Bubble detection method in long sheath tube based on acoustic field detection

Abstract

The method for detecting bubbles in a long sheath tube based on acoustic field detection comprises the following steps: step 1: acoustic wave propagation and preliminary detection; step 2: analysis of bubble scattering characteristics; step 3: acoustic field feature extraction, extracting the frequency domain features of the scattered acoustic field, establishing the frequency domain acoustic spectrum of the bubble, and extracting the size and velocity parameters of the bubble; step 4: bubble detection based on support vector machine, using support vector machine to build a bubble detection model, using labeled samples to train the model to detect the generation and disappearance transients of bubbles; step 5: photoacoustic imaging assisted detection, using acoustic field laser to detect bubble aggregation areas, and monitoring the movement of bubbles in the lumen; step 6: bubble motion parameter solution, combining acoustic and optical multi-frame intelligence streams, calculating the detailed motion trajectory of the bubble, setting multi-level warning and suction thresholds, and enabling corresponding measures according to the severity of the bubble; step 7: closed-loop control, realizing closed-loop control through a feedback mechanism until the bubble is completely cleared.

Description

Technical Field

[0001] The invention relates to the technical field of dynamic detection of bubbles in a long sheath tube, and more specifically to a method for detecting bubbles in a long sheath tube based on sound field detection. Background Art

[0002] In the field of cardiovascular interventional surgery, the use of long sheaths is key to achieving precise treatment. However, there is a significant risk factor during these procedures - the generation and introduction of bubbles. Bubbles entering the vascular system can cause serious complications, such as air embolism, posing a major threat to the patient's health. Therefore, accurately detecting and managing bubbles generated during surgery has become a vital issue.

[0003] In recent years, a variety of technologies and methods have been developed to address this challenge. These methods mainly include:

[0004] Acoustic monitoring technology: uses ultrasound to detect the presence of bubbles, but this method is limited in accuracy and sensitivity, especially in complex hemodynamic environments.

[0005] Optical monitoring technology: Bubbles are observed through optical equipment, but this method is limited by the resolution of the equipment and the field of view of operation.

[0006] Physical isolation methods: Use filters or other physical barriers to prevent bubbles from entering the blood vessels, but this may affect the operability of surgical tools and hemodynamics.

[0007] Although existing technologies can detect or prevent bubbles to a certain extent, they still have the following major drawbacks:

[0008] Insufficient detection accuracy: Existing technologies have difficulty in accurately detecting tiny bubbles, especially in dynamic and complex blood flow environments.

[0009] Limited real-time response capabilities: Existing methods have limitations in real-time response and processing after bubble detection, and cannot adjust surgical strategies in time to deal with detected bubbles.

[0010] Operation complexity: Some technologies, such as physical barrier methods, may increase the complexity and difficulty of surgical operations.

[0011] Safety issues: Existing technologies cannot completely eliminate the risks posed by bubbles, especially during long or complex surgeries.

[0012] In view of these drawbacks of the prior art, the development of the present invention is particularly necessary. Summary of the invention

[0013] 1. Technical problems to be solved

[0014] In view of the problems existing in the prior art, the purpose of the present invention is to provide a method for detecting bubbles in a long sheath tube based on acoustic field detection. By comprehensively applying acoustics, optics, fluid dynamics and closed-loop control technology, not only the accuracy and sensitivity of bubble detection are improved, but also real-time monitoring and dynamic response of bubbles are realized; this multi-dimensional, high-precision detection method can identify tiny bubbles in a timely manner, and adjust surgical operations through an intelligent feedback mechanism, thereby significantly improving the safety and efficiency of the operation; therefore, the application of the present invention in cardiovascular interventional surgery is of great significance for improving surgical safety, reducing the risk of complications and improving surgical efficiency.

[0015] 2. Technical solution

[0016] To solve the above problems, the present invention adopts the following technical solutions:

[0017] The method for detecting bubbles in a long sheath tube based on acoustic field detection is characterized by comprising the following steps:

[0018] Step 1: Sound wave propagation and preliminary detection

[0019] Embed a MHz-class miniaturized ultrasonic array in the long sheath tube wall to improve the coverage of the acoustic field scanning, realize comprehensive monitoring of the fluid in the tube, simulate the propagation of sound waves in the long sheath tube, and perform preliminary bubble detection;

[0020] Step 2: Analysis of bubble scattering characteristics

[0021] Based on the preliminary detection results of step 1, further analyze the scattering characteristics of the bubbles;

[0022] Step 3: Sound field feature extraction

[0023] Extract the frequency domain characteristics of the scattered sound field, establish the frequency domain acoustic spectrum of the bubble, and extract the size and velocity parameters of the bubble;

[0024] Step 4: Bubble detection based on support vector machine

[0025] A bubble detection model is constructed using support vector machine, and the model is trained using labeled samples to detect the generation and disappearance transients of bubbles.

[0026] Step 5: Photoacoustic imaging-assisted detection

[0027] Use acoustic field laser to detect bubble accumulation areas and monitor the movement of bubbles in the lumen;

[0028] Step 6: Bubble motion parameter calculation

[0029] Combined with the multi-frame intelligence stream of sound and light, the detailed movement trajectory of the bubble is calculated, multi-level warning and suction thresholds are set, and corresponding measures are initiated according to the severity of the bubble;

[0030] Step 7: Closed Loop Control

[0031] Closed-loop control is achieved through a feedback mechanism until the bubbles are completely removed.

