Kidney puncture method and system based on artificial intelligence

Abstract

The invention relates to the technical field of medical equipment, in particular to a kidney puncture method and system based on artificial intelligence, and the system can accurately identify a kidney lesion area and a contour thereof through a high-resolution imaging technology and an advanced image processing algorithm, and can accurately identify a kidney lesion area by analyzing a puncture position coordinate and a sampling position coordinate. The accurate and detailed movement track of the puncture needle is calculated, a corresponding control instruction is generated and sent to puncture equipment, the risk of tissue damage or insufficient sampling caused by inaccurate puncture can be effectively reduced through high-precision positioning, meanwhile, the automation degree of the system is remarkably improved, and the manual operation steps of doctors are reduced; through full-process automation of image acquisition, analysis, control and execution, the workload of medical staff is reduced, and the overall puncture efficiency is also improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical equipment, and more specifically, to a kidney puncture method and system based on artificial intelligence. Background Art

[0002] In clinical medicine, renal puncture sampling is a commonly used diagnostic method, widely used in pathological examination, histological analysis and diagnosis of renal lesions. However, the traditional renal puncture method relies on the doctor's experience, has extremely high technical requirements for the doctor, and is difficult to achieve standardized operation. Due to the lack of accurate target positioning, the puncture doctor can only rely on ultrasound guidance or CT imaging, which may cause inaccurate puncture position problems. The physiological differences between different patients increase the complexity of the puncture operation. Inexperienced doctors may fail to maintain the appropriate angle and depth during the puncture process, and the puncture needle angle and depth are difficult to control, resulting in tissue damage or insufficient sampling; therefore, the purpose of the present invention is to provide an artificial intelligence-based renal puncture system and method thereof, which solves the problems in the prior art by automating image acquisition and analysis and accurately controlling the movement of the puncture needle, and improves the accuracy and safety of renal puncture. Summary of the invention

[0003] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a kidney puncture method and system based on artificial intelligence to solve the problems existing in the above-mentioned background technology.

[0004] The above technical objectives of the present invention are achieved through the following technical solutions: A kidney puncture method based on artificial intelligence, comprising:

[0005] Step 1, obtaining a preliminary sample image of the patient: scanning the patient's kidney area to obtain a preliminary sample image of the patient;

[0006] Step 2: Preliminary analysis of the preliminary sample image to obtain the patient's kidney lesion image: filter the preliminary sample image to obtain a smooth sample image, and then segment the smooth sample image to remove the blank part to obtain the patient's kidney lesion binary image; the filtering process includes Gaussian filtering and bilateral filtering;

[0007] Step 3: Secondary contour feature analysis and processing of the binary image of the patient's kidney lesions: by increasing the image contrast of the patient's kidney lesions image using histogram equalization, so as to correct the kidney contour on the patient's kidney lesions image and obtain a contour image of the lesion kidney;

[0008] Step 4, obtaining a blood vessel area map of the patient's diseased kidney: performing a rectangle analysis on the contour map of the diseased kidney to obtain a blood vessel area map of the patient's diseased kidney;

[0009] Step 5. Calculate the travel distance and movement angle of the puncture needle: According to the vascular region map of the patient's diseased kidney, calculate the travel distance and movement angle of the puncture needle, and plan the movement trajectory of the puncture needle;

[0010] Step 6. Perform the sampling operation on the patient's kidney: According to the calculated results, control the puncture needle to perform puncture sampling on the patient's kidney and then reset it to complete the puncture sampling operation on the patient's kidney.

[0011] Optionally, the process of filtering the preliminary sample image in Step 2 to obtain a smooth sample image is as follows:

[0012] The preliminary sample image obtained in Step 1 is denoted as G(x, y). The collected preliminary sample image is denoised and smoothed by Gaussian filtering and bilateral filtering to obtain a smooth sample image where σ 2 is the standard deviation of the Gaussian kernel.

[0013] Optionally, the process of further segmenting the smooth sample image in Step 2 to remove the blank part and obtain the diseased image of the patient's kidney is as follows:

[0014] Set a threshold T for the smooth sample image G ’ (x, y), and convert the smooth sample image into a binary image, [B(x,y) = {1 if I smooth (x,y) ≥ T, 0 if I smooth (x,y) < T], where the diseased part is labeled as 1 and the blank part is labeled as 0. Use morphological operations to remove the blank part to obtain the processed binary image of the patient's kidney disease where S is the structuring element, ° is the erosion operation, is the dilation operation.

