Electrocardiographic imaging inversion method with real-time constraints using intraoperative mapping data
By using individualized heart-trunk geometric models and real-time constraints on mapping data, the pathological problem in electrocardiogram (ECG) imaging inversion was solved, enabling real-time, adaptive correction and updating of inversion results, thus improving the accuracy and clinical applicability of ECG imaging.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHAANXI CHAOS YUAN MICROELECTRONICS CO LTD
- Filing Date
- 2026-05-14
- Publication Date
- 2026-06-19
AI Technical Summary
In the existing technology, the electrocardiogram imaging inversion method is mathematically ill-conditioned and ill-posed, which makes the inversion results susceptible to noise, errors and model differences, and cannot effectively use intraoperative mapping data for real-time correction and updating.
By establishing an individualized heart-trunk geometric model, acquiring surface potential signals and initial mapping data, generating a mapping weight matrix, constructing a joint inversion objective function, and dynamically updating the cardiac electrical activity reconstruction results through incremental solution, real-time constraints and corrections are achieved.
It significantly reduces model error and inverse problem pathology, improves the accuracy and clinical applicability of electrocardiogram imaging, and can correct and update inversion results in real time during surgery.
Smart Images
Figure CN122244199A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of cardiac electrophysiological imaging technology, and more specifically to a method for real-time constrained electrocardiogram imaging inversion using intraoperative mapping data. Background Technology
[0002] Electrocardiographic imaging (ECGI) is a key technique that non-invasively reconstructs the electrical activity on or inside the heart by using multi-lead electrical potential signals from the body surface, combined with individualized geometric and electrical models of the patient. This technique has shown significant clinical application potential in the localization diagnosis, mechanism analysis, and surgical planning of arrhythmias. Its core lies in solving an inverse problem of the forward transmission model of potentials from the cardiac source to the body surface. However, this inverse ECG problem is mathematically inherently pathological and ill-posed, making the inversion results highly susceptible to interference from various factors such as surface signal noise, electrode position errors, patient intraoperative respiration and positional changes, and differences between the anatomical model constructed from preoperative imaging and the actual intraoperative cardiac state. These factors can lead to localization deviations or morphological distortions in the reconstructed electrical activity, affecting the stability and clinical reliability of the results.
[0003] Intracardiac catheter mapping systems are typically used to acquire intraoperative local electrophysiological information, including the three-dimensional location of the mapping point, local activation time, local potential, and local electrogram waveform characteristics, providing a basis for arrhythmia mechanism analysis, lesion localization, and ablation decisions. This system can acquire electrophysiological information of local cardiac regions, such as local potential and local activation time, in real time and directly, while simultaneously recording the spatial position of the catheter. These intraoperative mapping data, due to their strong local realism and synchronization with the current cardiac rhythm, are considered one of the "gold standard" measurements of cardiac electrical activity. In existing technologies, the main use of this type of invasive mapping data is to construct high-precision intracardiac electroanatomical maps for postoperative result verification, or for offline comparison and registration with ECGI reconstruction results to evaluate the accuracy of non-invasive imaging. Furthermore, its real-time location and electrical signals are also frequently used independently for catheter navigation and manipulation.
[0004] Despite the significant value of intraoperative mapping data, effectively integrating it into the ECGI inversion framework remains a series of challenges. First, the mapping data itself exhibits heterogeneity in quality, such as differences in catheter contact stability and local signal-to-noise ratios, which can introduce errors if used indiscriminately. Second, accurately mapping discrete catheter mapping points onto the surface of an individualized cardiac anatomy model and handling local model deformations caused by cardiac activity, respiration, or catheter contact remains an unresolved registration and modeling problem. Crucially, current technical workflows lack a unified computational mechanism that can dynamically and online constrain and update the ECGI inversion process during surgery based on continuously added mapping points and their confidence levels. This results in a relative disconnect between the ECGI reconstruction process and real-time, high-quality invasive measurement data, failing to fully utilize the latter to promptly correct inversion uncertainties caused by model errors or the pathological nature of inverse problems. Summary of the Invention
[0005] To address the aforementioned problems in the existing technology, this invention provides a method for electrocardiogram imaging inversion using real-time constraints derived from intraoperative mapping data.
