Passive cavitation monitoring method and platform based on ultrasonic cavitation effect

Abstract

The invention discloses a passive cavitation monitoring method and platform based on an ultrasonic cavitation effect. Ultrasonic passive cavitation monitoring is realized by adopting a concave ultrasonic transducer. Through combination of high intensity focused ultrasound (HIFU) and a passive cavitation detection system, real-time monitoring of the cavitation effect and dynamic adjustment of treatment parameters are realized. According to the method, the ICD is used as a biological index of the cavitation intensity, the cavitation damage is successfully quantified, the accuracy of the method is verified through a fitting model, the result shows that the ICD value and the damage area have significant positive correlation, and the tissue damage degree can be effectively predicted. The method provides reliable technical support for cavitation monitoring in ultrasonic treatment, has wide application prospects, and can improve treatment accuracy and safety especially in the fields of tumor treatment and the like.

Description

[0001] A passive cavitation monitoring method and platform based on ultrasonic cavitation effect Technical Field

[0002] This invention relates to the field of ultrasonic cavitation monitoring technology, and in particular to a passive cavitation monitoring method and platform based on ultrasonic cavitation effect. Background Technology

[0003] Ultrasonic cavitation plays a crucial role in ultrasonic tissue destruction. Through treatment systems such as high-intensity focused ultrasound (HIFU), the expansion and rupture of endogenous bubble nuclei within the tissue release energy, thereby destroying the target tissue. The dynamic changes in cavitation and its impact on surrounding tissues require real-time monitoring and control. However, existing cavitation monitoring technologies suffer from insufficient accuracy and limited real-time feedback capabilities, making them unreliable indicators during treatment. Real-time monitoring and precise quantification of cavitation intensity are key to improving treatment accuracy and safety. Current ultrasonic cavitation monitoring technologies have the following shortcomings: Insufficient accuracy: Traditional cavitation monitoring methods are unable to accurately capture the dynamic changes in cavitation signals, resulting in inaccurate cavitation intensity quantification.

[0004] Limited real-time feedback capability: Lacking an efficient real-time data processing and feedback mechanism, it is impossible to dynamically adjust ultrasound parameters during treatment.

[0005] Low reliability: Existing technologies do not directly link cavitation dose to the degree of tissue damage, making them difficult to use as a bioindicator of treatment reliability.

[0006] This technical solution solves the above problems by integrating real-time monitoring, data analysis, and model fitting.

[0007] Therefore, a method that integrates real-time monitoring, data analysis, and model fitting is needed to solve the above problems. Summary of the Invention

[0008] In view of this, the purpose of the present invention is to provide a passive cavitation monitoring method and system based on ultrasonic cavitation effect. The method uses a concave ultrasonic transducer to generate acoustic energy in the focal region to obtain inertial cavitation dose, while analyzing the area of ​​the damaged region and the correlation.

[0009] To achieve the above objectives, the present invention provides the following technical solution: The passive cavitation monitoring method based on ultrasonic cavitation effect provided by this invention includes the following steps: Step S1: Emits high-intensity focused ultrasound pulses toward the target tissue region using a concave ultrasound transducer; Step S2: Use a passive cavitation detection probe to capture the cavitation signal generated in the target tissue region in real time, and collect the time domain data of the cavitation signal; Step S3: Perform a Fast Fourier Transform on the time-domain data to obtain the spectrum data; Step S4: Apply digital band-stop filtering to the spectrum data to remove the fundamental frequency and its harmonic components, and obtain the filtered spectrum; Step S5: Calculate the inertial cavitation dose (ICD) based on the filtered spectrum, wherein the inertial cavitation dose (ICD) is quantized by the total spectral power of the filtered spectrum; Step S6: Establish a nonlinear fitting model based on the inertial cavitation dose (ICD) and tissue damage area. The nonlinear fitting model is used to calculate the cavitation intensity and tissue damage area. Step S7: Input the real-time inertial cavitation dose (ICD) value into the nonlinear fitting model to calculate the current tissue damage area. Compare the real-time calculated tissue damage area with the preset treatment target. Based on the comparison result, calculate and adjust at least one emission parameter of the concave ultrasonic transducer in real time through an optimization algorithm. The emission parameters include acoustic power, pulse repetition frequency, and duty cycle until the real-time calculated tissue damage area reaches the preset treatment target.

[0010] Furthermore, in step S4, the fundamental frequency of the digital band-stop filter is 0.982 MHz, removing harmonics from the 1st to the 9th order. The stopband bandwidth of the 1st to 7th order harmonics is ±40 kHz, and the stopband bandwidth of the 8th to 9th order harmonics is ±20 kHz.

[0011] Furthermore, the formula for calculating the total spectral power in step S5 is as follows: Where N is the number of sampling points, To the amplitude after harmonic removal, Frequency component index of the signal.

[0012] Furthermore, the nonlinear fitting model in step S6 is a power-law model, in the form of: ;in, ICD value, The area of ​​tissue damage. and These are the model coefficients.

[0013] Furthermore, the correlation analysis in step S6 employs the Spearman test, and the tissue damage area is obtained by quantifying the damage contour of the experimental sample using image processing software.

[0014] The passive cavitation monitoring platform based on ultrasonic cavitation effect provided by the present invention includes an ultrasonic therapy system, a real-time ultrasonic imaging probe, a data acquisition system, a passive cavitation detection probe, and a computer control system. The passive cavitation detection probe is used to capture cavitation signals generated in the target tissue area in real time; The data acquisition system is connected to the passive cavitation detection probe and is used to acquire time-domain data of the cavitation signal. The computer control system is configured to perform the following operations: Perform Fast Fourier Transform and Digital Band-Stop Filtering on the time-domain data; Inertial cavitation dose (ICD) is calculated based on the filtered spectrum. The ICD is fitted to the tissue damage area to establish a nonlinear model.

[0015] Furthermore, the ultrasound therapy system includes a concave ultrasound transducer with an outer diameter of 220 mm, an inner diameter of 80 mm, a focal length of 170 mm, and an operating frequency of 0.982 MHz.

[0016] Furthermore, the passive cavitation detection probe is a focusing transducer with a center frequency of 10 MHz and is connected to a digital oscilloscope with a sampling rate of 50 MSamples / s.

[0017] Furthermore, the computer control system is implemented using MATLAB programming and is connected to an external power amplifier to adjust the ultrasonic output power.

[0018] Furthermore, the platform also includes an image processing module for quantifying the area of ​​tissue damage through scale calibration and contour analysis.

[0019] The beneficial effects of this invention are as follows: This invention provides a passive cavitation monitoring method and platform based on ultrasonic cavitation effects, employing a concave ultrasonic transducer to achieve passive ultrasonic cavitation monitoring. By combining high-intensity focused ultrasound (HIFU) with a passive cavitation detection system, real-time monitoring of cavitation effects and dynamic adjustment of treatment parameters are achieved. Inertial cavitation dose (ICD), as a bioindicator of cavitation intensity, successfully quantifies cavitation damage, and the accuracy of the method is verified through a fitting model. The results show a significant positive correlation between ICD value and damage area, effectively predicting the degree of tissue damage. This method provides reliable technical support for cavitation monitoring in ultrasound therapy and has broad application prospects, particularly in fields such as tumor treatment, where it can improve treatment accuracy and safety.