[0032] The step 1 uses formula (1) to quickly identify possible bubbles and provide basic data for subsequent more accurate analysis. The specific formula is as follows:

[0033] ψ(x,t)=Ae i(kx-ωt) (1)

[0034] Among them, ψ(x, t): is the wave function of the sound wave, which represents the state of the sound wave at position x and time t; A is the amplitude of the sound wave, which represents the intensity of the sound wave; k is the wave number, which is related to the wavelength and frequency of the sound wave; ω is the angular frequency, which is related to the frequency of the sound wave; first, a sound wave of specific frequency and amplitude is emitted at one end of the long sheath; then the sensor array in the pipeline is used to monitor the propagation of the sound wave; the wave function data ψ(x, t) of the sound wave at different positions is collected, and the collected data, especially the changes in amplitude and frequency, are analyzed to identify possible bubbles.

[0035] Step 1 uses the propagation characteristics of sound waves in the medium to perform preliminary bubble detection. By analyzing the amplitude, frequency and phase changes of the sound waves in the long sheath tube, the presence of bubbles can be effectively identified. This step provides key preliminary data for subsequent more accurate bubble detection and analysis.

[0036] In step 2, formula (2) is used to further analyze the scattering characteristics of the bubbles to more accurately identify and locate the bubbles. The specific formula is as follows:

[0037]

[0038] Among them, σ is the scattering cross section, which represents the effective area where the sound wave is scattered by the bubble; k represents the wave number, which is related to the wavelength and frequency of the sound wave; δ0 is the phase shift, which is related to the acoustic characteristics of the bubble; first, data reception: receive the sound wave propagation data from step 1; based on formula (2), analyze the scattering characteristics in the sound wave data, extract the size, shape and material characteristics of the bubble from the scattering data, and finally fuse the scattering characteristic data with the data from step 1 to improve the accuracy of bubble detection.

[0039] Step 2 further improves the accuracy of bubble detection by analyzing the scattering properties of sound waves. This step not only identifies the presence of bubbles, but also provides detailed information about the size, shape, and material properties of the bubbles, providing important data support for subsequent steps. In this way, bubbles in long sheaths can be more accurately located and identified, thereby improving the safety of cardiovascular interventional procedures.

[0040] The step 3 uses formula (3) to further extract the frequency domain characteristics of the sound field in order to more accurately identify and classify bubbles. The specific formula is as follows:

[0041]

[0042] Among them, F(ω) is the frequency domain representation of the signal; f(t) is the time domain signal; ω is the angular frequency. First, the scattering characteristic data of the bubble is received from step 2, and then the Fourier transform is applied to convert the time domain sound wave data into frequency domain data. The frequency domain data is analyzed to identify the specific frequency components and modes caused by the bubble. Finally, key features such as the main frequency and energy distribution are extracted from the frequency domain data.

[0043] Step 3 converts the acoustic wave data from the time domain to the frequency domain through Fourier transform, and conducts in-depth analysis of the frequency domain features to more accurately identify and classify bubbles. This step is key to understanding the impact of bubbles on the acoustic field and provides an important data basis for subsequent bubble detection and processing. In this way, the size, type, and dynamic behavior of bubbles can be more accurately assessed, thereby improving the safety and efficiency of cardiovascular interventional procedures.

[0044] The step 4 adopts formula (4) to perform accurate bubble detection and classification through the SVM algorithm. The specific formula is as follows:

[0045]

[0046] Among them, f(x) is the decision function used for classification; α i is the Lagrange multiplier; y i is the category label of the training sample; K(x i ,x) is the kernel function used to map the data to a high-dimensional space, b is the bias term, first receive the frequency domain feature data: receive the frequency domain feature data of the bubble from step 3, then select the most informative features for bubble detection, and then use the selected features and labeled training data to train the SVM model; finally, use the trained SVM model to perform bubble detection and classification on the new data.

[0047] Step 4 achieves accurate detection and classification of bubbles by applying the support vector machine (SVM) algorithm combined with the frequency domain features extracted from the acoustic field data. This step is a key link in the entire bubble detection process and improves the accuracy and reliability of detection. In this way, bubbles and non-bubble areas can be effectively distinguished, providing important information support for subsequent processing and decision-making.

[0048] The step 5 adopts formula (5) and uses photoacoustic imaging technology to assist in detecting the spatial distribution of bubbles, so as to enhance the estimation of the bubble detection result based on the support vector machine in step 4. The specific formula is as follows:

[0049]

[0050] Where p is the sound pressure, that is, the intensity of the photoacoustic signal; Γ is the Grunesen parameter, which indicates the photoacoustic conversion efficiency of the medium; β is the thermal expansion coefficient, which indicates the response of the medium to heat; is the time derivative of light intensity, indicating the rate of change of the laser pulse. First, a laser pulse is emitted to the target area. Then, an ultrasonic probe is used to receive the photoacoustic signal generated by the bubble. The received photoacoustic signal is then preprocessed, including amplification, filtering, and denoising. Finally, the spatial distribution of the bubble is analyzed based on the intensity and time delay of the photoacoustic signal.

