Processing method and system applied to ultrasound contrast image in abdominal drainage operation

Abstract

The invention discloses an ultrasonic contrast image processing method and system applied to an abdominal drainage operation, and relates to the technical field of image processing, the ultrasonic contrast image in the abdominal drainage operation is acquired, and the ultrasonic contrast image is preprocessed by using an image processing technology; the preprocessed ultrasound contrast image is analyzed, a drainage safety area is delimited by combining the focus area position, and image data are obtained; establishing a coordinate system for the obtained image data, and carrying out digital processing on the image data; establishing a guide path prediction model based on historical guide operation data of the medical staff, and correcting the guide path prediction model in combination with the real-time image data; the method comprises the following steps: digitally processing an ultrasound contrast image, establishing a focus function and a safety function, establishing a guide path prediction model according to historical guide operation data of a medical worker, and cooperating with a real-time guide path of a guide tube to realize real-time supervision of operation behaviors of the medical worker.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method and system for processing ultrasonic contrast imaging used in abdominal drainage surgery. Background Art

[0002] Abdominal drainage surgery is an operation that uses surgical means to drain the fluid, pus, blood or other abnormal fluids in the abdominal cavity out of the body. It is often used to treat or prevent intra-abdominal infection, excessive fluid accumulation or postoperative complications. During abdominal drainage surgery, medical staff usually need to cooperate with ultrasound contrast imaging to introduce the drainage tube to achieve the effect of draining the abdominal fluid.

[0003] Existing ultrasound contrast imaging processing is only processing performed on the image itself. For example, a Chinese invention patent (CN116681790A) discloses a "training method for ultrasound contrast imaging image generation model and image generation method", which discloses: obtaining a training data set; wherein the training data set includes grayscale ultrasound sample images, ultrasound contrast real images, and blood flow sample images; using the training data set to train a generative adversarial network to obtain a trained generative adversarial network, and determining the generator in the trained generative adversarial network as an ultrasound contrast imaging image generation model; wherein the generative adversarial network includes a generator and a first discriminator, the generator is used to generate ultrasound contrast imaging prediction images based on grayscale ultrasound sample images and blood flow sample images, and the first discriminator is used to identify the authenticity of ultrasound contrast imaging real images and ultrasound contrast imaging prediction images. In the above scheme, ultrasound contrast imaging is generated using the model obtained after training, which can reduce the cost of acquiring ultrasound contrast imaging;

[0004] In the above-mentioned public scheme, only the ultrasound contrast imaging image itself is processed, and no coordinated analysis is performed in combination with the operation of the medical staff themselves, which results in the medical staff needing to judge the introduction of the drainage tube based on their own experience, which is prone to accidents due to external factors. Therefore, there is an urgent need for an intelligent and automated ultrasound contrast imaging image processing method in abdominal drainage surgery to solve the above-mentioned technical problems. Summary of the invention

[0005] The object of the present invention is to provide a method and system for processing ultrasound contrast images used in abdominal drainage surgery to solve the problems raised in the prior art.

[0006] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a method for processing ultrasound contrast images in abdominal drainage surgery, acquiring ultrasound contrast images in abdominal drainage surgery, and preprocessing the ultrasound contrast images using image processing technology;

[0007] The pre-processed ultrasound contrast images are analyzed, and the drainage safety area is delineated in combination with the location of the lesion area to obtain the image data;

[0008] When analyzing ultrasound contrast-enhanced images, the location of the lesion area has been determined. At this time, in the preprocessing of ultrasound contrast-enhanced images, dangerous areas such as blood vessel areas and fat areas will be marked. Then other areas that are not marked as dangerous areas are safe drainage areas.

[0009] A coordinate system is established for the obtained image data, and the image data is digitally processed;

[0010] This step establishes a coordinate system in order to digitally represent the lesion area and the drainage safety area. The present invention uses a function expression to represent the boundary of the lesion area and the drainage safety area, so that it can be more intuitively judged by digital calculation whether the drainage tube introduction operation of the medical staff meets expectations, so as to make a timely warning;

[0011] A guidance path prediction model is established based on the historical guidance operation data of medical staff. Each medical staff will leave data during the historical guidance tube introduction process. The historical guidance operation data is extracted, and the guidance path prediction model is established based on the historical guidance operation data. It is completely personalized to match the behavioral habits of each medical staff, and the guidance path prediction model is corrected in combination with real-time imaging data. The reason for the correction here is that in the actual drainage tube introduction operation process, there will be a certain degree of deviation from the predicted path of the guidance path prediction model. The correction is made in order to better predict the path or determine whether there are safety hazards in the current operation, and use the corrected guidance path prediction model to provide early warning of drainage introduction risks.

