Ultrasound imaging optimization method and system based on multi-angle deflection feature points for real-time correction

Abstract

The application provides an ultrasonic imaging optimization method and system based on multi-angle deflection feature points for real-time correction, acquires a historical ultrasonic image of an electronic phased array probe, extracts historical feature points of the historical ultrasonic image, provides accurate feature points for further analysis of the ultrasonic image, classifies and labels the historical feature points based on a deep learning network, trains a sound beam distribution model based on the labeling result, realizes feature learning of a sound beam emitted by the electronic phased array probe, dynamically focuses and controls the current sound beam emitted by the electronic phased array probe from different angles to a target region based on the sound beam distribution model and a sound ray tracing method, obtains multi-angle target ultrasonic imaging, improves the spatial resolution and imaging quality of the image, and provides a high-quality image basis for further medical diagnosis.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ultrasonic imaging, in particular to an ultrasonic imaging optimization method and system based on multi-angle deflection feature points for real-time correction. BACKGROUND

[0002] Through electronic phased array, real-time correction algorithm and hardware innovation, ultrasonic imaging technology has developed from single two-dimensional imaging to a diagnostic tool with high resolution, multi-modal fusion and intelligent analysis. Based on the dynamic optimization technology of multi-angle deflection feature points, the problems of motion artifacts and insufficient resolution of traditional ultrasound are further solved, which promotes its precise application in the fields of heart, obstetrics and intervention. In the future, with the deep integration of AI and hardware technology, ultrasound is expected to become one of the core imaging means of precision medicine.

[0003] In the process of applying ultrasonic imaging to medical diagnosis, how to optimize ultrasonic imaging based on multi-angle deflection feature points for real-time correction to improve image spatial resolution and image quality is an urgent problem to be solved. SUMMARY

[0004] The present application provides an ultrasonic imaging optimization method and system based on multi-angle deflection feature points for real-time correction to solve the problems raised in the background art.

[0005] An ultrasonic imaging optimization method based on multi-angle deflection feature points for real-time correction, comprising:

[0006] S1: obtaining a historical ultrasonic image of an electronic phased array probe and extracting a historical feature point of the historical ultrasonic image;

[0007] S2: classifying and labeling the historical feature point based on a deep learning network, and training a sound beam distribution model based on the labeling result;

[0008] S3: based on the sound beam distribution model, combining a sound ray tracing method, dynamically focusing and deflection controlling the current sound beam emitted by the electronic phased array probe from different angles to the target region to obtain multi-angle target ultrasonic imaging.

[0009] Preferably, in S1, obtaining a historical ultrasonic image of an electronic phased array probe and extracting a historical feature point of the historical ultrasonic image, comprising:

[0010] Collecting historical ultrasonic images of an electronic phased array probe and obtaining historical extraction information of the historical ultrasonic images;

[0011] Determining the historical feature points of the historical ultrasonic images based on the historical extraction information, and determining the feature types of the historical feature points.

[0012] Preferably, in the S2, the historical feature points are classified and labeled based on a deep learning network, including:

[0013] Based on the feature types of the historical feature points, the historical feature points are classified by combining the deep learning network, to obtain a plurality of historical feature point groups;

[0014] Based on the feature attributes of the historical feature points in each historical feature point group, the historical feature points are labeled by combining the deep learning network, to obtain a labeling result;

[0015] Based on the labeling result, all historical feature point groups are integrated to obtain an integrated result.

[0016] Preferably, in the S2, the sound beam distribution model is trained based on the labeling result, including:

[0017] Based on the target requirements, an iteration termination condition is set;

[0018] The historical feature points under the labeling result are used as training data to iteratively optimize and train the initial distribution model, until the iteration termination condition is met, to obtain the sound beam distribution model.

[0019] Preferably, in the S3, based on the sound beam distribution model, the current sound beam emitted by the electronic phased array probe from different angles to the target region is dynamically focused and deflected controlled by combining the sound ray tracing method, to obtain multi-angle target ultrasonic imaging, including:

[0020] Based on the sound beam distribution model, the energy distribution characteristics, propagation direction characteristics and focusing characteristics of the sound beam emitted by the electronic phased array probe are determined;

[0021] Based on the energy distribution characteristics, propagation direction characteristics and focusing characteristics, the sound beam emission characteristics of the current sound beam emitted by the electronic phased array probe from different angles to the target region are determined by combining the sound ray tracing method, to determine the energy distribution characteristics, propagation direction characteristics and focusing characteristics of the current sound beam;

[0022] Based on the energy distribution characteristics, propagation direction characteristics and focusing characteristics of the current sound beam, the dynamic focusing parameters and deflection control parameters of the current sound beam are determined by combining the acquisition target of the target region;

[0023] Based on the dynamic focusing parameters and deflection control parameters, multi-angle target ultrasonic imaging is obtained.

