Kidney stone in-vitro lithotripsy detection system and detection method based on artificial intelligence ultrasonic image

Abstract

The invention discloses a kidney stone in-vitro lithotripsy detection system and method based on an artificial intelligence ultrasonic image, and belongs to the technical field of kidney stone in-vitro lithotripsy detection. In order to improve the kidney stone in-vitro lithotripsy detection accuracy of an ultrasonic image, the method comprises the steps of obtaining a kidney B-ultrasonic image of a patient with calculus, performing data processing, constructing a kidney stone recognition model of the ultrasonic image by adopting an HRNe model in combination with mean filtering and Sobel operator processing technologies, and determining the kidney stone in-vitro lithotripsy detection accuracy of the ultrasonic image. Evaluating the trained kidney stone recognition model of the ultrasonic image by using the verification set, calculating the accuracy, sensitivity and specificity, evaluating the accuracy and reliability of the trained kidney stone recognition model of the ultrasonic image in a kidney stone recognition task, and further optimizing the trained kidney stone recognition model of the ultrasonic image to obtain the kidney stone recognition model of the ultrasonic image. Calculus features output by the optimized kidney calculus recognition model of the ultrasonic image are obtained, kidney calculus in-vitro lithotripsy energy is calculated, and a kidney calculus in-vitro lithotripsy scheme is planned. The invention provides a more accurate scheme.

Description

Technical Field

[0001] The present invention belongs to the technical field of extracorporeal lithotripsy detection of kidney stones, and in particular relates to an extracorporeal lithotripsy detection system and a detection method for kidney stones based on artificial intelligence ultrasonic images. Background Art

[0002] ESWL is a non-invasive treatment method that uses high-energy shock waves to focus outside the body and act on the stones, gradually exfoliating and breaking them from the surface, and eventually excreting them from the body with urine. This technology has been widely used in the treatment of kidney stones and has become one of the preferred methods for urolithiasis. B-ultrasound image acquisition: During the treatment process, B-ultrasound equipment is used to collect images of the kidneys and stones. These images provide key information such as the location, size and shape of the stones, which is essential for guiding the positioning and emission of shock waves.

[0003] Although artificial intelligence has shown great potential in the extracorporeal lithotripsy diagnosis and treatment system for kidney stones based on B-ultrasound images, the existing technology still has some defects;

[0004] 1. Image quality limitations: The quality of B-ultrasound images is affected by many factors, such as the patient's body position, abdominal wall fat thickness, stone composition and size, etc. For smaller or deeper stones, the B-ultrasound image may not be clear enough, making it difficult for the AI ​​algorithm to accurately identify and analyze them.

[0005] 2. Algorithm accuracy: Although AI algorithms perform well in image analysis, there is still a possibility of misjudgment and missed judgment. This may be due to the limitations of the algorithm itself, insufficient training data, or differences in image quality.

[0006] 3. Operational complexity: Although AI algorithms can simplify the interpretation of B-ultrasound images and the positioning of shock waves, the entire diagnosis and treatment system still requires professional operators to monitor and adjust. In addition, the operation of AI algorithms also requires high-performance computing equipment and professional technical support.

[0007] 4. Technology Popularity: Currently, not all medical institutions have the conditions and capabilities to introduce artificial intelligence-based B-ultrasound image-based extracorporeal lithotripsy diagnosis and treatment systems for kidney stones, which limits the popularity and application scope of this technology.

[0008] In summary, artificial intelligence has broad application prospects in the extracorporeal lithotripsy diagnosis and treatment system for kidney stones based on B-ultrasound images, but some technical challenges and limitations still need to be overcome. Summary of the invention

[0009] The problem to be solved by the present invention is to improve the accuracy of extracorporeal lithotripsy detection of kidney stones using ultrasonic images, and to propose an extracorporeal lithotripsy detection system and method for kidney stones using ultrasonic images based on artificial intelligence.