[0020] The above and other objects, advantages, and features of the present invention will be more fully set forth and demonstrated through the following detailed description of specific embodiments in conjunction with the accompanying drawings. Those skilled in the art, upon referring to the following detailed description and the accompanying drawings, will be able to better understand and realize the above advantages of the present invention. Other objects, features, and advantages of the present invention will become clearer after being described in detail in the detailed description section in conjunction with the accompanying drawings. Attached Figure Description

[0021] To make the objectives, technical solutions, and beneficial effects of this invention clearer, the following drawings are provided for illustration.

[0022] Figure 1 The flowchart shows a passive cavitation monitoring method based on ultrasonic cavitation effect. Figure 2 This is a schematic diagram of the experimental platform; Figure 3 This is a schematic diagram of the connection method for the signal acquisition system; Figure 4 Here is a detailed component diagram of the concave transducer; Figure 5 This is a schematic diagram of a passive cavitation monitoring platform. Figure 6 This is a schematic diagram of biological tissue damage under the action of gradient energy. Figure 7 The time and frequency domain diagrams show the results of a single cavitation signal. Figure 8 This is a statistical graph showing the tissue damage area and passive cavitation dose value under gradient energy.

[0023] In the figure, the control host is 1, the matching box is 2, the power amplifier is 3, the concave ultrasound therapy transducer assembly is 4, the B-ultrasound imaging probe is 5, the focusing probe is 6, the oscilloscope is 7; the acquisition probe fixing device is 8, the ring experimental fixture is 9, the simulated therapy transducer device is 10; the concave metal shell is 41, the concave piezoelectric transducer element and the acoustic matching layer is 42, the imaging probe mounting hole is 43, and the high voltage interface is 44. Detailed Implementation

[0024] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand and implement the present invention. However, the embodiments described are not intended to limit the present invention.

[0025] Example 1 like Figure 1 As shown, Figure 1 This is a flowchart of a passive cavitation monitoring method based on ultrasonic cavitation effect. The passive cavitation monitoring method based on ultrasonic cavitation effect provided in this embodiment includes the following steps: Step S1: Emits high-intensity focused ultrasound pulses toward the target tissue region using a concave ultrasound transducer; Step S2: Use a passive cavitation detection probe to capture the cavitation signal generated in the target tissue region in real time, and collect the time domain data of the cavitation signal; In this embodiment, a B-mode ultrasound image of the target area can be acquired using a confocal real-time ultrasound imaging probe. In this embodiment, at least three probes can also be used to acquire signals, and the arrival time difference or amplitude difference of the acquired cavitation signals can be used to calculate the three-dimensional spatial coordinates of cavitation in real time using a cavitation source localization algorithm. The calculated three-dimensional spatial coordinates of cavitation are compared with the preset theoretical focal coordinates of the concave ultrasound transducer. If the spatial deviation exceeds a threshold, a focal calibration prompt is generated or the treatment head posture adjustment is automatically triggered.

[0026] In this embodiment, cavitation source localization is achieved using a passive localization method based on time-of-arrival (TOA), specifically as follows: An array of at least three broadband hydrophones with known spatial coordinates is arranged around the sample. Cavitation signals from each channel are acquired synchronously. Using the channel with the highest signal-to-noise ratio as a reference, the TOA of signals from other channels is calculated using a generalized cross-correlation algorithm. Based on the sound velocity, time difference, and the coordinates of each hydrophone, a series of hyperboloid equations are established with each pair of hydrophones as the focus. The overdetermined system of equations is solved using the least squares method to obtain the estimated three-dimensional spatial coordinates of the cavitation event. These coordinates can be displayed in real-time as a cavitation cloud map and compared with a preset target area for subsequent spatially selective manipulation.

[0027] Step S3: Perform a Fast Fourier Transform on the time-domain data to obtain the spectrum data; Step S4: Apply digital band-stop filtering to the spectrum data to remove the fundamental frequency and its harmonic components, and obtain the filtered spectrum; In this embodiment, the fundamental frequency of the digital band-stop filter in step S4 is 0.982 MHz, which removes harmonics from the 1st to the 9th order. The stopband bandwidth of the 1st to 7th order harmonics is ±40 kHz, and the stopband bandwidth of the 8th to 9th order harmonics is ±20 kHz.

[0028] The specific implementation process of digital bandstop filtering in this embodiment is as follows: 1. Spectrum Calculation: The acquired time-domain signal is subjected to a Fast Fourier Transform (FFT) to obtain its spectral representation in the frequency domain. This spectrum contains all frequency components of the signal, and the characteristics of the cavitation signal are further evaluated by calculating the amplitude of each frequency component.

[0029] 2. Band-stop filter design: To remove narrowband interference caused by the fundamental frequency and its harmonics, a band-stop filter was used.

[0030] The specific design process for the band-stop filter is as follows. Fundamental frequency: The selected fundamental frequency is... 0 = 982kHz, which is the fundamental frequency of the ultrasound therapy system used in the experiment; Harmonic order: In the processing, the 1st to 9th harmonic components of the fundamental frequency were considered, each corresponding to an integer multiple of the fundamental frequency. For each harmonic order... , = 1,2,…,9, the corresponding frequencies in the frequency spectrum are = 0; Bandwidth setting: For 1st to 7th order harmonics ( = 1-7), with a set bandwidth of ±40 kHz, for 8th to 9th order harmonics ( = 8 - 9), the bandwidth is set to ± 20 kHz. This bandwidth setting is based on the frequency distribution characteristics of each harmonic to ensure that the filter can effectively remove noise in this frequency band without interfering with the valid signal.

[0031] 3. Filtering process: For each harmonic order, the corresponding frequency range (including positive and negative frequency components) is first calculated. Then, a band-stop filter is used to set the signal components in these frequency ranges to zero, removing harmonic components and retaining the effective components of other frequencies in the spectrum. These effective components reflect the inertial part of the cavitation signal.

[0032] Step S5: Calculate the inertial cavitation dose (ICD) based on the filtered spectrum. The ICD is quantized by the total spectral power of the filtered spectrum. In this embodiment, the formula for calculating the total spectral power in step S5 is: ; Where N is the number of sampling points, To the amplitude after harmonic removal, Indicates the frequency component index of the signal; This indicates the total power of the signal.

[0033] The calculation process for the inertial cavitation dose (ICD) in this embodiment is as follows: Time-domain voltage signal acquired using a single PCD channel ( For example, (unit V), the sampling interval is... The sampling frequency is: = 1 / ≈ 41.7MHz;) The PCD probe is connected to the oscilloscope via a BNC cable, and the signal is acquired through a high-speed data acquisition card. The data acquisition card has a sampling period of: = 0.02398 × 10−6 s, sampling the cavitation emission signal, yields a discrete sequence: = , = 0,1,…, -1; Perform a Fast Fourier Transform (FFT) on each file / segment of the PCD signal: ; The corresponding discrete frequency is: ; Let the fundamental frequency be: ; The harmonic orders that need to be filtered out are: ; For each harmonic, a stepped interval is set to construct a symmetrical band-stop interval: right :bandwidth ; right :bandwidth ; The center frequency of each harmonic is: ; and its folding position at the negative frequency. ; Design with a barrier mask :exist ; as well as ; The filtered spectrum is obtained: ; The spectrum mainly retains the broadband noise component generated by cavitation bubble collapse, while removing the fundamental wave and its harmonic components that drive the ultrasound.