[0024] Preferably, based on the energy distribution characteristics, propagation direction characteristics and focusing characteristics of the current sound beam, the dynamic focusing parameters of the current sound beam are determined by combining the acquisition target of the target region, including:

[0025] acquire echo time difference of each focal point at transmitting focusing and initial focusing position, determine central energy region based on energy distribution feature, determine direction convergence region based on propagation direction feature, determine focusing region based on focusing feature, determine multiple target focusing positions and focusing time nodes at different depths based on the echo time difference, combining central energy region, direction convergence region and focusing region;

[0026] determine focusing position adjustment information of transmitting focusing based on the target focusing positions and focusing time nodes, combining initial focusing position, and set initial excitation delay time of each array based on the focusing position adjustment information and element distribution of electronic phased array probe;

[0027] establish depth sequence in order of depth of focal point from small to large, and determine difference sequence of the depth sequence, determine participating element number and participating element distribution based on the difference sequence;

[0028] determine phase compensation sequence of electronic phased array probe based on the participating element number and participating element distribution, and determine initial phase dynamic compensation value based on the phase compensation sequence;

[0029] acquire ultrasonic imaging environment, determine temperature feature and sound velocity feature based on ultrasonic imaging environment, and establish temperature compensation value and sound velocity uniform compensation value based on temperature feature and sound velocity feature respectively;

[0030] determine correction value of initial excitation delay time and initial phase dynamic compensation value based on the temperature compensation value and sound velocity uniform compensation value respectively, and determine target excitation delay time and target phase dynamic compensation value based on the correction value;

[0031] use the target excitation delay time and target phase dynamic compensation value as dynamic focusing parameters of current sound beam.

[0032] Preferably, the deflection control parameters of current sound beam are determined based on energy distribution feature, propagation direction feature and focusing feature of current sound beam, combining acquisition target of target region, including:

[0033] determine target element phase difference sequence of electronic phased array probe based on energy distribution feature, propagation direction feature and focusing feature of current sound beam, combining acquisition target of target region;

[0034] determine deflection angle of adjacent elements of current sound beam in transmitting process based on deflection angle calculation formula;

[0035] determine whether the deflection angle meets the target element phase difference sequence;

[0036] if yes, determine to keep current deflection control parameters;

[0037] Otherwise, based on a PID feedback control algorithm, a phase adjustment difference of adjacent elements is determined, and based on the phase adjustment difference, a current deflection control parameter is determined.

[0038] An ultrasonic imaging optimization system based on multi-angle deflection feature points for real-time correction, comprising:

[0039] A feature point acquisition module is configured to acquire historical ultrasonic images of an electronic phased array probe and extract historical feature points of the historical ultrasonic images.

[0040] A model establishment module is configured to classify and label the historical feature points based on a deep learning network and train a sound beam distribution model based on the labeling results.

[0041] A dynamic control module is configured to dynamically focus and control current sound beams emitted by the electronic phased array probe from different angles to a target region based on the sound beam distribution model and a sound ray tracing method, and obtain multi-angle target ultrasonic imaging.

[0042] Preferably, the feature point acquisition module comprises:

[0043] A collection unit is configured to collect historical ultrasonic images of an electronic phased array probe and acquire historical extraction information of the historical ultrasonic images.

[0044] A determination unit is configured to determine historical feature points of the historical ultrasonic images based on the historical extraction information and determine feature types of the historical feature points.

[0045] Preferably, the dynamic control module comprises:

[0046] A characteristic determination unit is configured to determine energy distribution characteristics, propagation direction characteristics and focusing characteristics of sound beams emitted by the electronic phased array probe based on the sound beam distribution model.

[0047] A feature determination unit is configured to determine sound beam emission characteristics of current sound beams emitted by the electronic phased array probe from different angles to a target region based on the energy distribution characteristics, the propagation direction characteristics and the focusing characteristics, in combination with the sound ray tracing method, and determine energy distribution characteristics, propagation direction characteristics and focusing characteristics of the current sound beams.

[0048] A parameter determination unit is configured to determine dynamic focusing parameters and deflection control parameters of the current sound beams based on the energy distribution characteristics, the propagation direction characteristics and the focusing characteristics of the current sound beams, in combination with an acquisition target of the target region.

[0049] An imaging unit is configured to obtain multi-angle target ultrasonic imaging based on the dynamic focusing parameters and the deflection control parameters.