[0015] Optionally, the process of increasing the image contrast of the patient's kidney disease image by using histogram equalization to correct the kidney contour on the patient's kidney disease image to obtain a diseased kidney contour map is as follows:

[0016] C1. Extract the upper part region of the binary image B of the patient's kidney disease open and perform morphological operations: Perform morphological operations on B top (x,y) where S is the structuring element, ° is the erosion operation, is the dilation operation;

[0017] C2. Extract the largest kidney contour region of the lower boundary region in this upper part region: The preliminary kidney contour region C of the lower boundary region max ,

[0018] C3, extract the lower boundary point set of the maximum kidney contour area in the lower boundary area: P down =(x,y),(x,y)∈C max andy=max(y);

[0019] C3. Extract the y value from the lower boundary point set and sort the set: Y = y | (x, y) ∈ P down , sort Y to get the array set Y sorted ;

[0020] C4. Find the median of the lower boundary point set: If N is an odd number, If N is an even number, Where N is the length of Y;

[0021] C5. Correct the y value that is less than the middle value M: Y corrected =M,if,y <M,

[0022] C6. Maximum kidney contour area based on the updated lower boundary area of ​​the corrected lower boundary: C′ max =(x,y new ),|,(x,y)∈C max ,y new =min(y,M);

[0023] C7. Extract the middle area based on the maximum kidney contour area in the updated lower boundary area and perform morphological operations: extract the binary image B of the middle area middle (x,y)=1,y∈[y lower ,y upper ] and (x,y)∈C′ max , 0, otherwise; for the binary image B in the middle area middle The (x,y) binary image is subjected to morphological operations to obtain the processed binary image. Where S is the structural element, ° is the erosion operation, For expansion operation;

[0024] C8. Extract the largest contour area in the middle area, i.e. the contour of the patient's diseased kidney:

[0025]

[0026] Optionally, the rectangle analysis is performed on the contour map of the diseased kidney in step 4 to obtain a blood vessel area map of the patient's diseased kidney. The specific implementation steps are as follows:

[0027] D1. Find the rectangle that can completely enclose the outline of the patient's diseased kidney: (x, y, w, h) = boundingRect(C middle,max ), where (x, y) is the coordinate of the upper left corner of the rectangle, w is the width of the rectangle, and h is the height of the rectangle;

[0028] D2. The calculated rectangular image is the vascular area of ​​the patient's diseased kidney: Vascular_Area = (x ′ ,y ′ )|x≤x ′ <x+w,y≤y ′ <y+h。

[0029] Optionally, the step 5 calculates the travel distance and movement angle of the puncture needle according to the vascular area map of the patient's diseased kidney, and plans the movement trajectory of the puncture needle. The specific implementation process is as follows:

[0030] E1. According to the blood vessel area map in step 4, obtain the target blood vessel area coordinates A = (x i ,y i ), where i is the key position in the target vascular region;

[0031] E2. Select the sampling position coordinate P from the target blood vessel area coordinate A target =(x t ,y t );

[0032] E3. Obtain the initial insertion position coordinates P of the puncture needle start =(x s ,y s );

[0033] E4. Obtain the parameters of the puncture needle used, define the effective length of the puncture needle as L, and the tip radius of the puncture needle as r;

[0034] E5. Calculate the travel distance D of the puncture needle:

[0035] E6. Calculate the moving angle θ of the puncture needle:

[0036] E7. Plan the motion trajectory according to the travel distance D of the puncture needle and the moving angle θ of the puncture needle: P(t) = (x s +(x t -x s )·t,y s +(y t -y s )·t), where t=[0,1], represents the linear motion from the starting point to the target point.