[0034] Calculate the total power (approximately mV) from the filtered spectrum. 2 s): ; In the formula, frequency Explanation of the range of values: The sampling interval is: ; Therefore, the sampling frequency is: ; The discrete frequencies output by the FFT are: ; In the frequency domain, frequency variable The overall range is (Approximately 0–41.7 MHz in this embodiment). In actual ICD calculation, 0–10 MHz is selected as the effective analysis bandwidth, and a narrow band-stop filter range is set near the 1st–9th harmonics at 0.982 MHz to remove the fundamental wave and its harmonic components. Only the remaining broadband spectrum energy is integrated to obtain the inertial cavitation dose ICD.

[0035] Step S6: Perform correlation analysis between the inertial cavitation dose (ICD) and the tissue damage area, and establish a nonlinear fitting model. The nonlinear fitting model is used to quantify the relationship between cavitation intensity and tissue damage area, that is, to analyze the relationship between cavitation intensity and the degree of tissue damage. Step S7: Input the real-time inertial cavitation dose (ICD) value into the nonlinear fitting model to calculate the current tissue damage area. Compare the real-time calculated tissue damage area with the preset treatment target. Based on the comparison result, calculate and adjust at least one emission parameter of the concave ultrasonic transducer in real time through an optimization algorithm. The emission parameters include acoustic power, pulse repetition frequency, and duty cycle until the real-time calculated tissue damage area reaches the preset treatment target.

[0036] Select based on target damage area The method for setting the parameters was described, and the relationship between the nonlinear fitting model and the inertial cavitation dose (ICD) was established using experimental data. In this method, the target damage area can be determined in several ways: 1. Single-point damage target: Directly set the target damage area. For example, the target damage area for isolated tissue is set as Alternatively, the appropriate range (50-180 mm2) can be set according to the actual application.

[0037] 2. Treatment area coverage ratio: Based on the planned treatment area area. Set target damage area ,in For coverage percentage, it is usually set to 90% or higher.

[0038] 3. Mapping based on the fitted model: through a nonlinear power-law model The target damage area is then correlated with the target ICD value, and the required target damage area is calculated in reverse. The calculated damage area is compared with the preset target damage area, and the calculation error is accounted for. ,in This represents the actual area of ​​damage.

[0039] Set the allowable error threshold Generally, it is 5% of the target damage area (e.g. When the error is greater than the threshold (i.e.) This indicates insufficient lesion area, and the system will improve the treatment effect by increasing the acoustic power, pulse repetition frequency, or duty cycle; while when the error is less than the threshold (i.e., This indicates that the damaged area is too large, and the system will avoid overtreatment by reducing the acoustic power, pulse repetition frequency, or duty cycle. When the error... When the target damage area has reached the preset value, the system will stop adjusting or enter maintenance mode to ensure stable treatment.

[0040] Acoustic power (P) is a key parameter affecting ultrasound energy transfer. Adjusting the acoustic power directly controls the cavitation intensity in the treatment area. When the actual lesion area is smaller than the target lesion area, increasing the acoustic power enhances the ultrasound energy output, thus strengthening the cavitation effect. Conversely, when the lesion area exceeds the target value, the acoustic power is reduced to decrease cavitation intensity and avoid overtreatment. The adjustment range of acoustic power is usually limited by equipment performance, therefore upper and lower limits need to be set to ensure safe operation. Pulse repetition frequency (PRF) determines the number of ultrasound pulses emitted per unit time, thus affecting the formation and collapse of cavitation bubbles. Increasing the PRF helps enhance the cavitation effect and is suitable for situations requiring increased treatment intensity; while decreasing the PRF slows down the cavitation process and is suitable for reducing treatment intensity or avoiding excessive damage. PRF adjustment also needs to be controlled within the technical limitations of the equipment to ensure safety and effectiveness. Duty cycle (DC) controls the ratio of ultrasound emission duration to rest time, directly affecting energy accumulation and distribution. A higher duty cycle increases energy transfer per cycle, promoting cavitation bubble activation; a lower duty cycle helps reduce thermal effects and excessive cavitation. When the damaged area is too large, the duty cycle should be reduced appropriately to avoid excessive damage; when the damaged area is insufficient, the treatment effect can be improved by increasing the duty cycle.

[0041] Specifically, the real-time inertial cavitation dose (ICD) value is input into a pre-stored ICD-tissue damage area nonlinear fitting model to estimate the current tissue damage progression status in real time. Finally, based on the real-time estimated tissue damage progression status and the preset treatment target, at least one transmission parameter of the concave ultrasonic transducer is calculated and adjusted in real time through an optimization algorithm. The transmission parameters include acoustic power, pulse repetition frequency, and duty cycle. This process continues until the real-time estimated tissue damage progression status reaches the preset treatment endpoint.

[0042] This embodiment uses ImageJ software to quantify the damage area.

[0043] In this embodiment, the nonlinear fitting model described in step S6 is a power-law model, in the form of: ;in, ICD value, The area of ​​tissue damage. and For model coefficients, a= b = 1.82163 ± 0.08161.

[0044] In this embodiment, the fitting model uses GraphPad Prism software and employs a nonlinear regression method to fit the relationship between cavitation intensity (ICD value) and tissue damage area. Specifically, the fitting model is a power-law model, in the following form: ; in, Indicates the ICD value. Lesion Area and These are the model coefficients.

[0045] 1. Data Collection and Preparation Different tissue damage areas were collected through experiments. ) and the corresponding ICD value ( Data. ICD values ​​for each experimental data point were acquired using a cavitation detection system, and the tissue damage area for each experimental group was calculated using image analysis software (such as ImageJ).

[0046] 2. Use GraphPad Prism software for nonlinear fitting. In GraphPad Prism, a nonlinear regression method was employed, and a power-law model was selected for fitting. GraphPad Prism software has powerful fitting algorithms that can automatically estimate the model's parameter values ​​and calculate the optimal model coefficients by minimizing the sum of squared errors between the data points and the fitted curve. and .

[0047] The specific fitting process is as follows: Input the experimental data into GraphPad Prism, select the power-law model for fitting, and the software optimizes the model parameters using the least squares method. and That is, by minimizing the following objective function: ; in and The first ICD values ​​and damage area for each data point.

[0048] 3. Coefficient and Calculation GraphPad Prism software automatically provides fitting results, including coefficients. and The optimal estimate is obtained by least squares optimization, which involves taking the derivative of the objective function and minimizing the error. GraphPad Prism software calculates the optimal estimate. and Values. Finally, the coefficients obtained by fitting using GraphPad Prism software are: These coefficients reflect the nonlinear relationship between cavitation intensity and tissue damage area in the experimental data.

[0049] In this embodiment, the correlation analysis in step S6 uses the Spearman test, and the tissue damage area is obtained by quantifying the damage contour of the experimental sample using image processing software.

[0050] In the real-time treatment phase, the power-law model provided in this embodiment is used to estimate the corresponding tissue damage area based on the real-time measured ICD value. Specifically, by solving the inverse function of the above model, the formula for calculating the damage area is obtained as follows: In the computer control system, calibrated parameters a and b are pre-stored. The system substitutes the ICD values, which are collected and calculated in real time, into the aforementioned inverse function formula to calculate the estimated tissue damage area in real time. In this embodiment, the real-time inertial cavitation dose (ICD) value can be substituted into the inverse function of the nonlinear fitting model to calculate the current estimated tissue damage area. This embodiment uses ImageJ to calculate the tissue damage area, obtains real data, establishes a model, determines parameters, and calculates the real-time damage area according to the inverse function, thereby achieving real-time prediction.

