Ultrasonic signal processing system and method, and brain-like calculation module

Abstract

The invention is suitable for the technical field of medical imaging, and provides an ultrasonic signal processing system and method and a brain-like calculation module.The system comprises an ultrasonic transducer, an ultrasonic transmitting module and the brain-like calculation module.The ultrasonic transducer is coupled with the ultrasonic transmitting module and the brain-like calculation module; the ultrasonic transmitting module is used for controlling the ultrasonic transducer to transmit ultrasonic waves; and the brain-like calculation module is used for processing the echo signal received by the ultrasonic transducer based on brain-like calculation.

Description

Technical Field

[0001] This application belongs to the field of medical imaging, and in particular relates to an ultrasound signal processing system, method, and brain-like computing module. Background Technology

[0002] Ultrasound imaging systems use ultrasound waves as information carriers of the tissue structure of the object being examined (such as the human body). Because different tissue structures have impedance specificity, when ultrasound waves are sent to the object being examined, different echo signals are generated when the ultrasound waves pass through different parts. By building an image based on the echo signals, the structural information of the corresponding parts can be observed.

[0003] In related technologies, the received echo signal is an analog signal, while subsequent signal processing requires a digital signal. Therefore, an analog-to-digital converter (ADC) is needed to convert the echo signal into a digital signal before further processing. This conversion process can also be called sampling or quantization. A low bit depth of the ADC limits the sensitivity of ultrasound imaging, while a high bit depth ADC, although ensuring sensitivity, significantly increases the amount of sampled data. This places a heavy burden on the system for transmitting and processing digital ultrasound data, easily reaching a bottleneck. The ADC speed also limits the frame rate of ultrasound imaging. It can be seen that the method of first sampling the echo signal into a multi-bit digital signal before processing limits the effectiveness of ultrasound imaging. Summary of the Invention

[0004] This application provides an ultrasound signal processing system, method, and neuromorphic computing module, which can solve the problem that the processing method of echo signals in related technologies limits the effect of ultrasound imaging.

[0005] In a first aspect, embodiments of this application provide an ultrasonic signal processing system, which includes: an ultrasonic transducer, an ultrasonic transmitting module, and a neuromorphic computing module. The ultrasonic transducer is coupled to the ultrasonic transmitting module and the neuromorphic computing module respectively. The ultrasonic transmitting module is used to control the ultrasonic transducer to emit ultrasonic waves. The neuromorphic computing module is used to process the echo signals received by the ultrasonic transducer based on neuromorphic computing.

[0006] Secondly, embodiments of this application provide an ultrasonic signal processing method applied to a brain-like computing module. The method includes: receiving an echo signal from an ultrasonic transducer, wherein the echo signal is received by the ultrasonic transducer and generated by the reflection of ultrasonic waves emitted by the ultrasonic transducer by the object being detected; and processing the echo signal based on brain-like computing.

[0007] Thirdly, embodiments of this application provide a brain-like computing module for executing the ultrasound signal processing method described in the second aspect above.

[0008] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the ultrasonic signal processing method described in the second aspect above.

[0009] Fifthly, embodiments of this application provide a computer program product that, when run on an ultrasound imaging system, causes the ultrasound imaging system to execute the ultrasound signal processing method described in the second aspect above.

[0010] The beneficial effects of this application embodiment compared with the prior art are as follows: By providing an ultrasonic signal processing system, including an ultrasonic transducer, an ultrasonic transmitting module, and a neuromorphic computing module, the ultrasonic transducer is coupled to the ultrasonic transmitting module and the neuromorphic computing module respectively; the ultrasonic transmitting module is used to control the ultrasonic transducer to emit ultrasonic waves; the neuromorphic computing module is used to process the echo signal received by the ultrasonic transducer based on neuromorphic computing; the object of neuromorphic computing is not a traditional digital signal, but a pulse signal, so the analog-to-digital converter can be discarded, and the effect of ultrasonic imaging is no longer limited by the performance and cost of the analog-to-digital converter, which helps to improve the effect of ultrasonic imaging; at the same time, the possible values ​​of the pulse signal are generally 0 or 1. Compared with 8-bit or even more-bit digital signals, the pulse signal has obvious sparsity, thereby greatly reducing the bandwidth, power consumption (the large reduction in the number of bits significantly reduces the power consumption caused by level switching) and other resources required for data transmission and processing. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a schematic diagram of the structure of an ultrasound imaging system in related technologies;