[0010] To achieve the above object, the present invention is implemented through the following technical solutions:

[0011] An artificial intelligence-based ultrasound image-based extracorporeal lithotripsy detection method for kidney stones comprises the following steps:

[0012] S1. Obtain renal B-ultrasound images of patients with stones, classify the renal B-ultrasound images using image annotation tools, and construct a data set;

[0013] S2. Using a data processing unit to process the data set obtained in step S1, perform data cleaning, image interpolation, and standardization to obtain a data set after data processing, and divide it into a training set and a validation set;

[0014] S3. The artificial intelligence algorithm unit uses the HRNe model combined with mean filtering and Sobel operator processing technology to construct a kidney stone recognition model for ultrasound images, uses the training set of step S2 to train the constructed kidney stone recognition model for ultrasound images, obtains a trained kidney stone recognition model for ultrasound images, and outputs the kidney stone recognition result for ultrasound images;

[0015] S4. The model optimization unit evaluates the trained ultrasound image kidney stone recognition model obtained in step S3 using the validation set, calculates the accuracy, sensitivity and specificity, evaluates the accuracy and reliability of the trained ultrasound image kidney stone recognition model in the kidney stone recognition task, further optimizes the trained ultrasound image kidney stone recognition model, and obtains an optimized ultrasound image kidney stone recognition model;

[0016] S5. The result interaction unit calculates the extracorporeal lithotripsy energy and plans an extracorporeal lithotripsy plan based on the stone features output by the kidney stone recognition model of the ultrasound image optimized in step S4.

[0017] Furthermore, the specific implementation method of step S2 image interpolation includes the following steps:

[0018] S2.1. For the image data in the data set obtained in step S1, first Direction linear interpolation:

[0019]

[0020] Among them, f is the interpolation function, R1 is the coordinate of the first pixel point on the image, Q 11 of Axis coordinates, Q 21 of Axis coordinates, is the coordinate of the first pixel Axis coordinate, Q11 For R1 The first interpolation point in the direction, Q 21 For R1 The second interpolation point in the direction, is the coordinate of the first pixel Axis coordinates,

[0021]

[0022] Among them, Q 12 For R2 The first interpolation point in the direction, Q 22 For R2 The second interpolation point in the direction, R2 is the coordinate of the second pixel point on the image, The coordinates of the second point Axis coordinates,

[0023] S2.2. Then Direction linear interpolation:

[0024]

[0025] Where f(P) is the ratio of R1 and R2. Direction linear interpolation function;

[0026] S2.3. Combining steps S2.1 and S2.2, the image interpolation result is:

[0027]

[0028] Furthermore, the specific implementation method of step S3 includes the following steps:

[0029] S3.1. Constructing a HRNe model, including downsampling through two convolutional layers with a convolution kernel size of 3x3 and a stride of 2, and then repeatedly stacking through the Layer1 module, adding the stacked results and finally obtaining the fusion output of the downsampled 4-fold branch through the ReLU function; the fusion output of the downsampled 4-fold branch includes combining the output of the downsampled 8-fold branch and then upsampling by 2 times, and the output of the downsampled 16-fold branch and then upsampling by 4 times, and the outputs of the three branches are activated by ReLU again to form the final fusion output of the model;

[0030] The final expression of the HRNe model is:

[0031] F out =H HRNet (F in )

[0032] Among them, F out Represents the feature map output by the HRNe model, H HRNet Represents the processing of the HRNet algorithm, F in Represents the input convolution image;

[0033] S3.2. The feature map output by the HRNe model obtained in step S3.1 is smoothed using the mean filtering method. For each pixel in the image, the filtered The pixel value of the pixel The calculation formula is as follows:

[0034]

[0035] Among them, M represents the length of the mean filter window, and N represents the width of the mean filter window. is the feature map output by the HRNe model The pixel value of the point;

[0036] S3.3. Use the Sobel operator to process the edges of the image on the pixel values ​​filtered in step S3.2, and set the template in the horizontal direction to:

[0037]

[0038] For the pixels in the image Horizontal gradient magnitude for:

[0039]

[0040] Set the vertical template to:

[0041]

[0042] The vertical gradient amplitude is for:

[0043]

[0044] The total gradient magnitude is:

[0045]

[0046] Furthermore, the calculation formula of the accuracy acc in step S4 is:

[0047]

[0048] Among them, TP is the number of samples that actually have stones and are predicted to have stones by the model; TN is the number of samples that actually do not have stones and are predicted to have stones by the model; FP is the number of samples that actually do not have stones but are predicted to have stones by the model; FN is the number of samples that actually have stones but are predicted to have no stones by the model;

[0049] The calculation formula of sensitivity TPR is:

[0050]

[0051] The calculation formula of specificity TNR is:

[0052]

[0053] Furthermore, the specific implementation method of step S5 includes the following steps:

[0054] S5.1. Collect the stone features output by the kidney stone recognition model of the ultrasound image optimized in step S4, combined with the specific conditions of the patient, including stone size, location, and morphology;

[0055] S5.2. Apply SVM to build a classification model to comprehensively calculate the different features of the stones obtained in step S5.1. In the sample space of stones, the formula for dividing the hyperplane is as follows:

[0056] w T +b=y i

[0057] Where w=(w1,w2,……,w d ) is the parameter feature of the stone, b is the distance of the hyperplane, y i Classification of lithotripsy energy for patient selection;

[0058] S5.3. The physician selects the parameters of the extracorporeal lithotripsy device based on experience and the model parameters obtained in step S5.2, including shock wave intensity, frequency, and focal point, and plans an extracorporeal lithotripsy plan for kidney stones.