[0051] This embodiment uses ImageJ software to outline the damaged area, and the specific process of using the Spearman test is as follows: 1. ImageJ software operation steps A. Setting the scale Here are the steps to set a scale bar using ImageJ: (1) Open the image: Launch ImageJ software, select the "File" menu, click "Open" and select the image file to be analyzed.

[0052] (2) Select the scale tool: Select the Line Tool in the ImageJ toolbar and draw a line segment of known length (10mm) on the image according to the standard scale of the background at the time of shooting.

[0053] (3) Set the scale: In the ImageJ menu bar, click "Analyze" - "Set Scale". A settings dialog box will pop up. Enter the known length (enter 10) and select the unit ("mm").

[0054] (4) Verify the scale: After drawing a line segment of known length, select "Analyze" - "Measure" to confirm whether the set scale is effective and check whether the measurement value is accurate.

[0055] B. Outline drawing method Contour delineation is a common method in image analysis used to segment regions of interest (ROIs) and calculate quantitative metrics such as shape or area. The following are the steps for contour delineation using ImageJ: (1) Select tool: After opening the image, use the Polygon Tool in the ImageJ toolbar to outline the region of interest (ROI).

[0056] (2) Outline: The outline of the tissue cavity caused by the damage is drawn on the image. This embodiment can use automatic edge detection algorithms (such as the Canny operator) and image segmentation methods (such as thresholding and region growing) to achieve the outline drawing process.

[0057] (3) Area selection and measurement: After completing the outline, click the menu "Analyze" - "Measure", and ImageJ will automatically calculate the area, perimeter, and other geometric features of the outlined region.

[0058] (4) Save the outlined ROI: For easier future use, select "File" - "Save As" to save the outlined area.

[0059] (5) Output of results: All measurement results (such as area, perimeter, etc.) will be displayed in the "Results" window. You can export them to an Excel file by going to "File" and then "Save As" for easy data analysis later.

[0060] 2. The specific process of the Spearman test (1) Prepare data: Open the GraphPad Prism software, create a new project, select the “XY” data table type, enter the first set of data in column X, and enter the second set of data in column Y (x=damage area, y==ICD).

[0061] (2) Select correlation analysis: After entering the data in the table, click the “Analyze” button in the upper right corner. In the “Analyze Data” dialog box that pops up, select “Correlations”; then select “Spearman”, which is Spearman rank correlation coefficient analysis.

[0062] (3) Setting parameters: After selecting the Spearman test, GraphPad will ask you to select how to process the data, such as confirming the analysis type. At this time, you need to select the Spearman correlation coefficient.

[0063] (4) Run the analysis: After configuring the parameters, click "OK" or "Run" to start the Spearman correlation analysis. GraphPad Prism will then automatically calculate the Spearman rank correlation coefficient (SP) and the corresponding P value.

[0064] (5) View the analysis results: The results will be displayed in a new output table, including: Spearman correlation coefficient, which represents the monotonic correlation between the two variables. If SP is close to 1 or -1, it indicates a strong monotonic positive or negative correlation; if it is close to 0, it indicates no significant monotonic correlation; P-value, which is used to test whether the correlation is significant. Generally, when the p-value is less than 0.05, it indicates a significant correlation.

[0065] (6) Saving and exporting results: Once the correlation analysis is completed, GraphPad Prism will automatically save the data and analysis results. The results can be exported as PDF format via "File" -> "Export" for further use or publication.

[0066] The passive cavitation monitoring platform based on ultrasonic cavitation effect provided in this embodiment includes: Ultrasonic therapy system, real-time ultrasonic imaging probe, data acquisition system, passive cavitation detection probe, computer control system; The passive cavitation detection probe is used to capture cavitation signals generated in the target tissue area in real time; The data acquisition system is connected to the passive cavitation detection probe and is used to acquire time-domain data of the cavitation signal. The computer control system is configured to perform the following operations: Perform Fast Fourier Transform and Digital Band-Stop Filtering on the time-domain data; Inertial cavitation dose (ICD) is calculated based on the filtered spectrum. The ICD is fitted to the tissue damage area to establish a nonlinear model.

[0067] like Figure 3 As shown, Figure 3 This diagram illustrates the connection method of the signal acquisition system. In this embodiment, the passive cavitation detection probe is a focusing transducer with a center frequency of 10 MHz, connected to a digital oscilloscope, with a sampling rate of 50 MSamples / s. Connection method: Control host → Matching box → Power amplifier → Concave transducer → 10 MHz focusing probe → Oscilloscope.

[0068] In this embodiment, the PCD probe is connected to the oscilloscope via a BNC (Bayonet Neill-Concelman) cable, and signals are acquired through a high-speed data acquisition card (PicoTechnology, PicoScope). This connection method ensures that the PCD probe can acquire and transmit cavitation signals to the data acquisition system in real time, providing accurate dynamic monitoring of cavitation. During treatment, the PCD probe detects the activity of cavitation bubbles with high sensitivity and captures ultrasound signals in real time, reflecting the expansion and collapse of the bubbles. By analyzing these signals, the inertial cavitation dose (ICD) can be quantitatively assessed, thus providing data support for real-time treatment feedback.

[0069] The ultrasound therapy system described in this embodiment includes a concave ultrasound transducer with an outer diameter of 220 mm, an inner diameter of 80 mm, a focal length of 170 mm, and an operating frequency of 0.982 MHz. The specific structure of the concave ultrasound transducer is as follows: Figure 4 As shown, Figure 4 This is a detailed component diagram of a concave transducer.

[0070] The computer control system described in this embodiment is implemented using MATLAB programming and connected to an external power amplifier to adjust the ultrasonic output power.

[0071] The platform in this embodiment also includes an image processing module for quantifying the tissue damage area through scale calibration and contour analysis. The image processing module in this embodiment utilizes ImageJ software to quantify the actual damage area. Specific operation steps are as follows: A. Setting the scale Here are the steps to set a scale bar using ImageJ: 1. Open the image: Launch ImageJ software, select the "File" menu, click "Open" and select the image file to be analyzed.

[0072] 2. Select the scale tool: Select the Line Tool in the ImageJ toolbar and draw a line segment of known length (10mm) on the image according to the standard scale of the background at the time of shooting.

[0073] 3. Set the scale: In the ImageJ menu bar, click "Analyze" - "Set Scale". A settings dialog box will pop up. Enter the known length (enter 10) and select the unit ("mm").

[0074] 4. Verify the scale: After drawing a line segment of known length, select "Analyze" - "Measure" to confirm whether the set scale is effective and check whether the measurement value is accurate.

[0075] B. Outline drawing method Contour delineation is a common method in image analysis used to segment regions of interest (ROIs) and calculate quantitative metrics such as shape or area. The following are the steps for contour delineation using ImageJ: 1. Select a tool: After opening the image, use the Polygon Tool in the ImageJ toolbar to outline the region of interest (ROI).

[0076] 2. Outline: Draw the outline of the tissue cavity caused by the damage on the image.

[0077] 3. Area selection and measurement: After completing the outline, click the menu "Analyze" - "Measure", and ImageJ will automatically calculate the area, perimeter, and other geometric features of the outlined region.

[0078] 4. Save the drawn ROI: For easier future use, select "File" - "Save As" to save the outlined area.

[0079] 5. Output Results: All measurement results (such as area, perimeter, etc.) will be displayed in the "Results" window. You can export them to an Excel file by going to "File" and then "Save As" for easy data analysis later.

