Ultrasound-guided three-dimensional modeling system and method in heart cavity

Abstract

The invention discloses an intracardiac ultrasound-guided three-dimensional modeling system and method, relates to the technical field of medical imaging and three-dimensional modeling, and is used for solving the problems of insufficient dynamic target modeling precision and poor robustness to a complex acoustic window environment. Based on a built-in acceleration sensor, probe speed and displacement parameters are calculated in real time, sampling frequency analysis and spectrum feature extraction are carried out on ultrasonic signals, tissue motion attributes are reflected, on the basis of frequency domain processing, spectrum components are analyzed, and dynamic and static components are identified. And constructing a modeling mode selection standard value based on the difference value of the comprehensive dynamic index and the sampling frequency coefficient, and dynamically switching the modeling mode according to a preset rule to adapt to model construction requirements under different tissue states, thereby optimizing a model updating process and realizing dynamic regulation and control and error suppression.

Description

Technical Field

[0001] The present invention relates to the technical field of medical imaging and three-dimensional modeling, and more particularly to an intracardiac ultrasound-guided three-dimensional modeling system and method. Background Art

[0002] 3D modeling technology, an advanced tool based on medical imaging, has attracted considerable attention in the medical field for its intuitiveness and precision. In particular, 3D modeling under intracardiac ultrasound guidance can provide physicians with detailed anatomical information, assisting with surgical planning and navigation. However, existing 3D modeling technology has limitations in adapting to the complex intracardiac environment, ensuring accurate data acquisition, and enabling real-time modeling, hindering its widespread adoption in clinical practice.

[0003] The existing technology has the following deficiencies:

[0004] Currently, existing 3D modeling technologies mainly rely on the stacking or voxel interpolation of 2D ICE images. The rapid motion characteristics of intracardiac tissue structures lead to insufficient accuracy in dynamic target modeling and poor robustness to complex acoustic window environments. This makes it difficult to meet the real-time requirements for modeling results during surgery. Therefore, an intracardiac ultrasound-guided 3D modeling system and method are proposed.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an intracardiac ultrasound-guided three-dimensional modeling system and method. By performing multi-dimensional decomposition of the ultrasound signal and extracting dynamic features during the data acquisition stage, a high-precision three-dimensional model is generated in combination with a real-time modeling algorithm. At the same time, the model update process is optimized by utilizing the time correlation between adjacent frames, thereby solving the problems mentioned in the background technology such as poor adaptability to dynamic targets, insufficient modeling accuracy, and low robustness in complex environments.

[0007] To achieve the above objectives, the present invention provides the following technical solution: an intracardiac ultrasound-guided three-dimensional modeling system and method, comprising the following steps:

[0008] Step S1: Before receiving the ultrasound signal, the relative motion state between the ultrasound probe and the target tissue is detected, the sampling frequency and signal intensity distribution of the ultrasound signal are obtained, the ultrasound signal is divided into multiple time windows, and the signal in each time window is subjected to frequency domain analysis, and the spectrum characteristics in each window are recorded;

[0009] Step S2: extract the dynamic and static components of the signal based on the spectral characteristics, set a dynamic threshold to filter the dynamic components, evaluate the comprehensive dynamic index of the signal, and select a suitable modeling mode based on the sampling frequency and the comprehensive dynamic index;

[0010] Step S3: Before each model update, the ultrasound signal is processed according to the current modeling mode, the historical database is accessed to obtain the historical signal spectrum data under the corresponding mode, the standard spectrum interval is calculated and the model quality threshold is set, the deviation coefficient between the current signal spectrum and the standard spectrum interval is collected, and the model quality threshold and the deviation coefficient are compared to determine whether the model is abnormal and mark the abnormality;

[0011] Step S4: Detect the changing trend of the signal spectrum in adjacent time windows, calculate the spectrum change rate, and adjust the quality threshold of the current modeling mode according to the spectrum change rate to ensure that the model can adapt to the dynamic changes of the signal.

[0012] In a preferred embodiment, in step S1, the ultrasonic probe is a device for transmitting and receiving ultrasonic waves, and the displacement and velocity information of the probe is obtained through a built-in acceleration sensor, and the sampling frequency of the ultrasonic signal is analyzed in combination with the signal intensity distribution;

[0013] Frequency domain analysis uses the fast Fourier transform algorithm to convert the signal within the time window from the time domain to the frequency domain and extract the spectrum characteristics; the length of the time window is 50ms to 100ms; the spectrum characteristics include the main frequency, bandwidth and energy distribution.