[0037] An artificial intelligence-based renal puncture system for executing the artificial intelligence-based renal puncture method described above comprises:

[0038] Image acquisition module: used for scanning the patient's kidney area, obtaining a preliminary sample image of the patient's diseased kidney area, and transmitting the preliminary sample image of the diseased kidney area;

[0039] Image analysis module: used for filtering the preliminary sample image collected by the image acquisition module to obtain a smooth sample image, segmenting the smooth sample image to remove the blank part, and obtaining a binary image of the patient's kidney lesions; performing secondary contour feature analysis on the binary image of the patient's kidney lesions to correct the kidney contour on the patient's kidney lesions image to obtain a contour map of the kidney lesions; performing rectangle analysis on the contour map of the kidney lesions to obtain a vascular area map of the patient's kidney lesions;

[0040] Control and analysis module: according to the vascular area map of the patient's diseased kidney analyzed and processed by the image analysis module, calculate the travel distance and movement angle of the puncture needle, plan the movement trajectory of the puncture needle, generate corresponding control instructions and transmit them;

[0041] Sampling execution module: executes the puncture sampling operation on the patient's diseased kidney according to the control instructions generated by the control analysis module.

[0042] Optionally, the image acquisition module includes a B-ultrasound machine, which is used to locate and scan the patient's lesion site to generate video images and transmit the images.

[0043] Optionally, the image analysis module includes a computer workstation; the computer workstation is provided with image processing software for processing a large amount of image data and performing complex image analysis operations to achieve filtering, segmentation, and contour analysis.

[0044] Optionally, the sampling execution module includes a robotic arm; a multi-axis high-precision motion controller is arranged on one end of the robotic arm, and the multi-axis high-precision motion controller is used to receive control instructions from the control and analysis module to drive the robotic arm to move; a puncture device is arranged on the other end of the robotic arm; the puncture device includes a puncture needle and an infrared sensor for locating the puncture position.

[0045] In summary, the present invention has the following beneficial effects:

[0046] 1. The system uses high-resolution imaging technology and advanced image processing algorithms to accurately identify the kidney lesion area and its contour. By analyzing the insertion position coordinates and sampling position coordinates, the system calculates the precise and detailed motion trajectory of the puncture needle and generates corresponding control instructions for the puncture device. High-precision positioning can effectively reduce the risk of tissue damage or insufficient sampling caused by inaccurate puncture. At the same time, the system's automation level is significantly improved, reducing the number of manual operation steps for doctors. The full process automation of image acquisition, analysis, control and execution not only reduces the workload of medical staff, but also improves the overall puncture efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a schematic flow chart of the method steps of the present invention;

[0048] Figure 2 It is a flow logic diagram of the system implementation method of the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings. Several embodiments of the present invention are shown in the accompanying drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein.

[0050] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal connection of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. The terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features.

[0051] In the present invention, unless otherwise clearly specified and limited, the first feature being "above" or "below" the second feature may include the first and second features being in direct contact, or may include the first and second features being in contact not directly but through another feature between them. Moreover, the first feature being "above", "above" and "above" the second feature includes the first feature being directly above and obliquely above the second feature, or simply indicates that the first feature is higher in level than the second feature. The first feature being "below", "below" and "below" the second feature includes the first feature being directly below and obliquely below the second feature, or simply indicates that the first feature is lower in level than the second feature. The terms "vertical", "horizontal", "left", "right", "above", "below" and similar expressions are for illustrative purposes only, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0052] The present invention is described in detail below in conjunction with the accompanying drawings and embodiments.

[0053] The present invention provides a method and system for kidney puncture based on artificial intelligence, such as Figure 1 As shown, including:

[0054] Step 1, obtaining a preliminary sample image of the patient: scanning the patient's kidney area to obtain a preliminary sample image of the patient;

[0055] Step 2: Preliminary analysis of the preliminary sample image to obtain the patient's kidney lesion image: filter the preliminary sample image to obtain a smooth sample image, and then segment the smooth sample image to remove the blank part to obtain the patient's kidney lesion binary image; the filtering process includes Gaussian filtering and bilateral filtering;

[0056] Step 3: Secondary contour feature analysis and processing of the binary image of the patient's kidney lesions: by increasing the image contrast of the patient's kidney lesions image using histogram equalization, so as to correct the kidney contour on the patient's kidney lesions image and obtain a contour image of the lesion kidney;

[0057] Step 4, obtaining a blood vessel area map of the patient's diseased kidney: performing a rectangle analysis on the contour map of the diseased kidney to obtain a blood vessel area map of the patient's diseased kidney;

[0058] Step 5: Calculate the travel distance and moving angle of the puncture needle: Calculate the travel distance and moving angle of the puncture needle based on the vascular area map of the patient's diseased kidney, and plan the movement trajectory of the puncture needle;

[0059] Step 6, perform the sampling operation on the patient's kidney: according to the calculated result, control the puncture needle to puncture and sample the patient's kidney and then reset it to complete the puncture sampling operation on the patient's kidney.