[0080] Example 2 like Figure 2 As shown, Figure 2This is a schematic diagram of the experimental platform, illustrating the physical composition and connections of the entire passive cavitation monitoring platform. It includes a concave ultrasonic transducer, a real-time ultrasonic imaging probe (mounted in the transducer's central aperture), a passive cavitation detection probe, a data acquisition system (oscilloscope), and a computer control system. Arrows in the diagram indicate signal flow, for example, from the PCD probe to the oscilloscope and then to the computer.

[0081] This embodiment provides a concave ultrasound transducer system, including a control host 1, a matching box 2, a power amplifier 3, a concave ultrasound therapy transducer assembly 4, a B-mode ultrasound imaging probe 5, a 10 MHz focusing probe 6, and an oscilloscope 7.

[0082] The concave ultrasound transducer assembly 4 has an overall concave bowl-shaped structure, including: a concave metal shell 41, a concave piezoelectric transducer element fixed to the inner surface of the shell 41, and an acoustic matching layer 42. The acoustic matching layer is located outside the piezoelectric element. An imaging probe mounting hole 43 is positioned at the center of the transducer, and a high-voltage interface 44 connects to an external cable. An ultrasound imaging probe 5 can be inserted into the imaging probe mounting hole 43, with its probe surface basically flush with or slightly protruding from the acoustic matching layer 42, achieving real-time ultrasound imaging guidance coaxial with the treatment sound beam (see the detailed component diagram of the concave transducer).

[0083] The control host 1 outputs drive control signals to the power amplifier 3 through the matching box 2. The power amplifier 3 is electrically connected to the concave piezoelectric transducer 42 via a high-voltage coaxial cable, generating high-intensity focused ultrasound at the focal point of the concave sound field for the treatment of boiling tissue fragmentation. The B-mode imaging probe 5 is electrically connected to the B-mode host (not shown) via a separate data cable to acquire two-dimensional grayscale images of the treatment area in real time, enabling lesion localization and monitoring of the lesion liquefaction range.

[0084] The 10 MHz focusing probe 6 is used as a passive cavitation detection (PCD) probe. Its acoustic focus is located in the focal region of the concave ultrasonic transducer to receive the broadband acoustic emission signal generated by the cavitation bubble. The focusing probe 6 is connected to the oscilloscope 7 via a BNC cable. The received signal is sampled at high speed by an advanced acquisition card and output to the data acquisition and analysis module to realize the calculation of inertial cavitation dose (ICD), real-time monitoring of cavitation intensity, and preliminary signal indication functions (replacing the oscilloscope).

[0085] The real-time monitoring platform for cavitation effect of concave ultrasonic transducers provided in this embodiment includes an ultrasonic therapy system, a real-time ultrasonic imaging probe, a data acquisition system, a passive cavitation detection system, a computer control system, and an image processing module. The specific configuration is as follows: Ultrasound therapy system: Utilizing a customized commercial concave ultrasound transducer (model: JC200, Chongqing Haifu Medical Co., Ltd.), the transducer has an outer diameter of 220 mm, an inner diameter of 80 mm, a focal length of 170 mm, and an operating frequency of 0.982 MHz. This system can generate high-intensity ultrasound waves, precisely focused on the experimental tissue area for treatment.

[0086] Real-time ultrasound imaging probe: Equipped with a DC-8 model ultrasound imaging probe manufactured by Shenzhen Mindray Co., Ltd., which is installed in the center hole of the treatment transducer for real-time monitoring of tissue ultrasound images within the treatment area.

[0087] Data Acquisition System: Data acquisition is performed through a MATLAB control system. MATLAB is connected to an external power amplifier (model: EBS, Chongqing Ronghai Ultrasonic Medical Engineering Research Center) to adjust the ultrasound output power in real time. All ultrasound pulse exposure parameters are automatically controlled via MATLAB programming to ensure the stability and accuracy of the treatment process.

[0088] Passive cavitation detection probe (PCD): A focusing transducer (Olympus, Waltham, MA, USA, 16mm diameter) with a center frequency of 10 MHz is used as a cavitation detector to capture the acoustic signal generated by the random motion of boiling cavitation bubbles in real time. The signal is acquired using a Picoscope 5444D digital oscilloscope (Pico Technology, UK) at a sampling rate of 50 MSamples / s, recording the time-domain characteristics of the cavitation signal.

[0089] In this embodiment, the connection relationship and signal flow of the PCD probe are as follows: Figure 3 As shown, the arrows indicate the signal flow. The 10MHz focusing probe is connected to the oscilloscope via a BNC cable. Signals are acquired through a high-speed acquisition card, and the signals acquired by the oscilloscope are output to the data acquisition and analysis module via a data transmission interface. The signal flow is as follows: PCD probe (10 MHz focusing probe) → Oscilloscope → Data acquisition and analysis module (host computer or dedicated acquisition card) → ICD calculation and real-time monitoring. Computer control system: performs fast Fourier transform and digital bandstop filtering on the time-domain data; calculates inertial cavitation dose (ICD) based on the filtered spectrum; fits the ICD with the tissue damage area to establish a nonlinear model.

[0090] The computer control system described in this embodiment is configured to perform the following operations: Receive multi-channel cavitation time-domain signals from a passive cavitation detection probe; The multi-channel cavitation time-domain signal or its transformed frequency-domain signal is input into a pre-trained deep learning model for cavitation feature recognition. The deep learning model for cavitation feature recognition is trained to: adaptively output the inertial cavitation dose value, steady-state cavitation intensity value, and spatiotemporal distribution map of cavitation activity in the target region based on the input signal; and identify and quantify potential areas of tissue damage through an image processing module based on the spatiotemporal distribution map of cavitation activity output by the deep learning model.

[0091] The cavitation feature recognition deep learning model described in this embodiment can be implemented using a multi-task learning model based on a convolutional neural network (CNN) combined with an attention mechanism. This model is configured to simultaneously perform regression of inertial cavitation dose (ICD) values ​​and generation of a spatiotemporal distribution map of cavitation activity. The model specifically includes the following parts: Input and Feature Extraction: The model receives preprocessed cavitation signal data as input, such as frequency domain data after Fast Fourier Transform and filtering, or its time-spectrum. Features of the input signal are extracted through an encoder network containing multiple convolutional layers.

[0092] Attention mechanisms: Introduce attention mechanisms, such as channel attention and spatial attention, into the extracted features so that the model can adaptively focus on the feature channels and spatiotemporal regions of the signal most relevant to cavitation activity.

[0093] Multi-task output head: ICD regression head: Based on attention-weighted features, it outputs a scalar through global pooling and a fully connected layer as the predicted ICD value.

[0094] Distribution map generation head: Through a decoder network containing upsampling or transposed convolutional layers, features are mapped to a scale that matches the spatiotemporal dimension of the input signal, and a two-dimensional matrix is ​​output. After being processed by an activation function, the value at each position of this matrix represents the intensity or probability of cavitation activity occurring at that location, thus forming a spatiotemporal distribution map of cavitation activity.

[0095] The training data and methods for the model are as follows: Training data composition: The training dataset was obtained experimentally and contains multiple sets of synchronously recorded data samples. Each set of samples includes: The raw signal collected by the passive cavitation detection probe.

[0096] The corresponding true ICD value is calculated using the method described above according to the present invention.

[0097] Reference maps of the spatiotemporal distribution of real cavitation activity, obtained through high-speed photography, acoustic imaging, or tissue section analysis, are used to supervise the learning of the distribution maps.