[0050] Compared with the prior art, the present application has the following beneficial effects:

[0051] By acquiring historical ultrasound images of the electronic phased array probe, and extracting historical feature points of the historical ultrasound images, accurate feature points are provided for further analysis of the ultrasound images, the historical feature points are classified and labeled based on a deep learning network, and a sound beam distribution model is trained based on the labeling result, feature learning of the emission beam of the electronic phased array probe is realized, based on the sound beam distribution model, in combination with a sound ray tracing method, current sound beams emitted by the electronic phased array probe from different angles to a target region are dynamically focused and deflected, multi-angle target ultrasound imaging is obtained, the spatial resolution and imaging quality of the image are improved, and high-quality image basis is provided for further medical diagnosis.

[0052] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application can be realized and obtained by the structure particularly pointed out in the application file.

[0053] The technical solutions of the present application will be further described in detail below with the help of the accompanying drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0054] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, used to explain the present application, and do not constitute a limitation on the present application. In the drawings:

[0055] Figure 1 A flowchart of the ultrasound imaging optimization method based on multi-angle deflection feature points for real-time correction in the embodiments of the present application is shown in the figure.

[0056] Figure 2 A flowchart of the historical feature points of the historical ultrasound images extracted in the embodiments of the present application is shown in the figure.

[0057] Figure 3 A structure diagram of the ultrasound imaging optimization system based on multi-angle deflection feature points for real-time correction in the embodiments of the present application is shown in the figure. DETAILED DESCRIPTION

[0058] The preferred embodiments of the present application will be described below in conjunction with the accompanying drawings, and it should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application, and do not limit the present application.

[0059] Example 1:

[0060] The embodiments of the present application provide an ultrasound imaging optimization method based on multi-angle deflection feature points for real-time correction, as shown in the figure, which includes: Figure 1 ​

[0061] S1: Obtain historical ultrasound images of an electronic phased array probe, and extract historical feature points of the historical ultrasound images;

[0062] S2: Classify and label the historical feature points based on a deep learning network, and train a sound beam distribution model based on the labeling results;

[0063] S3: Based on the sound beam distribution model, combine a sound ray tracing method to dynamically focus and deflect control the current sound beam emitted by the electronic phased array probe from different angles to the target region, to obtain multi-angle target ultrasound imaging.

[0064] In this embodiment, the historical feature points are points in the relevant region of the identification task in the historical ultrasound images, which are usually anatomical structure or tissue feature points that are stable in spatial position and have discriminability.

[0065] In this embodiment, the sound beam distribution model is a mathematical model describing the energy distribution, propagation direction and focusing characteristics of the sound waves emitted by the ultrasound probe in space.

[0066] In this embodiment, dynamic focusing and deflection control of the current sound beam emitted to the target region realizes dynamic optimization of the sound beam direction, focal point position and resolution.

[0067] In this embodiment, the sound ray tracing method simulates the propagation process of sound rays in the indoor environment, including reflection, absorption and other phenomena, finally synthesizes the sound field of the listening point, and simulates various acoustic characteristics.

[0068] The beneficial effects of the above design scheme are: by obtaining historical ultrasound images of an electronic phased array probe, and extracting historical feature points of the historical ultrasound images, accurate feature points are provided for further analysis of the ultrasound images, based on a deep learning network, the historical feature points are classified and labeled, and a sound beam distribution model is trained based on the labeling results, realizing feature learning of the sound beam emitted by the electronic phased array probe, based on the sound beam distribution model, combining a sound ray tracing method, the current sound beam emitted by the electronic phased array probe from different angles to the target region is dynamically focused and deflected controlled, to obtain multi-angle target ultrasound imaging, improving the spatial resolution and imaging quality of the image, and providing a high-quality image basis for further medical diagnosis.

[0069] Embodiment 2:

[0070] Based on the basis of embodiment 1, the present embodiment provides an ultrasound imaging optimization method for real-time correction based on multi-angle deflection feature points, as shown in Figure 1 In S1, the historical ultrasound images of the electronic phased array probe are obtained, and the historical feature points of the historical ultrasound images are extracted, which include:

[0071] Collect historical ultrasound images of the electronic phased array probe, and obtain historical extraction information of the historical ultrasound images;

[0072] Determine historical feature points of the historical ultrasound images based on the historical extraction information, and determine feature types of the historical feature points.

[0073] In this embodiment, the feature types of the historical feature points are, for example, texture features, contour features, and the like.

[0074] The above design scheme has the beneficial effect that: by collecting historical ultrasound images of the electronic phased array probe, and obtaining historical extraction information of the historical ultrasound images, determining historical feature points of the historical ultrasound images based on the historical extraction information, and determining feature types of the historical feature points, accurate feature points are provided for further analysis of the ultrasound images.