[0006] The technical problem to be solved by this invention is achieved through the following technical solution: In a first aspect, the present invention provides a method for electrocardiogram imaging inversion using real-time constraints of intraoperative mapping data, comprising: A personalized heart-trunk geometric model was established using medical images, and the forward transfer matrix from cardiac electrical activity to body surface potential was calculated based on the personalized heart-trunk geometric model. Acquire multi-lead potential signals from the body surface and initial intraoperative mapping data; The coordinate system of the initial intraoperative mapping data is registered to the coordinate system of the individualized heart-trunk geometry model to obtain the spatial mapping relationship; Generate a mapping weight matrix based on the data quality of the initial intraoperative mapping data; Based on the multi-lead potential signals of the body surface, the forward transfer matrix, the initial intraoperative mapping data, the spatial mapping relationship, and the mapping weight matrix, a joint inversion objective function is constructed. The newly added intraoperative mapping data is obtained and input into the joint inversion objective function for incremental solution to obtain the final cardiac electrical activity reconstruction results. The final cardiac electrical activity reconstruction results are visualized and processed to obtain the final inversion imaging results.
[0007] In a second aspect, the present invention provides an electrocardiogram imaging inversion device constrained by intraoperative mapping data in real time. The electrocardiogram imaging inversion device constrained by intraoperative mapping data in real time includes: a calculation unit, an acquisition unit, a mapping unit, a function construction unit, and an imaging unit. The computing unit is used to establish a personalized heart-trunk geometry model through medical images and to calculate the forward transfer matrix from cardiac electrical activity to body surface potential based on the personalized heart-trunk geometry model. The acquisition unit is used to acquire multi-lead potential signals on the body surface and initial intraoperative mapping data; The mapping unit is used to register the coordinate system of the initial intraoperative mapping data to the coordinate system of the individualized heart-trunk geometry model to obtain the spatial mapping relationship; The computing unit is also used to generate a mapping weight matrix based on the data quality of the initial intraoperative mapping data; The function construction unit is used to construct a joint inversion objective function based on multi-lead potential signals from the body surface, forward transfer matrix, initial intraoperative mapping data, spatial mapping relationship, and mapping weight matrix. The computing unit is also used to acquire new intraoperative mapping data, input the new intraoperative mapping data into the joint inversion objective function for incremental solution, and obtain the final cardiac electrical activity reconstruction results; The imaging unit is used to visualize the final cardiac electrical activity reconstruction results and obtain the final inversion imaging results.
[0008] Thirdly, the present invention provides an electrocardiogram imaging inversion device constrained by intraoperative mapping data in real time, comprising: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electrocardiogram imaging inversion device constrained by intraoperative mapping data in real time is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the electrocardiogram imaging inversion method constrained by intraoperative mapping data in real time as described in the first aspect above.
