Adaptive beam adjustment system and method for handheld ultrasonic equipment
By introducing an adaptive beam adjustment system into ultrasonic equipment, using image processing models and adaptive adjustment models to optimize beam parameters in real time, the problem of ultrasonic imaging relies on manual adjustment in the existing technology is solved, and efficient and flexible ultrasonic imaging and diagnosis are achieved .
Patent Information
- Application Number
- CN202410992118.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-11-08
AI Technical Summary
The existing ultrasound equipment relies on manual adjustment of beam shape in in vivo imaging, resulting in image quality dependent on operator skills, lack of flexibility and efficiency, and is unable to adapt to changes in different in vivo structures, especially in complex or non-standardized anatomical structures. .
Adaptive beam adjustment system using handheld ultrasound equipment, through pre-trained image processing model and adaptive adjustment model, the beam parameters are optimized in real time based on ultrasound image evaluation results and historical adjustment data, and the working parameters of the ultrasound probe are automatically adjusted to reduce operation. Reliance on technical level of staff and improve imaging flexibility.
Ultrasound imaging with strong automatic adaptability, stable image quality and reliable in different in vivo structures is achieved, reducing operational complexity and optimization parameter repetition, and significantly improving imaging efficiency and diagnostic accuracy.
Smart Images

Figure CN118902489A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of ultrasound acquisition and processing, and in particular to an adaptive beam adjustment system and method for a handheld ultrasound device. Background Art
[0002] Existing ultrasound devices typically use fixed beam shapes for in vivo imaging. Image quality is highly dependent on the operator's skills and experience, and the operator manually adjusts the probe and device settings to obtain the best image of the desired area, which is a complex and time-consuming process. Due to the complex, variable and diverse internal structures of the human body, fixed beam shapes and manual adjustments limit the flexibility and accuracy of imaging.
[0003] This approach, which relies on manual tuning, has several limitations:
[0004] (1) The operator's skill level can affect the image quality. For example, due to fatigue or lack of experience, the operator may not be able to consistently achieve the optimal imaging angle and position, resulting in images that are sometimes not clear or detailed enough. Operators need to undergo long periods of training and practice to master precise control skills, which are not easy to standardize.
[0005] (2) Multiple fine-tuning is required for different internal structures, which is time-consuming and affects clinical work efficiency. The body shape, tissue density and internal structure of different patients vary greatly, and fixed beam parameters cannot adapt to this diversity.
[0006] (3) Fixed beam shapes cannot automatically adapt to changes in body structures, limiting their applicability and flexibility. Operators can only make adjustments based on real-time displayed images, lacking deeper automated analysis assistance and insufficient real-time feedback, which increases operational difficulty.
[0007] Some known technologies attempt to solve the above problems by introducing adjustable beam settings, such as using multiple preset beam modes and allowing the operator to select the best matching mode based on the structure seen. However, these still cannot meet the rapidly changing clinical needs because they do not have real-time adaptive capabilities and still require operator intervention. Traditional equipment is usually unable to flexibly adjust beam parameters during the imaging process, resulting in poor image quality in complex or non-standardized anatomical structures. Especially in dynamic or high-demand clinical environments, existing technologies often show slow response and insufficient imaging speed, and cannot provide high-quality diagnostic images in a timely manner, affecting the efficiency and accuracy of clinical decision-making.
[0008] It can be seen that traditional ultrasound imaging has significant deficiencies in flexibility, operational complexity, imaging efficiency and intelligence. Summary of the invention
[0009] Based on the above problems, the present invention provides an adaptive beam adjustment system and method for a handheld ultrasonic device, which aims to solve the problem that the portable ultrasonic device in the prior art is not sufficiently intelligent in operation.
[0010] An adaptive beam adjustment system for a handheld ultrasound device, comprising:
[0011] An ultrasonic probe, used for generating an ultrasonic transmission beam and receiving an ultrasonic echo signal;
[0012] A signal processing module, connected to the ultrasonic probe, for preprocessing the ultrasonic echo signal;
[0013] An image forming module, connected to the signal processing module, for processing the pre-processed ultrasonic echo signal to form an ultrasonic image;
[0014] An adaptive adjustment module, connected to the image forming module, is used to process the ultrasound image based on a pre-trained image processing model and output an image evaluation result, and to process the current image evaluation result and historical adjustment data based on the pre-trained adaptive adjustment model and output a beam optimization parameter;
[0015] A beam control module, connected to the adaptive adjustment module, for adjusting the working parameters of the ultrasound probe according to the beam optimization parameters;
[0016] The ultrasonic probe generates a new ultrasonic transmission beam and receives an ultrasonic echo signal according to the adjusted working parameters;
[0017] The historical adjustment data includes the working parameters, image evaluation results and beam optimization parameters corresponding to the historical scanning period from the start of the scanning to the current time.