[0075] Embodiment 3

[0076] Based on the basis of Embodiment 1, the present embodiment provides an ultrasound imaging optimization method for real-time correction based on multi-angle deflection feature points, and in S2, the historical feature points are classified and labeled based on a deep learning network, including:

[0077] Based on the feature types of the historical feature points, the historical feature points are classified in combination with the deep learning network to obtain a plurality of historical feature point groups;

[0078] Based on the feature attributes of the historical feature points in each historical feature point group, the historical feature points are labeled in combination with the deep learning network to obtain a labeling result;

[0079] Based on the labeling result, all the historical feature point groups are integrated to obtain an integration result.

[0080] In this embodiment, the feature attributes are, for example, lesion features, healthy features, and the like.

[0081] In this embodiment, integrating all the historical feature point groups is to integrate historical feature points of a combination of medical diagnoses together.

[0082] The above design scheme has the beneficial effect that: by classifying the historical feature points based on the feature types of the historical feature points in combination with the deep learning network to obtain a plurality of historical feature point groups, labeling the historical feature points based on the feature attributes of the historical feature points in each historical feature point group in combination with the deep learning network to obtain a labeling result, and integrating all the historical feature point groups based on the labeling result to obtain an integration result, accurate training data is provided for model training and construction.

[0083] Embodiment 4

[0084] Based on the basis of embodiment 1, the embodiment of the application provides an ultrasonic imaging optimization method for real-time correction based on multi-angle deflection feature points, in S2, the sound beam distribution model is trained based on the labeling result, including:

[0085] The iteration termination condition is set based on the target demand;

[0086] The historical feature points under the labeling result are used as training data to iteratively optimize and train the initial distribution model, until the iteration termination condition is met, and the sound beam distribution model is obtained.

[0087] In this embodiment, the iteration termination condition is that the model performance meets the target demand or the iteration number reaches the maximum value.

[0088] The beneficial effects of the above design scheme are: by setting the iteration termination condition based on the target demand, using the historical feature points under the labeling result as training data to iteratively optimize and train the initial distribution model, until the iteration termination condition is met, and the sound beam distribution model is obtained, realizing feature learning of the electronic phased array probe transmitting beam, and improving the accuracy of the ultrasonic image based on the model.

[0089] Embodiment 5:

[0090] Based on the basis of embodiment 1, the embodiment of the application provides an ultrasonic imaging optimization method for real-time correction based on multi-angle deflection feature points, in S3, based on the sound beam distribution model, the current sound beam emitted by the electronic phased array probe from different angles to the target region is dynamically focused and deflected controlled by combining the sound ray tracing method, and multi-angle target ultrasonic imaging is obtained, including:

[0091] Based on the sound beam distribution model, the energy distribution characteristics, propagation direction characteristics and focusing characteristics of the sound beam emitted by the electronic phased array probe are determined;

[0092] Based on the energy distribution characteristics, propagation direction characteristics and focusing characteristics, and combining the sound ray tracing method, the sound beam emission characteristics of the current sound beam emitted by the electronic phased array probe from different angles to the target region are determined, and the energy distribution characteristics, propagation direction characteristics and focusing characteristics of the current sound beam are determined;

[0093] Based on the energy distribution characteristics, propagation direction characteristics and focusing characteristics of the current sound beam, and combining the acquisition target of the target region, the dynamic focusing parameters and deflection control parameters of the current sound beam are determined;

[0094] Based on the dynamic focusing parameters and deflection control parameters, multi-angle target ultrasonic imaging is obtained.

[0095] In this embodiment, the energy distribution characteristic, the propagation direction characteristic and the focusing characteristic are rules of the electronic phased array probe emitting the sound beam, and based on the rules, the specific characteristic values of the energy distribution characteristic, the propagation direction characteristic and the focusing characteristic of the current sound beam are determined in combination with the sound beam emission characteristic of the current sound beam.

[0096] In this embodiment, the sound beam emission characteristic is, for example, the emission direction, the emission angle, the emission quantity, the excitation time and the amplitude, etc.

[0097] In this embodiment, the acquisition target of the target region is, for example, the range of the acquisition region, the definition requirement of the acquired image, etc.

[0098] The beneficial effects of the above design scheme are that: by determining the dynamic focusing parameter and the deflection control parameter of the current sound beam based on the energy distribution characteristic, the propagation direction characteristic and the focusing characteristic of the current sound beam in combination with the acquisition target of the target region, and based on the dynamic focusing parameter and the deflection control parameter, the multi-angle target ultrasonic imaging is obtained, and the spatial resolution and the imaging quality of the image are improved.