[0009] This invention provides a method for real-time constrained electrocardiogram (ECG) imaging inversion using intraoperative mapping data, comprising: establishing an individualized heart-trunk geometric model through medical images, and calculating the forward transfer matrix from cardiac electrical activity to body surface potentials based on the individualized heart-trunk geometric model; acquiring body surface multi-lead potential signals and initial intraoperative mapping data; registering the coordinate system of the initial intraoperative mapping data to the coordinate system of the individualized heart-trunk geometric model to obtain a spatial mapping relationship; generating a mapping weight matrix based on the data quality of the initial intraoperative mapping data; constructing a joint inversion objective function based on body surface multi-lead potential signals, the forward transfer matrix, the initial intraoperative mapping data, the spatial mapping relationship, and the mapping weight matrix; acquiring newly added intraoperative mapping data, inputting the newly added intraoperative mapping data into the joint inversion objective function for incremental solution, and obtaining the final cardiac electrical activity reconstruction result; and performing visualization imaging processing on the final cardiac electrical activity reconstruction result to obtain the final inversion imaging result. In this invention, firstly, a mapping weight matrix is generated based on the data quality of the initial intraoperative mapping data, addressing the potential error introduced by heterogeneity in mapping data quality, thereby distinguishing and weighting data of different quality. Secondly, by registering the coordinate system of the initial intraoperative mapping data to the coordinate system of an individualized heart-trunk geometric model, a spatial mapping relationship is obtained, solving the registration and modeling problems of mapping discrete mapping points to the surface of the heart model and handling local deformations caused by heartbeat, respiration, or catheter contact. Finally, by constructing a joint inversion objective function combining surface potential, forward transfer matrix, mapping data, mapping relationship, and weight matrix, and supporting incremental solution, the inversion process is dynamically updated when new intraoperative mapping data is acquired, addressing the lack of a dynamic online constraint mechanism and achieving immediate correction and fusion of ECGI inversion with real-time high-quality invasive measurement data. In summary, this invention achieves real-time, adaptive constraints of intraoperative mapping data on the ECGI inversion process, significantly reducing inversion uncertainty caused by model errors and the pathological nature of the inverse problem, and improving the accuracy and clinical applicability of electrocardiogram imaging.
[0010] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description
[0011] Figure 1 This is a flowchart illustrating an electrocardiogram imaging inversion method using real-time constraints of intraoperative mapping data, provided as an embodiment of the present invention. Figure 2 This is a schematic diagram of an electrocardiogram imaging inversion device that utilizes intraoperative mapping data for real-time constraint, provided in an embodiment of the present invention. Figure 3 This is a schematic diagram of an electrocardiogram imaging inversion device that utilizes intraoperative mapping data for real-time constraint, as provided in an embodiment of the present invention. Detailed Implementation
[0012] The purpose of this invention is to provide a real-time constrained ECG imaging inversion method using intraoperative mapping data, addressing the problems of existing ECGI inversion results relying primarily on surface potential signals, lacking sufficient local reliability, and being difficult to dynamically correct based on real intraoperative electrophysiological information. Specifically, by establishing a unified inversion framework, surface potential observation, preoperative individualized anatomical models, and intraoperative mapping information can jointly participate in the reconstruction of cardiac electrical activity. Furthermore, with the continuous addition of new mapping points, the ECGI results are updated incrementally in real time, thereby improving the accuracy, stability, timeliness, and clinical interpretability of the inversion results.
[0013] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.
[0014] To reduce inversion uncertainty caused by model errors and the ill-conditioned nature of inverse problems, and to improve the accuracy and clinical applicability of electrocardiogram (ECG) imaging, this invention provides an ECG imaging inversion method that utilizes real-time constraints from intraoperative mapping data. Figure 1 This is a flowchart illustrating a method for real-time constrained electrocardiogram imaging inversion using intraoperative mapping data, as provided in an embodiment of the present invention. Figure 1 As shown, it includes: S101. Establish an individualized heart-trunk geometric model using medical images, and calculate the forward transfer matrix from cardiac electrical activity to body surface potential based on the individualized heart-trunk geometric model.
[0015] Optionally, S101 includes: Establish individualized heart-trunk geometric models using medical imaging; Based on an individualized heart-trunk geometry model, the forward transfer matrix from cardiac electrical activity to body surface potential is generated using the boundary element method, finite element method, or meshless method.
[0016] The medical images can be CT, MRI, or other medical images. Specifically, firstly, medical image data of the heart and trunk of the subject or patient are acquired. The locations of the heart, trunk, and surface electrodes in the medical image data are reconstructed and their coordinates are unified to establish an individualized heart-trunk geometric model. Furthermore, the heart region is discretized into multiple nodes, surface elements, units, or sampling points, and a forward transfer matrix from cardiac electrical activity to surface potential is constructed using boundary element method, finite element method, meshless method, or a combination thereof.
[0017] S102. Acquire multi-lead potential signals on the body surface and initial intraoperative mapping data.
[0018] Optionally, both the initial intraoperative mapping data and the newly added intraoperative mapping data include at least one of the following: local activation time, local unipolar potential, local bipolar potential, or local electrogram waveform characteristics.