[0012] According to the above technical solution, in the process of analyzing the pre-processed ultrasound contrast imaging image, the contour line of the lesion area is marked, and the boundary of the drainage introduction safe area is determined to obtain the guided safe area;

[0013] The purpose of confirming the safe drainage area here is to ensure that the medical staff is in the safe drainage area during the entire process of guiding the catheter. Once there is a possibility of breaking through the safe drainage area, it is determined that there is a risk of drainage introduction;

[0014] A plane rectangular coordinate system is established on the obtained image data, and coordinate values ​​are assigned to pixel points of the image data;

[0015] The contour line of the lesion area is expressed as the lesion function F (x) The lesion function is theoretically a closed area, which is represented as the lesion function by digital processing; the boundary of the guiding safety area is represented as the safety function G(x) ; The boundary function of the guidance safety area is determined here to prevent the end of the guidance tube from breaking through the boundary and entering the dangerous area during the actual guidance process. Through digital representation, the guidance process can be analyzed and judged more accurately and intuitively;

[0016] Finally, digital image data is obtained.

[0017] According to the above technical solution, when establishing the guidance path prediction model, the following steps are included:

[0018] Step 1: Conduct medical staff identity verification;

[0019] Step 2: Obtain the historical guidance operation data H of medical staff k , H k =[l 1 ,a 1 ,l 2 ,a 2 ,…,a n-1 ,l n ,a min ], where k represents the kth group of historical guidance operation data of medical staff, l represents the length value of the drainage tube without bending during insertion, a represents the bending angle of the drainage tube in the patient's abdomen during insertion, and a min It represents the minimum value of the bending angle of the drainage tube in the patient's abdomen during insertion, and n represents the number of sections of the drainage tube without bending in the patient's abdomen;

[0020] Step 3, using high-order Markov chain to establish a guidance path prediction model;

[0021] The historical guidance operation data of medical staff are integrated as a training set to train the guidance path prediction model. The guidance path prediction model is used to predict the guidance path based on the minimally invasive incision coordinate points of medical staff to obtain the predicted guidance path function Y (x) .

[0022] According to the above technical solution, during the process of inserting the guide tube into the patient's abdomen, the real-time ultrasound contrast image is preprocessed using image processing technology, the real-time insertion guide path of the guide tube is marked, and the real-time guide path function E is established through digital processing. (x) .

[0023] Specifically, the digital processing process is to assign coordinates to the image of the guide tube in the patient's abdomen. At the same time, combined with the coordinates, the image of the guide tube in the patient's abdomen is converted into a function representation. The purpose is to facilitate the later comparison of the real-time guidance path with the predicted guidance path generated by the guidance path prediction model, and to determine whether the guidance path generated by the medical staff's operating behavior is consistent with the predicted guidance path, so as to make timely adjustments.

[0024] According to the above technical solution, the real-time guidance path function E (x) The guide end point coordinate value is extracted to obtain (x j ,y j ), the predicted guided path function Y (x) Extract the pixel coordinates, get the coordinate set Q, and calculate the coordinate value (x j ,y j ) calculates the distance between any coordinate value of the coordinate set Q and takes the minimum value of the distance as the deviation p;

[0025] The distance between any two coordinate values ​​is calculated here in order to determine the coordinate value (x j ,y j ) and the predicted guided path function Y (x) If the real-time guidance path deviates, the minimum value of the distance is the offset;

[0026] When the deviation p is less than the set threshold q, according to the real-time guidance path function E (x) Correct the guidance path prediction model; because in theory, if the offset p meets the set threshold, that is, within the error range, the final drainage tube introduction can be completed through the operation of medical staff;

[0027] When the deviation p is greater than or equal to the set threshold q, the following steps are performed:

[0028] S1. Determine the coordinate value of the guiding end point (x j ,y j ), the minimum bending angle a of the drainage tube during insertion into the patient's abdomen min As a real-time guidance path function E (X) The next turning point of the simulation guidance path function E is obtained (X) ’ ;

[0029] In this process, the minimum bending angle of the guide tube is a minIt is the last chance for medical staff to make a remedy, because once the minimum value of the guide tube bending angle cannot prevent the guide tube from being directed to the dangerous area, then during the current operation of the medical staff, the end of the guide tube will definitely be led into the dangerous area. Therefore, by analyzing this extreme behavior, it is possible to determine whether there will be danger.