[0099] Embodiment 6:

[0100] Based on the basis of embodiment 5, the embodiment of the application provides an ultrasonic imaging optimization method for real-time correction based on multi-angle deflection feature points, the dynamic focusing parameter of the current sound beam is determined based on the energy distribution characteristic, the propagation direction characteristic and the focusing characteristic of the current sound beam in combination with the acquisition target of the target region, and the method comprises the following steps:

[0101] The echo time difference of each focus point at the time of emission focusing and the initial focusing position is obtained, the central energy region is determined based on the energy distribution characteristic, the direction convergence region is determined based on the propagation direction characteristic, the focusing region is determined based on the focusing characteristic, and the multiple target focusing positions and the focusing time nodes at different depths are determined based on the echo time difference in combination with the central energy region, the direction convergence region and the focusing region;

[0102] Based on the target focusing position and the focusing time node, the focusing position adjustment information of the emission focusing is determined in combination with the initial focusing position, and based on the focusing position adjustment information, the initial excitation delay time of each array is set in combination with the element distribution of the electronic phased array probe;

[0103] The depth sequence is established in the order from small to large according to the depth of the focus point, and the difference sequence of the depth sequence is determined, the participating element quantity and the participating element distribution are determined based on the difference sequence;

[0104] The phase compensation sequence of the electronic phased array probe is determined based on the participating element quantity and the participating element distribution, and the initial phase dynamic compensation value is determined based on the phase compensation sequence;

[0105] acquire an ultrasonic imaging environment, determine a temperature feature and a sound speed feature based on the ultrasonic imaging environment, establish a temperature compensation value and a sound speed uniform compensation value based on the temperature feature and the sound speed feature respectively;

[0106] determine a correction value of an initial excitation delay time and an initial phase dynamic compensation value based on the temperature compensation value and the sound speed uniform compensation value respectively, and determine a target excitation delay time and a target phase dynamic compensation value based on the correction value;

[0107] use the target excitation delay time and the target phase dynamic compensation value as dynamic focusing parameters for a current acoustic beam.

[0108] In this embodiment, the dynamic focusing parameters control both the focusing transmission and the focusing reception.

[0109] In this embodiment, the temperature compensation value and the sound speed uniform compensation value are established based on the temperature feature and the sound speed feature respectively, which eliminates the influence of the external environment on the imaging and guarantees the imaging quality.

[0110] In this embodiment, the excitation delay time makes the sound waves converge at a certain depth to form a focus point.

[0111] In this embodiment, in the signal receiving stage, the phase compensation of each array element is dynamically adjusted according to the echo arrival time to realize dynamic focusing control.

[0112] The beneficial effects of the above design scheme are: by acquiring the echo time difference of each focus at the time of transmitting focusing and the initial focusing position, determining the central energy region based on the energy distribution characteristics, determining the direction convergence region based on the propagation direction characteristics, determining the focusing region based on the focusing characteristics, determining the multiple target focusing positions and focusing time nodes at different depths based on the echo time difference, combining the central energy region, the direction convergence region and the focusing region, determining the focusing position adjustment information of the transmitting focusing based on the target focusing positions and the focusing time nodes, combining the initial focusing position, setting the initial excitation delay time of each array based on the focusing position adjustment information and the element distribution of the electronic phased array probe, realizing the control of dynamic focusing in the transmitting stage, establishing the depth sequence based on the order of the depth of the focus from small to large, and determining the difference sequence of the depth sequence, determining the number of participating elements and the distribution of participating elements based on the difference sequence, determining the phase compensation sequence of the electronic phased array probe based on the number of participating elements and the distribution of participating elements, determining the initial phase dynamic compensation value based on the phase compensation sequence, acquiring the ultrasonic imaging environment, determining the temperature characteristics and the sound velocity characteristics based on the ultrasonic imaging environment, respectively establishing the temperature compensation value and the sound velocity uniform compensation value based on the temperature characteristics and the sound velocity characteristics, realizing the control of dynamic focusing in the receiving stage, acquiring the ultrasonic imaging environment, determining the temperature characteristics and the sound velocity characteristics based on the ultrasonic imaging environment, respectively establishing the temperature compensation value and the sound velocity uniform compensation value based on the temperature characteristics and the sound velocity characteristics, respectively determining the correction value of the initial excitation delay time and the initial phase dynamic compensation value based on the temperature compensation value and the sound velocity uniform compensation value, determining the target excitation delay time and the target phase dynamic compensation value based on the correction value, realizing the elimination of the interference of the external environment on the imaging, finally ensuring the accuracy and comprehensiveness of the acquisition information of the ultrasonic image, and improving the spatial resolution of the image.

[0113] Embodiment 7:

[0114] Based on the basis of embodiment 5, the embodiment of the application provides an ultrasonic imaging optimization method for real-time correction based on multi-angle deflection feature points, which determines the deflection control parameter of the current sound beam based on the energy distribution characteristics, the propagation direction characteristics and the focusing characteristics of the current sound beam, and the acquisition target of the target region, including:

[0115] Determine the target element phase difference sequence of the electronic phased array probe based on the energy distribution characteristics, the propagation direction characteristics and the focusing characteristics of the current sound beam, and the acquisition target of the target region.