[0019] In addition, initial intraoperative mapping data may further include: contact force, contact quality index, mapping stability, local signal quality, low voltage zone label, ablation label or lesion area marker, etc.
[0020] S103. Register the coordinate system of the initial intraoperative mapping data to the coordinate system of the individualized heart-trunk geometric model to obtain the spatial mapping relationship.
[0021] Optionally, S103 includes: Based on coordinate registration algorithms and geometric mapping methods, the coordinate system of the initial intraoperative mapping data is registered to the coordinate system of the individualized heart-trunk geometric model to obtain the spatial mapping relationship; Coordinate registration algorithms include: iterative nearest point registration or dynamic compensation; Geometric mapping methods include: nearest neighbor mapping, barycenter coordinate interpolation, local surface projection, or projection matrix construction methods.
[0022] Specifically, in this implementation, a coordinate registration algorithm is first used to register the coordinate system of the initial intraoperative mapping data to the coordinate system of the individualized cardiac model, in order to establish the correspondence between the mapping points and discrete cardiac nodes, local surface regions, or volumetric state variables. The registration method can employ any of the following: rigid registration based on anatomical landmarks, iterative nearest-point registration based on surface point clouds, non-rigid correction based on the electroanatomical shell and the imaging model, or dynamic compensation combining respiratory phase, cardiac cycle, or catheter motion state. After registration, the spatial mapping relationship between the mapping observations and cardiac state variables is obtained through nearest-neighbor mapping, barycentric coordinate interpolation, local surface projection, surface interpolation, or projection matrix construction.
[0023] S104. Generate a mapping weight matrix based on the data quality of the initial intraoperative mapping data.
[0024] Optionally, S104 includes: A mapping weight matrix is generated based on at least one of the following: catheter contact force, catheter contact quality, catheter position stability, local electrogram signal-to-noise ratio, or registration residual, corresponding to the initial intraoperative mapping data.
[0025] In addition, the calibration weight matrix can be determined based on temporal consistency, consistency of repeated sampling, and consistency with the neighborhood propagation law. High-quality calibration points are given higher constraint strength, while low-quality calibration points are downweighted or removed to reduce the interference of low-confidence calibration information on the overall inversion results.
[0026] S105. Based on the multi-lead potential signals of the body surface, the forward transfer matrix, the initial intraoperative mapping data, the spatial mapping relationship, and the mapping weight matrix, a joint inversion objective function is constructed.
[0027] Optionally, S105 includes: Construct the main body of the objective function; the main body of the objective function should at least include a surface potential fitting term and an intraoperative mapping data constraint term; Based on the multi-lead potential signals of the body surface and the forward transfer matrix, a body surface potential fitting term is constructed. Based on the initial intraoperative mapping data, spatial mapping relationships, and mapping weight matrix, intraoperative mapping data constraint terms are constructed. The combined surface potential fitting term and the intraoperative mapping data constraint term are used to obtain the joint inversion objective function.
[0028] Furthermore, the objective function body in this embodiment of the invention can also be configured with a regularization term, a time smoothing term, or a state increment constraint term. This regularization term, time smoothing term, or state increment constraint term can be set based on the mathematical stability requirements of the inverse problem, prior physiological knowledge of cardiac electrical activity, and the time-series characteristics of multi-lead potential signals from the body surface.
[0029] At the mathematical solution level, let the multi-lead potential signal on the body surface be... The distribution of electrophysiological states inside the heart is as follows The forward transfer matrix is The initial intraoperative mapping data were The projection matrix (spatial mapping relationship) is: The joint inversion objective function of this invention It can be represented as: ; The ECGI reconstruction consists of four terms: First, a surface potential fitting term, ensuring consistency between the reconstructed results and measured surface signals; second, a regularization term, designed to mitigate ill-posedness in the inverse problem, which can employ zero-order Tikhonov, first- or second-order smoothing terms, graphical Laplace constraints, total variational constraints, sparsity constraints, propagation prior constraints, or physiological model constraints; third, an intraoperative mapping constraint term, ensuring consistency between the ECGI reconstruction results and locally measured electrophysiological information at the mapping points; and fourth, a time update term, maintaining continuity between reconstructed results at adjacent time points during real-time updates to prevent abrupt changes in the overall solution due to the addition of a small number of mapping points. This is the body surface lead weight matrix. This is a weight matrix (standardized weight matrix) constructed based on the standard confidence level. For time smoothing or state increment matrix, , , These are the first weight parameter, the second weight parameter, and the third weight parameter, respectively.