[0098] Training Process: The model is trained in a supervised manner using the training dataset. A composite loss function is used during training, which simultaneously measures the error between the predicted ICD value and the true ICD value (e.g., using mean squared error loss) and the difference between the predicted distribution map and the true reference map (e.g., using cross-entropy loss or Dice loss). Training is performed using a common gradient descent optimization algorithm (e.g., the Adam optimizer). To prevent overfitting, conventional regularization techniques such as dropout and data augmentation can be used during training, and early stopping is used to determine the final model based on the validation set performance.

[0099] The image processing module in this embodiment is used to quantify the area of ​​tissue damage through scale calibration and contour analysis.

[0100] This method is based on the ultrasonic cavitation effect, utilizing a concave ultrasonic transducer to generate high-intensity focused ultrasound (HIFU), combined with a passive cavitation detection (PCD) system to monitor cavitation signals in real time. The time-domain data of the cavitation signals are recorded through a data acquisition system, and Fast Fourier Transform (FFT) and power spectrum analysis are performed to calculate the inertial cavitation dose (ICD) as a biometric indicator quantifying cavitation intensity. A nonlinear fitting model (such as a power-law model) between ICD and tissue damage area is established using experimental data to achieve real-time monitoring, dynamic adjustment, and prediction of treatment effects related to cavitation.

[0101] This method integrates acquisition, analysis, quantification, and model fitting. It utilizes real-time ultrasound imaging and passive cavitation detection technology to capture cavitation signals in real time during treatment, demonstrating that passive cavitation dose can serve as a biological indicator for measuring the specific degree of damage in cavitation injury experiments. Furthermore, a model is fitted to these two indicators to quantify the corresponding damage intensity.

[0102] like Figure 5 As shown, Figure 5 This is a schematic diagram of a passive cavitation monitoring platform, illustrating the layout of the passive cavitation detection system. It shows the relative positions of the PCD probe and the treatment transducer, as well as the signal acquisition path (e.g., the transmission of acoustic signals from the target area to the PCD probe). The system includes a probe fixing device 8, a ring-shaped experimental fixture 9, and a simulated treatment transducer device 10. The probe fixing device securely fixes the cavitation signal acquisition probe, ensuring the stability and accuracy of the probe's position during the experiment. Its main function is to precisely position the probe for real-time capture of cavitation signals from the target area. The ring-shaped experimental fixture serves as the supporting structure for the entire experiment, primarily supporting and fixing other equipment required for the experiment, such as the treatment transducer and probe. Its design ensures accurate alignment of all equipment and provides the necessary operating space during the experiment. The simulated treatment transducer device simulates the function of an ultrasound transducer in actual treatment, responsible for emitting high-intensity focused ultrasound pulses towards the target area. Figure 5 The sketch simulates the ultrasound energy transfer structure in actual treatment equipment, ensuring the accuracy and functional simulation of the experimental platform and providing an intuitive reference for the research. These three components work together to ensure the smooth progress of the experiment and guarantee the accuracy of the collected data and the controllability of the treatment effect.

[0103] The experimental platform operation procedure is explained in detail below: The key processes of controlling the ultrasound therapy transducer with MATLAB are as follows: A MATLAB program controls an ultrasound therapy transducer to emit frequency pulsed ultrasound waves. A real-time ultrasound imaging probe monitors the grayscale changes in B-mode ultrasound images during the treatment process, capturing cavitation signals in the treatment area in real time. The signal intensity and changes are transmitted to a data acquisition system. 1. Connecting MATLAB to system hardware: This method connects an external computer controlled by a MATLAB program to the ultrasonic transducer power amplifier via a matching box, thereby controlling the transducer's acoustic output power. The matching box converts the control signals sent by the MATLAB program into signals suitable for the power amplifier input, thus precisely controlling the ultrasonic transducer's drive output.

[0104] 2. Ultrasonic pulse parameter settings: The MATLAB program is used to set parameters such as the frequency, pulse width, and pulse repetition frequency (PRF) of the ultrasonic output. The specific process is as follows: 1) Operating frequency setting: Set the required ultrasonic fundamental frequency (982kHz) using a MATLAB script.

[0105] 2) Pulse width and duration: The duration of the ultrasonic pulse is set by the pulsewidth parameter in the control program, usually in the microsecond / millisecond range.

[0106] 3) Pulse Repetition Frequency (PRF): Sets the number of pulses generated per second, controlling the number of treatment pulses per second.

[0107] 3. Controlling ultrasound output power via a matching box: The MATLAB control program precisely adjusts the transducer's power output through a matching box connected to the power amplifier. Specifically, MATLAB sends a frequency-modulated signal and adjusts the power output according to the amplitude of the input signal.

[0108] 1) Set the desired ultrasonic output power value in MATLAB (e.g., 1484W), and convert the control signal into a power amplifier input signal through a matching box.

[0109] 2) When configuring a power amplifier to drive a transducer, adjust the amplitude of the voltage and current signals according to the required power output.

[0110] The Picoscope signal acquisition card synchronously records the time-domain data of the cavitation signal. Then, through Fast Fourier Transform (FFT) and power spectrum analysis, the spectral characteristics of the cavitation signal are extracted, and key parameters such as cavitation intensity and cavitation distribution are quantified, providing a basis for dynamic adjustment of treatment.

[0111] ICD (Inertial Cavitation Dose) is a key indicator used to quantify the intensity of acoustic cavitation. Its core idea is to reflect the strength of the cavitation effect by analyzing the accumulation of broadband noise energy generated during the cavitation process over time.

[0112] Passive cavitation detection (PCD) analyzes the dual-channel signal in the frequency domain to quantify the broadband noise energy generated by inertial cavitation.

[0113] First, a Fast Fourier Transform (FFT) is performed on the time-domain signal to obtain its spectrum. To eliminate narrowband components caused by the fundamental frequency and its harmonics, a digital band-stop filter is applied to the signal. The fundamental frequency is set to f0 = 982 kHz, and harmonics up to the 9th order (n = 1-9) are removed.

[0114] For each harmonic order n, its stopband center is located at nf0, where the stopband width of the 1st to 7th harmonics is ±40 kHz, and the stopband width of the 8th to 9th harmonics is ±20 kHz.

[0115] The specific process of the filtering algorithm in this embodiment is as follows: 1. Spectrum Calculation and FFT Processing First, the time-domain signal is converted to the frequency domain using the Fast Fourier Transform (FFT), which enables the processing of the spectrum.

[0116] Signal acquisition: The raw signals acquired from the data acquisition system are first stored in the variable signal.

[0117] FFT calculation: The signal is subjected to a fast Fourier transform using the fft function in MATLAB to obtain the frequency domain signal fftsignal.

[0118] 2. Construct a band-stop filter The core of this filtering algorithm is to construct a band-stop filter to filter out harmonics (such as the fundamental frequency of ultrasound and its harmonics) within the target frequency range. The band-stop filter works by setting a mask to reduce the amplitude of the harmonic frequency band to zero, thereby removing signals from these frequency bands.

[0119] Frequency range setting: Assuming the fundamental frequency is f0=982kHz, the filter needs to remove the 1st to 9th harmonics (including positive and negative frequencies).

[0120] Frequency bandwidth setting: Set different bandwidths for different harmonic orders; for harmonics from the 1st to the 7th order, set the bandwidth to 40 kHz; for harmonics from the 8th to the 9th order, set the bandwidth to 20 kHz.

[0121] Mask application: Create a mask of the same size as the FFT result `fftsignal`, initialized with all values ​​of 1. For each harmonic order (1 to 9), set the corresponding frequency range values ​​in the mask to 0. This way, only the broadband cavitation signal is retained after filtering.