[0051] Step 5 provides a powerful auxiliary means for bubble detection by combining photoacoustic imaging technology. This step can provide spatial distribution information of bubbles, enhance the detection results of the previous steps, and make the location of bubbles more accurate. In this way, the distribution of bubbles in the long sheath can be more fully understood, providing important auxiliary information for subsequent processing and decision-making.

[0052] The step 6 uses formula (6) and combines the photoacoustic imaging data in step 5 to calculate the detailed motion trajectory and dynamic characteristics of the bubble, so as to more accurately understand the behavior of the bubble in the long sheath tube. The specific formula is as follows:

[0053]

[0054] Wherein, ρ is the fluid density; v is the fluid velocity field; p is the fluid pressure; μ is the dynamic viscosity of the fluid; and t is time. First, the photoacoustic imaging data is received, and the position and size information of the bubble is obtained from step 5. According to the fluid characteristics in the long sheath, such as blood, a fluid dynamics model is established. Formula (6) is used to numerically solve the fluid dynamics equation and calculate the motion trajectory of the bubble in the fluid. Finally, the motion speed, direction and change trend of the bubble are analyzed according to the motion trajectory.

[0055] Step 6 calculates the trajectory and dynamic characteristics of the bubble in the long sheath by applying fluid dynamics principles and numerical solution techniques. This step provides important information for understanding the behavior of bubbles in blood vessels and helps to evaluate the possible impact of bubbles on blood flow and surrounding tissues. This method can more comprehensively monitor and manage the risk of bubbles in cardiovascular interventional surgery, improving the safety and effectiveness of the surgery.

[0056] In step 7, formula (7) is used to combine the bubble motion trajectory and dynamic characteristics calculated in step 6 to implement a closed-loop control system to adjust the operation during the surgery to ensure that the bubbles are effectively managed and removed. The specific formula is as follows:

[0057] Δ(s)=det(sI-A)=0 (7)

[0058] Wherein, Δ(s) is the characteristic equation of the system; s is the complex variable in Laplace transform; I is the unit matrix; A is the system matrix; First, the motion trajectory and dynamic characteristics of the bubble are obtained from step 6, and then a closed-loop control model is established according to the behavior of the bubble and the surgical environment. Formula (7) is used to analyze the stability of the system to ensure the effectiveness of the control strategy, and finally the surgical operation is adjusted according to the behavior of the bubble.

[0059] Step 7 implements a closed-loop control strategy to adjust the surgical operation according to the movement trajectory and dynamic characteristics of the bubbles, thereby effectively managing and removing bubbles. This step is a key link in ensuring surgical safety, and can respond to the behavior of bubbles in real time to reduce surgical risks. In this way, the bubble problem during surgery can be more effectively controlled, improving the safety and success rate of surgery.

[0060] In step 4, the kernel function is a radial basis function kernel, and the data is mapped to a high-dimensional space using formula (8). The specific formula is as follows:

[0061] K(x i ,x j )=exp(-γ∥x i -x j ∥ 2 ) (8)

[0062] Among them, γ is the parameter of the kernel function, which controls the smoothness of the mapped feature space. In step 4, by properly selecting and applying the kernel function, the support vector machine can effectively process nonlinearly separable data, such as the frequency domain characteristics of bubbles. The choice of kernel function has an important impact on the performance of SVM, so it needs to be adjusted according to the specific data characteristics and problem requirements. In this way, bubble detection and classification can be performed more accurately, providing important support for decision-making in cardiovascular interventional surgery.

[0063] 3. Beneficial effects

[0064] Compared with the prior art, the advantages of the present invention are:

[0065] 1. Improve the accuracy and sensitivity of bubble detection: By combining the characteristics of acoustic wave propagation, acoustic wave scattering, frequency domain feature extraction, support vector machine classification and photoacoustic imaging technology, the present invention can accurately identify and locate bubbles in long sheath tubes from multiple dimensions. Each step targets different aspects of bubble detection, ensuring comprehensive coverage from preliminary detection to precise positioning, greatly improving the accuracy and sensitivity of detection.

[0066] 2. Real-time monitoring and dynamic response: Through the fluid dynamics model and closed-loop control system, the present invention can monitor the movement trajectory and dynamic characteristics of bubbles in real time and adjust the surgical operation accordingly. This real-time monitoring and dynamic response mechanism makes the surgical process safer and can prevent and deal with potential risks caused by bubbles in a timely manner.

[0067] 3. Enhance surgical safety and efficiency: Through precise bubble detection and real-time dynamic control, the present invention significantly reduces the risk of bubbles in cardiovascular interventional surgery and improves surgical safety. At the same time, it reduces surgical interruptions or delays caused by bubble problems and improves surgical efficiency.

[0068] 4. Multidisciplinary technology integration: This invention combines advanced technologies from multiple disciplines such as acoustics, optics, fluid dynamics, and control theory, demonstrating the great potential of interdisciplinary integration in solving complex medical problems. This multidisciplinary technology integration not only improves the efficiency and effectiveness of problem solving, but also provides valuable experience for the development of similar technologies in the future.