[0030] S2. Solve the simulation guidance path function E (X) ’ With the security function G (x) If there is a solution, a risk warning of drainage import is performed; if there is no solution, step S3 is executed;

[0031] In this process, if a solution exists, it means that the end of the guide tube will at least touch the edge of the dangerous area, because during the digitization process, the boundary of the dangerous area coincides with the boundary of the guided safe area;

[0032] S3. Solve the simulation guidance path function E (X) ’ and the lesion function F (X) If there is a solution, the current guide tube introduction step is safe. If there is no solution, the simulation guide path function E (X) ’ Modify the guidance path prediction model.

[0033] In the above technical solution, two judgments are made to finally confirm whether there is an introduction risk, that is, whether the end of the guide tube after digital processing is consistent with the safety function G (x) between them, and whether it is related to the lesion function F (X) There is a solution between them, because when with the security function G (x) When there is no solution, the end of the guide tube may have been inserted into the lesion area. Therefore, through the above judgment logic, while issuing a risk warning for drainage introduction, it is possible to avoid false alarms during the imaging surgery process.

[0034] An ultrasound contrast imaging image processing system, the system comprising an image processing module, a model building module, a real-time guidance processing module and a risk warning judgment module;

[0035] The image processing module is used to pre-process and digitize the ultrasound contrast imaging images; the model building module establishes a guidance path prediction model based on the historical guidance operation data of medical staff; the real-time guidance processing module is used to digitize the real-time guidance path of the guide tube; the risk warning judgment model is used to make risk judgments on the real-time guidance process of the guide tube and issue warnings.

[0036] According to the above technical solution, the image processing module includes an image acquisition unit, an image preprocessing unit, a digital processing unit and a function establishment unit;

[0037] The image acquisition unit is used to acquire real-time ultrasound contrast images during abdominal drainage surgery; the image preprocessing unit is used to preprocess the real-time ultrasound contrast images acquired by the image acquisition unit; the digital processing unit is used to digitally process the preprocessed ultrasound contrast images; and the function establishment unit is used to analyze the digitally processed images and establish lesion functions and safety functions.

[0038] According to the above technical solution, the model building module includes an identity authentication unit, a historical data retrieval unit, a model training unit and a predicted path output unit;

[0039] The identity authentication unit is used to identify and authenticate the identity information of the user; the historical data retrieval unit is used to retrieve the historical guidance operation data of the medical staff after the identity information is identified; the model training unit is used to train the guidance path prediction model using the historical operation guidance data of the medical staff as a training set; the predicted path output unit uses the trained guidance path prediction model to predict and output the guidance path of the guide tube.

[0040] According to the above technical solution, the real-time guidance processing module includes a guidance path marking unit and a path function establishment unit;

[0041] The guide path marking unit is used to mark the real-time guide path of the guide tube; the path function establishment unit is used to digitally process the marked real-time guide path to obtain a real-time guide path function.

[0042] According to the above technical solution, the risk warning judgment module includes a deviation calculation unit, a simulation guidance path unit, a function solving unit, a model correction unit and a risk warning unit;

[0043] The deviation calculation unit is used to calculate the offset between the real-time guidance path and the predicted guidance path; the simulation guidance path unit is used to take the minimum value of the bending angle in the patient's abdomen during the insertion of the drainage tube as the next bending point of the real-time guidance path to obtain the simulation guidance path; the function solving unit is used to solve the solutions between the simulation guidance path function and the safety function and the lesion function respectively; the model correction unit is used to correct the guidance path prediction model when there is no safety risk but there is a deviation between the real-time guidance path and the predicted guidance path; the risk warning unit is used to provide risk warning reminders to medical staff for their operating behaviors when there is a safety risk.