[0018] Furthermore, the ultrasound probe includes a plurality of independent transmitting elements and independent receiving elements;
[0019] Each transmitting element is independently configured with a control unit, and each control unit is connected to a beam control module;
[0020] The working parameters of the ultrasound probe adjusted by the beam control module include the number of transmitting elements, the positions of the transmitting elements, and the transmitting parameters;
[0021] The beam control module sends corresponding transmission parameters to the control units of the transmission elements in the adjusted working parameters of the ultrasound probe respectively;
[0022] The control unit controls the connected transmitting element to transmit ultrasonic waves according to the received transmitting parameters.
[0023] Furthermore, the image processing model includes a pre-trained image segmentation model and an image evaluation model;
[0024] The adaptive adjustment module includes:
[0025] An image segmentation unit, used for performing segmentation processing on the ultrasound image based on the image segmentation model to obtain an image segmentation result;
[0026] An image evaluation unit, connected to the image segmentation unit, is used to evaluate the image quality index of the image segmentation result based on the image evaluation model to obtain an image evaluation result;
[0027] The parameter optimization unit is connected to the image evaluation unit and is used to process the current image evaluation result and the historical adjustment data based on the pre-trained adaptive adjustment model and output the beam optimization parameters.
[0028] Furthermore, the image processing model also includes a pre-trained image classification model;
[0029] Different scanning parts have their own corresponding image segmentation models and image evaluation models;
[0030] The adaptive adjustment module also includes:
[0031] A part acquisition unit, used to acquire preset scanning part information;
[0032] An image classification unit, used to classify and identify the ultrasound image based on the image classification model when no scan part information is set in advance, and identify the scan part of the ultrasound image;
[0033] The model calling unit is connected to the part acquisition unit and the image classification unit respectively, and is used for:
[0034] When the preset scanning part information is obtained, the corresponding image segmentation model and image evaluation model are called according to the preset scanning part information; and
[0035] When the scanning part information is not set in advance, the corresponding image segmentation model and image evaluation model are called according to the scanning part identified by the image classification model.
[0036] Furthermore, the image quality index includes at least one of resolution, contrast, clarity, and signal-to-noise ratio.
[0037] Furthermore, the beam optimization parameters include at least one of depth, range, frequency, edge enhancement, focusing, main lobe width, and side lobe level.
[0038] Furthermore, each emitting element is equipped with a temperature sensor for collecting temperature data of the emitting element;
[0039] The temperature sensor in each transmitting element is connected to the control unit, and the control unit transmits the temperature data of the transmitting element to the beam control module in real time;
[0040] The beam control module is also used to adjust the working parameters of the ultrasound probe in combination with the temperature data of each transmitting element and the beam optimization parameters.
[0041] Furthermore, it also includes a display module, which is connected to the image forming module and is used to display the formed ultrasound image in real time.
[0042] An adaptive beam adjustment method for a handheld ultrasonic device, using the aforementioned adaptive beam adjustment system for a handheld ultrasonic device, comprises:
[0043] Step A1, generating an ultrasonic transmission beam and receiving an ultrasonic echo signal;
[0044] Step A2, preprocessing the ultrasonic echo signal;
[0045] Step A3, processing the preprocessed ultrasonic echo signal to form an ultrasonic image;
[0046] Step A4, processing the ultrasound image based on the pre-trained image processing model and outputting the image evaluation result, processing the current image evaluation result and the historical adjustment data based on the pre-trained adaptive adjustment model and outputting the beam optimization parameters;
[0047] Step A5, adjusting the working parameters of the ultrasound probe according to the beam optimization parameters;
[0048] Step A6, the ultrasonic probe generates a new ultrasonic transmission beam according to the adjusted working parameters and receives an ultrasonic echo signal, and then executes step A2;
[0049] The historical adjustment data includes the working parameters of the ultrasound probe, the image evaluation results, and the beam optimization parameters corresponding to the historical scanning period from the time when the scanning starts to the current time.
[0050] Furthermore, the ultrasound probe includes a plurality of independent transmitting elements and independent receiving elements;
[0051] Each transmitting element is independently configured with a control unit;
[0052] The working parameters of the ultrasonic probe generated in step A5 include the number of transmitting elements, the positions of the transmitting elements, and the transmitting parameters;
[0053] Step A5 also includes: sending corresponding transmission parameters to the control units of the transmission elements in the adjusted working parameters of the ultrasound probe respectively;
[0054] In step A6, the control unit controls the connected transmitting element to transmit ultrasonic waves according to the received transmitting parameters.
[0055] The beneficial technical effect of the present invention is that it can automatically identify and learn the optimal beam settings from a large amount of image data through advanced data analysis based on machine learning, and adjust the beam in real time to adapt to different in vivo structures encountered during the scanning process without manual adjustment. It reduces the dependence on the operator's technical level, reduces the complexity of operation, and can provide more stable and reliable image quality, significantly improving the performance and application level of ultrasound imaging technology. In addition, not only the currently collected ultrasound image, but also the historical adjustment data before the scanning process are combined to form the overall beam optimization parameters. Due to the introduction of historical adjustment data, the optimization deficiencies brought about by the previous beam optimization parameters can be used as a reference, and the deficiencies can be further improved, reducing the overall duplication and proximity of the optimization parameters, and further speeding up the acquisition of the optimal ultrasound image. BRIEF DESCRIPTION OF THE DRAWINGS
[0056] Figure 1-3 A schematic diagram of a module of an adaptive beam adjustment system for a handheld ultrasound device according to the present invention;
[0057] Figure 4-5 The present invention is a flowchart of the steps of an adaptive beam adjustment method for a handheld ultrasonic device. DETAILED DESCRIPTION
[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.