[0060] Further, the filtering process of the preliminary sample image in step 2 to obtain a smoothed sample image is implemented as follows:

[0061] The preliminary sample image obtained in step 1 is denoted as G(x, y). The collected preliminary sample image is denoised and smoothed through Gaussian filtering and bilateral filtering to obtain a smoothed sample image. where σ 2 is the standard deviation of the Gaussian kernel.

[0062] Further, the smoothed sample image is segmented in step 2 to remove the blank part and obtain the patient's kidney lesion image. The specific implementation process is as follows:

[0063] After setting a threshold T for the smoothed sample image G ’ (x, y), the smoothed sample image is converted into a binary image [B(x, y) = {1 if I smooth (x, y) ≥ T, 0 if I smooth (x, y) < T], where the lesion part is marked as 1 and the blank part is marked as 0. The blank part is removed using morphological operations to obtain the processed binary image of the patient's kidney lesion. where S is the structuring element, ° is the erosion operation, is the dilation operation.

[0064] Further, in step 3, the image contrast of the patient's kidney lesion image is increased by using histogram equalization to correct the kidney contour on the patient's kidney lesion image, and a lesion kidney contour map is obtained. The specific implementation process is as follows:

[0065] C1. Extract the upper part region of the binary image B of the patient's kidney lesion open and perform morphological operations: Perform morphological operations on B top (x, y) where S is the structuring element, ° is the erosion operation, is the dilation operation;

[0066] C2. Extract the largest kidney contour region of the lower boundary region in this upper part region: The preliminary kidney contour region C of the lower boundary region max ,

[0067] C3. Extract the set of lower boundary points of the largest kidney contour region of the lower boundary region: P down = (x, y), (x, y) ∈ C max and y = max(y);

[0068] C3. Extract the y value from the lower boundary point set and sort the set: Y = y | (x, y) ∈ P down , sort Y to get the array set Y sorted ;

[0069] C4. Find the median of the lower boundary point set: If N is an odd number, If N is an even number, Where N is the length of Y;

[0070] C5. Correct the y values ​​that are less than the middle value M:

[0071] C6. Maximum kidney contour area based on the updated lower boundary area of ​​the corrected lower boundary: C′ max =(x,y new ),|,(x,y)∈C max ,y new =min(y,M);

[0072] C7. Extract the middle area based on the maximum kidney contour area in the updated lower boundary area and perform morphological operations: extract the binary image B of the middle area middle (x,y)=1,y∈[y lower ,y upper ] and (x,y)∈C′ max , 0, otherwise; for the binary image B in the middle area middle The (x,y) binary image is subjected to morphological operations to obtain the processed binary image. Where S is the structural element, For erosion operations, For expansion operation;

[0073] C8. Extract the largest contour area in the middle area, i.e. the contour of the patient's diseased kidney:

[0074]

[0075] Furthermore, the rectangle analysis is performed on the contour map of the diseased kidney in step 4 to obtain the vascular area map of the patient's diseased kidney. The specific implementation steps are as follows:

[0076] D1. Find the rectangle that can completely enclose the outline of the patient's diseased kidney: (x, y, w, h) = boundingRect(C middle,max ), where (x, y) is the coordinate of the upper left corner of the rectangle, w is the width of the rectangle, and h is the height of the rectangle;

[0077] D2. The calculated rectangular image is the vascular area of ​​the patient's diseased kidney: Vascular_Area = (x ′ ,y ′ )|x≤x ′ <x+w,y≤y ′ <y+h。

[0078] Furthermore, in step 5, according to the vascular area map of the patient's diseased kidney, the travel distance and movement angle of the puncture needle are calculated, and the movement trajectory of the puncture needle is planned. The specific implementation process is as follows:

[0079] E1. According to the blood vessel area map in step 4, obtain the target blood vessel area coordinates A = (x i ,y i ), where i is the key position in the target vascular region;

[0080] E2. Select the sampling position coordinate P from the target blood vessel area coordinate A target =(x t ,y t );

[0081] E3. Obtain the initial insertion position coordinates P of the puncture needle start =(x s ,y s );

[0082] E4. Obtain the parameters of the puncture needle used, define the effective length of the puncture needle as L, and the tip radius of the puncture needle as r;

[0083] E5. Calculate the travel distance D of the puncture needle:

[0084] E6. Calculate the moving angle θ of the puncture needle:

[0085] E7. Plan the motion trajectory according to the travel distance D of the puncture needle and the moving angle θ of the puncture needle: P(t) = (x s +(x t -x s )·t,y s +(y t -y s )·t), where t=[0,1], represents the linear motion from the starting point to the target point.