[0116] Determine the deflection angle of the adjacent elements of the current sound beam in the transmitting process based on the deflection angle calculation formula.

[0117] Determine whether the deflection angle meets the target element phase difference sequence.

[0118] If yes, determine to keep the current deflection control parameter;

[0119] Otherwise, determine the adjustment phase difference of the adjacent elements based on the PID feedback control algorithm, and determine the current deflection control parameter based on the adjustment phase difference.

[0120] In this embodiment, the phase difference of the adjacent elements is adjusted based on the deflection control parameter of the current sound beam, so that the sound beam is directed to different angles, thereby realizing multi-angle scanning and realizing omnidirectional acquisition of the target region.

[0121] In this embodiment, the target element phase difference sequence ensures that the phase difference of the adjacent elements meets the requirements.

[0122] In this embodiment, the deflection angle calculation formula is related to the sound speed, the delay difference of the adjacent elements, and the element spacing.

[0123] In this embodiment, the PID feedback control algorithm is obtained in advance according to the actual situation and is used for adjusting the phase difference.

[0124] The beneficial effects of the above design scheme are: by determining the deflection control parameter of the current sound beam based on the energy distribution characteristics, propagation direction characteristics and focusing characteristics of the current sound beam, and combining the acquisition target of the target region, the real-time control and adjustment of the element phase difference are realized, thereby realizing multi-angle scanning and ensuring the accuracy of imaging.

[0125] Embodiment 8:

[0126] The embodiment of the application provides an ultrasonic imaging optimization system for real-time correction based on multi-angle deflection feature points, as shown in Figure 3 , comprising:

[0127] The feature point acquisition module is configured to acquire a historical ultrasonic image of an electronic phased array probe and extract historical feature points of the historical ultrasonic image.

[0128] The model establishing module is configured to classify and label the historical feature points based on a deep learning network, and train a sound beam distribution model based on the labeling result.

[0129] The dynamic control module is configured to perform dynamic focusing and deflection control on a current sound beam emitted by the electronic phased array probe from different angles to the target region based on the sound beam distribution model and a ray tracing method, and obtain multi-angle target ultrasonic imaging.

[0130] In this embodiment, the historical feature points are points related to the recognition task in the target region of the historical ultrasonic image, which are usually anatomical structure or tissue feature points that are stable in spatial position and have discriminability.

[0131] In this embodiment, the acoustic beam distribution model is a mathematical model describing the energy distribution, propagation direction and focusing characteristics of the acoustic waves emitted by the ultrasonic probe in space.

[0132] In this embodiment, dynamic focusing and deflection control of the current acoustic beam emitted by the target region achieves dynamic optimization of the acoustic beam direction, focal point position and resolution.

[0133] The beneficial effects of the above design scheme are: by acquiring historical ultrasonic images of the electronic phased array probe and extracting historical feature points of the historical ultrasonic images, accurate feature points are provided for further analysis of the ultrasonic images, the historical feature points are classified and labeled based on a deep learning network, and a sound beam distribution model is trained based on the labeling result, realizing feature learning of the sound beam emitted by the electronic phased array probe, based on the sound beam distribution model, combining a sound ray tracing method, dynamic focusing and deflection control of the current acoustic beam emitted by the electronic phased array probe from different angles to the target region is realized, multi-angle target ultrasonic imaging is obtained, the spatial resolution and imaging quality of the image are improved, and high-quality images are provided for further medical diagnosis.

[0134] Embodiment 9:

[0135] Based on the basis of embodiment 8, the embodiment of the application provides an ultrasonic imaging optimization system for real-time correction based on multi-angle deflection feature points, the feature point acquisition module comprises:

[0136] The collection unit is configured to collect historical ultrasonic images of the electronic phased array probe and acquire historical extraction information of the historical ultrasonic images.

[0137] The determination unit is configured to determine historical feature points of the historical ultrasonic images based on the historical extraction information and determine the feature types of the historical feature points.

[0138] In this embodiment, the feature types of the historical feature points are, for example, texture features, contour features, etc.

[0139] The beneficial effects of the above design scheme are: by collecting historical ultrasonic images of the electronic phased array probe and acquiring historical extraction information of the historical ultrasonic images, historical feature points of the historical ultrasonic images are determined based on the historical extraction information, and the feature types of the historical feature points are determined, thereby providing accurate feature points for further analysis of the ultrasonic images.