[0059] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.
[0060] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.
[0061] See also Figure 1 The present invention provides an adaptive beam adjustment system for a handheld ultrasonic device, comprising:
[0062] An ultrasonic probe (1) is used to generate an ultrasonic transmission beam and receive an ultrasonic echo signal;
[0063] A signal processing module (2), connected to the ultrasonic probe (1), for preprocessing the ultrasonic echo signal;
[0064] An image forming module (3) is connected to the signal processing module (2) and is used to process the pre-processed ultrasonic echo signal to form an ultrasonic image;
[0065] An adaptive adjustment module (4) is connected to the image forming module (3) and is used to process the ultrasound image based on a pre-trained image processing model and output an image evaluation result, and to process the current image evaluation result and historical adjustment data based on the pre-trained adaptive adjustment model and output a beam optimization parameter;
[0066] A beam control module (5) connected to the adaptive adjustment module (4) and used to adjust the working parameters of the ultrasound probe (1) according to the beam optimization parameters;
[0067] The ultrasonic probe (1) generates a new ultrasonic transmission beam and receives an ultrasonic echo signal according to the adjusted working parameters;
[0068] The historical adjustment data includes the working parameters of the ultrasound probe, the image evaluation results, and the beam optimization parameters corresponding to the historical scanning period from the time when the scanning starts to the current time.
[0069] The adaptive adjustment module usually continuously adjusts parameters to find the best imaging quality. Set the stop condition of the adaptive model, such as converging to a certain image quality index or reaching a preset number of iterations. Once the stop condition is met, the adaptive adjustment process will automatically terminate. After the adaptive model is adjusted, the system can issue an alarm or prompt to notify the operator that the next step can be taken. After the adaptive model is adjusted, the display interface of the display module can refresh and display the final imaging results, while locking the current ultrasound image to prevent misoperation.
[0070] Specifically, the handheld ultrasound device is a handheld ultrasound medical wearable device.
[0071] By integrating machine learning technology and adaptive beam forming technology, the optimal beam settings are automatically identified and learned from a large amount of image data, and the beam is adjusted in real time to adapt to different in vivo structures encountered during the scanning process without manual adjustment. It reduces the dependence on the operator's technical level, reduces the complexity of operation, and can provide more stable and reliable image quality to significantly improve image quality, imaging efficiency and intelligence. In addition, not only the currently acquired ultrasound images, but also the historical adjustment data before the scanning process are combined to form the overall beam optimization parameters. Due to the introduction of historical adjustment data, the optimization deficiencies brought about by the previous beam optimization parameters can be learned, and the deficiencies can be further improved, reducing the overall duplication and proximity of the optimization parameters, and further speeding up the acquisition of the optimal ultrasound image.
[0072] See also Figure 2 , further, the ultrasonic probe (1) includes a plurality of independent transmitting elements (101) and independent receiving elements (102);
[0073] Each transmitting element (101) is independently configured with a control unit (103), and each control unit (103) is connected to a beam control module (5);
[0074] The operating parameters of the ultrasonic probe (1) generated by the beam control module (5) include the number of transmitting elements (101), the positions of the transmitting elements (101) and the transmitting parameters;
[0075] The beam control module (5) sends corresponding transmission parameters to the control unit (103) of the transmission element (101) in the adjusted working parameters of the ultrasonic probe respectively;
[0076] The control unit (101) controls the connected transmitting element (101) to transmit ultrasonic waves according to the received transmitting parameters.
[0077] The operating parameters of the ultrasonic probe (1) generated by the beam control module (5) also include transmission frequency, phase, etc. Transmission parameters are, for example:
[0078] Transmit voltage: affects the strength of the transmitted signal. A higher transmit voltage can enhance the signal, but it may also increase noise.
[0079] Transmit waveform: Set the waveform of the transmit signal, such as long pulse, short pulse, etc., to adapt to different application scenarios.
[0080] Delay Time: The relative delay setting of each transmitting element helps to correctly direct the beam.
[0081] The ultrasonic probe comprises a plurality of transmitting elements (101) and receiving elements (102), which are used to generate and receive ultrasonic signals respectively.
[0082] The transmitting elements (101) can be designed in an array, and each transmitting element (101) can be controlled independently. The receiving elements can be designed in an array, and each receiving element can be controlled independently.
[0083] Furthermore, the control unit (103) of the transmitting unit (101) includes an FPGA. As the system complexity and functional requirements increase, the FPGA can be introduced to achieve dynamic control. The FPGA provides parallel processing capabilities and can dynamically adjust the operating parameters of each transmitting element (101) (such as the phase and amplitude of the transmitting signal), thereby achieving more complex signal processing, such as adaptive beamforming and real-time imaging.