[0059] An artificial intelligence-based ultrasonic image-based extracorporeal lithotripsy detection system for kidney stones is implemented based on the artificial intelligence-based ultrasonic image-based extracorporeal lithotripsy detection method for kidney stones, and includes a data processing unit, an artificial intelligence algorithm unit, a model optimization unit, and a result interaction unit. The data processing unit is connected to the artificial intelligence algorithm unit, the artificial intelligence algorithm unit is connected to the model optimization unit, and the model optimization unit is connected to the result interaction unit.

[0060] Beneficial effects of the present invention:

[0061] The present invention discloses an artificial intelligence-based extracorporeal lithotripsy detection method for kidney stones using ultrasound images, which significantly improves the resolution and feature extraction accuracy of ultrasound images through deep fusion image interpolation technology, a high-resolution network (HRNet) algorithm, and a mean filtering method. A support vector machine is then used to select the setting of the lithotripsy energy for the stones, providing a more accurate and personalized solution for the extracorporeal lithotripsy treatment of kidney stones.

[0062] The present invention discloses an artificial intelligence-based extracorporeal lithotripsy detection method for kidney stones using ultrasound images, which uses cutting-edge technologies such as deep learning to automatically analyze and extract features from ultrasound images. By training and optimizing the algorithm model, the system can automatically identify and locate kidney stones, while accurately measuring and analyzing key features such as their size, shape, and position. These key features will serve as an important basis for extracorporeal lithotripsy treatment decisions, providing doctors with more accurate and reliable diagnostic results and treatment recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 This is a flow chart of an artificial intelligence-based ultrasound image-based extracorporeal lithotripsy detection method for kidney stones according to the present invention;

[0064] Figure 2 It is a structural block diagram of an extracorporeal kidney stone lithotripsy detection method based on artificial intelligence ultrasound images according to the present invention;

[0065] Figure 3 This is a connection relationship diagram of an artificial intelligence-based ultrasonic image-based extracorporeal kidney stone lithotripsy detection system described in the present invention. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical solution and advantages of the present invention more clear, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention, that is, the specific embodiments described are only part of the embodiments of the present invention, rather than all of the specific embodiments. The components of the specific embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations, and the present invention can also have other embodiments.

[0067] Therefore, the following detailed description of the specific embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents the selected specific embodiments of the present invention. Based on the specific embodiments of the present invention, all other specific embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0068] In order to further understand the content, features and effects of the present invention, the following specific implementation methods are given as examples, and the attached Figure 1 -Attached Figure 3 The detailed instructions are as follows:

[0069] Embodiment 1:

[0070] An artificial intelligence-based ultrasound image-based extracorporeal lithotripsy detection method for kidney stones comprises the following steps:

[0071] S1. Obtain renal B-ultrasound images of patients with stones, classify the renal B-ultrasound images using image annotation tools, and construct a data set;

[0072] S2. Using a data processing unit to process the data set obtained in step S1, perform data cleaning, image interpolation, and standardization to obtain a data set after data processing, and divide it into a training set and a validation set;

[0073] Furthermore, the specific implementation method of step S2 image interpolation includes the following steps:

[0074] S2.1. For the image data in the data set obtained in step S1, first Direction linear interpolation:

[0075]

[0076] Among them, f is the interpolation function, R1 is the coordinate of the first pixel point on the image, Q 11 of Axis coordinates, Q 21 of Axis coordinates, is the coordinate of the first pixel Axis coordinate, Q 11 For R1 The first interpolation point in the direction, Q 21 For R1 The second interpolation point in the direction, is the coordinate of the first pixel Axis coordinates,

[0077]

[0078] Among them, Q 12 For R2 The first interpolation point in the direction, Q 22 For R2 The second interpolation point in the direction, R2 is the coordinate of the second pixel point on the image, The coordinates of the second point Axis coordinates,

[0079] S2.2. Then Direction linear interpolation:

[0080]

[0081] Where f(P) is the ratio of R1 and R2. Direction linear interpolation function;

[0082] S2.3. Combining steps S2.1 and S2.2, the image interpolation result is:

[0083]

[0084] Furthermore, the ultrasound image is preprocessed by using an image interpolation method. Image interpolation can improve the resolution of the image by increasing the number of pixels in the image, thereby showing the detailed structure of the stone more clearly.