[0013] Figure 2 This is a schematic diagram of the structure of an ultrasound imaging system provided in an embodiment of this application;

[0014] Figure 3 This is a schematic diagram of the structure of an ultrasound imaging system provided in a specific embodiment of this application;

[0015] Figure 4 This is a schematic diagram of the structure of an ultrasound imaging system provided in another specific embodiment of this application;

[0016] Figure 5 This is a schematic diagram of an excitation-inhibition model provided in a specific example of this application;

[0017] Figure 6 This is a schematic diagram of the structure of a spiking neural network obtained by pulsedizing a convolutional neural network, provided in a specific example of this application;

[0018] Figure 7 This is a schematic flowchart of an embodiment of the ultrasonic signal processing method provided in this application;

[0019] Figure 8 yes Figure 7 A specific process diagram of S2 in China;

[0020] Figure 9 Based on Figure 8 The results of the experiment were compared using the procedure shown in the figure. Detailed Implementation

[0021] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0022] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0023] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0024] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0025] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0026] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0027] For ease of understanding, the following description, in conjunction with the accompanying drawings, illustrates the principle and specific structure of an ultrasound imaging system in the related art.

[0028] like Figure 1 As shown, an ultrasound imaging system in the related technology includes an ultrasound transducer 10, an ultrasound transmitting module 20, an ultrasound receiving module 30, a data preprocessing module 40, an information extraction module 50, a host computer 60, and an ultrasound transmitting wavefront module 70.

[0029] The ultrasonic transducer 10 is a key energy conversion device in the ultrasonic imaging system, responsible for converting between electrical energy and mechanical energy (acoustic energy). Specifically, when the ultrasonic transducer 10 receives an electrical signal, it converts the electromagnetic field change caused by the electrical signal into mechanical vibration based on the direct piezoelectric effect, thereby generating ultrasonic waves. Conversely, when the ultrasonic transducer 10 receives echoes generated by the reflection / scattering of ultrasound from tissue within the object being examined, it converts the mechanical vibration into electromagnetic field change based on the inverse piezoelectric effect, thereby outputting an echo signal.

[0030] The number of ultrasonic transducers 10 can be greater than or equal to one. In practical applications, in order to improve the imaging speed, ultrasonic imaging systems often include multiple arrays of ultrasonic transducers 10. Each ultrasonic transducer 10 can independently transmit / receive ultrasonic waves, and a single ultrasonic transducer 10 can also be referred to as an array element or channel.

[0031] The ultrasonic transmitting module 20 is used to control the ultrasonic transducer 10 to emit ultrasonic waves. Specifically, the ultrasonic transmitting module 20 can generate and provide the electrical signals required to emit ultrasonic waves to each ultrasonic transducer 10 according to the received control instructions. The ultrasonic waves can be emitted in a pulsed manner. The control instructions can include the emission parameters of each ultrasonic pulse to be emitted next, such as emission time, pulse width, sound intensity, direction, frequency, phase, etc.

[0032] The ultrasonic receiving module 30, also known as an analog-to-digital converter (ADC), is used to convert the analog echo signal received by the ultrasonic transducer 10 into a digital signal and send it to the data preprocessing module 40.

[0033] The data preprocessing module 40 is used to preprocess the digital echo signal and send it to the information extraction module 50, such as time gain compensation (TGC) and filtering. The filtering may include at least one of low-pass filtering, band-pass filtering and decimation filtering.

[0034] The information extraction module 50 is used to extract the feature information of the echo signal from the preprocessed echo signal and send it to the host computer 60. The feature information may include at least one of amplitude, phase, frequency, etc.

[0035] The host computer 60 is used for imaging based on feature information. In addition, the host computer 60 is also used to calculate the overall parameters of the next ultrasonic pulse to be emitted based on the feature information of the current echo signal and send them to the ultrasonic wavefront module 70.

[0036] The ultrasonic wavefront module 70 is used to decompose and calculate the overall parameters of the next ultrasonic pulse to be emitted (e.g., using beamforming technology), obtain the emission parameters of each ultrasonic pulse to be emitted required by each array element, and send them to the ultrasonic emission module 20.

[0037] Optionally, if beamforming technology is not required, the ultrasonic wavefront module 70 can be omitted.