[0014] In a preferred embodiment, in step S2, the specific steps of extracting the dynamic component and the static component are as follows:

[0015] The main frequency and bandwidth in the spectrum feature are used as input to construct a spectrum feature matrix. The mean and variance of each element in the matrix are calculated. The variance is multiplied by the preset variance ratio coefficient and then summed with the mean value to obtain the result as the dynamic threshold.

[0016] When the main frequency or bandwidth in the spectrum feature exceeds the dynamic threshold, the spectrum feature is judged to belong to the dynamic component;

[0017] Otherwise, it is judged to be a static component;

[0018] When calculating the comprehensive dynamic index, the ratio of the spectral characteristics of the dynamic component to the dynamic threshold is used as the weight of a single dynamic component, and the average of all dynamic component weights is used as the comprehensive dynamic index.

[0019] In a preferred embodiment, in step S2, the sampling frequency is normalized using a normalization algorithm, the normalized result is used as the sampling frequency coefficient, and the difference between the sampling frequency coefficient and the comprehensive dynamic index is used as the standard value for selecting the modeling mode;

[0020] When the selection criterion value of the modeling mode is less than 0, the high-precision modeling mode is selected; when the selection criterion value of the modeling mode is greater than 0 and less than the preset stability threshold, the balanced modeling mode is selected;

[0021] When the selection criterion value of the modeling mode exceeds the preset stability threshold, the fast modeling mode is selected.

[0022] In a preferred embodiment, in step S3, when processing the ultrasonic signal according to the modeling mode, if the current mode is the high-precision modeling mode, the ultrasonic signal is subjected to full-band analysis and all spectral features are retained;

[0023] If the current mode is the fast modeling mode, only the spectral features of the dynamic components in the ultrasonic signal are extracted; if the current mode is the balanced modeling mode, the ultrasonic signal is segmented and the spectral features of the high frequency band and the low frequency band are extracted respectively.

[0024] In a preferred embodiment, in step S3, the historical database is accessed to obtain the historical signal spectrum data in the corresponding mode and merged into a historical spectrum data set. The support vector machine classification algorithm is used to set the model quality threshold and calculate the standard spectrum interval. The specific steps are as follows:

[0025] Set the classification benchmark and use the mean of the main frequency in the historical spectrum data set as the classification benchmark;

[0026] Positive and negative sample labeling: Data in the historical spectrum dataset that exceeds the classification benchmark is labeled as a positive sample, and data that is below the classification benchmark is labeled as a negative sample;

[0027] True and false classification: positive samples and negative samples exceeding the preset spectrum judgment threshold are marked as true samples, and positive samples and negative samples below the preset spectrum judgment threshold are marked as false samples;

[0028] Calculate the classification rate: Calculate the true sample rate and false sample rate using the classification rate formula. For the classification rate increment, set M different spectrum judgment thresholds to obtain multiple true sample rates and false sample rates. Subtract the obtained true sample rates from large to small to obtain M-1 true sample rate differences. Subtract the obtained false sample rates from large to small to obtain M-1 false sample rate differences.

[0029] Calculate the model quality threshold: Calculate the model quality threshold using the obtained true sample rate difference and false sample rate difference;

[0030] Calculate the standard spectrum interval: select the maximum and minimum values ​​of the true sample rate and the false sample rate respectively, use the maximum value of the true sample rate and the false sample rate to calculate the upper interval of the standard spectrum, use the minimum value of the true sample rate and the false sample rate to calculate the lower interval of the standard spectrum, and use the range between the upper interval of the standard spectrum and the lower interval of the standard spectrum as the standard spectrum interval.

[0031] In a preferred embodiment, in step S3, when determining whether the model is abnormal, the current signal spectrum is first compared with the standard spectrum interval. If the current signal spectrum is within the standard spectrum interval, it is determined that the model is not abnormal.

[0032] If the current signal spectrum is outside the standard spectrum range, the deviation coefficient is obtained by subtracting the center value of the current signal spectrum from the center value of the standard spectrum range;

[0033] When the deviation coefficient is lower than the model quality threshold, the model is judged to be normal;

[0034] When the deviation coefficient exceeds the model quality threshold, the model is judged to be abnormal and marked as abnormal.