[0085] S3. The artificial intelligence algorithm unit uses the HRNe model combined with mean filtering and Sobel operator processing technology to construct a kidney stone recognition model for ultrasound images, uses the training set of step S2 to train the constructed kidney stone recognition model for ultrasound images, obtains a trained kidney stone recognition model for ultrasound images, and outputs the kidney stone recognition result for ultrasound images;

[0086] Furthermore, the specific implementation method of step S3 includes the following steps:

[0087] S3.1. Constructing a HRNe model, including downsampling through two convolutional layers with a convolution kernel size of 3x3 and a stride of 2, and then repeatedly stacking through the Layer1 module, adding the stacked results and finally obtaining the fusion output of the downsampled 4-fold branch through the ReLU function; the fusion output of the downsampled 4-fold branch includes combining the output of the downsampled 8-fold branch and then upsampling by 2 times, and the output of the downsampled 16-fold branch and then upsampling by 4 times, and the outputs of the three branches are activated by ReLU again to form the final fusion output of the model;

[0088] The final expression of the HRNe model is:

[0089] F out =H HRNet (F in )

[0090] Among them, F out Represents the feature map output by the HRNe model, H HRNet Represents the processing of the HRNet algorithm, F in Represents the input convolution image;

[0091] Furthermore, HRNet (High Resolution Network) is used as the core algorithm for B-ultrasound image feature extraction. HRNet can maintain high-resolution features throughout the entire network process through parallel multi-resolution paths and repeated multi-scale feature fusion, thereby effectively capturing the subtle structures in ultrasound images.

[0092] Furthermore, the process of upsampling or downsampling Xi from resolution i to resolution k is included in the function a(Xi, k), which is expressed as:

[0093] X i =a(X i , k)

[0094] The calculation formula of loss M is:

[0095]

[0096] Where n represents the number of samples, y i Indicates the actual value, y p represents the predicted value, and ∑ represents the sum of all samples.

[0097] S3.2. The feature map output by the HRNe model obtained in step S3.1 is smoothed using the mean filtering method. For each pixel in the image, the filtered The pixel value of the pixel The calculation formula is as follows:

[0098]

[0099] Among them, M represents the length of the mean filter window, and N represents the width of the mean filter window. is the feature map output by the HRNe model The pixel value of the point;

[0100] After the HRNet algorithm is processed, the extracted feature map is smoothed using the mean filter method. Mean filtering is a simple image processing technique that removes noise by calculating the average value of the neighborhood around the pixel. The window size of the mean filter can be adjusted according to the actual situation to remove noise while ensuring image clarity.

[0101] S3.3. Use the Sobel operator to process the edge of the image after filtering the pixel values ​​in step S3.2. Set

[0102] The template for the horizontal direction is:

[0103]

[0104] For the pixels in the image Horizontal gradient magnitude for:

[0105]

[0106] Set the vertical template to:

[0107]

[0108] The vertical gradient amplitude is for:

[0109]

[0110] The total gradient magnitude is:

[0111]

[0112] The main purpose of mean filtering is to remove noise and smooth images to make model recognition more accurate, while the main purpose of Sobel operator is to detect edges in images. Mean filtering achieves smoothing by calculating the mean of pixels in a window, while Sobel operator detects edges by calculating the approximate value of image gradients. Both mean filtering and Sobel operator use convolution kernels, but their convolution kernels are designed and function differently. The convolution kernel of mean filtering is usually square, and each element has an equal weight; while the convolution kernel of Sobel operator is specially designed to capture grayscale changes in images and highlight edge features. Combining the two methods can help improve the accuracy of image recognition.

