acquiring a three-dimensional cardiac anatomical model containing anatomical boundary information of the heart and its four chambers, and discrete cardiac magnetic signal data with spatial coordinates and signal intensity varying over time; converting the discrete cardiac magnetic signal data into electrical signals characterizing cardiac electrical excitation activity using a source reconstruction algorithm; initializing a three-dimensional volume data space using the spatial bounding box information of the three-dimensional cardiac anatomical model, and calculating initial electrical signal intensities for voxel points in the three-dimensional volume data space based on the converted electrical signals, obtaining three-dimensional volume data containing the initial electrical signal intensities; smoothing the three-dimensional volume data to generate a smoothed electrical signal field; cropping the smoothed electrical signal field using the anatomical boundaries provided by the three-dimensional cardiac anatomical model, setting the signal intensity in the region outside the anatomical boundaries to zero, generating final electrical signal volume data; and performing volume rendering on the final electrical signal volume data to generate a visualization image showing the dynamic propagation process of the electrical signals within the three-dimensional cardiac anatomical model.

[0060] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0061] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0062] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for imaging and visualization of cardiac electrical excitation signals reconstructed from a magnetic field inverse source, characterized in that, include: A three-dimensional cardiac anatomical model containing information on the anatomical boundaries of the heart and its four chambers is obtained, along with discrete cardiac magnetic signal data with spatial coordinates and signal intensity varying over time. The discrete cardiac magnetic signal data is then converted into electrical signals characterizing cardiac electrical excitation activity using a source reconstruction algorithm. The three-dimensional volume data space is initialized using the spatial bounding box information of the three-dimensional cardiac anatomy model, and the initial electrical signal intensity is calculated for the voxel points in the three-dimensional volume data space based on the converted electrical signal, so as to obtain three-dimensional volume data containing the initial electrical signal intensity. The calculation of the initial electrical signal intensity for voxel points in the three-dimensional volume data space includes: For any voxel point to be calculated in the three-dimensional volume data space, the kd-tree algorithm is used to retrieve the K nearest sampling points in the converted electrical signal; the electrical signal contribution intensity of each of the K sampling points to the voxel point is calculated, the electrical signal contribution intensity is positively correlated with the electrical signal intensity of the sampling point at the corresponding time, and negatively correlated with the distance from the voxel point to the sampling point; the maximum value of the electrical signal contribution intensity among the K sampling points is taken as the initial electrical signal intensity of the voxel point; The three-dimensional volume data is smoothed to generate a smooth electrical signal field; Using the anatomical boundaries provided by the three-dimensional cardiac anatomy model, the smoothed electrical signal field is cropped, and the signal intensity in the region outside the anatomical boundaries is set to zero to generate the final electrical signal volume data; the cropping of the smoothed electrical signal field using the anatomical boundaries provided by the three-dimensional cardiac anatomy model includes: The internal space of the three-dimensional cardiac anatomy model is filled to generate a binary solid heart mask image; the solid heart mask image is multiplied on a voxel-by-voxel basis with the smoothed electrical signal field to preserve the electrical signal intensity inside the mask. Volume rendering is performed on the final electrical signal volume data to generate a visualization image showing the dynamic propagation process of the electrical signal within the three-dimensional cardiac anatomical model; the volume rendering of the final electrical signal volume data includes: Logarithmic transformation mapping is performed on the electrical signal intensity values ​​in the final electrical signal volume data to enhance the display dynamics of weak electrical signals; according to the preset transparency mapping table and color mapping table, color and transparency are assigned to the electrical signal intensity values ​​after the logarithmic transformation mapping, and the maximum density projection method is used for rendering to generate the visualization image.

2. The method for imaging and visualizing cardiac electrical excitation signals reconstructed from a magnetic field reverse source according to claim 1, characterized in that, The formula for calculating the contribution intensity of the electrical signal is: ， Where I represents the electrical signal contribution intensity of any of the K sampling points corresponding to the voxel point to be calculated to that voxel point. This represents the electrical signal strength value at that sampling point. The distance from the voxel to the sampling point. The radius of influence of the sampling point.