[0035] In a preferred embodiment, in step S4, the changing trend of the signal spectrum in adjacent time windows is detected, the spectrum characteristics of the current time window are compared with the spectrum characteristics of the previous time window, and the spectrum change rate is calculated;

[0036] The formula for calculating the spectrum change rate is the difference between the main frequency of the current time window and the main frequency of the previous time window divided by the main frequency of the previous time window;

[0037] If the spectrum change rate exceeds the preset evaluation threshold, the quality threshold of the current modeling mode is adjusted based on the modeling mode and spectrum change rate of the adjacent time window. The specific steps are as follows:

[0038] If the current modeling mode is high-precision modeling mode or fast modeling mode, and is the same as the modeling mode of the previous time window, the quality threshold of the current modeling mode will not be adjusted;

[0039] If the current modeling mode is the balanced modeling mode and is different from the modeling mode of the previous time window, the product of the quality threshold in the current modeling mode and the spectrum change rate is used as the adjusted quality threshold to re-judge whether the model is abnormal.

[0040] An intracardiac ultrasound-guided three-dimensional modeling system includes a signal acquisition module, a mode selection module, an anomaly detection module, and a quality optimization module;

[0041] The signal acquisition module is used to detect the relative motion state between the ultrasound probe and the target tissue and to collect the sampling frequency and spectrum characteristics of the ultrasound signal;

[0042] The mode selection module is used to extract dynamic and static components based on spectrum characteristics, set dynamic thresholds to filter dynamic components, and select the current modeling mode after evaluating the comprehensive dynamic index;

[0043] The anomaly detection module calculates the standard spectrum interval based on historical signal spectrum data and sets the model quality threshold. It then determines whether the model has an anomaly based on the deviation coefficient between the current signal spectrum and the standard spectrum interval.

[0044] The quality optimization module adjusts the quality threshold of the current modeling mode according to the modeling mode and spectrum change rate of adjacent time windows.

[0045] In an intracardiac ultrasound-guided three-dimensional modeling system, a signal acquisition module obtains probe displacement and velocity information through a built-in acceleration sensor and analyzes the sampling frequency of the ultrasound signal in combination with the signal intensity distribution;

[0046] The mode selection module calculates the dynamic threshold and evaluates the comprehensive dynamic index through the spectrum feature matrix;

[0047] The anomaly detection module uses the support vector machine classification algorithm to generate standard spectrum intervals and calculate the model quality threshold;

[0048] The quality optimization module adjusts the quality threshold of the current modeling mode according to the spectrum change rate.

[0049] The technical effects and advantages of the present invention are as follows:

[0050] The present invention obtains the relative motion state information between the ultrasonic probe and the target tissue, calculates the velocity and displacement parameters of the probe in real time based on the built-in acceleration sensor, performs sampling frequency analysis and spectrum feature extraction on the ultrasonic signal, wherein the spectrum features include main frequency and bandwidth indicators to reflect the tissue motion properties contained in the signal, and performs component analysis on the extracted spectrum features on the basis of frequency domain signal processing to identify the dynamic components and static components contained in the signal, and constructs a modeling mode selection standard value based on the difference between the comprehensive dynamic index and the sampling frequency coefficient, and dynamically switches the current modeling mode according to preset rules to adapt to the model construction requirements under different tissue states, thereby optimizing the model update process and realizing dynamic regulation and error suppression of the modeling process. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a schematic diagram of the module structure of an intracardiac ultrasound-guided three-dimensional modeling system of the present invention.

[0052] Figure 2 This is a flow chart of intracardiac ultrasound signal processing in an intracardiac ultrasound-guided three-dimensional modeling method of the present invention.

[0053] Figure 3This is a schematic diagram of the calculation of the deviation coefficient between the standard spectrum interval and the current signal spectrum in an intracardiac ultrasound-guided three-dimensional modeling method of the present invention. DETAILED DESCRIPTION

[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0055] Example 1

[0056] The present invention provides an intracardiac ultrasound guided three-dimensional modeling system and method, the specific implementation of which is combined with Figures 1 to 3 Provide detailed explanation; Figure 1 A schematic diagram of the system's module structure is shown, including the signal acquisition module, mode selection module, anomaly detection module, and quality optimization module.

[0057] These modules are interconnected through data flow and logic control, and together complete the entire process from ultrasonic signal acquisition to three-dimensional modeling.

[0058] The signal acquisition module is responsible for acquiring the relative motion state between the ultrasound probe and the target tissue and extracting the sampling frequency and spectrum characteristics of the ultrasound signal;

[0059] The mode selection module separates the dynamic component and the static component according to the spectrum characteristics and selects the current modeling mode;

[0060] The anomaly detection module calculates the standard spectrum interval based on historical data and sets the model quality threshold to determine whether the model has anomalies;

[0061] The quality optimization module adjusts the quality threshold based on the modeling patterns and spectrum change rates of adjacent time windows to optimize the model updating process.