[0084] The transmitting element / receiving element is a piezoelectric element. The appropriate piezoelectric material (such as PZT or PMN-PT) is selected to ensure that it has good electroacoustic conversion efficiency in the required frequency range. The arrangement can be linear or matrix, using a two-dimensional matrix array consisting of 128 elements. The density and arrangement of piezoelectric elements will directly affect the imaging resolution and depth. High-density arrangement will provide higher resolution, but it also requires more complex electronic control and higher processing power.
[0085] Common piezoelectric materials include PZT (lead zirconate titanate) and PMN-PT (barium zirconium titanium lanthanum niobate).
[0086] The arrangement of piezoelectric elements has a direct impact on imaging quality. Common arrangements include linear arrays and matrix arrays. A highly integrated 128-element two-dimensional matrix array is used. This array allows the probe to collect data in multiple directions simultaneously, thereby improving imaging resolution and depth. The matrix arrangement can generate three-dimensional images, which is suitable for more complex imaging needs.
[0087] Piezoelectric elements are arranged linearly in one dimension, for example, 128 elements are arranged along one axis. This arrangement is simple and suitable for linear scanning and imaging.
[0088] The working parameters of the ultrasonic probe generated by the beam control module (5) also include the selected receiving element (102). Each receiving element (102) is independently configured with a controller, and the beam control module (5) sends a receiving control signal to the controller of the selected receiving element (102), and the controller of the selected receiving element (102) receives the ultrasonic echo signal according to the receiving control signal.
[0089] The working parameters of the ultrasonic probe (1) include frequency, and the commonly used ultrasonic frequency range is 1-15MHz. The choice of frequency is closely related to the depth of imaging and tissue characteristics. Low-frequency signals (1-5MHz) are suitable for imaging deeper tissues because low-frequency signals have better penetration ability, but the resolution is lower. Medium-frequency signals (5-10MHz) are suitable for imaging at medium depths, taking into account both penetration depth and resolution. High-frequency signals (10-15MHz): Suitable for imaging superficial tissues, with higher resolution, but poorer penetration ability.
[0090] The working parameters of the ultrasound probe also include the waveform type. Short Pulse can achieve higher time resolution by generating extremely short high-frequency pulse signals, because short pulses can accurately locate echo signals in time. Chirp Signal can optimize the signal spectrum, increase energy transfer efficiency, and improve image contrast and resolution by adjusting the frequency (such as linear frequency modulation) during the transmission process.
[0091] The control unit (103) also includes a signal generator and a power amplifier. The signal generator generates a high-precision, adjustable transmission signal waveform. The power amplifier amplifies the generated signal to a required amplitude and drives the transmitting element to emit an ultrasonic beam.
[0092] The received ultrasonic echo signal is very weak, contains a lot of background noise and unnecessary frequency components, and needs to be processed in a complex manner before it can be used for imaging. Therefore, a signal processing module (2) is set to pre-process the received ultrasonic echo signal and extract the effective ultrasonic signal.
[0093] The signal processing module (2) comprises a preamplifier, a filter and an analog-to-digital converter.
[0094] The pre-amplifier amplifies the received ultrasonic echo signal to a processable level while minimizing the introduced electrical noise.
[0095] The filter performs filtering on the amplified ultrasonic echo signal. The band-pass filter is used to remove non-target frequency components in the received ultrasonic echo signal and retain only the signal within the frequency range of interest. This frequency range is usually the same as the frequency of the ultrasonic transmission signal or its harmonic frequency is close. The low-pass filter further filters out high-frequency noise to ensure the signal is pure.
[0096] The analog-to-digital converter (ADC) converts the filtered ultrasound echo signal from an analog signal to a digital signal. The high-precision, high-speed ADC converts the processed analog signal into a digital signal to ensure the fidelity and details of the signal.
[0097] The signal preprocessed by the signal processing module (2) serves as the basis for subsequent high-quality image generation and analysis. The precise design and execution of these steps directly affect the imaging performance and diagnostic accuracy of the entire ultrasound system.
[0098] The signal processing module (2) selects a high-performance signal processing chip or FPGA (Field Programmable Gate Array). The high-performance signal processing chip is such as an application-specific integrated circuit (ASIC) or a high-performance computing platform (such as GPU).
[0099] See also Figure 3 ,Furthermore, the image processing model includes a pre-trained image segmentation model and an image evaluation ,model;
[0100] The adaptive adjustment module (4) comprises:
[0101] An image segmentation unit (41) is used to perform segmentation processing on the ultrasound image based on the image segmentation model to obtain an image segmentation result;
[0102] The image evaluation unit (42) is connected to the image segmentation unit (41) and is used to evaluate the image quality index of the image segmentation result based on the image evaluation model to obtain an image evaluation result.
[0103] The parameter optimization unit (43) is connected to the image evaluation unit (42) and is used to process the current image evaluation result and the historical adjustment data based on the pre-trained adaptive adjustment model and output the beam optimization parameters.
[0104] The adaptive adjustment module (4) uses an embedded processor to run a machine learning model (such as a convolutional neural network) to optimize beam parameters in real time.