[0044] Compared with the prior art, the present invention has the following beneficial effects: the present invention establishes a lesion function and a safety function by digitally processing ultrasound contrast images, and establishes a guidance path prediction model based on the historical guidance operation data of medical staff, and cooperates with the real-time guidance path of the guide tube to monitor the operation behavior of medical staff in real time, and in the process of real-time monitoring, the guidance path prediction model can be corrected according to the monitoring data to improve the prediction accuracy of the guidance path prediction model;

[0045] At the same time, when there is a deviation between the real-time guidance path and the predicted guidance path, the threshold is determined and solved. Through the secondary solution, the rigor of risk warning is improved to avoid the impact of erroneous risk warning on the operation of medical staff;

[0046] During the entire abdominal drainage operation, digital processing is used to make the medical staff's operations more precise and the supervision of operational behaviors more rigorous, ensuring the normal progress of the abdominal drainage operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 The figure is a logical relationship diagram of the method for processing ultrasound contrast imaging used in abdominal drainage surgery according to the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] like Figure 1 As shown, the present invention provides a method and system technical solution for processing ultrasound contrast images in abdominal drainage surgery, which acquires ultrasound contrast images in abdominal drainage surgery and pre-processes the ultrasound contrast images using image processing technology;

[0050] The preprocessing of ultrasound contrast-enhanced images includes dynamic sequence noise reduction, motion artifact correction, microbubble signal enhancement, contrast optimization and local enhancement, and ROI segmentation;

[0051] Specifically: Dynamic sequence denoising includes spatial filtering and temporal filtering, that is, using anisotropic diffusion filtering, retaining edges, and performing sliding average or median filtering on consecutive frames;

[0052] Motion artifact correction uses extraction of vascular branches or anatomical landmarks and calculation of the affine transformation matrix;

[0053] Microbubble signal enhancement includes phase difference enhancement and frequency domain high-pass filtering, that is, by subtracting the arterial phase and venous phase images, the dynamic changes of microbubbles are highlighted, the low-frequency background is suppressed, and the high-frequency signals of microbubbles are retained;

[0054] Contrast optimization and local enhancement include adaptive histogram equalization and Retinex enhancement, which improve local contrast, avoid global overexposure, separate illumination and reflection components, and optimize dark area details;

[0055] ROI segmentation is to analyze the lesion area or the effusion area to obtain the lesion area;

[0056] The pre-processed ultrasound contrast images are analyzed, and the drainage safety area is delineated in combination with the location of the lesion area to obtain the image data;

[0057] When analyzing ultrasound contrast-enhanced images, the location of the lesion area has been determined. At this time, in the preprocessing of ultrasound contrast-enhanced images, dangerous areas such as blood vessel areas and fat areas will be marked. Then other areas that are not marked as dangerous areas are safe drainage areas.

[0058] Specifically, in the process of analyzing the pre-processed ultrasound contrast-enhanced images, the contour of the lesion area is marked, and the boundary of the drainage introduction safety area is determined to obtain the guided safety area;

[0059] The purpose of confirming the safe drainage area here is to ensure that the medical staff is in the safe drainage area during the entire process of guiding the catheter. Once there is a possibility of breaking through the safe drainage area, it is determined that there is a risk of drainage introduction;

[0060] A coordinate system is established for the obtained image data, and the image data is digitally processed;

[0061] This step establishes a coordinate system in order to digitally represent the lesion area and the drainage safety area. The present invention uses a function expression to represent the boundary of the lesion area and the drainage safety area, so that it can be more intuitively judged by digital calculation whether the drainage tube introduction operation of the medical staff meets expectations, so as to make a timely warning;

[0062] Specifically, a plane rectangular coordinate system is established on the obtained image data, and coordinate values ​​are assigned to pixel points of the image data;

[0063] The contour line of the lesion area is expressed as the lesion function F (x) The lesion function is theoretically a closed area, which is represented as the lesion function by digital processing; the boundary of the guiding safety area is represented as the safety function G (x); The boundary function of the guidance safety area is determined here to prevent the end of the guidance tube from breaking through the boundary and entering the dangerous area during the actual guidance process. Through digital representation, the guidance process can be analyzed and judged more accurately and intuitively;

[0064] Finally, digital image data is obtained.

[0065] The guidance path prediction model is established based on the historical guidance operation data of medical staff. Each medical staff will leave data during the historical guidance tube introduction process. The historical guidance operation data is extracted and the guidance path prediction model is established based on the historical guidance operation data. It is a completely personalized match for the behavior habits of each medical staff.