[0140] Embodiment 10:

[0141] Based on the basis of embodiment 8, the embodiment of the application provides an ultrasonic imaging optimization system for real-time correction based on multi-angle deflection feature points, the dynamic control module comprises:

[0142] The characteristic determining unit is configured to determine energy distribution characteristics, propagation direction characteristics and focusing characteristics of the sound beam emitted by the electronic phased array probe based on the sound beam distribution model;

[0143] The characteristic determining unit is configured to determine sound beam emission characteristics of the current sound beam emitted by the electronic phased array probe from different angles to the target region, and determine the energy distribution characteristics, the propagation direction characteristics and the focusing characteristics of the current sound beam based on the sound ray tracing method and the energy distribution characteristics, the propagation direction characteristics and the focusing characteristics.

[0144] The parameter determining unit is configured to determine the dynamic focusing parameter and the deflection control parameter of the current sound beam based on the energy distribution characteristics, the propagation direction characteristics and the focusing characteristics of the current sound beam and the acquisition target of the target region.

[0145] The imaging unit is configured to obtain multi-angle target ultrasonic imaging based on the dynamic focusing parameter and the deflection control parameter.

[0146] In this embodiment, the energy distribution characteristics, the propagation direction characteristics and the focusing characteristics are the rules of the sound beam emitted by the electronic phased array probe, and based on the rules, the specific characteristic values of the energy distribution characteristics, the propagation direction characteristics and the focusing characteristics of the current sound beam are determined in combination with the sound beam emission characteristics of the current sound beam.

[0147] In this embodiment, the sound beam emission characteristics are, for example, emission direction, emission angle, emission quantity, excitation time and amplitude.

[0148] In this embodiment, the acquisition target of the target region is, for example, the range of the acquisition region, the definition requirement of the acquired image and the like.

[0149] The beneficial effects of the above design scheme are that the dynamic focusing parameter and the deflection control parameter of the current sound beam are determined based on the energy distribution characteristics, the propagation direction characteristics and the focusing characteristics of the current sound beam and the acquisition target of the target region, and multi-angle target ultrasonic imaging is obtained based on the dynamic focusing parameter and the deflection control parameter, thereby improving the spatial resolution and imaging quality of the image.

[0150] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. An ultrasound imaging optimization method based on multi-angle deflection feature points for real-time correction, characterized in that, The method comprises the following steps: S1: obtaining historical ultrasound images of an electronic phased array probe, and extracting historical feature points of the historical ultrasound images; S2: classifying and labeling the historical feature points based on a deep learning network, and training a sound beam distribution model based on the labeling results; S3: based on the sound beam distribution model, combining a sound ray tracing method, dynamically focusing and deflecting control of the current sound beam emitted by the electronic phased array probe from different angles to the target region is performed to obtain multi-angle target ultrasound imaging.

2. The method of claim 1, wherein, In the S1, the historical ultrasound images of the electronic phased array probe are obtained, and the historical feature points of the historical ultrasound images are extracted, comprising: Collecting historical ultrasound images of the electronic phased array probe, and obtaining historical extraction information of the historical ultrasound images; Based on the historical extraction information, the historical feature points of the historical ultrasound images are determined, and the feature types of the historical feature points are determined.

3. The method of claim 1, wherein, In the S2, the historical feature points are classified and labeled based on a deep learning network, comprising: Based on the feature types of the historical feature points, the historical feature points are classified by combining a deep learning network to obtain a plurality of historical feature point groups; Based on the feature attributes of the historical feature points in each historical feature point group, the historical feature points are labeled by combining a deep learning network to obtain labeling results; Based on the labeling results, all historical feature point groups are integrated to obtain integration results.

4. The method of claim 1, wherein, In the S2, the sound beam distribution model is trained based on the labeling results, comprising: Setting an iteration termination condition based on target requirements; The historical feature points under the labeling results are used as training data to iteratively optimize and train the initial distribution model until the iteration termination condition is met, and the sound beam distribution model is obtained.

5. The method of claim 1, wherein, In the S3, based on the sound beam distribution model, combining a sound ray tracing method, dynamic focusing and deflection control of the current sound beam emitted by the electronic phased array probe from different angles to the target region is performed to obtain multi-angle target ultrasound imaging, comprising: Based on the sound beam distribution model, the energy distribution characteristics, propagation direction characteristics and focusing characteristics of the sound beam emitted by the electronic phased array probe are determined; Based on the energy distribution characteristics, propagation direction characteristics and focusing characteristics, combining a sound ray tracing method, the sound beam emission characteristics of the current sound beam emitted by the electronic phased array probe from different angles to the target region are determined, and the energy distribution characteristics, propagation direction characteristics and focusing characteristics of the current sound beam are determined; Based on the energy distribution characteristics, propagation direction characteristics and focusing characteristics of the current sound beam, combining the acquisition target of the target region, the dynamic focusing parameters and deflection control parameters of the current sound beam are determined; Based on the dynamic focusing parameters and deflection control parameters, multi-angle target ultrasound imaging is obtained.