[0105] The adaptive adjustment model can be CNN / deep learning, random forest or support vector machine, etc. The system provides real-time evaluation values and feedback of image quality indicators, which are displayed together with the ultrasound image on the display module in real time, so that the operator can intuitively see the progress and effect of the adaptive adjustment. In addition to displaying the current ultrasound image and its image quality indicators in real time, the display module also shows the changing trend of each image quality indicator during the scanning process.
[0106] The image segmentation model is mainly based on the artificial intelligence network model to identify the key anatomical structures and lesion characteristics of the human body scanning area.
[0107] The image quality assessment of the image assessment model focuses on the evaluation of these key anatomical structures and lesion characteristics.
[0108] Regarding the optimization combined with historical adjustment data, for example, the image resolution is related to the ultrasonic frequency, the number and position of the transmitting elements. If the ultrasonic frequency, the number and position of the transmitting elements already exist in the historical working parameters, resulting in the image resolution not being optimized and improved, then in order to improve the image resolution, when adjusting the current working parameters, it will avoid repeating the same settings of the ultrasonic frequency, the number and position of the transmitting elements that appeared in history, and comprehensively select other ultrasonic frequencies, the number and position of the transmitting elements, and determine these parameters that can improve the image resolution based on the trend of historical settings. It can be seen that the introduction of historical adjustment data can, on the one hand, avoid repeated parameter settings and improve adjustment efficiency, and on the other hand, it can determine the optimization trend and balance of parameters based on the trend of historical parameters and the quality of scanning results. For example, assuming that the more concentrated the transmitting elements are in history, the higher the resolution is, the ultrasonic frequency can be further increased to achieve efficient and reliable parameter settings. It can be seen that the addition of historical adjustment data can avoid repeated parameter settings and accelerate the acquisition of desired image quality.
[0109] Furthermore, the image processing model also includes a pre-trained image classification model;
[0110] Different scanning parts have their own corresponding image segmentation models and image evaluation models;
[0111] The adaptive adjustment module (4) further comprises:
[0112] A part acquisition unit (44) is used to acquire preset scanning part information;
[0113] An image classification unit (45) connected to a part acquisition unit (44) is used to classify and identify the ultrasound image based on an image classification model when no scan part information is set in advance, and identify the scan part of the ultrasound image;
[0114] The model calling unit (46) is connected to the part acquisition unit (44) and the image classification unit (45) respectively, and is used for:
[0115] When the preset scanning part information is obtained, the corresponding image segmentation model and image evaluation model are called according to the preset scanning part information; and
[0116] When the scanning part information is not set in advance, the corresponding image segmentation model and image evaluation model are called according to the scanning part identified by the image classification model.
[0117] Further, the image quality index includes at least one of resolution, contrast, edge definition, signal-to-noise ratio, etc. When all of these indexes reach a set threshold, the adaptive adjustment is stopped.
[0118] Edge definition, such as edge definition of key anatomical structures and lesion areas.
[0119] Furthermore, the beam optimization parameters include at least one of depth, range, frequency, edge enhancement, focusing, main lobe width, side lobe level, etc.
[0120] Depth: Imaging depth, usually related to the distance of the detected target.
[0121] Range: The horizontal and vertical detection range also reflects the image coverage area. If the coverage area of the key anatomical structure or lesion area in the entire ultrasound image is small, the range optimization parameters can be used to increase the coverage area of the key anatomical structure or lesion area.
[0122] Frequency: The frequency at which ultrasound waves are emitted, which is generally related to image resolution and penetration depth. High frequency provides higher resolution, and low frequency provides better penetration.
[0123] Edge Enhancement: Improves the clarity of image edge details. This parameter can be used to further scan and improve the edge clarity of key anatomical structures and lesion areas.
[0124] Focusing: Focuses at a specified depth to optimize resolution.
[0125] Main Lobe Width: The width of the main energy concentration area of the beam, which affects the resolution and target detection capabilities.
[0126] Side Lobe Level: Compared with the side lobe strength of the main lobe, a lower side lobe level can reduce interference to side signals and enhance image quality.
[0127] Furthermore, each emitting element (101) is provided with a temperature sensor (104) for collecting temperature data of the emitting element;
[0128] The temperature sensor (104) in each transmitting element (101) is connected to the control unit (103), and the control unit (104) transmits the temperature data of the transmitting element (101) to the beam control module (5) in real time;
[0129] The beam control module (5) is also used to adjust the working parameters of the ultrasound probe in combination with the temperature data of each transmitting element (101) and the beam optimization parameters.
[0130] The ultrasonic probe (1) generates heat during use. If the heat is not dissipated in time, it may affect the performance of the probe or even damage the internal components. Therefore, effective temperature control and heat dissipation design must be adopted. A thermistor is used as a temperature sensor to monitor the temperature in real time, and the beam control module controls the temperature by, for example, adjusting the frequency and power of the transmission signal. When the temperature is too high, the transmission power can be reduced or the duty cycle of the transmission pulse can be reduced to reduce heat generation.