[0066] Specifically, when establishing the guidance path prediction model, the following steps are included:

[0067] Step 1: Perform medical staff identification and authentication. Specifically, fingerprint recognition, face recognition or ID card recognition can be used;

[0068] Step 2: Obtain the historical guidance operation data H of medical staff k , H k =[l 1 ,a 1 ,l 2 ,a 2 ,…,a n-1 ,l n ,a min ], where k represents the kth group of historical guidance operation data of medical staff, l represents the length value of the drainage tube without bending during insertion, a represents the bending angle of the drainage tube in the patient's abdomen during insertion, and a min It represents the minimum value of the bending angle of the drainage tube in the patient's abdomen during insertion, and n represents the number of sections of the drainage tube without bending in the patient's abdomen;

[0069] For example: Historical guidance operation data of medical staff H k , H k =[l 1 ,a 1 ,l 2 ,a 2 ,…,a n-1 ,l n ,a min ]=[12,165°,20,170°,5,168°,10,161°]

[0070] Step 3, using high-order Markov chain to establish a guidance path prediction model;

[0071] Specifically, the steps to build a model using a high-order Markov chain are as follows:

[0072] Discretize the patient's abdominal anatomy into a meshed state plane:

[0073] Status S t : represents the coordinates (x, y) and direction θ of the tip of the drainage tube in the plane space;

[0074] State transition: From S t-k ,S t-k+1 ,...,S t-1 →S t ;

[0075] Construct a k-order Markov chain and define the transition probability matrix P(S t ∣S t-k ,...,S t-1 ), whose value is determined as follows:

[0076] Anatomical constraints: Based on the distribution of organs and blood vessels segmented by CT / MRI images, high-risk areas are assigned a low probability of metastasis;

[0077] Mechanical constraints: Tissue hardness affects penetration difficulty and adjusts probability weights;

[0078] Historical path smoothness: angle mutation penalty between consecutive nodes;

[0079] Obtaining a guidance path prediction model;

[0080] The historical guidance operation data of medical staff are integrated as a training set to train the guidance path prediction model. The guidance path prediction model is used to predict the guidance path based on the minimally invasive incision coordinate points of medical staff to obtain the predicted guidance path function Y (x) .

[0081] Specifically, the Viterbi algorithm is used to solve the optimal path:

[0082] Input: obstacle mask for real-time ultrasound / CT image segmentation, starting point S 0 、Target point S end ;

[0083] State initialization: Starting from the starting point, generate all possible k-step historical state sequences;

[0084] Recursive calculation: For each candidate state S t , calculate the cumulative maximum probability path;

[0085] Backtrack to the optimal path: select the path sequence with the highest global probability.

[0086] The guidance path prediction model is corrected in combination with real-time image data. The reason for the correction here is that during the actual drainage tube introduction operation, there will be a certain degree of deviation from the predicted path of the guidance path prediction model. The correction is made in order to better predict the path or determine whether there are safety hazards in the current operation, and use the corrected guidance path prediction model to provide early warning of drainage introduction risks.

[0087] During the process of inserting the guide tube into the patient's abdomen, the real-time ultrasound contrast images are preprocessed using image processing technology to mark the real-time insertion guidance path of the guide tube, and the real-time guidance path function E is established through digital processing. (x) .

[0088] Specifically, the digital processing process is to assign coordinates to the image of the guide tube in the patient's abdomen. At the same time, combined with the coordinates, the image of the guide tube in the patient's abdomen is converted into a function representation. The purpose is to facilitate the later comparison of the real-time guidance path with the predicted guidance path generated by the guidance path prediction model, and to determine whether the guidance path generated by the medical staff's operating behavior is consistent with the predicted guidance path, so as to make timely adjustments.