[0086] An artificial intelligence-based renal puncture system for executing the artificial intelligence-based renal puncture method described above comprises:

[0087] Image acquisition module: used for scanning the patient's kidney area, obtaining a preliminary sample image of the patient's diseased kidney area, and transmitting the preliminary sample image of the diseased kidney area;

[0088] Image analysis module: used for filtering the preliminary sample image collected by the image acquisition module to obtain a smooth sample image, segmenting the smooth sample image to remove the blank part, and obtaining a binary image of the patient's kidney lesions; performing secondary contour feature analysis on the binary image of the patient's kidney lesions to correct the kidney contour on the patient's kidney lesions image to obtain a contour map of the kidney lesions; performing rectangle analysis on the contour map of the kidney lesions to obtain a vascular area map of the patient's kidney lesions;

[0089] Control and analysis module: according to the vascular area map of the patient's diseased kidney analyzed and processed by the image analysis module, calculate the travel distance and movement angle of the puncture needle, plan the movement trajectory of the puncture needle, generate corresponding control instructions and transmit them;

[0090] Sampling execution module: executes the puncture sampling operation on the patient's diseased kidney according to the control instructions generated by the control analysis module.

[0091] Optionally, the image acquisition module includes a B-ultrasound machine, which is used to locate and scan the patient's lesion site to generate video images and transmit the images.

[0092] Optionally, the image analysis module includes a computer workstation; the computer workstation is provided with image processing software for processing a large amount of image data and performing complex image analysis operations to achieve filtering, segmentation, and contour analysis.

[0093] Optionally, the sampling execution module includes a robotic arm; a multi-axis high-precision motion controller is arranged on one end of the robotic arm, and the multi-axis high-precision motion controller is used to receive control instructions from the control and analysis module to drive the robotic arm to move; a puncture device is arranged on the other end of the robotic arm; the puncture device includes a puncture needle and an infrared sensor for locating the puncture position.

[0094] In an embodiment, the robotic arm has multi-axis high-precision motion control and can receive instructions transmitted from the image processing system, and then execute the motion trajectory according to the instructions; a puncture device is installed on the robotic arm, and the puncture device can optionally be installed with a B-ultrasound probe. The B-ultrasound probe is equipped with a pressure sensor, which can automatically detect the pressure of the B-ultrasound probe and set a pressure threshold. When the pressure threshold is exceeded, the operation will be automatically stopped and an alarm will be prompted. At the same time, the angle of the puncture needle can be adjusted. After adjustment, the angle can be stably maintained during the operation. At the same time, it is equipped with an infrared sensor to locate the target area for puncture, thereby further improving the accuracy of puncture sampling.

[0095] During the specific implementation process, the image acquisition module scans the patient's kidneys to obtain a preliminary sample image of the diseased area, and the image analysis module uses image processing software to filter the collected preliminary sample image to eliminate noise to obtain a smooth sample image; the blank area is removed by image segmentation technology, and a binary image of the kidney lesion is obtained; the contour feature analysis is performed on the binary image, and the contour of the diseased kidney is corrected to obtain an accurate contour image of the diseased kidney; then, the patient's vascular area map is solved by a geometric analysis method; the control analysis module calculates the travel amount and movement angle of the puncture needle according to the kidney lesion vascular area map generated by the image analysis module, and plans a precise motion trajectory, and the generated control instructions are transmitted to the sampling execution module through wireless methods such as the network to ensure the automatic control of the puncture operation; the sampling execution module controls the high-precision puncture needle to automatically reach the predetermined position according to the instructions generated by the control analysis module, and performs puncture sampling operations, thereby ensuring the accuracy and effectiveness of sampling, reducing the workload of doctors, and ensuring the safety of patients.