[0030] It should be noted that this invention is applicable to the reconstruction of various cardiac state quantities. The state to be reconstructed can be epicardial potential, endocardial potential, myocardial cell transmembrane potential, or local activation time field, conduction delay distribution, transmembrane potential-related state quantities, or intermediate states projected from a cardiac in vivo dynamic model. For LAT (Local Activation Time) type constraints, intraoperative mapping values can directly constrain the activation time field; for unipolar or bipolar potential type constraints, constraint terms can be constructed based on waveform characteristics, peak amplitude, slope characteristics, or local electrogram similarity; for ablated areas, scar areas, or low-voltage areas, they can also be used as regional priors to limit the conduction velocity, amplitude, or waveform characteristics of the corresponding areas, thereby improving the ability to identify complex lesion structures.
[0031] S106. Obtain the newly added intraoperative mapping data, input the newly added intraoperative mapping data into the joint inversion objective function for incremental solution, and obtain the final cardiac electrical activity reconstruction result.
[0032] Optionally, S106 includes: The newly added intraoperative mapping data were acquired, and based on the individualized heart-trunk geometry model and spatial mapping relationship, the new correspondence of the newly added intraoperative mapping data in the individualized heart-trunk geometry model was determined. New weighting information is generated based on the data quality of the newly added intraoperative mapping data; The newly added intraoperative mapping data, the newly added correspondence, and the newly added weight information are added to the intraoperative mapping data constraint term of the joint inversion objective function to obtain the updated joint inversion objective function; The final cardiac electrical activity reconstruction result is obtained by solving the updated joint inversion objective function.
[0033] Optionally, the final cardiac electrical activity reconstruction results are obtained by solving the updated joint inversion objective function, including: Based on the updated joint inversion objective function, the recursive least squares method, incremental ADMM algorithm, Kalman filter or rolling time window optimization are used to solve the problem and obtain the final cardiac electrical activity reconstruction results.
[0034] In addition, in this embodiment, local strong constraint priority update can be performed for the neighborhood of the newly added calibration point, while global smooth consistency correction is performed for the region far away from the newly added calibration point, so as to take into account both real-time performance and the continuous stability of the overall solution.
[0035] S107. Visualize the final cardiac electrical activity reconstruction results to obtain the final inversion imaging results.
[0036] It should be noted that the visualization imaging processing method in the embodiments of the present invention is not limited to: three-dimensional rendering and interaction, isopotential diagram or isochronous diagram method.
[0037] Through the above technical solutions, this invention forms a unified and integrated ECGI inversion framework that integrates "preoperative model + surface observation + intraoperative mapping + dynamic updating". Compared with traditional ECGI methods that rely solely on surface signals, this method can significantly improve inversion accuracy in local areas, making the reconstruction results closer to the actual intraoperative electrophysiological state. Compared with methods that use mapping information only for preoperative prediction or postoperative verification, this invention can achieve online correction and continuous updating during the operation, making it more suitable for serving real-time ablation decisions. Compared with methods that use all mapping data equally, this invention reduces the adverse impact of low-quality mapping points on the overall inversion results through mapping confidence weighting and local update mechanisms, improving the robustness and reliability of the results. Because this invention is compatible with multiple intraoperative observation types and multiple ECGI state quantities, it has strong versatility and scalability, and can directly serve various clinical electrophysiological scenarios such as ventricular fibrillation, atrial fibrillation, premature ventricular contractions, and ventricular tachycardia.