[0131] During operation, the temperature of the transmitting element is continuously monitored to ensure that it operates within a safe range. Based on the temperature monitoring data, the transmitting parameters are adaptively adjusted. For example, under high temperature conditions, the transmitting voltage or transmitting frequency may need to be reduced to reduce heat generation.
[0132] The beam control module (5) has an overheat protection mechanism, a temperature threshold and a protection mechanism. When the temperature exceeds a safe range, the working state of the transmitting element is automatically adjusted, or the transmission is suspended to prevent damage.
[0133] Use efficient heat dissipation materials and structural designs, such as embedding metal heat sinks in the probe housing, or combining small fans and heat pipe technology to improve heat dissipation efficiency.
[0134] Furthermore, it also includes a display module (6), which is connected to the image forming module (3) and is used to display the formed ultrasound image in real time.
[0135] The display module uses VR / AR technology. VR (Virtual Reality) and AR (Augmented Reality) are two advanced digital technologies that have changed the way we interact with digital content and the real world. Using VR / AR technology instead of traditional display screens provides more intuitive and multi-dimensional imaging display.
[0136] To train the model, a large amount of high-quality training data is required. This data includes ultrasound signals of different angles, depths, and tissue types. A large amount of signal data is collected under various conditions using an ultrasound probe, and the relevant working parameters are recorded. Data preprocessing includes signal denoising, normalization, and feature extraction. Preliminary processing is performed using existing signal processing algorithms to ensure data consistency and quality.
[0137] First, collect ultrasound signals at different angles, depths, and tissue types. Choose the right ultrasound probe: Probes of different frequencies are suitable for different scanning scenarios. Choosing the right probe helps obtain high-quality signals. Set a variety of acquisition conditions, such as angle, depth, and tissue type. Angle Scan from different angles to obtain comprehensive information about the sample. Depth Adjust the depth of the probe to capture tissue information at different levels. Tissue type covers various types of tissues, such as muscle, fat, bone, etc. These acquisition parameters should be recorded each time data is collected to facilitate subsequent data annotation and classification.
[0138] Secondly, the collected signal data is preprocessed by denoising, normalizing and feature extraction. Use filters (such as low-pass, high-pass, and band-pass filters) to remove noise. Wavelet denoising technology can be considered. Normalization is to standardize the signal data to a specific range (such as between 0 and 1) to facilitate subsequent processing and model training. Feature extraction is to extract signal features using existing signal processing algorithms (such as FFT and Hilbert transform). Extract features in the time domain and frequency domain, such as mean, variance, peak, spectral features, etc.
[0139] Again, the preprocessed data is labeled and stored securely for subsequent training. Labels are annotated according to the working parameters and the diagnostic results of the ultrasound image. Experts classify and label the data according to image quality and clarity. The labeling process includes identifying characteristic areas in the signal (such as tissue boundaries, lesions, etc.), and scoring the image quality to obtain the values of image quality indicators (such as clarity, contrast). The labeled data can be used to train and verify the machine learning model to ensure that the model can automatically adjust the beamforming parameters according to the input data to optimize the imaging effect.
[0140] Finally, design and select a suitable machine learning model architecture, such as a convolutional neural network (CNN) or other deep learning models, and train and validate the model.
[0141] The system includes an online learning module that continuously improves the image processing model and adaptively adjusts the model performance during operation to adapt to the new environment, such as online stochastic gradient descent (Online SGD) or Bayesian online learning (Bayesian Online Learning). The online learning module needs to run efficiently under the computing power limitations of the device, extract new features from real-time collected data and update model weights to ensure that the model always maintains optimal performance in the new operating environment.
[0142] The system also includes a model training and update module that is not in the handheld ultrasound device, which regularly updates the image processing model and the adaptive adjustment model. The model training and update module can be in the cloud, using more powerful computing resources for large-scale model training and optimization. The new model trained in the cloud is downloaded to the handheld ultrasound device through OTA (Over-the-Air) upgrade. In order to reduce service interruptions during model updates, model version control and hot update technology can be used to deploy new models without interrupting system operation.
[0143] The system of the present invention can be applied to various parts of the body such as heart examination, liver examination, pregnancy monitoring, carotid artery detection, etc. In remote areas or during emergency treatment, handheld ultrasound equipment can provide high-quality images and provide accurate imaging information for doctors to remotely diagnose and deal with emergencies.
[0144] See also Figure 4 The present invention also provides an adaptive beam adjustment method for a handheld ultrasonic device, using the aforementioned adaptive beam adjustment system for a handheld ultrasonic device, comprising:
[0145] Step A1, generating an ultrasonic transmission beam and receiving an ultrasonic echo signal;
[0146] Step A2, preprocessing the ultrasonic echo signal;
[0147] Step A3, processing the preprocessed ultrasonic echo signal to form an ultrasonic image;
[0148] Step A4, processing the ultrasound image based on the pre-trained image processing model and outputting the image evaluation result, processing the current image evaluation result and the historical adjustment data based on the pre-trained adaptive adjustment model and outputting the beam optimization parameters;
[0149] Step A5, adjusting the working parameters of the ultrasound probe according to the beam optimization parameters;
[0150] Step A6, the ultrasonic probe generates a new ultrasonic transmission beam according to the adjusted working parameters and receives an ultrasonic echo signal, and then executes step A2;
[0151] The historical adjustment data includes the working parameters of the ultrasound probe, the image evaluation results, and the beam optimization parameters corresponding to the historical scanning period from the time when the scanning starts to the current time.