[0038] Optionally, the ultrasonic wavefront module 70 can be integrated with the ultrasonic wavefront module 20 or the host computer 60.

[0039] Figure 2 The diagram shown is a block diagram of a portion of the structure of an ultrasound imaging system provided in one embodiment of this application. (Reference) Figure 2 The ultrasound imaging system includes: an ultrasound transducer 10, an ultrasound transmission module 20, a pulse encoder 80, and a brain-like computing module 90.

[0040] Those skilled in the art will understand that Figure 2 The structure of the ultrasound imaging system shown does not constitute a limitation on the ultrasound imaging system, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0041] The following is combined Figure 2 This section provides a detailed introduction to the various components of an ultrasound imaging system, including those related to... Figure 1 The parts of the ultrasound imaging system shown are identical and will not be described again here. For ease of description, this embodiment uses the integration of the ultrasound wavefront module 70 and the ultrasound transmission module 20 as an example, but in practice, the two can be independent devices.

[0042] The neuromorphic computing module 90 can be a neuromorphic chip. Neuromorphic chips are designed for neuromorphic computing, and their architecture is based on the biological nervous system. This architecture is different from the von Neumann architecture widely used in non-neuromorphic chips, resulting in a significant reduction in power consumption compared to non-neuromorphic chips.

[0043] Alternatively, the neuromorphic computing module 90 can be a non-neuromorphic chip capable of performing neuromorphic computing, specifically a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. In some embodiments, the neuromorphic computing module 90 may include an AI (Artificial Intelligence) processor for handling computational operations related to neural networks.

[0044] In the case where the neuromorphic computing module 90 is not a neuromorphic chip, the ultrasound signal processing system also includes a memory (not shown in the figure). The memory can be independent of the neuromorphic computing module 90 or integrated within it. The memory is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of a computer program. The memory can also be used to temporarily store data required for and generated by the executing program. The memory can include high-speed random access memory and non-volatile memory, such as flash memory, hard disk, multimedia card, card-type memory, etc. The memory can include storage units located inside the neuromorphic computing module 90, such as the hard disk of the neuromorphic computing module 90, and / or removable external storage units, such as portable hard drives, USB flash drives, smart media cards (SMC), secure digital cards (SD cards), etc.

[0045] The neuromorphic computing module 90 is used to process the echo signal received by the ultrasonic transducer 10 based on neuromorphic computing. Specifically, the neuromorphic computing module 90 is used to extract feature information of the echo signal based on neuromorphic computing. (Reference) Figure 1 The ultrasound imaging system shown in this application embodiment has a neuromorphic computing module 90 that at least replaces the data preprocessing module 40 and the information extraction module 50.

[0046] The input to the neuromorphic computing module 90 is a pulse signal, which can be either analog or digital. Figure 1 The analog-to-digital converter in the ultrasound imaging system shown can be omitted.

[0047] The pulse encoder 80 is coupled to the ultrasonic transducer 10 and the neuromorphic computing module 90. The pulse encoder 80 is used to pulse encode the echo signal and send the echo signal to the neuromorphic computing module 90 in the form of a pulse signal.

[0048] The specific pulse encoder 80 can be a pulse width encoder, used to sample the echo signal and encode it in the form of pulse width. Its specific circuit structure is not limited here.

[0049] Alternatively, the pulse encoder 80 can be a pulse frequency encoder, used to sample the echo signal and encode it in the form of a pulse transmission frequency. This encoding method can also be called rate encoding. The circuit structure of the pulse frequency encoder can be obtained by modifying the pulse width encoder. Specifically, a voltage-controlled oscillator (VCO) can be added to the output of the pulse width encoder.

[0050] Optionally, when the neuromorphic computing module 90 is a neuromorphic chip, the echo signal can be input to the neuromorphic computing module 90 in the form of an analog signal. In this case, the pulse encoder 80 can be omitted.

[0051] In some embodiments, the neuromorphic computing module 90 can be further replaced Figure 1 At least some of the functions of the host computer 60 in the ultrasound imaging system shown, particularly in the case that the neuromorphic computing module 90 is not a neuromorphic chip.