[0062] During actual operation, the signal acquisition module first detects the relative motion state between the ultrasound probe and the target tissue. This process relies on the built-in acceleration sensor to obtain the displacement and velocity information of the probe.

[0063] At the same time, the signal acquisition module analyzes the sampling frequency and signal intensity distribution of the ultrasonic signal and divides the ultrasonic signal into multiple time windows. The length of each time window is usually set to 50ms to 100ms to balance resolution and computational efficiency.

[0064] The sampling frequency acquisition logic is to obtain the actual sampling timing based on the internal clock control mechanism of the ultrasound system. By measuring the time interval between two consecutive sampling points and calculating the number of data points collected per unit time, the current sampling frequency is determined.

[0065] In each time window, the fast Fourier transform (FFT) algorithm is used to convert the time domain signal into the frequency domain signal and extract the spectrum characteristics, including the main frequency and bandwidth parameters.

[0066] It should be noted that the spectral characteristics include but are not limited to the main frequency, spectral bandwidth, and spectral energy density distribution. The specific parameters are set by the experimenters based on the motion characteristics of the target tissue, the signal-to-noise ratio of the ultrasound signal, and the frequency stability.

[0067] These spectral features are then passed to the mode selection module for further processing.

[0068] The core function of the mode selection module is to separate the dynamic component from the static component and evaluate the comprehensive dynamic index based on the separation results to select the current modeling mode.

[0069] Specifically, the mode selection module takes the main frequency and bandwidth in the spectrum feature as input, constructs the spectrum feature matrix, and calculates the mean and variance of each element in the matrix.

[0070] The result obtained by multiplying the variance by the preset variance ratio coefficient and then summing the result with the average value is used as the dynamic threshold.

[0071] When the main frequency or bandwidth in the spectrum feature exceeds the dynamic threshold, the spectrum feature is judged to belong to the dynamic component; otherwise, it is judged to belong to the static component.

[0072] To evaluate the comprehensive dynamic index, the mode selection module uses the ratio of the spectral characteristics of the dynamic component to the dynamic threshold as the weight of the single dynamic component and takes the average of all dynamic component weights as the comprehensive dynamic index.

[0073] In addition, the mode selection module also uses a normalization algorithm to standardize the sampling frequency, uses the normalized result as the sampling frequency coefficient, and uses the difference between the sampling frequency coefficient and the comprehensive dynamic index as the standard value for selecting the modeling mode.

[0074] When the selection criterion value of the modeling mode is less than 0, the high-precision modeling mode is selected;

[0075] When the selection criterion value is greater than 0 and less than the preset stability threshold, the equilibrium modeling mode is selected;

[0076] When the selection criterion value exceeds the preset stability threshold, the fast modeling mode is selected.

[0077] It should be noted that the preset stability threshold was obtained by our experimenters based on the statistical interval of typical intracardiac tissue motion rate and the ultrasound signal sampling stability deviation model, which will not be elaborated here;

[0078] The choice of different modeling modes directly affects the subsequent signal processing strategy, thereby adapting to different modeling needs.

[0079] The function of the anomaly detection module is to obtain the historical signal spectrum data under the corresponding mode through the historical database, and calculate the standard spectrum interval and model quality threshold to determine whether the model has an anomaly.

[0080] Specifically, the anomaly detection module first accesses the historical database, obtains the historical signal spectrum data under the corresponding mode and merges them into a historical spectrum dataset.

[0081] Subsequently, the historical spectrum dataset is processed using the support vector machine classification algorithm.

[0082] When setting the classification benchmark, the mean of the main frequency in the historical spectrum dataset is used as the classification benchmark, and the data in the historical spectrum dataset that exceeds the classification benchmark is marked as a positive sample, and the data that is lower than the classification benchmark is marked as a negative sample.

[0083] Then, the positive and negative samples are classified as true or false by the preset spectrum judgment threshold. The positive and negative samples exceeding the spectrum judgment threshold are marked as true samples, and the positive and negative samples below the spectrum judgment threshold are marked as false samples.

[0084] The true sample rate and false sample rate are calculated respectively by the classification rate formula, and M different spectrum judgment thresholds are set to obtain multiple true sample rates and false sample rates.

[0085] The obtained true sample rates are sequentially subtracted from large to small to obtain M-1 true sample rate differences, and the obtained false sample rates are sequentially subtracted from large to small to obtain M-1 false sample rate differences.