[0113] S4. The model optimization unit evaluates the trained ultrasound image kidney stone recognition model obtained in step S3 using the validation set, calculates the accuracy, sensitivity and specificity, evaluates the accuracy and reliability of the trained ultrasound image kidney stone recognition model in the kidney stone recognition task, further optimizes the trained ultrasound image kidney stone recognition model, and obtains an optimized ultrasound image kidney stone recognition model;

[0114] Furthermore, the calculation formula of the accuracy acc in step S4 is:

[0115]

[0116] Among them, TP is the number of samples that actually have stones and are predicted to have stones by the model; TN is the number of samples that actually do not have stones and are predicted to have stones by the model; FP is the number of samples that actually do not have stones but are predicted to have stones by the model; FN is the number of samples that actually have stones but are predicted to have no stones by the model;

[0117] The calculation formula of sensitivity TPR is:

[0118]

[0119] The calculation formula of specificity TNR is:

[0120]

[0121] S5. The result interaction unit calculates the extracorporeal lithotripsy energy and plans an extracorporeal lithotripsy plan based on the stone features output by the kidney stone recognition model of the ultrasound image optimized in step S4.

[0122] Furthermore, the specific implementation method of step S5 includes the following steps:

[0123] S5.1. Collect the stone features output by the kidney stone recognition model of the ultrasound image optimized in step S4, combined with the specific conditions of the patient, including stone size, location, and morphology;

[0124] S5.2. Apply SVM to build a classification model to comprehensively calculate the different features of the stones obtained in step S5.1. In the sample space of stones, the formula for dividing the hyperplane is as follows:

[0125] w T +b=y i

[0126] Where w=(w1,w2,……,w d ) is the parameter feature of the stone, b is the distance of the hyperplane, y i Classification of lithotripsy energy for patient selection;

[0127] S5.3. The physician selects the parameters of the extracorporeal lithotripsy device based on experience and the model parameters obtained in step S5.2, including shock wave intensity, frequency, and focal point, and plans an extracorporeal lithotripsy plan for kidney stones.

[0128] Embodiment 2:

[0129] An artificial intelligence-based ultrasonic image-based extracorporeal lithotripsy detection system for kidney stones is implemented based on an artificial intelligence-based ultrasonic image-based extracorporeal lithotripsy detection method for kidney stones described in Example 1, and is characterized in that it includes a data processing unit, an artificial intelligence algorithm unit, a model optimization unit, and a result interaction unit, the data processing unit is connected to the artificial intelligence algorithm unit, the artificial intelligence algorithm unit is connected to the model optimization unit, and the model optimization unit is connected to the result interaction unit.

[0130] It should be noted that relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0131] Although the present application has been described above with reference to specific embodiments, various modifications may be made thereto and parts thereof may be replaced with equivalents without departing from the scope of the present application. In particular, as long as there is no structural conflict, the various features in the specific embodiments disclosed in the present application may be used in combination with each other in any manner, and the fact that these combinations are not exhaustively described in this specification is only for the sake of omitting space and saving resources. Therefore, the present application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A method for detecting extracorporeal lithotripsy of kidney stones based on ultrasound images using artificial intelligence, characterized in that: The steps include: S1. Obtain renal B-ultrasound images of patients with stones, classify the renal B-ultrasound images using image annotation tools, and construct a data set; S2. Using a data processing unit to process the data set obtained in step S1, perform data cleaning, image interpolation, and standardization to obtain a data set after data processing, and divide it into a training set and a validation set; S3. The artificial intelligence algorithm unit uses the HRNe model combined with mean filtering and Sobel operator processing technology to construct a kidney stone recognition model for ultrasound images, uses the training set of step S2 to train the constructed kidney stone recognition model for ultrasound images, obtains a trained kidney stone recognition model for ultrasound images, and outputs the kidney stone recognition result for ultrasound images; S4. The model optimization unit evaluates the trained ultrasound image kidney stone recognition model obtained in step S3 using the validation set, calculates the accuracy, sensitivity and specificity, evaluates the accuracy and reliability of the trained ultrasound image kidney stone recognition model in the kidney stone recognition task, further optimizes the trained ultrasound image kidney stone recognition model, and obtains an optimized ultrasound image kidney stone recognition model; S5. The result interaction unit calculates the extracorporeal lithotripsy energy and plans an extracorporeal lithotripsy plan based on the stone features output by the kidney stone recognition model of the ultrasound image optimized in step S4.

2. The method for detecting extracorporeal lithotripsy of kidney stones based on ultrasonic images using artificial intelligence according to claim 1, characterized in that: The specific implementation method of step S2 image interpolation includes the following steps: S2.