[0052] Optionally, the neuromorphic computing module 90 is also used to replace Figure 1 The ultrasound imaging system shown illustrates the transmission control function of the host computer 60. Specifically, the neuromorphic computing module 90 is also coupled to the ultrasound transmission module 20. The neuromorphic computing module 90 is also used to generate and send control information to the ultrasound transmission module 20 based on feature information. The control information is used to control the ultrasound transmission module 20 to adjust the ultrasound waves. The neuromorphic computing module 90 can send control information directly to the ultrasound transmission module 20, or it can send control information to the ultrasound transmission module 20 through the ultrasound transmission wavefront module 70; no limitation is made here.

[0053] Control information can be used to modulate the ultrasound waves to be emitted, including but not limited to transcranial wavefront shaping modulation and region of interest (ROI) local enhancement modulation, thereby adaptively improving the quality of ultrasound imaging.

[0054] Optionally, the neuromorphic computing module 90 is also used to replace Figure 1 The imaging function of the host computer 60 in the ultrasound imaging system shown performs ultrasound imaging based on feature information.

[0055] Optionally, the ultrasound imaging system also includes a host computer 60, which performs functions not replaced by the neuromorphic computing module 90. For example, the neuromorphic computing module 90 does not have transmission control functionality. The host computer 60 is coupled to the neuromorphic computing module 90 and the ultrasound transmission module 20. The neuromorphic computing module 90 is also used to send feature information to the host computer 60. The host computer 60 is used to control the ultrasound transmission module 20 to adjust the ultrasound waves based on the feature information; that is, the host computer 60 generates control information based on the feature information and sends it to the ultrasound transmission module 20.

[0056] Optionally, the ultrasound imaging system also includes a host computer 60, which is coupled to a brain-like computing module 90 and an ultrasound transmission module 20. The brain-like computing module 90 can generate control information based on feature information and then send the feature information and / or control information to the host computer 60.

[0057] The structure of the ultrasound signal processing system can vary depending on the type and function of the neuromorphic computing module 90. For example, in a specific embodiment of this application, the neuromorphic computing module 90 is a neuromorphic chip, and the structure of the ultrasound signal processing system is as follows: Figure 3 As shown, the pulse encoder 80 is omitted, while the host computer 60 is retained and implements imaging and transmission control functions. In a specific embodiment of this application, the neuromorphic computing module 90 is a non-neuromorphic chip, and the structure of the ultrasound signal processing system is as follows. Figure 4 As shown, it includes a pulse encoder 80, while the host computer 60 is omitted and its functions are implemented by the neuromorphic computing module 90.

[0058] Regardless of whether the neuromorphic computing module 90 is a neuromorphic chip, it can run a neuromorphic network to process echo signals.

[0059] Brain-like networks include, but are not limited to, biological spiking neural network (SNN) computational models.

[0060] SNN (Spiral Neural Network) is an artificial neural network model that encodes signals based on the frequency and timing of neural impulses. Unlike traditional artificial neural networks (ANNs), SNNs use impulse signals as the basic information transmission method, simulating the firing process of biological neurons. When input data or neurons are stimulated by external stimuli, the data or stimuli are encoded into specific impulse sequences. These impulse sequences are transmitted and processed between neurons, and the processed impulse sequences are decoded to provide specific responses. SNNs offer higher biological realism, parallel processing capabilities, and energy efficiency.

[0061] The neuromorphic network includes an input layer, an output layer, and a network structure between them. The input layer is used to receive pulse signals. In this embodiment, the input layer includes at least one neuron, and the number of neurons corresponds one-to-one with the array elements in the ultrasonic transducer 10.

[0062] The output layer of the neuromorphic network can directly output pulse signals. For example, the output pulse signal using width encoding can be directly used as the transmission delay in the control information; or, the output pulse signal can be converted into a digital signal by a pulse counter with the corresponding encoding method and then transmitted to the host computer 60. Optionally, the output layer of the neuromorphic network can directly output digital signals.

[0063] The network structure of brain-like networks can be constructed in two ways: one is to pulse a traditional ANN; the other is to use a brain-like simulation model, such as the excitation-inhibition (EI) model.

[0064] For the first construction approach, pulsed ANNs include, but are not limited to, fully connected networks, convolutional neural networks (CNNs), recurrent neural networks (RNNs), residual networks (Res-Net), and U-Nets. Pulsation involves modifying the traditional activation function in the ANN to a pulsed activation function, such as the leakage current accumulation discharge (LIF) function, and changing the weight calculation from product accumulation to Boolean accumulation. For ANNs requiring memory, such as RNNs, Res-Nets, and U-Nets, since neuromorphic networks are temporally sequential and cannot provide their own memory, additional storage modules, such as oscillating circuits, need to be designed to store the network's features.