[0086] The model quality threshold is calculated using the obtained true sample rate difference and false sample rate difference, and the maximum and minimum values ​​of the true sample rate and false sample rate are selected respectively. The maximum value of the true sample rate and false sample rate is used to calculate the upper interval of the standard spectrum, and the minimum value of the true sample rate and false sample rate is used to calculate the lower interval of the standard spectrum.

[0087] Finally, the range between the upper interval of the standard spectrum and the lower interval of the standard spectrum is taken as the standard spectrum interval.

[0088] When determining whether the model is abnormal, the anomaly detection module compares the current signal spectrum with the standard spectrum interval.

[0089] If the current signal spectrum is within the standard spectrum range, the model is judged to be normal;

[0090] If the current signal spectrum is outside the standard spectrum interval, the deviation coefficient is obtained by subtracting the center value of the current signal spectrum from the center value of the standard spectrum interval.

[0091] If the deviation coefficient is lower than the model quality threshold, the model is judged to be normal;

[0092] If the deviation coefficient exceeds the model quality threshold, the model is judged to be abnormal and marked as abnormal.

[0093] The role of the quality optimization module is to adjust the quality threshold of the current modeling mode according to the modeling mode and spectrum change rate of adjacent time windows to ensure that the model can adapt to the dynamic changes of the signal.

[0094] Specifically, the quality optimization module first detects the changing trend of the signal spectrum in adjacent time windows, compares the spectrum characteristics of the current time window with the spectrum characteristics of the previous time window, and calculates the spectrum change rate.

[0095] The spectrum change rate is calculated as the difference between the main frequency of the current time window and the main frequency of the previous time window divided by the main frequency of the previous time window.

[0096] If the spectrum change rate exceeds a preset evaluation threshold, the quality threshold of the current modeling mode is adjusted in combination with the modeling mode and spectrum change rate of the adjacent time window.

[0097] The specific adjustment steps are as follows:

[0098] If the current modeling mode is high-precision modeling mode or fast modeling mode, and is the same as the modeling mode of the previous time window, the quality threshold of the current modeling mode will not be adjusted;

[0099] If the current modeling mode is the balanced modeling mode and is different from the modeling mode of the previous time window, the product of the quality threshold in the current modeling mode and the spectrum change rate is used as the adjusted quality threshold, and the model is re-judged to see if there is any abnormality.

[0100] Figure 2 The overall process of intracardiac ultrasound signal processing is further demonstrated.

[0101] Starting from the signal acquisition module, the ultrasonic signal is analyzed in the frequency domain to extract the spectrum characteristics, and then passed to the mode selection module to separate the dynamic component and the static component, and select the current modeling mode.

[0102] Subsequently, the anomaly detection module calculates the standard spectrum interval through historical data and sets the model quality threshold to determine whether the model has anomalies.

[0103] Finally, the quality optimization module adjusts the quality threshold according to the modeling mode and spectrum change rate of adjacent time windows to complete the model update process.

[0104] The entire process is closely connected, ensuring the efficient operation of the system.

[0105] Figure 3 The calculation process of the deviation coefficient between the standard spectrum interval and the current signal spectrum is described in detail.

[0106] After generating the standard spectrum interval through historical spectrum data, the deviation coefficient is obtained by subtracting the current signal spectrum from the center value of the standard spectrum interval, and the model is judged based on the deviation coefficient to see whether there is any abnormality.

[0107] This process intuitively reflects the working principle of the anomaly detection module.

[0108] The above is the specific implementation method of this design, covering the complete process of signal acquisition, mode selection, anomaly detection and quality optimization.

[0109] Through the collaborative work of various modules, the system can achieve high-precision three-dimensional modeling while having strong dynamic adaptability and robustness.

[0110] In order to better enable relevant personnel in this technical field to fully understand and implement this design, the specific implementation principles of this design are supplemented below with reference to a specific application scenario.

[0111] In actual application, the system obtains the echo signal of the target tissue in the heart cavity through an ultrasound probe and generates a three-dimensional model based on it.

[0112] First, the operator places the ultrasound probe into the patient's cardiac cavity area and starts the signal acquisition module to detect the relative motion state between the probe and the target tissue.

[0113] The built-in acceleration sensor in the signal acquisition module monitors the displacement and velocity information of the probe in real time, and analyzes the sampling frequency and signal intensity distribution of the ultrasonic signal.