[0069] 5. Scalability and adaptability: The design of the present invention takes into account long sheaths of different types and sizes, and has good scalability and adaptability. This means that the technology can be applied to different surgical environments and needs, and has broad application prospects.

[0070] In summary, through multi-dimensional technology integration and innovation, the present invention not only significantly improves the accuracy and efficiency of bubble detection in cardiovascular interventional surgery, but also enhances the safety and real-time response capability of the surgery, demonstrating the huge potential of interdisciplinary technology integration in the medical field. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 The present invention is a flow chart of a method for detecting bubbles in a long sheath tube based on acoustic field detection. DETAILED DESCRIPTION

[0072] The following will combine 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 the embodiments. All other embodiments obtained by ordinary technicians in this field without creative work based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0073] In the description of the present invention, it should be noted that the terms "upper", "lower", "inner", "outer", "top / bottom" indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, which are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance.

[0074] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "provided with", "mounted / connected", and "connected" should be understood in a broad sense. For example, "connected" means fixed connection, detachable connection, or integral connection; mechanical connection, electrical connection; direct connection, indirect connection through an intermediate medium, and internal communication between two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention are understood according to specific circumstances.

[0075] Example 1

[0076] Please refer to Figure 1 The method for detecting bubbles in a long sheath tube based on acoustic field detection comprises the following steps:

[0077] The method for detecting bubbles in a long sheath tube based on acoustic field detection is characterized by comprising the following steps:

[0078] Step 1: Sound wave propagation and preliminary detection

[0079] Embed a MHz-class miniaturized ultrasonic array in the long sheath tube wall to improve the coverage of the acoustic field scanning, realize comprehensive monitoring of the fluid in the tube, simulate the propagation of sound waves in the long sheath tube, and perform preliminary bubble detection;

[0080] Step 2: Analysis of bubble scattering characteristics

[0081] Based on the preliminary detection results of step 1, further analyze the scattering characteristics of the bubbles;

[0082] Step 3: Sound field feature extraction

[0083] Extract the frequency domain characteristics of the scattered sound field, establish the frequency domain acoustic spectrum of the bubble, and extract the size and velocity parameters of the bubble;

[0084] Step 4: Bubble detection based on support vector machine

[0085] A bubble detection model is constructed using support vector machine, and the model is trained using labeled samples to detect the generation and disappearance transients of bubbles.

[0086] Step 5: Photoacoustic imaging-assisted detection

[0087] Use acoustic field laser to detect bubble accumulation areas and monitor the movement of bubbles in the lumen;

[0088] Step 6: Bubble motion parameter calculation

[0089] Combined with the multi-frame intelligence stream of sound and light, the detailed movement trajectory of the bubble is calculated, multi-level warning and suction thresholds are set, and corresponding measures are initiated according to the severity of the bubble;

[0090] Step 7: Closed Loop Control

[0091] Closed-loop control is achieved through a feedback mechanism until the bubbles are completely removed.

[0092] The step 1 uses formula (1) to quickly identify possible bubbles and provide basic data for subsequent more accurate analysis. The specific formula is as follows:

[0093] ψ(x,t)=Ae i(kx-ωt) (1)

[0094] Among them, ψ(x, t): is the wave function of the sound wave, which represents the state of the sound wave at position x and time t; A is the amplitude of the sound wave, which represents the intensity of the sound wave; k is the wave number, which is related to the wavelength and frequency of the sound wave; ω is the angular frequency, which is related to the frequency of the sound wave; first, a sound wave of specific frequency and amplitude is emitted at one end of the long sheath; then the sensor array in the pipeline is used to monitor the propagation of the sound wave; the wave function data ψ(x, t) of the sound wave at different positions is collected, and the collected data, especially the changes in amplitude and frequency, are analyzed to identify possible bubbles.

[0095] When sound waves propagate in a medium, their propagation characteristics (such as speed, frequency, and amplitude) will change due to different media (such as the presence of bubbles). Using this principle, bubbles can be detected by analyzing the propagation characteristics of sound waves. The propagation of sound waves in a medium is affected by the characteristics of the medium. Wave functions are used to describe the state of sound waves at different positions and times, and the presence of bubbles will cause the amplitude and frequency of sound waves to change. The data type includes sound wave propagation data, including the amplitude, frequency, and phase of the wave function. The data comes from an array of sound wave sensors built into a long sheath. The preprocessing steps include denoising, normalization, and time window cutting. The key features of the sound wave data, such as amplitude change rate, frequency offset, etc., are extracted. The preliminary bubble detection results are output, and the possibility of bubble existence is evaluated based on the changes in the sound wave propagation characteristics. Finally, the preliminary analysis results are passed to the next step for a more in-depth analysis of bubble characteristics.

[0096] Step 1 uses the propagation characteristics of sound waves in the medium to perform preliminary bubble detection. By analyzing the amplitude, frequency and phase changes of the sound waves in the long sheath tube, the presence of bubbles can be effectively identified. This step provides key preliminary data for subsequent more accurate bubble detection and analysis.