[0065] For example, an SNN obtained by pulse-transforming a CNN is as follows: Figure 5 As shown, it includes convolutional layers, normalization layers, and LIF activation layers.

[0066] Convolutional layers employ convolutional kernels of different scales to extract features of different scales from the input signal.

[0067] The normalization layer primarily employs batch normalization, which involves normalizing the input of each mini-batch to a mean of 0 and a variance of 1 before scaling and translation. This technique is generally used to handle complex multi-scale ultrasonic feature extraction and can significantly improve model performance.

[0068] Different ultrasonic center frequencies have different attenuation coefficients in the medium, resulting in variations in the received echo amplitude. This poses a significant challenge to the scalability of networks in applications with different media. Two solutions exist: one, as mentioned above, is to incorporate normalization into the model to remove signal differences caused by attenuation in different media. The other solution is to use an adaptive attenuation compensator. Specifically, this involves single-channel transmission calibration. The compensator module is calibrated before network use. The calibration method involves transmitting ultrasonic waves through a single channel while simultaneously receiving them using all other array elements, followed by synthesis and fitting. This method is only suitable for simple scenarios with a small number of array elements.

[0069] Optionally, since ultrasound gradually attenuates as it passes through the internal structure of the object being detected, the echo signal gradually decreases with increasing detection depth. Without compensation, the resulting ultrasound image will exhibit a bright near-field and a dark far-field phenomenon. To address this issue, traditional ultrasound imaging systems employ Time Gain Compensation (TGC) technology during ultrasound signal processing. This involves gain compensation based on the time it takes for the echo signal to return to the array element (corresponding to the distance from the element), making the image appear relatively consistent at different depths. Similarly, in neuromorphic networks, a TGC compensation network can be added after the input layer. This network performs curve correction, specifically implemented using a linear impulse suppressor. The linear impulse suppressor samples the input impulse signal, and the sampling rate increases with the input data.

[0070] When an ultrasound imaging system is applied to different ultrasound frequencies and media, the threshold of the activation function varies, which negatively impacts the generalization performance of the network model. To address this issue, we propose a solution.

[0071] Since the frequency characteristics of the pulsed ultrasonic signal are reflected in the signal attenuation coefficient, the ultrasonic transmission signal (i.e., the electrical signal provided by the ultrasonic transmission module 20 to the ultrasonic transducer 10) can be further introduced as the model input. Specifically, the intensity of the ultrasonic transmission signal can be reduced by a voltage-controlled attenuator (VCAT) before pulsed operation, and then input into the network as prior information. This allows for adaptive adjustment of the activation function threshold, thereby improving the model's generalization ability.

[0072] The above construction method will be illustrated using a dynamic model skeleton based on the EI model as an example.

[0073] like Figure 6As shown, the EI model comprises two neuronal groups, E and I, where E represents the excitatory neuron group and I represents the inhibitory neuron group. Neurons are connected by synapses, and their states change under the influence of input signals, specifically including four cases: E2E, E2I, I2E, and I2I, representing changes from excitation to excitation, from excitation to inhibition, from inhibition to excitation, and from inhibition to inhibition, respectively. Decision information is output based on these changes. The EI model does not restrict the encoding method of the input signal.

[0074] Because ultrasonic beam synthesis involves a well-defined physical process, and this process is first-order without considering attenuation, the received echo signal is input into a homogeneous medium as an input signal, according to the ultrasonic reciprocity theorem. The non-homogeneous scattering bodies that generated the received signal appear sequentially in this medium, forming the ultrasonic image. Using this principle, the received echo signal (which can be in the form of an analog signal) is input into a brain-like network. This network consists of multiple neuron nodes, each containing a set of excitatory and inhibitory neurons. Each neuron node is considered a pixel in the imaging medium. Analogous to the finite element diffusion simulation method, according to the reciprocity theorem, the ultrasound will diffuse between different pixels. During training, the connection strength of each neuron node is determined based on finite element theory and the wave equation. Random coefficients are added according to the neuronal stochastic resonance theory to enhance the network's generalization ability.