[0089] The real-time guidance path function E (x) The guide end point coordinate value is extracted to obtain (x j ,y j ), the predicted guided path function Y (x) Extract the pixel coordinates, get the coordinate set Q, and calculate the coordinate value (x j ,y j ) calculates the distance between any coordinate value of the coordinate set Q and takes the minimum value of the distance as the deviation p;

[0090] The distance between any two coordinate values ​​is calculated here in order to determine the coordinate value (x j ,y j ) and the predicted guided path function Y (x) If the real-time guidance path deviates, the minimum value of the distance is the offset;

[0091] When the deviation p is less than the set threshold q, according to the real-time guidance path function E (x) Correct the guidance path prediction model; because in theory, if the offset p meets the set threshold, that is, within the error range, the final drainage tube introduction can be completed through the operation of medical staff;

[0092] When the deviation p is greater than or equal to the set threshold q, the following steps are performed:

[0093] S1. Determine the coordinate value of the guiding end point (x j,y j ), the minimum bending angle a of the drainage tube during insertion into the patient's abdomen min As a real-time guidance path function E (X) The next turning point of the simulation guidance path function E is obtained (X) ’ ;

[0094] In this process, the minimum bending angle of the guide tube is a min It is the last chance for medical staff to make a remedy, because once the minimum value of the guide tube bending angle cannot prevent the guide tube from being directed to the dangerous area, then during the current operation of the medical staff, the end of the guide tube will definitely be led into the dangerous area. Therefore, by analyzing this extreme behavior, it is possible to determine whether there will be danger.

[0095] S2. Solve the simulation guidance path function E (X) ’ With the security function G (x) If there is a solution, a risk warning of drainage import is performed; if there is no solution, step S3 is executed;

[0096] In this process, if a solution exists, it means that the end of the guide tube will at least touch the edge of the dangerous area, because during the digitization process, the boundary of the dangerous area coincides with the boundary of the guided safe area;

[0097] S3. Solve the simulation guidance path function E (X) ’ and the lesion function F (X) If there is a solution, the current guide tube introduction step is safe. If there is no solution, the simulation guide path function E (X) ’ Modify the guidance path prediction model.

[0098] In the above technical solution, two judgments are made to finally confirm whether there is an introduction risk, that is, whether the end of the guide tube after digital processing is consistent with the safety function G (x) between them, and whether it is related to the lesion function F (X) There is a solution between them, because when with the security function G (x) When there is no solution, the end of the guide tube may have been inserted into the lesion area. Therefore, through the above judgment logic, while issuing a risk warning for drainage introduction, it is possible to avoid false alarms during the imaging surgery process.

[0099] Embodiment 1:

[0100] The real-time guidance path function E (x) The guide end point coordinate value is extracted to obtain (x j ,y j), the predicted guided path function Y (x) Extract the pixel coordinates, get the coordinate set Q, and calculate the coordinate value (x j ,y j ) calculates the distance between any coordinate value of the coordinate set Q and takes the minimum value of the distance as the deviation p;

[0101] If the deviation p is greater than or equal to the set threshold q, perform the following steps:

[0102] S1. Determine the coordinate value of the guiding end point (x j ,y j ), the minimum bending angle a of the drainage tube during insertion into the patient's abdomen min = 161° as a function of the real-time guidance path E (X) The next turning point of the simulation guidance path function E is obtained (X) ’ ;

[0103] In this process, the minimum bending angle of the guide tube is a min =161° is the last chance for medical staff to make a remedy, because once the minimum value of the guide tube bending angle cannot prevent the guide tube from being directed to the dangerous area, then during the current operation of the medical staff, the end of the guide tube will definitely be directed to the dangerous area. Therefore, by analyzing this extreme behavior, it is possible to determine whether there will be danger.

[0104] S2. Solve the simulation guidance path function E (X) ’ With the security function G (x) There is a solution A between the two, and a risk warning for diversion import is issued.

[0105] Embodiment 2:

[0106] The real-time guidance path function E (x) The guide end point coordinate value is extracted to obtain (x j ,y j ), the predicted guided path function Y (x) Extract the pixel coordinates, get the coordinate set Q, and calculate the coordinate value (x j ,y j ) calculates the distance between any coordinate value of the coordinate set Q and takes the minimum value of the distance as the deviation p;

[0107] If the deviation p is greater than or equal to the set threshold q, perform the following steps:

[0108] S1. Determine the coordinate value of the guiding end point (x j ,y j ), the minimum bending angle a of the drainage tube during insertion into the patient's abdomenmin = 161° as a function of the real-time guidance path E (X) The next turning point of the simulation guidance path function E is obtained (X) ’ ;

[0109] In this process, the minimum bending angle of the guide tube is a min =161° is the last chance for medical staff to make a remedy, because once the minimum value of the guide tube bending angle cannot prevent the guide tube from being directed to the dangerous area, then during the current operation of the medical staff, the end of the guide tube will definitely be directed to the dangerous area. Therefore, by analyzing this extreme behavior, it is possible to determine whether there will be danger.