[0038] Furthermore, the present invention is not limited to the embodiments described above. In some embodiments, intraoperative mapping data can come from different brands of electroanatomical mapping systems, or from contact mapping, non-contact mapping, or multimodal electrophysiological measurements. Coordinate registration methods can employ surface point cloud registration, marker-based registration, or dynamic registration combining respiratory and cardiac compensation. Mapping constraints can apply to the cardiac surface state or to the myocardial volume state through a projection matrix. The inversion solver can use traditional regularized least squares methods, or it can be implemented using ADMM, extended Kalman filtering, Bayesian estimation, or physical information neural networks. Body surface observations can be high-density body surface potentials or body surface potential signals with fewer leads. The output results can be cardiac electrograms, activation time maps, local confidence maps, abnormal origin probability maps, or intraoperative navigation reference information.
[0039] This invention provides a method for ECG imaging inversion using real-time constraints based on intraoperative mapping data. First, a mapping weight matrix is generated based on the data quality of the initial intraoperative mapping data, addressing the potential error introduced by heterogeneity in mapping data quality, thereby distinguishing and weighting data of different quality. Then, by registering the coordinate system of the initial intraoperative mapping data to the coordinate system of an individualized heart-trunk geometric model, a spatial mapping relationship is obtained, solving the registration and modeling problems of mapping discrete mapping points to the surface of the heart model and handling local deformations caused by heartbeat, respiration, or catheter contact. Finally, a joint inversion objective function combining surface potential, forward transfer matrix, mapping data, mapping relationship, and weight matrix is constructed, supporting incremental solution. The inversion process is dynamically updated when new intraoperative mapping data is acquired, addressing the lack of a dynamic online constraint mechanism and achieving instantaneous correction and fusion of ECGI inversion with real-time high-quality invasive measurement data. In summary, this invention enables real-time and adaptive constraints of intraoperative mapping data on the ECGI inversion process, significantly reducing inversion uncertainty caused by model errors and the pathological nature of inverse problems, and improving the accuracy and clinical applicability of electrocardiogram imaging.
[0040] The method provided in this embodiment of the invention can be applied to electronic devices. Specifically, the electronic device can be a desktop computer, a portable computer, a smart mobile terminal, a server, etc., and this embodiment of the invention does not limit the application to such devices.
[0041] Based on the same inventive concept, embodiments of the present invention also provide an electrocardiogram imaging inversion device that utilizes intraoperative mapping data for real-time constraint. Figure 2 This is a schematic diagram of a cardiac imaging inversion device that utilizes intraoperative mapping data for real-time constraint, as provided in an embodiment of the present invention. Figure 2 As shown, it includes: a calculation unit 201, an acquisition unit 202, a mapping unit 203, a function construction unit 204, and an imaging unit 205; The computing unit 201 is used to establish an individualized heart-trunk geometric model through medical images and to calculate the forward transfer matrix from cardiac electrical activity to body surface potential based on the individualized heart-trunk geometric model. Acquisition unit 202 is used to acquire multi-lead potential signals on the body surface and initial intraoperative mapping data; The mapping unit 203 is used to register the coordinate system of the initial intraoperative mapping data to the coordinate system of the individualized heart-trunk geometric model to obtain the spatial mapping relationship; The calculation unit 201 is also used to generate a mapping weight matrix based on the data quality of the initial intraoperative mapping data; Function construction unit 204 is used to construct a joint inversion objective function based on multi-lead potential signals from the body surface, forward transfer matrix, initial intraoperative mapping data, spatial mapping relationship and mapping weight matrix; The computing unit 201 is also used to acquire the newly added intraoperative mapping data, input the newly added intraoperative mapping data into the joint inversion objective function for incremental solution, and obtain the final cardiac electrical activity reconstruction result; Imaging unit 205 is used to perform visualization imaging processing on the final cardiac electrical activity reconstruction results to obtain the final inversion imaging results.