[0122] 3. Apply filters The FFT result is filtered using the mask described above. Specifically, the frequency domain signal is multiplied by the mask to obtain the filtered spectrum. filteredfft = fftsignal ×mask; 4. Calculate the inertial cavitation dose (ICD). Calculating total power: In the frequency domain, the filtered signal represents the cavitation signal after harmonic removal. The inertial cavitation dose (ICD) is obtained by calculating the signal's energy (power). The calculation formula is as follows: Where Xfiltered is the filtered spectrum and N is the signal length; in MATLAB, it is expressed as sum(abs(filteredfft). 2 To calculate the total power of the filtered signal.

[0123] totalpower = sum(abs(filteredfft). 2 ) / (N 2 ); ICDdB = 10 * log10(totalpower); This filtering algorithm removes harmonic frequencies using a band-stop filter while preserving the broadband cavitation signal, effectively extracting its characteristics. The filter precisely adjusts the signal processing range by setting the bandwidth of different order harmonics, and finally quantifies the cavitation intensity using inertial cavitation dose (ICD) to provide real-time treatment feedback. The algorithm's implementation is clear, its parameters are reasonably set, and it effectively supports the analysis and processing of cavitation signals.

[0124] After filtering, the remaining broadband spectral components mainly reflect acoustic emissions generated by inertial cavitation. The total spectral power P of the filtered signal is calculated as follows: Where N is the number of sampling points, This is the amplitude after harmonics are removed.

[0125] Inertial cavitation dose (ICD) is defined as the broadband noise energy of the filtered spectrum, and is expressed in both linear and logarithmic forms: in, This represents the linear power of a broadband cavitation signal, measured in V. 2 This reflects the total energy of the signal; It represents the logarithmic form of the power, in decibels (dB), and is often used for signal amplitude comparison and data visualization to process and observe a wide range of signal changes.

[0126] The calculated inertial cavitation dose (ICD) characterizes the cumulative broadband acoustic power (or cumulative broadband noise energy) generated by the inertial cavitation event during each ultrasonic exposure.

[0127] Experimental data quantification and fitting model: Using isolated bovine liver tissue as an experimental sample, the correlation between inertial cavitation dose (ICD) and tissue damage area was investigated.

[0128] It should be noted that parameters a and b in the aforementioned power-law model were calibrated under specific experimental conditions, primarily reflecting the damage patterns of ex vivo bovine liver tissue under the stated ultrasound parameters. In actual clinical applications, the model can be initially selected or its parameters adjusted based on the type of target tissue (e.g., ex vivo / living tissue, liver, prostate, uterine fibroids, etc.). Preferably, before formal treatment, a low-energy test pulse can be applied to a safe area adjacent to the target tissue. Based on the measured initial ICD value and the potential minor damage (which can be assessed through ultrasound imaging), model parameters a or b can be fine-tuned to achieve a degree of personalized calibration and improve prediction accuracy.

[0129] The experiment used an ultrasound therapy system and a passive cavitation monitoring system to collect cavitation signal data in real time and quantified the ICD value for each experimental group. To ensure the accuracy of the experiment, each experimental group underwent multiple repeated trials to obtain stable experimental data. Experimental parameters: duty cycle 1%, applied acoustic energy ranging from 118.72 J to 1305.92 J.

[0130] The method for calculating acoustic energy in this embodiment is as follows: Q = p × t; Where Q represents the total energy (J), p represents the applied acoustic power (W), and t represents the effective duration (s).

[0131] The damaged area exhibits a teardrop-shaped cavity characteristic, a shape typically caused by focused ultrasound waves in the focal region. The damage is most severe in the center, gradually transitioning to unaffected tissue areas around the periphery. This teardrop-shaped feature indicates that the damage is caused by the action of ultrasound waves in a localized area of ​​high energy density within the tissue.

[0132] This embodiment can use materials such as isolated pig liver, bovine liver, or biomimetic tissue, or other similar tissue types for experiments, such as... Figure 6 As shown, Figure 6 This diagram illustrates the damage to biological tissue under gradient energy, showcasing the morphology of excised bovine liver tissue at different ultrasound energy levels (from 118.72 J to 1305.92 J). The damaged areas are teardrop-shaped, with the most severe damage at the center, gradually transitioning outwards to normal tissue, visually reflecting the focusing effect of ultrasound. The damaged tissue samples were photographed against a background marked with a scale bar. During image processing, the contours of the damaged areas were delineated using ImageJ software by comparing the pixel values ​​in the image with the scale bar.

[0133] Based on the previously calculated ratio of pixels to actual size, the software can automatically convert and calculate the actual area of ​​the damaged region, thereby obtaining accurate damaged area data (N=20). This embodiment uses ImageJ operations, and the specific steps are as follows: Specific steps for quantifying the actual damage area using ImageJ software: A. Setting the scale Here are the steps to set a scale bar using ImageJ: 1. Open the image: Launch ImageJ software, select the "File" menu, click "Open" and select the image file to be analyzed.

[0134] 2. Select the scale tool: Select the Line Tool in the ImageJ toolbar and draw a line segment of known length (10mm) on the image according to the standard scale of the background at the time of shooting.

[0135] 3. Set the scale: In the ImageJ menu bar, click "Analyze" - "Set Scale". A settings dialog box will pop up. Enter the known length (enter 10) and select the unit ("mm").

[0136] 4. Verify the scale: After drawing a line segment of known length, select "Analyze" - "Measure" to confirm whether the set scale is effective and check whether the measurement value is accurate.

[0137] B. Outline drawing method Contour delineation is a common method in image analysis used to segment regions of interest (ROIs) and calculate quantitative metrics such as shape or area. The following are the steps for contour delineation using ImageJ: 1. Select a tool: After opening the image, use the Polygon Tool in the ImageJ toolbar to outline the region of interest (ROI).

[0138] 2. Outline: Draw the outline of the tissue cavity caused by the damage on the image.

[0139] 3. Area selection and measurement: After completing the outline, click the menu "Analyze" - "Measure", and ImageJ will automatically calculate the area, perimeter, and other geometric features of the outlined region.

[0140] 4. Save the drawn ROI: For easier future use, select "File" - "Save As" to save the outlined area.

[0141] 5. Output Results: All measurement results (such as area, perimeter, etc.) will be displayed in the "Results" window. You can export them to an Excel file by going to "File" and then "Save As" for easy data analysis later.

[0142] Throughout the experiment, a concave focusing transducer was used as the ultrasonic generator, and the cavitation signal at the acoustic focal point was collected under each ultrasonic pulse using the passive cavitation system of the concave transducer.

[0143] Figure 7 This diagram illustrates the time and frequency domain representations of a single cavitation signal. The left image shows the time-domain waveform of the cavitation signal (amplitude varying over time), and the right image shows the frequency-domain spectrum after FFT transformation. This diagram is used to explain the signal processing procedure, including the distribution of the fundamental frequency and harmonic components. The blue box represents the fundamental frequency portion of the frequency domain signal, the red box represents the second harmonic (and so on up to the third to tenth harmonics), and the black box represents the broadband noise portion.