[0114] The signal acquisition module then divides the ultrasound signal into multiple time windows, each ranging from 50ms to 100ms. Within each time window, the Fast Fourier Transform (FFT) algorithm is used to convert the time-domain signal into the frequency domain, extracting spectral features such as dominant frequency, bandwidth, and energy distribution.

[0115] These spectral features are passed to the mode selection module for subsequent separation of dynamic and static components.

[0116] After receiving the spectrum features, the mode selection module inputs them into the spectrum feature matrix for processing.

[0117] By calculating the mean and variance of each element in the matrix and combining it with the preset variance ratio coefficient, a dynamic threshold is generated.

[0118] If the main frequency or bandwidth in the spectrum feature exceeds the dynamic threshold, it is determined to be a dynamic component; otherwise it is classified as a static component.

[0119] Furthermore, the mode selection module calculates a comprehensive dynamic index according to the ratio of the dynamic component to the dynamic threshold, and determines the current modeling mode in combination with the sampling frequency coefficient.

[0120] For example, when the comprehensive dynamic index is high and the sampling frequency coefficient is low, the system selects the high-precision modeling mode; and when both are in the middle range, the balanced modeling mode is selected.

[0121] This mode selection mechanism ensures that the system can flexibly adjust the modeling strategy according to the dynamic characteristics of the intracardiac environment.

[0122] Subsequently, the anomaly detection module intervenes in the process to determine whether the model has any anomalies.

[0123] The anomaly detection module accesses the historical database, obtains the historical signal spectrum data under the corresponding modeling mode, and processes it using the support vector machine classification algorithm.

[0124] By setting the classification benchmark and true and false sample labels, the true sample rate and false sample rate are calculated, and a standard spectrum interval is generated.

[0125] The deviation coefficient is obtained by subtracting the center value of the current signal spectrum from the center value of the standard spectrum interval. If the deviation coefficient is lower than the model quality threshold, the model is judged to be operating normally.

[0126] Otherwise, it is marked as abnormal. This process effectively avoids modeling errors caused by signal noise or environmental interference.

[0127] Finally, the quality optimization module adjusts the quality threshold according to the modeled patterns and spectrum change rates of adjacent time windows to adapt to the dynamic changes of the signal.

[0128] For example, if the current modeling mode is the balanced modeling mode and is different from the mode of the previous time window, the quality optimization module multiplies the current quality threshold by the spectrum change rate to generate a new quality threshold.

[0129] This adjustment mechanism ensures the robustness of the model in complex cardiac chamber environments while improving modeling accuracy.

[0130] Through the above steps, the system realizes the whole process from ultrasonic signal acquisition to three-dimensional modeling.

[0131] The signal acquisition module is responsible for acquiring basic data, the mode selection module dynamically adjusts the modeling strategy, the anomaly detection module monitors the model status in real time, and the quality optimization module continuously optimizes and updates the process.

[0132] The modules work together to form a complete closed-loop system, thus overcoming the shortcomings of existing technologies in dynamic adaptability, modeling accuracy and robustness in complex environments, and meeting the clinical demand for high-precision anatomical structures.

[0133] An intracardiac ultrasound-guided three-dimensional modeling method comprises the following steps:

[0134] Step S1: Before receiving the ultrasound signal, the relative motion state between the ultrasound probe and the target tissue is detected, the sampling frequency and signal intensity distribution of the ultrasound signal are obtained, the ultrasound signal is divided into multiple time windows, and the signal in each time window is subjected to frequency domain analysis, and the spectrum characteristics in each window are recorded;

[0135] Step S2: extract the dynamic and static components of the signal based on the spectral characteristics, set a dynamic threshold to filter the dynamic components, evaluate the comprehensive dynamic index of the signal, and select a suitable modeling mode based on the sampling frequency and the comprehensive dynamic index;

[0136] Step S3: Before each model update, the ultrasound signal is processed according to the current modeling mode, the historical database is accessed to obtain the historical signal spectrum data under the corresponding mode, the standard spectrum interval is calculated and the model quality threshold is set, the deviation coefficient between the current signal spectrum and the standard spectrum interval is collected, and the model quality threshold and the deviation coefficient are compared to determine whether the model is abnormal and mark the abnormality;

[0137] Step S4: Detect the changing trend of the signal spectrum in adjacent time windows, calculate the spectrum change rate, and adjust the quality threshold of the current modeling mode according to the spectrum change rate to ensure that the model can adapt to the dynamic changes of the signal;