[0042] Figure 3 This is a schematic diagram of a real-time constrained electrocardiogram (ECG) imaging inversion device based on intraoperative mapping data, provided in an embodiment of the present invention. The device includes a processor 310, a storage medium 320, and a bus 330. The storage medium 320 stores machine-readable instructions executable by the processor 310. When the ECG imaging inversion device is running, the processor 310 communicates with the storage medium 320 via the bus 330. The processor 310 executes the machine-readable instructions to perform the steps of the above-described method embodiment. Specific implementation methods and technical effects are similar and will not be described in detail here.
[0043] The storage medium may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the storage medium may also be at least one storage device located remotely from the aforementioned processor.
[0044] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.
[0045] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, those skilled in the art can combine and integrate the different embodiments or examples described in this specification.
[0046] Although the invention has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings and the disclosure, will understand and implement other variations of the disclosed embodiments in carrying out the claimed invention. In this description, the word "comprising" does not exclude other components or steps, "a" or "an" does not exclude a plurality, and "a plurality" means two or more, unless otherwise explicitly specified. Furthermore, while different embodiments may describe certain measures, this does not mean that these measures cannot be combined to produce good results.
[0047] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the inventive concept, and all such modifications and substitutions should be considered within the scope of protection of the present invention.
Claims
1. A method for electrocardiogram imaging inversion using real-time constraints of intraoperative mapping data, characterized in that, include: A personalized heart-trunk geometric model is established using medical images, and the forward transfer matrix from cardiac electrical activity to body surface potential is calculated based on the personalized heart-trunk geometric model. Acquire multi-lead potential signals from the body surface and initial intraoperative mapping data; The coordinate system of the initial intraoperative mapping data is registered to the coordinate system of the individualized heart-trunk geometry model to obtain the spatial mapping relationship; A mapping weight matrix is generated based on the data quality of the initial intraoperative mapping data; Based on the multi-lead potential signal of the body surface, the forward transfer matrix, the initial intraoperative mapping data, the spatial mapping relationship, and the mapping weight matrix, a joint inversion objective function is constructed; The newly added intraoperative mapping data is obtained, and the newly added intraoperative mapping data is input into the joint inversion objective function for incremental solution to obtain the final cardiac electrical activity reconstruction result; The final cardiac electrical activity reconstruction results are then visualized using imaging processing to obtain the final inversion imaging results.
2. The method for real-time constrained electrocardiogram imaging inversion using intraoperative mapping data as described in claim 1, characterized in that, Both the initial intraoperative mapping data and the newly added intraoperative mapping data include at least one of the following: local activation time, local unipolar potential, local bipolar potential, or local electrogram waveform characteristics.
3. The method for real-time constrained electrocardiogram imaging inversion using intraoperative mapping data as described in claim 1, characterized in that, The process of registering the coordinate system of the initial intraoperative mapping data to the coordinate system of the individualized heart-trunk geometry model to obtain the spatial mapping relationship includes: Based on coordinate registration algorithms and geometric mapping methods, the coordinate system of the initial intraoperative mapping data is registered to the coordinate system of the individualized heart-trunk geometric model to obtain the spatial mapping relationship; The coordinate registration algorithm includes: iterative nearest point registration or dynamic compensation; The geometric mapping methods include: nearest neighbor mapping, barycenter coordinate interpolation, local surface projection, or projection matrix construction methods.
4. The method for real-time constrained electrocardiogram imaging inversion using intraoperative mapping data as described in claim 1, characterized in that, The generation of the mapping weight matrix based on the data quality of the initial intraoperative mapping data includes: The mapping weight matrix is generated based on at least one of the following: catheter contact force, catheter contact quality, catheter position stability, local electrogram signal-to-noise ratio, or registration residual, corresponding to the initial intraoperative mapping data.
5. The method for real-time constrained electrocardiogram imaging inversion using intraoperative mapping data according to claim 1, characterized in that, The process of establishing a personalized heart-trunk geometry model using medical images and calculating the forward transmission matrix from cardiac electrical activity to body surface potential based on the personalized heart-trunk geometry model includes: The individualized heart-trunk geometric model was established using medical imaging. Based on the individualized heart-trunk geometry model, the forward transfer matrix from cardiac electrical activity to body surface potential is generated using the boundary element method, finite element method, or meshless method.