[0152] Furthermore, the ultrasound probe includes a plurality of independent transmitting elements and independent receiving elements;
[0153] Each transmitting element is independently configured with a control unit;
[0154] The operating parameters of the ultrasound probe adjusted in step A5 include the number of transmitting elements, the positions of the transmitting elements, and the transmitting parameters;
[0155] Step A5 also includes: sending corresponding transmission parameters to the control units of the transmission elements in the adjusted working parameters of the ultrasound probe respectively;
[0156] In step A6, the control unit controls the connected transmitting element to transmit ultrasonic waves according to the received transmitting parameters.
[0157] See also Figure 5 ,Furthermore, the image processing model also includes a pre-trained image classification model;
[0158] The process of forming the image evaluation result in step A4 includes:
[0159] Step A41, performing segmentation processing on the ultrasound image based on the image segmentation model to obtain an image segmentation result;
[0160] Step A42: performing image quality index evaluation on the image segmentation result based on the image evaluation model to obtain an image evaluation result.
[0161] Furthermore, the image processing model also includes a pre-trained image classification model; different scanning parts have their own corresponding image segmentation models and image evaluation models;
[0162] Before step A41, the following steps are also included:
[0163] Step A401, obtaining the preset scanning part information:
[0164] If obtained, execute step A403;
[0165] If not set in advance, execute step A402;
[0166] Step A402, classifying and identifying the ultrasound image based on the image classification model, identifying the scanned part of the ultrasound image, and then executing step A404;
[0167] Step 403, calling the corresponding image segmentation model and image evaluation model according to the preset scanning part information, and then executing step A41;
[0168] Step A404, calling the corresponding image segmentation model and image evaluation model according to the scanned part identified by the image classification model, and executing step A41.
[0169] Furthermore, in step A42, the image quality index includes at least one of resolution, contrast, clarity, signal-to-noise ratio, and the like.
[0170] Further, in step A4, the beam optimization parameters include at least one of depth, range, frequency, edge enhancement, focusing, main lobe width, side lobe level, etc.
[0171] Furthermore, each emitting element is equipped with a temperature sensor for collecting temperature data of the emitting element;
[0172] In step A5, the operating parameters of the ultrasound probe are adjusted in combination with the temperature data of each transmitting element and the beam optimization parameters.
[0173] Furthermore, step A3 further includes step A7, which displays the formed ultrasonic image in real time.
[0174] The beneficial effects of the present invention are as follows:
[0175] The present invention introduces an intelligent adaptive adjustment mechanism to enable the device to automatically optimize beam parameters and imaging angles, thereby obtaining consistent and high-quality images without relying on operator skills;
[0176] Using advanced machine learning algorithms, beam parameters are dynamically adjusted based on the patient’s specific body shape and tissue characteristics to achieve optimal personalized imaging results;
[0177] Through real-time data processing and automated adjustments, the imaging speed and response sensitivity of the equipment are improved, so that high-quality images can be captured more quickly and accurately, improving diagnostic efficiency;
[0178] With the help of machine learning technology, a large amount of historical imaging data is used for training to develop an intelligent adjustment mechanism, which enables the equipment to automatically optimize the imaging process based on real-time data;
[0179] Through automated and intelligent technical solutions, operators can reduce the need to adjust complex equipment parameters, making equipment operation easier and lowering the learning and usage threshold;
[0180] The above are only preferred embodiments of the present invention, and are not intended to limit the implementation methods and protection scope of the present invention. Those skilled in the art should be aware that all solutions obtained by equivalent substitutions and obvious changes made using the description and illustrations of the present invention should be included in the protection scope of the present invention.
Claims
1. An adaptive beam adjustment system for a handheld ultrasound device, It is characterized in that include: An ultrasonic probe, used for generating an ultrasonic transmission beam and receiving an ultrasonic echo signal; A signal processing module, connected to the ultrasonic probe, for preprocessing the ultrasonic echo signal; An image forming module, connected to the signal processing module, for processing the pre-processed ultrasonic echo signal to form an ultrasonic image; An adaptive adjustment module, connected to the image forming module, for processing the ultrasound image based on a pre-trained image processing model and outputting an image evaluation result, and for processing the current image evaluation result and historical adjustment data based on a pre-trained adaptive adjustment model and outputting a beam optimization parameter; A beam control module, connected to the adaptive adjustment module, used to adjust the working parameters of the ultrasound probe according to the beam optimization parameters; The ultrasonic probe generates a new ultrasonic transmission beam and receives an ultrasonic echo signal according to the adjusted working parameters; The historical adjustment data includes the working parameters of the ultrasound probe corresponding to the historical scanning period from the start of the scanning to the current time, the image evaluation results and the beam optimization parameters.