[0097] In step 2, formula (2) is used to further analyze the scattering characteristics of the bubbles to more accurately identify and locate the bubbles. The specific formula is as follows:

[0098]

[0099] Among them, σ is the scattering cross section, which represents the effective area where the sound wave is scattered by the bubble; k represents the wave number, which is related to the wavelength and frequency of the sound wave; δ0 is the phase shift, which is related to the acoustic characteristics of the bubble; first, data reception: receive the sound wave propagation data from step 1; based on formula (2), analyze the scattering characteristics in the sound wave data, extract the size, shape and material characteristics of the bubble from the scattering data, and finally fuse the scattering characteristic data with the data from step 1 to improve the accuracy of bubble detection.

[0100] Sound waves scatter when they encounter bubbles. The size, shape, and material properties of bubbles affect the scattering properties of sound waves. By analyzing these scattering properties, bubbles can be identified more accurately. The scattering phenomenon of sound waves when they encounter bubbles. The scattering cross section is used to quantify the degree to which sound waves are scattered by bubbles. The effect of bubbles on the phase of sound waves reflects the characteristics of bubbles.

[0101] The data types include acoustic wave scattering characteristic data; the data comes from the acoustic wave propagation data provided in step 1; the preprocessing steps include data filtering, denoising, and feature extraction of scattering characteristics; extracting key features in the scattering characteristic data, such as scattering intensity, phase change, etc. Finally, the bubble characteristic analysis results are output: including the size, shape and material characteristics of the bubble; and bubble positioning information: based on the scattering characteristic analysis, the precise location information of the bubble is provided.

[0102] Step 2 further improves the accuracy of bubble detection by analyzing the scattering properties of sound waves. This step not only identifies the presence of bubbles, but also provides detailed information about the size, shape, and material properties of the bubbles, providing important data support for subsequent steps. In this way, bubbles in long sheaths can be more accurately located and identified, thereby improving the safety of cardiovascular interventional procedures.

[0103] The step 3 uses formula (3) to further extract the frequency domain characteristics of the sound field in order to more accurately identify and classify bubbles. The specific formula is as follows:

[0104]

[0105] Among them, F(ω) is the frequency domain representation of the signal; f(t) is the time domain signal; ω is the angular frequency. First, the scattering characteristic data of the bubble is received from step 2, and then the Fourier transform is applied to convert the time domain sound wave data into frequency domain data. The frequency domain data is analyzed to identify the specific frequency components and modes caused by the bubble. Finally, key features such as the main frequency and energy distribution are extracted from the frequency domain data.

[0106] Fourier transform is a mathematical tool that converts time domain signals into frequency domain signals, which can reveal the frequency components of the signal. By analyzing the frequency domain characteristics of the sound field, we can have a deeper understanding of the impact of bubbles on sound waves. Frequency domain analysis is used to analyze the distribution and characteristics of signals in frequency; Fourier transform is a method to convert time domain signals into frequency domain signals; the purpose of signal feature extraction is to extract key information about bubble characteristics from frequency domain data; the data type is the frequency domain feature data of sound waves; the data comes from the sound wave scattering characteristic data in step 2; the preprocessing steps include windowing, denoising and normalization of the signal. Extract key features from the frequency domain data, such as peak frequency, energy spectral density, etc. Finally, output frequency domain feature data: including the frequency domain characteristics of the sound field under the influence of bubbles; and bubble feature identification: identification of bubble size, type and dynamic behavior based on frequency domain features.

[0107] Step 3 converts the acoustic wave data from the time domain to the frequency domain through Fourier transform, and conducts in-depth analysis of the frequency domain features to more accurately identify and classify bubbles. This step is key to understanding the impact of bubbles on the acoustic field and provides an important data basis for subsequent bubble detection and processing. In this way, the size, type, and dynamic behavior of bubbles can be more accurately assessed, thereby improving the safety and efficiency of cardiovascular interventional procedures.

[0108] The step 4 adopts formula (4) to perform accurate bubble detection and classification through the SVM algorithm. The specific formula is as follows:

[0109]

[0110] Among them, f(x) is the decision function used for classification; α i is the Lagrange multiplier; y i is the category label of the training sample; K(x i ,x) is the kernel function used to map the data to a high-dimensional space, b is the bias term, first receive the frequency domain feature data: receive the frequency domain feature data of the bubble from step 3, then select the most informative features for bubble detection, and then use the selected features and labeled training data to train the SVM model; finally, use the trained SVM model to perform bubble detection and classification on the new data.

[0111] Support vector machine is a powerful supervised learning algorithm for classification and regression analysis. It distinguishes different categories of data by finding the best segmentation hyperplane in the data set. Classification is the division of data into different categories; hyperplane refers to the interface that separates different categories in the feature space; kernel function is used to process nonlinearly separable data. The data type is the frequency domain feature data of sound waves. The data comes from the frequency domain feature analysis in step 3. The preprocessing step includes standardization to ensure that the data is on the same scale. Select and extract the features that are most helpful for classification. Output the classification results, the classification of whether bubbles exist for each data point; and bubble characteristics, the classification of the size, type and other characteristics of the detected bubbles.