3. The method for imaging and visualizing cardiac electrical excitation signals reconstructed from a magnetic field reverse source according to claim 1, characterized in that, Before calculating the initial electrical signal intensity for voxel points in the three-dimensional volume data space, the method further includes: The surface vertices and spatial coordinates of the converted electrical signals of the three-dimensional cardiac anatomy model are obtained and mapped to foreground points in the three-dimensional volume data space. Calculate the distance field from each voxel point in the three-dimensional volume data space to the foreground point, and extract voxel points whose distance field values ​​are less than a preset threshold to form narrowband candidate regions; The initial signal intensity is calculated only for voxel points located within the narrowband candidate region, while the signal intensity of voxel points outside the narrowband candidate region is set to zero.

4. The method for imaging and visualization of cardiac electrical excitation signals reconstructed from a magnetic field reverse source according to claim 1, characterized in that, The smoothing process employs a Gaussian smoothing algorithm, and the Gaussian smoothing parameters sigma, the influence radius when calculating electrical signals, and the transparency settings when drawing volumes are dynamically adjusted based on the feedback from the visualization effect.

5. A magnetic field reverse source reconstruction imaging visualization system for cardiac electrical excitation signals, characterized in that, include: The acquisition module is used to acquire a three-dimensional cardiac anatomical model containing anatomical boundary information of the heart and its four chambers, as well as discrete cardiac magnetic signal data with spatial coordinates and signal intensity varying over time; and to convert the discrete cardiac magnetic signal data into electrical signals characterizing cardiac electrical excitation activity through a source reconstruction algorithm. The initial signal intensity calculation module is used to initialize the three-dimensional volume data space using the spatial bounding box information of the three-dimensional cardiac anatomy model, and calculate the initial electrical signal intensity for the voxel points in the three-dimensional volume data space based on the converted electrical signal, so as to obtain three-dimensional volume data containing the initial electrical signal intensity. The calculation of the initial electrical signal intensity for voxel points in the three-dimensional volume data space includes: For any voxel point to be calculated in the three-dimensional volume data space, the kd-tree algorithm is used to retrieve the K nearest sampling points in the converted electrical signal; the electrical signal contribution intensity of each of the K sampling points to the voxel point is calculated, the electrical signal contribution intensity is positively correlated with the electrical signal intensity of the sampling point at the corresponding time, and negatively correlated with the distance from the voxel point to the sampling point; the maximum value of the electrical signal contribution intensity among the K sampling points is taken as the initial electrical signal intensity of the voxel point; A smoothing module is used to smooth the three-dimensional volume data and generate a smooth electrical signal field. A clipping module is used to clip the smoothed electrical signal field using the anatomical boundaries provided by the three-dimensional cardiac anatomy model, setting the signal intensity in the region outside the anatomical boundaries to zero, and generating the final electrical signal volume data; the clipping of the smoothed electrical signal field using the anatomical boundaries provided by the three-dimensional cardiac anatomy model includes: The internal space of the three-dimensional cardiac anatomy model is filled to generate a binary solid heart mask image; the solid heart mask image is multiplied on a voxel-by-voxel basis with the smoothed electrical signal field to preserve the electrical signal intensity inside the mask. A visualization module is used to perform volume rendering on the final electrical signal volume data, generating a visualization image showing the dynamic propagation process of the electrical signal within the three-dimensional cardiac anatomical model; the volume rendering of the final electrical signal volume data includes: Logarithmic transformation mapping is performed on the electrical signal intensity values ​​in the final electrical signal volume data to enhance the display dynamics of weak electrical signals; according to the preset transparency mapping table and color mapping table, color and transparency are assigned to the electrical signal intensity values ​​after the logarithmic transformation mapping, and the maximum density projection method is used for rendering to generate the visualization image.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the magnetic field reverse source reconstruction imaging visualization method for cardiac electrical excitation signals as described in any one of claims 1 to 4.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the magnetic field reverse source reconstruction imaging visualization method for cardiac electrical excitation signals as described in any one of claims 1 to 4.

Images