[0075] This embodiment provides an ultrasonic signal processing system, including an ultrasonic transducer 10, an ultrasonic transmitting module 20, and a neuromorphic computing module 90. The ultrasonic transducer 10 is coupled to the ultrasonic transmitting module 20 and the neuromorphic computing module 90. The ultrasonic transmitting module 20 controls the ultrasonic transducer 10 to emit ultrasonic waves. The neuromorphic computing module 90 processes the echo signals received by the ultrasonic transducer 10 based on neuromorphic computing. The object of neuromorphic computing is not a traditional digital signal, but a pulse signal. Therefore, the analog-to-digital converter can be discarded, and the effect of ultrasonic imaging is no longer limited by the performance and cost of the analog-to-digital converter, which helps to improve the effect of ultrasonic imaging. At the same time, the possible values ​​of the pulse signal are generally 0 or 1. Compared with 8-bit or more digital signals, the pulse signal has obvious sparsity, which greatly reduces the bandwidth, power consumption (the large reduction in the number of bits significantly reduces the power consumption caused by level switching) and other resources required for data transmission and processing.

[0076] The ultrasound signal processing method provided in this application embodiment can be applied to a neuromorphic computing module 90, which can be a neuromorphic chip or a non-neuromorphic chip. This application embodiment does not impose any restrictions on the specific type of the neuromorphic computing module 90.

[0077] The ultrasound signal processing method provided in this application can be implemented as a computer software program. For example, an embodiment of this application provides a computer program product including a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network, and / or installed from a removable external storage unit. When the computer program is executed by the neuromorphic computing module 90, it implements the various functions defined in the ultrasound signal processing method provided in this application.

[0078] Figure 7 A schematic flowchart of an ultrasound signal processing method provided in an embodiment of this application is shown. It is an example and not a limitation. The method can be applied to the above-mentioned brain-like computing module. The parts that are the same as or similar to those in the foregoing embodiments will not be described again.

[0079] S1: Receives echo signals from the ultrasonic transducer.

[0080] The echo signal is received by the ultrasonic transducer and is generated by the reflection of ultrasonic waves emitted by the ultrasonic transducer by the object being detected.

[0081] S2: Processing echo signals based on neuromorphic computing.

[0082] Feature information of echo signals can be extracted based on neuromorphic computing. Specifically, a neuromorphic network can be run to process the echo signals. The neuromorphic network includes an input layer, in which the number of neurons is the same as the number of array elements in an ultrasonic transducer.

[0083] Optionally, the neuromorphic computing module can generate control information based on the feature information, and the control information is used to control the ultrasonic emission module to adjust the ultrasonic waves.

[0084] like Figure 8 As shown, in a specific embodiment of this application, the control information includes the ultrasonic wave emission delay, and S2 specifically includes the following parts.

[0085] S21: Divide the echo signal into multiple small blocks in the time domain.

[0086] Uniform partitioning, correlation partitioning, maximum entropy partitioning, etc., can be used, and there are no restrictions here.

[0087] S22: Select at least one ROI region from multiple small blocks.

[0088] Each small block can be fed into a corresponding neuron operation node. Then, the corresponding small block region is used as a sub-region input. The outputs of all neuron operation nodes in that region are fed into a comprehensive neuron. The comprehensive neuron is then input into a comparison neuron node, and finally one or more ROI regions are selected.

[0089] S23: Generate the ultrasonic wave emission delay based on the ROI region.

[0090] Based on the corresponding ROI region, the emission delay is generated using existing emission delay algorithms or generators based on neuromorphic networks or ANN networks.

[0091] This specific embodiment draws on brain region theory to determine the Region of Interest (ROI). The ROI represents the frequency information containing the echo signal, thus enabling more accurate generation of the transmission delay to achieve closed-loop modulation of the transmitted ultrasound waves, specifically through local enhancement modulation of the ROI region. The experimental results obtained based on the ultrasound imaging method provided in this specific embodiment are as follows: Figure 9 As shown in the figure, the left side of the image is the ultrasound image obtained without ROI region local enhancement modulation, and the right side is the ultrasound image obtained with ROI region local enhancement modulation. It can be seen that ROI region local enhancement modulation significantly improves the quality of ultrasound imaging (brightness, contrast, etc.).