6. The method for real-time constrained electrocardiogram imaging inversion using intraoperative mapping data as described in claim 1, characterized in that, The joint inversion objective function is constructed based on the multi-lead potential signal from the body surface, the forward transfer matrix, the initial intraoperative mapping data, the spatial mapping relationship, and the mapping weight matrix, including: Construct the main body of the objective function; the main body of the objective function includes at least a surface potential fitting term and an intraoperative mapping data constraint term. Based on the multi-lead potential signal of the body surface and the forward transfer matrix, the body surface potential fitting term is constructed. Based on the initial intraoperative mapping data, the spatial mapping relationship, and the mapping weight matrix, the intraoperative mapping data constraint terms are constructed. The joint inversion objective function is obtained by combining the surface potential fitting term with the intraoperative mapping data constraint term.
7. The method for real-time constrained electrocardiogram imaging inversion using intraoperative mapping data according to claim 6, characterized in that, The process of acquiring newly added intraoperative mapping data, inputting the newly added intraoperative mapping data into the joint inversion objective function for incremental solution, and obtaining the final cardiac electrical activity reconstruction result includes: The newly added intraoperative mapping data is obtained, and based on the individualized heart-trunk geometry model and the spatial mapping relationship, the new correspondence of the newly added intraoperative mapping data in the individualized heart-trunk geometry model is determined. New weighting information is generated based on the data quality of the newly added intraoperative mapping data; The newly added intraoperative mapping data, the newly added correspondence, and the newly added weight information are added to the intraoperative mapping data constraint term of the joint inversion objective function to obtain the updated joint inversion objective function; The final cardiac electrical activity reconstruction result is obtained by solving the updated joint inversion objective function.
8. The method for real-time constrained electrocardiogram imaging inversion using intraoperative mapping data according to claim 7, characterized in that, The calculation based on the updated joint inversion objective function yields the final cardiac electrical activity reconstruction result, including: Based on the updated joint inversion objective function, the final cardiac electrical activity reconstruction result is obtained by solving the problem using recursive least squares method, incremental ADMM algorithm, Kalman filter or rolling time window optimization.
9. A device for real-time constrained electrocardiogram imaging inversion using intraoperative mapping data, characterized in that, The ECG imaging inversion device that utilizes intraoperative mapping data for real-time constraint includes: a calculation unit, an acquisition unit, a mapping unit, a function construction unit, and an imaging unit; The computing unit is used to establish an individualized heart-trunk geometric model through medical images, and to calculate the forward transfer matrix from cardiac electrical activity to body surface potential based on the individualized heart-trunk geometric model. The acquisition unit is used to acquire multi-lead potential signals on the body surface and initial intraoperative mapping data; The mapping unit is used to register the coordinate system of the initial intraoperative mapping data to the coordinate system of the individualized heart-trunk geometric model to obtain a spatial mapping relationship; The computing unit is also used to generate a mapping weight matrix based on the data quality of the initial intraoperative mapping data; The function construction unit is used to construct a joint inversion objective function based on the multi-lead potential signal of the body surface, the forward transfer matrix, the initial intraoperative mapping data, the spatial mapping relationship, and the mapping weight matrix. The calculation unit is also used to acquire newly added intraoperative mapping data, input the newly added intraoperative mapping data into the joint inversion objective function for incremental solution, and obtain the final cardiac electrical activity reconstruction result. The imaging unit is used to perform visualization imaging processing on the final cardiac electrical activity reconstruction results to obtain the final inversion imaging results.
10. An electrocardiogram imaging inversion device that utilizes intraoperative mapping data for real-time constraint, characterized in that, include: The device includes a processor, a storage medium, and a bus. The storage medium stores machine-readable instructions executable by the processor. When the ECG imaging inversion device using intraoperative mapping data with real-time constraints is running, the processor communicates with the storage medium via the bus. The processor executes the machine-readable instructions to perform the steps of the ECG imaging inversion method using intraoperative mapping data with real-time constraints as described in any one of claims 1-8.