[0138] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0139] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0140] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0141] In this application, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0142] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0143] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0144] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0145] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0146] The units described as separate components may or may not be physically separate, and 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 these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0147] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0148] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0149] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for intracardiac ultrasound-guided three-dimensional modeling, characterized by: The following steps are involved: Step S1: Before receiving the ultrasound signal, the relative motion state between the ultrasound probe and the target tissue is detected, the sampling frequency and signal intensity distribution of the ultrasound signal are obtained, the ultrasound signal is divided into multiple time windows, and the signal in each time window is subjected to frequency domain analysis, and the spectrum characteristics in each window are recorded; Step S2: extract the dynamic and static components of the signal based on the spectral characteristics, set a dynamic threshold to filter the dynamic components, evaluate the comprehensive dynamic index of the signal, and select a suitable modeling mode based on the sampling frequency and the comprehensive dynamic index; Step S3: Before each model update, the ultrasound signal is processed according to the current modeling mode, the historical database is accessed to obtain the historical signal spectrum data under the corresponding mode, the standard spectrum interval is calculated and the model quality threshold is set, the deviation coefficient between the current signal spectrum and the standard spectrum interval is collected, and the model quality threshold and the deviation coefficient are compared to determine whether the model is abnormal and mark the abnormality; Step S4: Detect the changing trend of the signal spectrum in adjacent time windows, calculate the spectrum change rate, and adjust the quality threshold of the current modeling mode according to the spectrum change rate to ensure that the model can adapt to the dynamic changes of the signal.

2. The intracardiac ultrasound-guided three-dimensional modeling method according to claim 1, characterized in that: In step S1, the ultrasonic probe is a device for transmitting and receiving ultrasonic waves. The displacement and velocity information of the probe are obtained through the built-in acceleration sensor, and the sampling frequency of the ultrasonic signal is analyzed in combination with the signal intensity distribution. Frequency domain analysis uses the fast Fourier transform algorithm to convert the signal within the time window from the time domain to the frequency domain and extract the spectrum characteristics; the length of the time window is 50ms to 100ms; Spectral characteristics include main frequency, bandwidth and energy distribution.

3. The intracardiac ultrasound-guided three-dimensional modeling method according to claim 2, characterized in that: In step S2, the specific steps of extracting the dynamic component and the static component are as follows: The main frequency and bandwidth in the spectrum feature are used as input to construct a spectrum feature matrix. The mean and variance of each element in the matrix are calculated. The variance is multiplied by the preset variance ratio coefficient and then summed with the mean value to obtain the result as the dynamic threshold. When the main frequency or bandwidth in the spectrum feature exceeds the dynamic threshold, the spectrum feature is judged to belong to the dynamic component; Otherwise, it is judged to be a static component; When calculating the comprehensive dynamic index, the ratio of the spectral characteristics of the dynamic component to the dynamic threshold is used as the weight of a single dynamic component, and the average of all dynamic component weights is used as the comprehensive dynamic index.

4. The intracardiac ultrasound-guided three-dimensional modeling method according to claim 3, characterized in that: In step S2, the sampling frequency is normalized using a normalization algorithm, the normalized result is used as the sampling frequency coefficient, and the difference between the sampling frequency coefficient and the comprehensive dynamic index is used as the standard value for selecting the modeling mode; When the selection criterion value of the modeling mode is less than 0, the high-precision modeling mode is selected; when the selection criterion value of the modeling mode is greater than 0 and less than the preset stability threshold, the balanced modeling mode is selected; When the selection criterion value of the modeling mode exceeds the preset stability threshold, the fast modeling mode is selected.

5. The intracardiac ultrasound-guided three-dimensional modeling method according to claim 4, characterized in that: In step S3, when processing the ultrasonic signal according to the modeling mode, if the current mode is the high-precision modeling mode, the ultrasonic signal is subjected to full-band analysis and all spectral features are retained; If the current mode is the fast modeling mode, only the spectral features of the dynamic components in the ultrasonic signal are extracted; if the current mode is the balanced modeling mode, the ultrasonic signal is segmented and the spectral features of the high frequency band and the low frequency band are extracted respectively.