[0112] Step 4 achieves accurate detection and classification of bubbles by applying the support vector machine (SVM) algorithm combined with the frequency domain features extracted from the acoustic field data. This step is a key link in the entire bubble detection process and improves the accuracy and reliability of detection. In this way, bubbles and non-bubble areas can be effectively distinguished, providing important information support for subsequent processing and decision-making.

[0113] The step 5 adopts formula (5) and uses photoacoustic imaging technology to assist in detecting the spatial distribution of bubbles, so as to enhance the estimation of the bubble detection result based on the support vector machine in step 4. The specific formula is as follows:

[0114]

[0115] Where p is the sound pressure, that is, the intensity of the photoacoustic signal; Γ is the Grunesen parameter, which indicates the photoacoustic conversion efficiency of the medium; β is the thermal expansion coefficient, which indicates the response of the medium to heat; is the time derivative of light intensity, indicating the rate of change of the laser pulse. First, a laser pulse is emitted to the target area. Then, an ultrasonic probe is used to receive the photoacoustic signal generated by the bubble. The received photoacoustic signal is then preprocessed, including amplification, filtering, and denoising. Finally, the spatial distribution of the bubble is analyzed based on the intensity and time delay of the photoacoustic signal.

[0116] Photoacoustic imaging is an imaging technique that combines optics and acoustics. When tissue or material absorbs a laser pulse, it causes thermal expansion, which in turn generates ultrasound waves. This technique can be used to detect and image structures inside tissues, including bubbles. Photoacoustic imaging is a technique that uses the photoacoustic effect for imaging. The acoustic pressure signal is the ultrasound signal generated by the photoacoustic effect. Thermal expansion is the change in volume caused by the absorption of light energy by a material. Photoacoustic imaging parameters: Determine the wavelength, energy and pulse width of the laser pulse, as well as the frequency and sensitivity of the ultrasound probe.

[0117] Signal processing algorithms: Select appropriate signal processing algorithms, such as Fast Fourier Transform (FFT) for signal analysis and advanced algorithms for image reconstruction.

[0118] The data type is photoacoustic signal data; the data comes from the photoacoustic imaging device. The preprocessing steps include signal amplification, bandpass filtering, and denoising. The time-space characteristics of the photoacoustic signal are extracted, such as signal intensity, waveform characteristics, etc. Finally, the bubble spatial distribution map is output to show the distribution of bubbles in the detection area; as well as bubble positioning information: providing the precise location and size information of the bubble.

[0119] Step 5 provides a powerful auxiliary means for bubble detection by combining photoacoustic imaging technology. This step can provide spatial distribution information of bubbles, enhance the detection results of the previous steps, and make the location of bubbles more accurate. In this way, the distribution of bubbles in the long sheath can be more fully understood, providing important auxiliary information for subsequent processing and decision-making.

[0120] The step 6 uses formula (6) and combines the photoacoustic imaging data in step 5 to calculate the detailed motion trajectory and dynamic characteristics of the bubble, so as to more accurately understand the behavior of the bubble in the long sheath tube. The specific formula is as follows:

[0121]

[0122] Wherein, ρ is the fluid density; v is the fluid velocity field; p is the fluid pressure; μ is the dynamic viscosity of the fluid; and t is time. First, the photoacoustic imaging data is received, and the position and size information of the bubble is obtained from step 5. According to the fluid characteristics in the long sheath, such as blood, a fluid dynamics model is established. Formula (6) is used to numerically solve the fluid dynamics equation and calculate the motion trajectory of the bubble in the fluid. Finally, the motion speed, direction and change trend of the bubble are analyzed according to the motion trajectory.

[0123] Fluid dynamics is the science of fluid motion and its interaction with objects. In this step, we will apply the principles of fluid dynamics to simulate and calculate the motion of bubbles in a fluid such as blood. The velocity field can describe the velocity of the fluid at each point. The pressure field is used to describe the pressure of the fluid at each point.

[0124] The data types are photoacoustic imaging data and fluid dynamic parameters. The data comes from the photoacoustic imaging results and fluid property data in step 5. The preprocessing steps include data standardization, parameter estimation and calibration. The key features of bubble motion, such as velocity, acceleration, etc., are extracted. Finally, the bubble motion trajectory is output, including the motion path of the bubble in the long sheath. And dynamic characteristics analysis, including the velocity, acceleration and motion trend of the bubble.

[0125] Step 6 calculates the trajectory and dynamic characteristics of the bubble in the long sheath by applying fluid dynamics principles and numerical solution techniques. Selecting an appropriate numerical solution algorithm, such as the finite difference method or the finite element method, to solve the fluid dynamics equations provides important information for understanding the behavior of bubbles in blood vessels and helps to evaluate the possible impact of bubbles on blood flow and surrounding tissues. This method can more comprehensively monitor and manage the risk of bubbles in cardiovascular interventional surgery, improving the safety and effectiveness of the surgery.