[0096] The present invention provides an artificial intelligence-based kidney puncture method and system. The system can accurately identify the kidney lesion area and its contour through high-resolution imaging technology and advanced image processing algorithms, calculate the precise and detailed motion trajectory of the puncture needle by analyzing the insertion position coordinates and the sampling position coordinates, and generate corresponding control instructions to the puncture device. High-precision positioning can effectively reduce the risk of tissue damage or insufficient sampling caused by inaccurate puncture. At the same time, the system's automation level is significantly improved, reducing the number of manual operation steps for doctors. Through the full process automation of image acquisition, analysis, control and execution, not only the workload of medical staff is reduced, but also the overall puncture efficiency is improved.

[0097] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A method for renal puncture based on artificial intelligence, characterized in that: include: Step 1, obtaining a preliminary sample image of the patient: scanning the patient's kidney area to obtain a preliminary sample image of the patient; Step 2: Preliminary analysis of the preliminary sample image to obtain the patient's kidney lesion image: filter the preliminary sample image to obtain a smooth sample image, and then segment the smooth sample image to remove the blank part to obtain the patient's kidney lesion binary image; the filtering process includes Gaussian filtering and bilateral filtering; Step 3: Secondary contour feature analysis and processing of the binary image of the patient's kidney lesions: by increasing the image contrast of the patient's kidney lesions image using histogram equalization, so as to correct the kidney contour on the patient's kidney lesions image and obtain a contour image of the lesion kidney; Step 4, obtaining a blood vessel area map of the patient's diseased kidney: performing a rectangle analysis on the contour map of the diseased kidney to obtain a blood vessel area map of the patient's diseased kidney; Step 5: Calculate the travel distance and moving angle of the puncture needle: Calculate the travel distance and moving angle of the puncture needle based on the vascular area map of the patient's diseased kidney, and plan the movement trajectory of the puncture needle; Step 6, perform the sampling operation on the patient's kidney: according to the calculated result, control the puncture needle to puncture and sample the patient's kidney and then reset it to complete the puncture sampling operation on the patient's kidney.

2. The method for renal puncture based on artificial intelligence according to claim 1, characterized in that: The filtering process of the preliminary sample image in step 2 to obtain a smoothed sample image is specifically implemented as follows: The preliminary sample image obtained in step 1 is represented as G(x, y). The preliminary sample image is subjected to denoising and smoothing by Gaussian filtering and bilateral filtering to obtain a smoothed sample image. where σ 2 is the standard deviation of the Gaussian kernel.

3. The method for renal puncture based on artificial intelligence according to claim 1, characterized in that: The step 2 is to segment the smoothed sample image, remove the blank part, and obtain the patient's kidney lesion image. The specific implementation process is as follows: For the smoothed sample image G ’ After setting the threshold T for (x, y), the smoothed sample image is converted into a binary image, [B(x, y) = {1 if I smooth (x, y) ≥ T, 0 if I smooth (x, y) < T], where the lesion part is labeled as 1 and the blank part is labeled as 0. The blank part is removed using morphological operations to obtain the processed binary image of the patient's kidney lesion where S is the structuring element, ° is the erosion operation, is the dilation operation.

4. The method for renal puncture based on artificial intelligence according to claim 1, characterized in that: In step 3, the image contrast of the patient's renal lesion image is increased by using histogram equalization to correct the renal contour on the patient's renal lesion image to obtain a lesion kidney contour map. The specific implementation process is as follows: C1. Extract the patient's kidney lesion binary image B open The upper part of the area and perform morphological operations: To B top (x,y) performs morphological operations Where S is the structural element, ° is the erosion operation, For expansion operation; C2. Extract the maximum kidney contour area of ​​the lower boundary area in the upper area: the preliminary kidney contour area C of the lower boundary area max , C3, extract the lower boundary point set of the maximum kidney contour area in the lower boundary area: P down =(x,y),(x,y)∈C max andy=max(y); C3. Extract the y value from the lower boundary point set and sort the set: Y = y | (x, y) ∈ P down , sort Y to get the array set Y sorted ; C4. Find the median of the lower boundary point set: If N is an odd number, If N is an even number, Where N is the length of Y; C5. Correct the y value that is less than the middle value M: Y corrected =M,if,y <M, C6. Maximum kidney contour area based on the updated lower boundary area of ​​the corrected lower boundary: C′ max =(x,y new ),|,(x,y)∈C max ,y new =min(y,M); C7. Extract the middle area based on the maximum kidney contour area in the updated lower boundary area and perform morphological operations: extract the binary image B of the middle area middle (x,y)=1,y∈[y lower ,y upper ] and (x,y)∈C′ max , 0, otherwise; for the binary image B in the middle area middle The (x,y) binary image is subjected to morphological operations to obtain the processed binary image. Where S is the structural element, ° is the erosion operation, For expansion operation; C8. Extract the largest contour area in the middle area, i.e. the contour of the patient's diseased kidney:

5. The artificial intelligence-based renal puncture according to claim 1, characterized in that: The specific implementation steps of performing rectangle analysis on the contour map of the diseased kidney in step 4 to obtain the vascular area map of the patient's diseased kidney are as follows: D1. Find the rectangle that can completely enclose the outline of the patient's diseased kidney: (x, y, w, h) = boundingRect(C middle,max ), where (x, y) is the coordinate of the upper left corner of the rectangle, w is the width of the rectangle, and h is the height of the rectangle; D2. The calculated rectangular image is the vascular area of ​​the patient's diseased kidney: Vascular_Area = (x ′ ,y ′ )|x≤x ′ <x+w,y≤y ′ <y+h。 6. The method for renal puncture based on artificial intelligence according to claim 1, characterized in that: In step 5, the travel distance and movement angle of the puncture needle are calculated based on the vascular area map of the patient's diseased kidney, and the movement trajectory of the puncture needle is planned. The specific implementation process is as follows: E1. According to the blood vessel area map in step 4, obtain the target blood vessel area coordinates A = (x i ,y i ), where i is the key position in the target vascular region; E2. Select the sampling position coordinate P from the target blood vessel area coordinate A target =(x t ,y t ); E3. Obtain the initial insertion position coordinates P of the puncture needle start =(x s ,y s ); E4. Obtain the parameters of the puncture needle used, define the effective length of the puncture needle as L, and the tip radius of the puncture needle as r; E5. Calculate the travel distance D of the puncture needle: E6. Calculate the moving angle θ of the puncture needle: E7. Plan the motion trajectory according to the travel distance D of the puncture needle and the moving angle θ of the puncture needle: P(t) = (x s +(x t -x s )·t,y s +(y t -y s )·t), where t=[0,1], represents the linear motion from the starting point to the target point.

7. An artificial intelligence-based renal puncture system for executing the artificial intelligence-based renal puncture method according to any one of claims 1 to 6, characterized in that: include: Image acquisition module: used for scanning the patient's kidney area, obtaining a preliminary sample image of the patient's diseased kidney area, and transmitting the preliminary sample image of the diseased kidney area; Image analysis module: used for filtering the preliminary sample image collected by the image acquisition module to obtain a smooth sample image, segmenting the smooth sample image to remove the blank part, and obtaining a binary image of the patient's kidney lesions; performing secondary contour feature analysis on the binary image of the patient's kidney lesions to correct the kidney contour on the patient's kidney lesions image to obtain a contour map of the kidney lesions; performing rectangle analysis on the contour map of the kidney lesions to obtain a vascular area map of the patient's kidney lesions; Control and analysis module: according to the vascular area map of the patient's diseased kidney analyzed and processed by the image analysis module, calculate the travel distance and movement angle of the puncture needle, plan the movement trajectory of the puncture needle, generate corresponding control instructions and transmit them; Sampling execution module: executes the puncture sampling operation on the patient's diseased kidney according to the control instructions generated by the control analysis module.

8. The artificial intelligence-based renal puncture system according to claim 7, characterized in that: The image acquisition module includes a B-ultrasound machine, which is used to locate and scan the patient's lesion site to generate video images and transmit the images.

9. The artificial intelligence-based renal puncture system according to claim 7, characterized in that: The image analysis module includes a computer workstation; the computer workstation is provided with image processing software for processing a large amount of image data and performing complex image analysis operations to achieve filtering, segmentation, and contour analysis.

10. The artificial intelligence-based renal puncture system according to claim 7, characterized in that: The sampling execution module includes a robotic arm; a multi-axis high-precision motion controller is arranged on one end of the robotic arm, and the multi-axis high-precision motion controller is used to receive control instructions from the control and analysis module to drive the robotic arm to move; a puncture device is arranged on the other end of the robotic arm; the puncture device includes a puncture needle and an infrared sensor for locating the puncture position.

Images