6. The intracardiac ultrasound-guided three-dimensional modeling method according to claim 5, characterized in that: In step S3, the historical database is accessed to obtain the historical signal spectrum data under the corresponding mode and merged into a historical spectrum data set. The support vector machine classification algorithm is used to set the model quality threshold and calculate the standard spectrum interval. The specific steps are as follows: Set the classification benchmark and use the mean of the main frequency in the historical spectrum data set as the classification benchmark; Positive and negative sample labeling: Data in the historical spectrum dataset that exceeds the classification benchmark is labeled as a positive sample, and data that is below the classification benchmark is labeled as a negative sample; True and false classification: positive samples and negative samples exceeding the preset spectrum judgment threshold are marked as true samples, and positive samples and negative samples below the preset spectrum judgment threshold are marked as false samples; Calculate the classification rate: Calculate the true sample rate and false sample rate respectively through the classification rate formula; Classification rate increment: set M different spectrum judgment thresholds to obtain multiple true sample rates and false sample rates, and make the difference between the obtained true sample rates from large to small to obtain M-1 true sample rate differences, and make the difference between the obtained false sample rates from large to small to obtain M-1 false sample rate differences; Calculate the model quality threshold: Calculate the model quality threshold using the obtained true sample rate difference and false sample rate difference; Calculate the standard spectrum interval: select the maximum and minimum values ​​of the true sample rate and the false sample rate respectively, use the maximum value of the true sample rate and the false sample rate to calculate the upper interval of the standard spectrum, use the minimum value of the true sample rate and the false sample rate to calculate the lower interval of the standard spectrum, and use the range between the upper interval of the standard spectrum and the lower interval of the standard spectrum as the standard spectrum interval.

7. The intracardiac ultrasound-guided three-dimensional modeling method according to claim 6, characterized in that: In step S3, when determining whether the model is abnormal, the current signal spectrum is first compared with the standard spectrum interval. If the current signal spectrum is within the standard spectrum interval, it is determined that the model is normal. If the current signal spectrum is outside the standard spectrum range, the deviation coefficient is obtained by subtracting the center value of the current signal spectrum from the center value of the standard spectrum range; When the deviation coefficient is lower than the model quality threshold, the model is judged to be normal; When the deviation coefficient exceeds the model quality threshold, the model is judged to be abnormal and marked as abnormal.

8. The intracardiac ultrasound-guided three-dimensional modeling method according to claim 7, characterized in that: In step S4, the changing trend of the signal spectrum in adjacent time windows is detected, the spectrum characteristics of the current time window are compared with the spectrum characteristics of the previous time window, and the spectrum change rate is calculated; The formula for calculating the spectrum change rate is the difference between the main frequency of the current time window and the main frequency of the previous time window divided by the main frequency of the previous time window; If the spectrum change rate exceeds the preset evaluation threshold, the quality threshold of the current modeling mode is adjusted based on the modeling mode and spectrum change rate of the adjacent time window. The specific steps are as follows: If the current modeling mode is high-precision modeling mode or fast modeling mode, and is the same as the modeling mode of the previous time window, the quality threshold of the current modeling mode will not be adjusted; If the current modeling mode is the balanced modeling mode and is different from the modeling mode of the previous time window, the product of the quality threshold in the current modeling mode and the spectrum change rate is used as the adjusted quality threshold to re-judge whether the model is abnormal.

9. An intracardiac ultrasound-guided 3D modeling system, configured to implement the intracardiac ultrasound-guided 3D modeling method according to any one of claims 1 to 8, characterized in that: It includes signal acquisition module, mode selection module, anomaly detection module and quality optimization module; The signal acquisition module is used to detect the relative motion state between the ultrasound probe and the target tissue and to collect the sampling frequency and spectrum characteristics of the ultrasound signal; The mode selection module is used to extract dynamic and static components based on spectrum characteristics, set dynamic thresholds to filter dynamic components, and select the current modeling mode after evaluating the comprehensive dynamic index; The anomaly detection module calculates the standard spectrum interval based on historical signal spectrum data and sets the model quality threshold. It then determines whether the model has an anomaly based on the deviation coefficient between the current signal spectrum and the standard spectrum interval. The quality optimization module adjusts the quality threshold of the current modeling mode according to the modeling mode and spectrum change rate of adjacent time windows.

10. The intracardiac ultrasound-guided three-dimensional modeling system according to claim 9, characterized in that: The signal acquisition module obtains the displacement and velocity information of the probe through the built-in acceleration sensor, and analyzes the sampling frequency of the ultrasonic signal based on the signal intensity distribution; The mode selection module calculates the dynamic threshold and evaluates the comprehensive dynamic index through the spectrum feature matrix; The anomaly detection module uses the support vector machine classification algorithm to generate standard spectrum intervals and calculate the model quality threshold; The quality optimization module adjusts the quality threshold of the current modeling mode according to the spectrum change rate.

Images