1. For the image data in the data set obtained in step S1, first perform linear interpolation in the X direction: Among them, f is the interpolation function, R1 is the coordinate of the first pixel point on the image, Q 11 of Axis coordinates, Q 21 of Axis coordinates, is the coordinate of the first pixel Axis coordinate, Q 11 For R1 The first interpolation point in the direction, Q 21 For R1 The second interpolation point in the direction, is the coordinate of the first pixel Axis coordinates, Among them, Q 12 For R2 The first interpolation point in the direction, Q 22 For R2 The second interpolation point in the direction, R2 is the coordinate of the second pixel point on the image, The coordinates of the second point Axis coordinates, S2.

2. Then Direction linear interpolation: Where f(P) is the ratio of R1 and R2. Direction linear interpolation function; S2.

3. Combining steps S2.1 and S2.2, the image interpolation result is:

3. The method for detecting extracorporeal lithotripsy of kidney stones based on ultrasonic images using artificial intelligence according to claim 2, characterized in that: The specific implementation method of step S3 includes the following steps: S3.

1. Constructing a HRNe model, including downsampling through two convolutional layers with a convolution kernel size of 3x3 and a stride of 2, and then repeatedly stacking through the Layer1 module, adding the stacked results and finally obtaining the fusion output of the downsampled 4-fold branch through the ReLU function; the fusion output of the downsampled 4-fold branch includes combining the output of the downsampled 8-fold branch and then upsampling by 2 times, and the output of the downsampled 16-fold branch and then upsampling by 4 times, and the outputs of the three branches are activated by ReLU again to form the final fusion output of the model; The final expression of the HRNe model is: F out =H HRNet (F in ) Among them, F out Represents the feature map output by the HRNe model, H HRNet Represents the processing of the HRNet algorithm, F in Represents the input convolution image; S3.

2. The feature map output by the HRNe model obtained in step S3.1 is smoothed using the mean filtering method. For each pixel in the image, the filtered The pixel value of the pixel The calculation formula is as follows: Among them, M represents the length of the mean filter window, and N represents the width of the mean filter window. is the feature map output by the HRNe model The pixel value of the point; S3.

3. Use the Sobel operator to process the edges of the image on the pixel values ​​filtered in step S3.2, and set the template in the horizontal direction to: For the pixels in the image Horizontal gradient magnitude for: Set the vertical template to: The vertical gradient magnitude is for: The total gradient magnitude is:

4. The method for detecting extracorporeal lithotripsy of kidney stones based on ultrasonic images using artificial intelligence according to claim 3, characterized in that: The calculation formula of the accuracy acc in step S4 is: Among them, TP is the number of samples that actually have stones and are predicted to have stones by the model; TN is the number of samples that actually do not have stones and are predicted to have stones by the model; FP is the number of samples that actually do not have stones but are predicted to have stones by the model; FN is the number of samples that actually have stones but are predicted to have no stones by the model; The calculation formula of sensitivity TPR is: The calculation formula of specificity TNR is:

5. The method for detecting extracorporeal lithotripsy of kidney stones based on ultrasonic images using artificial intelligence according to claim 4, characterized in that: The specific implementation method of step S5 includes the following steps: S5.

1. Collect the stone features output by the kidney stone recognition model of the ultrasound image optimized in step S4, combined with the specific conditions of the patient, including stone size, location, and morphology; S5.

2. Apply SVM to build a classification model to comprehensively calculate the different features of the stones obtained in step S5.

1. In the sample space of stones, the formula for dividing the hyperplane is as follows: w T +b=y i Where w=(w1,w2,......,w d ) is the parameter feature of the stone, b is the distance of the hyperplane, y i Classification of lithotripsy energy for patient selection; S5.

3. The physician selects the parameters of the extracorporeal lithotripsy device based on experience and the model parameters obtained in step S5.2, including shock wave intensity, frequency, and focal point, and plans an extracorporeal lithotripsy plan for kidney stones.

6. An artificial intelligence-based ultrasound image-based extracorporeal lithotripsy detection system for kidney stones, implemented by an artificial intelligence-based ultrasound image-based extracorporeal lithotripsy detection method for kidney stones as described in any one of claims 1 to 5, characterized in that: It includes a data processing unit, an artificial intelligence algorithm unit, a model optimization unit, and a result interaction unit. The data processing unit is connected to the artificial intelligence algorithm unit, the artificial intelligence algorithm unit is connected to the model optimization unit, and the model optimization unit is connected to the result interaction unit.

Images