[0110] S2. Solve the simulation guidance path function E (X) ’ With the security function G (x) If the solution between , does not exist, then execute step S3;

[0111] S3. Solve the simulation guidance path function E (X) ’ and the lesion function F (X) There is a solution B between the two solutions, and the current guide tube introduction step is safe.

[0112] Embodiment three:

[0113] The real-time guidance path function E (x) The guide end point coordinate value is extracted to obtain (x j ,y j ), the predicted guided path function Y (x) Extract the pixel coordinates, get the coordinate set Q, and calculate the coordinate value (x j ,y j ) calculates the distance between any coordinate value of the coordinate set Q and takes the minimum value of the distance as the deviation p;

[0114] If the deviation p is greater than or equal to the set threshold q, perform the following steps:

[0115] S1. Determine the coordinate value of the guiding end point (x j ,y j ), the minimum bending angle a of the drainage tube during insertion into the patient's abdomen min = 161° as a function of the real-time guidance path E (X) The next turning point of the simulation guidance path function E is obtained (X) ’ ;

[0116] In this process, the minimum bending angle of the guide tube is a min=161° is the last chance for medical staff to make a remedy, because once the minimum value of the guide tube bending angle cannot prevent the guide tube from being directed to the dangerous area, then during the current operation of the medical staff, the end of the guide tube will definitely be directed to the dangerous area. Therefore, by analyzing this extreme behavior, it is possible to determine whether there will be danger.

[0117] S2. Solve the simulation guidance path function E (X) ’ With the security function G (x) If the solution between , does not exist, then execute step S3;

[0118] S3. Solve the simulation guidance path function E (X) ’ and the lesion function F (X) The solution between the two is that there is no solution. According to the simulation guidance path function E (X) ’ Modify the guidance path prediction model.

[0119] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A method for processing ultrasound contrast images used in abdominal drainage surgery, characterized in that: Acquire ultrasound contrast images during abdominal drainage surgery and pre-process the ultrasound contrast images using image processing technology; The pre-processed ultrasound contrast images are analyzed, and the drainage safety area is delineated in combination with the location of the lesion area to obtain the image data; A coordinate system is established for the obtained image data, and the image data is digitally processed; A guidance path prediction model is established based on the historical guidance operation data of medical staff, and the guidance path prediction model is corrected in combination with real-time imaging data. The corrected guidance path prediction model is used to provide early warning of drainage introduction risks.

2. The method for processing ultrasound contrast images in abdominal drainage surgery according to claim 1, characterized in that: During the analysis of the pre-processed ultrasound contrast-enhanced images, the contour of the lesion area is marked, and the boundary of the drainage introduction safe area is determined to obtain the guided safe area; A plane rectangular coordinate system is established on the obtained image data, and coordinate values ​​are assigned to pixel points of the image data; The contour line of the lesion area is expressed as the lesion function F (x) The boundary of the guided safety area is represented by the safety function G (x) ; Finally, digital image data is obtained.

3. The method for processing ultrasound contrast images in abdominal drainage surgery according to claim 2, characterized in that: When establishing the guidance path prediction model, the following steps are included: Step 1: Conduct medical staff identity verification; Step 2: Obtain the historical guidance operation data H of medical staff k , H k =[l1,a1,l2,a2,…,a n-1 ,l n ,a min ], where k represents the kth group of historical guidance operation data of medical staff, l represents the length value of the drainage tube without bending during insertion, a represents the bending angle of the drainage tube in the patient's abdomen during insertion, and a min It represents the minimum value of the bending angle of the drainage tube in the patient's abdomen during insertion, and n represents the number of sections of the drainage tube without bending in the patient's abdomen; Step 3, using high-order Markov chain to establish a guidance path prediction model; The historical guidance operation data of medical staff are integrated as a training set to train the guidance path prediction model. The guidance path prediction model is used to predict the guidance path based on the minimally invasive incision coordinate points of medical staff to obtain the predicted guidance path function Y (x) .