[0126] In step 7, formula (7) is used to combine the bubble motion trajectory and dynamic characteristics calculated in step 6 to implement a closed-loop control system to adjust the operation during the surgery to ensure that the bubbles are effectively managed and removed. The specific formula is as follows:

[0127] Δ(s)=det(sI-A)=0 (7)

[0128] Wherein, Δ(s) is the characteristic equation of the system; s is the complex variable in Laplace transform; I is the unit matrix; A is the system matrix; First, the motion trajectory and dynamic characteristics of the bubble are obtained from step 6, and then a closed-loop control model is established according to the behavior of the bubble and the surgical environment. Formula (7) is used to analyze the stability of the system to ensure the effectiveness of the control strategy, and finally the surgical operation is adjusted according to the behavior of the bubble.

[0129] Closed-loop control (also known as feedback control) is a control system design that uses the output of a system to adjust its input to achieve the desired performance. In this step, we will apply closed-loop control theory to adjust the surgical process in response to the behavior of bubbles. Closed-loop control is a control method that uses the output of a system to adjust the input. System stability refers to the ability of a system to maintain or return to a stable state when disturbed. Feedback mechanism refers to the adjustment mechanism of the system output to the system input. The data types include bubble motion data and control system parameters. The data comes from the bubble dynamic analysis results in step 6. The preprocessing steps include data normalization and parameter estimation. Extract key features related to control decisions. Finally, output the control strategy, adjustment instructions for bubble behavior, and system adjustment results, including adjustments to surgical operations and bubble removal status.

[0130] Step 7 implements a closed-loop control strategy to adjust the surgical operation according to the movement trajectory and dynamic characteristics of the bubbles, thereby effectively managing and removing bubbles. This step is a key link in ensuring surgical safety, and can respond to the behavior of bubbles in real time to reduce surgical risks. In this way, the bubble problem during surgery can be more effectively controlled, improving the safety and success rate of surgery.

[0131] In step 4, the kernel function is a radial basis function kernel, and the data is mapped to a high-dimensional space using formula (8). The specific formula is as follows:

[0132] K(x i ,x j )=exp(-γ∥x i -x j ∥ 2 )(8)

[0133] Among them, γ is the parameter of the kernel function, which controls the smoothness of the mapped feature space. In step 4, by properly selecting and applying the kernel function, the support vector machine can effectively process nonlinearly separable data, such as the frequency domain characteristics of bubbles. The choice of kernel function has an important impact on the performance of SVM, so it needs to be adjusted according to the specific data characteristics and problem requirements. In this way, bubble detection and classification can be performed more accurately, providing important support for decision-making in cardiovascular interventional surgery.

[0134] A kernel function is a function that can calculate the inner product of data in a high-dimensional space without explicitly defining the high-dimensional space; the radial basis function (RBF) kernel is a commonly used kernel function, which is particularly suitable for processing nonlinearly separable data.

[0135] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the scope of protection of the invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on these embodiments, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in this field can still combine, add, delete or make other adjustments to the features in the various embodiments of the present invention according to the circumstances without conflict, without making creative work, so as to obtain different other technical solutions that do not deviate from the concept of the present invention in essence, and these technical solutions also belong to the scope of protection of the present invention.

Claims

1. A method for detecting bubbles in a long sheath tube based on acoustic field detection, characterized in that: The following steps are involved: Step 1: Sound wave propagation and preliminary detection Embed a MHz-class miniaturized ultrasonic array in the long sheath tube wall to improve the coverage of the acoustic field scanning, realize comprehensive monitoring of the fluid in the tube, simulate the propagation of sound waves in the long sheath tube, and perform preliminary bubble detection; Step 2: Analysis of bubble scattering characteristics Based on the preliminary detection results of step 1, further analyze the scattering characteristics of the bubbles; Step 3: Sound field feature extraction Extract the frequency domain characteristics of the scattered sound field, establish the frequency domain acoustic spectrum of the bubble, and extract the size and velocity parameters of the bubble; Step 4: Bubble detection based on support vector machine A bubble detection model is constructed using support vector machine, and the model is trained using labeled samples to detect the generation and disappearance transients of bubbles. Step 5: Photoacoustic imaging-assisted detection Use acoustic field laser to detect bubble accumulation areas and monitor the movement of bubbles in the lumen; Step 6: Bubble motion parameter calculation Combined with the multi-frame intelligence stream of sound and light, the detailed movement trajectory of the bubble is calculated, multi-level warning and suction thresholds are set, and corresponding measures are initiated according to the severity of the bubble; Step 7: Closed Loop Control Closed-loop control is achieved through a feedback mechanism until the bubbles are completely removed; In step 2, formula (2) is used to further analyze the scattering characteristics of the bubbles to more accurately identify and locate the bubbles. The specific formula is as follows: Among them, σ is the scattering cross section, which represents the effective area where the sound wave is scattered by the bubble; k represents the wave number, which is related to the wavelength and frequency of the sound wave; δ0 is the phase shift, which is related to the acoustic characteristics of the bubble; first, data reception: receive the sound wave propagation data from step 1; based on formula (2), analyze the scattering characteristics in the sound wave data, extract the size, shape and material characteristics of the bubble from the scattering data, and finally fuse the scattering characteristic data with the data from step 1 to improve the accuracy of bubble detection.