[0092] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0093] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0094] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0095] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to a photographic device / ultrasound imaging system, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0096] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0097] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0098] In the embodiments provided in this application, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0099] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0100] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. An ultrasonic signal processing system, characterized in that, The system includes an ultrasonic transducer, an ultrasonic transmitting module, and a brain-like computing module, wherein the ultrasonic transducer is coupled to the ultrasonic transmitting module and the brain-like computing module respectively. The ultrasonic transmitting module is used to control the ultrasonic transducer to emit ultrasonic waves. The neuromorphic computing module is used to process the echo signal received by the ultrasonic transducer based on neuromorphic computing.

2. The system as described in claim 1, characterized in that, The system also includes a pulse encoder, and the ultrasonic transducer is coupled to the neuromorphic computing module through the pulse encoder. The pulse encoder is used to pulse encode the echo signal and send the echo signal to the neuromorphic computing module in the form of a pulse signal.

3. The system as described in claim 2, characterized in that, The pulse encoder is a pulse width encoder.

4. The system as described in claim 2, characterized in that, The pulse encoder is a pulse frequency encoder.

5. The system as described in claim 1, characterized in that, The neuromorphic computing module is a neuromorphic chip, and the echo signal is input to the neuromorphic computing module in the form of an analog signal.

6. The system as described in claim 1, characterized in that, The neuromorphic computing module is used to run a neuromorphic network to process the echo signal. The neuromorphic network includes an input layer, the number of neurons in the input layer being the same as the number of array elements in the ultrasonic transducer.

7. The system as described in any one of claims 1-6, characterized in that, The neuromorphic computing module is used to extract feature information of the echo signal based on neuromorphic computing.

8. The system as described in claim 7, characterized in that, The system also includes a host computer, which is coupled to the neuromorphic computing module and the ultrasonic transmission module. The neuromorphic computing module is also used to send the feature information to the host computer, and the host computer is used to control the ultrasonic transmission module to adjust the ultrasonic waves according to the feature information.

9. The system as described in claim 7, characterized in that, The neuromorphic computing module is also used to generate control information based on the feature information, and the control information is used to control the ultrasonic emission module to adjust the ultrasonic waves.

10. The system as described in claim 9, characterized in that, The system also includes a host computer, which is coupled to the neuromorphic computing module and the ultrasonic transmission module. The neuromorphic computing module is also used to send the feature information and / or the control information to the host computer.

11. The system as described in claim 9, characterized in that, The neuromorphic computing module is also coupled to the ultrasonic transmission module, and the neuromorphic computing module is also used to send the control information to the ultrasonic transmission module.

12. The system as described in claim 9, characterized in that, The control information includes the transmission delay of the ultrasound. The neuromorphic computing module is used to divide the echo signal into multiple small blocks in the time domain, select at least one region of interest from the multiple small blocks, and generate the transmission delay based on the region of interest.

13. An ultrasound signal processing method, wherein the ultrasound signal processing method is applied to a neuromorphic computing module, the method comprising: Receive echo signals from an ultrasonic transducer, the echo signals being received by the ultrasonic transducer and generated by the reflection of ultrasonic waves emitted by the ultrasonic transducer by the object being detected; The echo signal is processed using neuromorphic computing.

14. The method as described in claim 13, characterized in that, The brain-like computing-based processing of the echo signal includes: A neuromorphic network is run to process the echo signal. The neuromorphic network includes an input layer, the number of neurons in which is the same as the number of array elements in the ultrasonic transducer.

15. The method as described in claim 13, characterized in that, The operation of the neuromorphic network to process the echo signal includes: The feature information of the echo signal is extracted based on neuromorphic computing.

16. The method as described in claim 15, characterized in that, The process of running the neuromorphic network to process the echo signal also includes: Control information is generated based on the feature information, and the control information is used to control the ultrasonic transmitting module to adjust the ultrasonic waves.

17. The method as described in claim 16, characterized in that, The control information includes the transmission delay of the ultrasonic wave, and the generation of control information based on the feature information includes: The echo signal is divided into multiple small blocks in the time domain; Select at least one region of interest from the plurality of small blocks; The emission delay of the ultrasonic wave is generated based on the region of interest.

18. A neuromorphic computing module, characterized in that, The module is used to execute any one of the ultrasonic signal processing methods 13-17.

19. The module as described in claim 18, characterized in that, The echo signal received by the module is an analog signal.

20. The module as described in claim 18, characterized in that, The echo signal received by the module is a pulse signal.

Images