2. The adaptive beam adjustment system for a handheld ultrasound device according to claim 1, It is characterized in that The ultrasonic probe includes a plurality of independent transmitting elements and independent receiving elements; Each transmitting element is independently configured with a control unit, and each of the control units is connected to the beam control module; The working parameters of the ultrasound probe adjusted by the beam control module include the number of transmitting elements, the positions of the transmitting elements and the transmitting parameters; The beam control module sends corresponding transmission parameters to the control unit of the transmitting element in the adjusted working parameters of the ultrasound probe respectively; The control unit controls the connected transmitting element to transmit ultrasonic waves according to the received transmitting parameters.
3. The adaptive beam adjustment system for a handheld ultrasound device according to claim 1, It is characterized in that The image processing model includes a pre-trained image segmentation model and an image evaluation model; The adaptive adjustment module comprises: An image segmentation unit, configured to perform segmentation processing on the ultrasound image based on the image segmentation model to obtain an image segmentation result; An image evaluation unit, connected to the image segmentation unit, and configured to perform image quality index evaluation on the image segmentation result based on the image evaluation model to obtain the image evaluation result; A parameter optimization unit is connected to the image evaluation unit and is used to process the current image evaluation result and historical adjustment data based on a pre-trained adaptive adjustment model and output beam optimization parameters.
4. The adaptive beam adjustment system for a handheld ultrasound device according to claim 3, It is characterized in that The image processing model also includes a pre-trained image classification model; Different scanning parts have their own corresponding image segmentation models and image evaluation models; The adaptive adjustment module also includes: A part acquisition unit, used to acquire preset scanning part information; An image classification unit, configured to classify and identify the ultrasound image based on the image classification model when the scanned part information is not set in advance, and identify the scanned part of the ultrasound image; The model calling unit is connected to the part acquisition unit and the image classification unit respectively, and is used for: When the preset scanning part information is obtained, calling the corresponding image segmentation model and image evaluation model according to the preset scanning part information; and When the scanning part information is not set in advance, the corresponding image segmentation model and image evaluation model are called according to the scanning part identified by the image classification model.
5. The adaptive beam adjustment system for a handheld ultrasound device according to claim 3, It is characterized in that The image quality index includes at least one of resolution, contrast, clarity, and signal-to-noise ratio.
6. The adaptive beam adjustment system for a handheld ultrasound device according to claim 3, It is characterized in that The beam optimization parameters include at least one of depth, range, frequency, edge enhancement, focusing, main lobe width, and side lobe level.
7. The adaptive beam adjustment system for a handheld ultrasound device according to claim 2, It is characterized in that Each of the emitting elements is equipped with a temperature sensor for collecting temperature data of the emitting element; The temperature sensor in each of the transmitting elements is connected to the control unit, and the control unit transmits the temperature data of the transmitting element to the beam control module in real time; The beam control module is further used to adjust the working parameters of the ultrasound probe in combination with the temperature data of each transmitting element and the beam optimization parameters.
8. The adaptive beam adjustment system for a handheld ultrasound device according to claim 2, It is characterized in that It also includes a display module, which is connected to the image forming module and is used to display the formed ultrasound image in real time.
9. An adaptive beam adjustment method for a handheld ultrasound device, It is characterized in that An adaptive beam adjustment system for a handheld ultrasound device according to any one of claims 1 to 8, comprising: Step A1, generating an ultrasonic transmission beam and receiving an ultrasonic echo signal; Step A2, preprocessing the ultrasonic echo signal; Step A3, processing the preprocessed ultrasonic echo signal to form an ultrasonic image; Step A4, processing the ultrasound image based on a pre-trained image processing model and outputting an image evaluation result, processing the current image evaluation result and historical adjustment data based on a pre-trained adaptive adjustment model and outputting a beam optimization parameter; Step A5, adjusting the working parameters of the ultrasound probe according to the beam optimization parameters; Step A6, the ultrasonic probe generates a new ultrasonic transmission beam according to the adjusted working parameters and receives an ultrasonic echo signal, and then executes step A2; The historical adjustment data includes the working parameters of the ultrasound probe corresponding to the historical scanning period from the start of the scanning to the current time, the image evaluation results and the beam optimization parameters.
10. The adaptive beam adjustment method for a handheld ultrasonic device according to claim 9, It is characterized in that The ultrasonic probe includes a plurality of independent transmitting elements and independent receiving elements; Each transmitting element is independently configured with a control unit; The working parameters of the ultrasonic probe generated in step A5 include the number of transmitting elements, the positions of the transmitting elements and the transmitting parameters; The step A5 further includes: sending corresponding transmission parameters to the control unit of the transmitting element in the adjusted working parameters of the ultrasonic probe respectively; In step A6, the control unit controls the connected transmitting element to transmit ultrasonic waves according to the received transmitting parameters.
Citation Information
Cited By
Ultrasonic imaging optimization method and system for real-time correction based on multi-angle deflection feature points
CN120471789A
Adaptive data determination method and equipment for liver cancer detection equipment database
CN122436259A