2. The method for detecting bubbles in a long sheath tube based on acoustic field detection according to claim 1, characterized in that: The step 1 uses formula (1) to quickly identify possible bubbles and provide basic data for subsequent more accurate analysis. The specific formula is as follows: ψ(x,t)=Ae i(kx-ωt) (1) Among them, ψ(x, t): is the wave function of the sound wave, which represents the state of the sound wave at position x and time t; A is the amplitude of the sound wave, which represents the intensity of the sound wave; k is the wave number, which is related to the wavelength and frequency of the sound wave; ω is the angular frequency, which is related to the frequency of the sound wave; first, a sound wave of specific frequency and amplitude is emitted at one end of the long sheath; then the sensor array in the pipeline is used to monitor the propagation of the sound wave; the wave function data ψ(x, t) of the sound wave at different positions is collected, and the collected data, especially the changes in amplitude and frequency, are analyzed to identify possible bubbles.

3. The method for detecting bubbles in a long sheath tube based on acoustic field detection according to claim 2, characterized in that: The step 3 uses formula (3) to further extract the frequency domain characteristics of the sound field in order to more accurately identify and classify bubbles. The specific formula is as follows: Among them, F(ω) is the frequency domain representation of the signal; f(t) is the time domain signal; ω is the angular frequency. First, the scattering characteristic data of the bubble is received from step 2, and then the Fourier transform is applied to convert the time domain sound wave data into frequency domain data. The frequency domain data is analyzed to identify the specific frequency components and modes caused by the bubble. Finally, key features such as the main frequency and energy distribution are extracted from the frequency domain data.

4. The method for detecting bubbles in a long sheath tube based on acoustic field detection according to claim 3 is characterized in that: The step 4 adopts formula (4) to perform accurate bubble detection and classification through the SVM algorithm. The specific formula is as follows: Among them, f(x) is the decision function used for classification; α i is the Lagrange multiplier; y i is the category label of the training sample; K(x i ,x) is the kernel function used to map the data to a high-dimensional space, b is the bias term, first receive the frequency domain feature data: receive the frequency domain feature data of the bubble from step 3, then select the most informative features for bubble detection, and then use the selected features and labeled training data to train the SVM model; finally, use the trained SVM model to perform bubble detection and classification on the new data.

5. The method for detecting bubbles in a long sheath tube based on acoustic field detection according to claim 4, characterized in that: The step 5 adopts formula (5) and uses photoacoustic imaging technology to assist in detecting the spatial distribution of bubbles, so as to enhance the estimation of the bubble detection result based on the support vector machine in step 4. The specific formula is as follows: Where p is the sound pressure, that is, the intensity of the photoacoustic signal; Γ is the Grunesen parameter, which indicates the photoacoustic conversion efficiency of the medium; β is the thermal expansion coefficient, which indicates the response of the medium to heat; is the time derivative of light intensity, indicating the rate of change of the laser pulse. First, a laser pulse is emitted to the target area. Then, an ultrasonic probe is used to receive the photoacoustic signal generated by the bubble. The received photoacoustic signal is then preprocessed, including amplification, filtering, and denoising. Finally, the spatial distribution of the bubble is analyzed based on the intensity and time delay of the photoacoustic signal.

6. The method for detecting bubbles in a long sheath tube based on acoustic field detection according to claim 5, characterized in that: The step 6 uses formula (6) and combines the photoacoustic imaging data in step 5 to calculate the detailed motion trajectory and dynamic characteristics of the bubble, so as to more accurately understand the behavior of the bubble in the long sheath tube. The specific formula is as follows: Wherein, ρ is the fluid density; v is the fluid velocity field; p is the fluid pressure; μ is the dynamic viscosity of the fluid; and t is time. First, the photoacoustic imaging data is received, and the position and size information of the bubble is obtained from step 5. According to the fluid characteristics in the long sheath, such as blood, a fluid dynamics model is established. Formula (6) is used to numerically solve the fluid dynamics equation, calculate the motion trajectory of the bubble in the fluid, and finally analyze the motion speed, direction and change trend of the bubble according to the motion trajectory.

7. The method for detecting bubbles in a long sheath tube based on acoustic field detection according to claim 6, characterized in that: In step 7, formula (7) is used to combine the bubble motion trajectory and dynamic characteristics calculated in step 6 to implement a closed-loop control system to adjust the operation during the surgery to ensure that the bubbles are effectively managed and removed. The specific formula is as follows: Δ(s)=det(sI-A)=0 (7) Wherein, Δ(s) is the characteristic equation of the system; s is the complex variable in Laplace transform; I is the unit matrix; A is the system matrix; First, the motion trajectory and dynamic characteristics of the bubble are obtained from step 6, and then a closed-loop control model is established according to the behavior of the bubble and the surgical environment. Formula (7) is used to analyze the stability of the system to ensure the effectiveness of the control strategy, and finally the surgical operation is adjusted according to the behavior of the bubble.

8. The method for detecting bubbles in a long sheath tube based on acoustic field detection according to claim 4, characterized in that: In step 4, the kernel function is a radial basis function kernel, and the data is mapped to a high-dimensional space using formula (8). The specific formula is as follows: K(x i ,x j )=exp(-γ||x i -x j || 2 ) (8) Among them, γ is the parameter of the kernel function, which controls the smoothness of the mapped feature space.

Images