[0144] Using the above calculation method, a statistical graph showing the change between the damaged area and the calculated accumulated passive cavitation dose (ICD) was obtained. The statistical law is explained as follows: This fitted model demonstrates a nonlinear power-law relationship between the damaged area and the inertial cavitation dose (ICD). Specifically, the exponential part of the model... This indicates a power-law increase in the relationship between the damaged area and the ICD value. This is a typical superlinear relationship. It is a proportionality constant, representing the proportional relationship between the damaged area and the ICD. This indicates the initial effect of the damaged area on the ICD value—when the damaged area is 1 mm. 2 At that time, the value of ICD is This value is closely related to experimental conditions, equipment configuration, and data processing methods. It is an index that controls the intensity of the effect of changes in the damaged area on the ICD. This indicates that the effect of damage area on ICD value is superlinear; the rate of increase in ICD value gradually accelerates with increasing damage area. This also suggests an increasing, rather than a nonlinear, relationship between cavitation intensity (ICD value) and tissue damage. If this index is greater than 1, it means that as the damaged area increases, the rate of increase in the ICD value will gradually accelerate.

[0145] Therefore, the lesion area shows a significant increasing trend with increasing gradient energy. In the initial stage, the lesion area begins to increase rapidly with the gradual increase of ultrasound energy, showing a relatively significant correlation. When the energy reaches a certain threshold, the lesion area tends to stabilize, indicating that the energy effect of ultrasound therapy has reached saturation. Within the corresponding treatment cycle, the ICD value shows a cumulative increasing trend with the increase of gradient energy. Especially in the initial stage, the ICD value is positively correlated with the increase of energy, and the cavitation intensity continuously increases with the advancement of the treatment cycle. As ultrasound therapy progresses, the ICD value accumulates continuously in each cycle and shows a gradual increasing trend, reflecting the gradually increasing intensity of the cavitation effect during the treatment process.

[0146] Figure 8 This chart presents statistical graphs of tissue damage area and passive cavitation dose under gradient energy. The left graph shows the change in tissue damage area with increasing energy, and the right graph shows the cumulative curve of ICD value with increasing energy. The two graphs together demonstrate the positive correlation between ICD and damage area, providing data support for the fitted model. Further scatter plots or fitted curves of ICD value versus damage area could be shown to verify the accuracy of the power-law model (e.g., adjusting R²). 2 (Value). In this embodiment, the Spearman test was used to analyze the correlation between the two sets of data.

[0147] The results showed that the Spearman correlation coefficient was 0.897, and the p-value was less than 0.001, indicating a significant and strong correlation between the two sets of data. A fitting model analysis was performed on these two sets of data, resulting in a nonlinear power-law model. The specific model form is shown in the figure. The adjusted R-squared value of the fitting results is shown in the figure. 2 The value is 0.82954, indicating that the model can describe the relationship between the data well. The nonlinear characteristics of the data were further verified by fitting the coefficients a and b in the model. x represents the lesion area, and y represents the cumulative passive dose value.

[0148] Table 1 Specific parameters of the fitting model Here, Spearman's correlation represents the correlation coefficient under the Spearman test; Adj.R 2 (Adjusted coefficient of determination): Adj.R 2 =0.82954 represents the goodness of fit of the model; the closer to 1, the better the model fits the data. Here, the value is 0.83, indicating that the model can explain 83% of the data variability, and the fit is good.

[0149] The above-described embodiments are merely preferred embodiments provided to fully illustrate the present invention, and the scope of protection of the present invention is not limited thereto. Equivalent substitutions or modifications made by those skilled in the art based on the present invention are all within the scope of protection of the present invention. The scope of protection of the present invention is defined by the claims.

Claims

1. A passive cavitation monitoring method based on ultrasonic cavitation effect, characterized in that: Includes the following steps: Step S1: Emits high-intensity focused ultrasound pulses toward the target tissue region using a concave ultrasound transducer; Step S2: Use a passive cavitation detection probe to capture the cavitation signal generated in the target tissue region in real time, and collect the time domain data of the cavitation signal; Step S3: Perform a Fast Fourier Transform on the time-domain data to obtain the spectrum data; Step S4: Apply digital band-stop filtering to the spectrum data to remove the fundamental frequency and its harmonic components, and obtain the filtered spectrum; Step S5: Calculate the inertial cavitation dose (ICD) based on the filtered spectrum, wherein the inertial cavitation dose (ICD) is quantized by the total spectral power of the filtered spectrum; Step S6: Establish a nonlinear fitting model based on the inertial cavitation dose (ICD) and tissue damage area. The nonlinear fitting model is used to calculate the cavitation intensity and tissue damage area. Step S7: Input the real-time inertial cavitation dose (ICD) value into the nonlinear fitting model to calculate the current tissue damage area, compare the real-time calculated tissue damage area with the preset treatment target, and adjust at least one emission parameter of the concave ultrasonic transducer based on the comparison result. The emission parameters include acoustic power, pulse repetition frequency, and duty cycle until the real-time calculated tissue damage area reaches the preset treatment target.

2. The passive cavitation monitoring method based on ultrasonic cavitation effect as described in claim 1, characterized in that: In step S4, the fundamental frequency of the digital band-stop filter is 0.982 MHz, which removes harmonics from the 1st to the 9th order. The stopband bandwidth of the 1st to 7th order harmonics is ±40 kHz, and the stopband bandwidth of the 8th to 9th order harmonics is ±20 kHz.

3. The passive cavitation monitoring method based on ultrasonic cavitation effect as described in claim 1, characterized in that: The formula for calculating the total spectral power in step S5 is as follows: Where N is the number of sampling points, To the amplitude after harmonic removal, Frequency component index of the signal.

4. The passive cavitation monitoring method based on ultrasonic cavitation effect as described in claim 1, characterized in that: The nonlinear fitting model in step S6 is a power-law model, in the form of: ;in, ICD value, The area of ​​tissue damage. and These are the model coefficients.

5. The passive cavitation monitoring method based on ultrasonic cavitation effect as described in claim 1, characterized in that: The correlation analysis in step S6 uses the Spearman test, and the tissue damage area is obtained through image processing.

6. A passive cavitation monitoring platform based on ultrasonic cavitation effect, characterized in that: Includes an ultrasound therapy system, a real-time ultrasound imaging probe, a data acquisition system, a passive cavitation detection probe, and a computer control system; The passive cavitation detection probe is used to capture cavitation signals generated in the target tissue area in real time; The data acquisition system is connected to the passive cavitation detection probe and is used to acquire time-domain data of the cavitation signal. The computer control system is configured to perform the following operations: Perform Fast Fourier Transform and Digital Band-Stop Filtering on the time-domain data; Calculation of inertial cavitation dose (ICD) based on the filtered spectrum; The ICD is fitted to the tissue damage area to establish a nonlinear model.

7. The passive cavitation monitoring platform based on ultrasonic cavitation effect as described in claim 6, characterized in that: The ultrasound therapy system includes a concave ultrasound transducer with an outer diameter of 220 mm, an inner diameter of 80 mm, a focal length of 170 mm, and an operating frequency of 0.982 MHz.

8. The passive cavitation monitoring platform based on ultrasonic cavitation effect as described in claim 6, characterized in that: The passive cavitation detection probe is a focusing transducer with a center frequency of 10 MHz and is connected to a digital oscilloscope with a sampling rate of 50 MSamples / s.

9. The passive cavitation monitoring platform based on ultrasonic cavitation effect as described in claim 6, characterized in that: The computer control system is implemented using MATLAB programming and is connected to an external power amplifier to adjust the ultrasonic output power.

10. The passive cavitation monitoring platform based on ultrasonic cavitation effect as described in claim 6, characterized in that: The platform also includes an image processing module for quantifying the area of ​​tissue damage through scale calibration and contour analysis.

Images