6. The method of claim 5, wherein, Based on the energy distribution characteristics, propagation direction characteristics and focusing characteristics of the current sound beam, combining the acquisition target of the target region, the dynamic focusing parameters of the current sound beam are determined, comprising: acquire echo time difference of each focal point at the time of transmitting focusing and initial focusing position, determine central energy region based on energy distribution feature, determine direction convergence region based on propagation direction feature, determine focusing region based on focusing feature, determine multiple target focusing positions and focusing time nodes at different depths based on the echo time difference, combine central energy region, direction convergence region and focusing region, determine focusing position adjustment information of transmitting focusing based on the target focusing positions and focusing time nodes, combine initial focusing position, set initial excitation delay time of each array based on the focusing position adjustment information and element distribution of electronic phased array probe; determine focusing position adjustment information of transmitting focusing based on the target focusing positions and focusing time nodes, combine initial focusing position, set initial excitation delay time of each array based on the focusing position adjustment information and element distribution of electronic phased array probe; establish depth sequence in the order of depth of focal point from small to large, and determine difference sequence of the depth sequence, determine participating element number and participating element distribution based on the difference sequence; determine phase compensation sequence of electronic phased array probe based on the participating element number and participating element distribution, determine initial phase dynamic compensation value based on the phase compensation sequence; acquire ultrasonic imaging environment, determine temperature feature and sound velocity feature based on ultrasonic imaging environment, respectively establish temperature compensation value and sound velocity uniform compensation value based on temperature feature and sound velocity feature; determine correction value of initial excitation delay time and initial phase dynamic compensation value based on the temperature compensation value and sound velocity uniform compensation value, respectively, determine target excitation delay time and target phase dynamic compensation value based on the correction value; set the target excitation delay time and target phase dynamic compensation value as dynamic focusing parameters of the current sound beam.

7. The method of claim 5, wherein, determine deflection control parameters of the current sound beam based on the energy distribution feature, propagation direction feature and focusing feature of the current sound beam, and the acquisition target of the target region, including: determine target element phase difference sequence of electronic phased array probe based on the energy distribution feature, propagation direction feature and focusing feature of the current sound beam, and the acquisition target of the target region; determine deflection angle of adjacent elements of the current sound beam in the transmitting process based on deflection angle calculation formula; determine whether the deflection angle meets the target element phase difference sequence; if yes, determine to keep the current deflection control parameters; otherwise, determine adjustment phase difference of adjacent elements based on PID feedback control algorithm, and determine the current deflection control parameters based on the adjustment phase difference.

8. An ultrasound imaging optimization system based on multi-angle deflection feature points for real-time correction, used in the ultrasound imaging optimization method of claim 1, characterized in that, including: feature point acquisition module, used for acquiring historical ultrasonic image of electronic phased array probe, and extracting historical feature points of the historical ultrasonic image; model establishing module, used for classifying and labeling historical feature points based on deep learning network, and training to obtain sound beam distribution model based on the labeling result; dynamic control module, used for dynamically focusing and deflecting the current sound beam emitted by electronic phased array probe from different angles to the target region based on the sound beam distribution model and sound ray tracing method, and obtaining multi-angle target ultrasonic imaging.

9. The system of claim 8, wherein, the feature point acquisition module, including: collection unit, used for collecting historical ultrasonic image of electronic phased array probe, and acquiring historical extraction information of the historical ultrasonic image; The determining unit is configured to determine a historical feature point of a historical ultrasound image based on the historical extraction information, and determine a feature type of the historical feature point.

10. The system of claim 8, wherein the system is configured to: The dynamic control module comprises: The characteristic determining unit is configured to determine, based on the sound beam distribution model, energy distribution characteristics, propagation direction characteristics and focusing characteristics of the sound beam emitted by the electronic phased array probe; The characteristic determining unit is configured to determine, based on the energy distribution characteristics, propagation direction characteristics and focusing characteristics, in combination with a sound ray tracing method, sound beam emission characteristics of the current sound beam emitted by the electronic phased array probe from different angles to the target region, determine energy distribution characteristics, propagation direction characteristics and focusing characteristics of the current sound beam; The parameter determining unit is configured to determine, based on the energy distribution characteristics, propagation direction characteristics and focusing characteristics of the current sound beam, in combination with the acquisition target of the target region, dynamic focusing parameters and deflection control parameters of the current sound beam; The imaging unit is configured to obtain multi-angle target ultrasound imaging based on the dynamic focusing parameters and the deflection control parameters.

Images