4. The method for processing ultrasound contrast images in abdominal drainage surgery according to claim 3, characterized in that: During the process of inserting the guide tube into the patient's abdomen, the real-time ultrasound contrast images are preprocessed using image processing technology to mark the real-time insertion guidance path of the guide tube, and the real-time guidance path function E is established through digital processing. (x) .

5. The method for processing ultrasound contrast images in abdominal drainage surgery according to claim 4, characterized in that: The real-time guidance path function E (x) The guide end point coordinate value is extracted to obtain (x j ,y j ), the predicted guided path function Y (x) The pixel coordinates are extracted to obtain the coordinate set Q and the coordinate value (x j ,y j ) calculates the distance between any coordinate value of the coordinate set Q and takes the minimum value of the distance as the deviation p; When the deviation p is less than the set threshold q, according to the real-time guidance path function E (x) Modify the guidance path prediction model; When the deviation p is greater than or equal to the set threshold q, the following steps are performed: S1. Determine the coordinate value of the guiding end point (x j ,y j ), the minimum bending angle a of the drainage tube during insertion into the patient's abdomen min As a real-time guidance path function E (X) The next turning point of the simulation guidance path function E is obtained (X) ’ ; S2. Solve the simulation guidance path function E (X) ’ With the security function G (x) If there is a solution, a risk warning of drainage import is performed; if there is no solution, step S3 is executed; S3. Solve the simulation guidance path function E (X) ’ and the lesion function F (X) If there is a solution, the current guide tube introduction step is safe. If there is no solution, the simulation guide path function E (X) ’ Modify the guidance path prediction model.

6. A system for processing ultrasound contrast-enhanced images for executing the ultrasound contrast-enhanced image processing method according to any one of claims 1 to 5, characterized in that: The system includes an image processing module, a model building module, a real-time guidance processing module and a risk warning judgment module; The image processing module is used to pre-process and digitize the ultrasound contrast imaging images; the model building module establishes a guidance path prediction model based on the historical guidance operation data of medical staff; the real-time guidance processing module is used to digitize the real-time guidance path of the guide tube; the risk warning judgment model is used to make risk judgments on the real-time guidance process of the guide tube and issue warnings.

7. The ultrasound contrast imaging processing system according to claim 6, characterized in that: The image processing module includes an image acquisition unit, an image preprocessing unit, a digital processing unit and a function establishment unit; The image acquisition unit is used to acquire real-time ultrasound contrast imaging images in abdominal drainage surgery; the image preprocessing unit is used to preprocess the real-time ultrasound contrast imaging images acquired by the image acquisition unit; and the digital processing unit is used to digitally process the preprocessed ultrasound contrast imaging images; The function establishing unit is used to analyze the digitally processed image and establish a lesion function and a safety function.

8. The ultrasound contrast imaging processing system according to claim 7, characterized in that: The model building module includes an identity authentication unit, a historical data retrieval unit, a model training unit and a predicted path output unit; The identity authentication unit is used to identify and authenticate the identity information of the user; the historical data retrieval unit is used to retrieve the historical guidance operation data of the medical staff after the identity information is identified; the model training unit is used to train the guidance path prediction model using the historical operation guidance data of the medical staff as a training set; the predicted path output unit uses the trained guidance path prediction model to predict and output the guidance path of the guide tube.

9. The ultrasound contrast imaging processing system according to claim 8, characterized in that: The real-time guidance processing module includes a guidance path marking unit and a path function establishment unit; The guide path marking unit is used to mark the real-time guide path of the guide tube; the path function establishment unit is used to digitally process the marked real-time guide path to obtain a real-time guide path function.

10. The ultrasound contrast imaging image processing system according to claim 9, characterized in that: The risk warning judgment module includes a deviation calculation unit, a simulation guidance path unit, a function solving unit, a model correction unit and a risk warning unit; The deviation calculation unit is used to calculate the offset between the real-time guidance path and the predicted guidance path; the simulation guidance path unit is used to take the minimum value of the bending angle in the patient's abdomen during the insertion of the drainage tube as the next turning point of the real-time guidance path to obtain the simulation guidance path; the function solving unit is used to solve the solutions between the simulation guidance path function and the safety function and the lesion function respectively; the model correction unit is used to correct the guidance path prediction model when there is no safety risk but there is a deviation between the real-time guidance path and the predicted guidance path; the risk warning unit is used to provide risk warning reminders to medical staff for their operating behaviors when there is a safety risk.

Images