Catheter Positioning Method, System and Storage Medium for Vascular Intervention

Abstract

The present invention belongs to the technical field of vascular intervention, and particularly relates to a catheter positioning method, system and storage medium for vascular intervention. The method includes obtaining a plurality of DSA images, taking internal images of blood vessels at preset time intervals and also obtaining the corresponding first position during shooting, transmitting the DSA images, internal images of blood vessels and the first position to a processing module, establishing a three-dimensional model of the blood vessels, obtaining the three-dimensional coordinate points corresponding to the first position in the three-dimensional model of the blood vessels, simulating and obtaining the simulated internal images taken at the positions of the three-dimensional coordinate points in the three-dimensional model, calculating the similarity between the internal images of blood vessels and the simulated internal images, obtaining the simulated internal image with the largest similarity to the internal images of blood vessels, obtaining the three-dimensional coordinate points corresponding to the simulated internal image, obtaining the center line of the current moving area, moving the catheter forward along the center line, and if encountering a special area, changing the moving direction of the catheter in advance. The present invention can improve the accuracy and success rate of vascular intervention surgery.

Description

Technical Field

[0001] The present invention belongs to the technical field of vascular intervention, and particularly relates to a catheter positioning method, system and storage medium for vascular intervention. Background Art

[0002] In vascular intervention surgery, accurately positioning the catheter in the target blood vessel is crucial. For example, the Chinese patent application with the publication number CN118476864A provides a method and system for automatically adjusting three-dimensional medical images from the perspective of a guide wire. This method processes the angiographic images collected by DSA to obtain a three-dimensional blood vessel model, matches and overlays the three-dimensional blood vessel model with the DSA fluoroscopic image to obtain a fused image; uses guide wire recognition technology to locate the tip of the guide wire in the fused image and determines whether the guide wire moves or rotates; if the guide wire moves, automatically adjusts the image position according to the moving position of the guide wire; if the guide wire rotates, automatically adjusts the image angle according to the rotating position of the guide wire; if the guide wire moves and rotates, simultaneously adjusts the image position and angle according to the moving position and rotating position of the guide wire. However, the accuracy of the image is not considered when acquiring the angiographic images collected by DSA. If inaccurate images are obtained, the subsequent generated three-dimensional models will also be inaccurate, affecting subsequent treatments. Another similar prior art is the Chinese patent application with the publication number CN113855239A, which discloses a guide wire navigation system and method in vascular intervention surgery. The system includes: a C-arm X-ray imaging device for generating angiographic images of the blood vessels at the lesion site of the patient and sending the angiographic images to an image processor; a first positioning sensor disposed on the C-arm X-ray imaging device; a second positioning sensor disposed at the end of the guide wire; a third positioning sensor disposed on the flat panel detector of the C-arm X-ray imaging device; a locator for obtaining the first spatial coordinates of the first positioning sensor, the second spatial coordinates of the second positioning sensor, and the third spatial coordinates of the third positioning sensor, and sending the first spatial coordinates, the second spatial coordinates, and the third spatial coordinates to the image processor; and an image processor for completing the navigation of the guide wire. However, the accuracy of the image during transmission is not considered when sending the angiographic image to the image processor. Therefore, the present invention provides a catheter positioning method, system and storage medium for vascular intervention. Summary of the Invention

[0003] The present invention collects and transmits DSA images, images inside blood vessels and corresponding two-dimensional positions, adds verification data at preset time intervals during transmission to ensure the accuracy of the images, establishes a three-dimensional model of the blood vessels, obtains three-dimensional coordinate points corresponding to the two-dimensional positions in the three-dimensional model, simulates and obtains simulated internal images taken at the positions of the three-dimensional coordinate points in the three-dimensional model, accurately reverses the three-dimensional position of the catheter tip based on the similarity between the simulated internal images and the corresponding real internal images of the blood vessels, obtains the center line of the current moving area during the movement of the catheter, and moves the catheter forward along the center line. If a special area is encountered, the moving direction of the catheter is changed in advance to prevent the catheter from mistakenly entering a non-target blood vessel or damaging the blood vessel wall.

[0004] In order to achieve the above-mentioned object of the invention, the present invention provides the following catheter positioning method for vascular intervention, which is implemented by performing the following steps:

[0005] Step S1, obtaining a plurality of DSA images, setting a micro camera at the top of the catheter, after the catheter is inserted into the blood vessel, as the top of the catheter moves in the blood vessel, the micro camera takes pictures of the inside of the blood vessel at preset time intervals to obtain images inside the blood vessel, and records a first position of the top of the catheter in the blood vessel while taking the images inside the blood vessel, and transmits the DSA images, the images inside the blood vessel and the first position to a processing module;

[0006] Step S2, verifying the DSA image, the internal image of the blood vessel and the first position that need to be verified. After successful verification, a three-dimensional model of the blood vessel is established based on the DSA image, the first position and the corresponding internal image of the blood vessel, a three-dimensional coordinate point corresponding to the first position is obtained in the three-dimensional model of the blood vessel, and a simulated internal image taken at the position of the three-dimensional coordinate point is simulated and obtained in the three-dimensional model;

[0007] Step S3, calculating the similarity between the blood vessel internal image and the corresponding simulated internal image, obtaining the simulated internal image having the greatest similarity to the blood vessel internal image, and obtaining the three-dimensional coordinate point corresponding to the simulated internal image, wherein the three-dimensional coordinate point is the second position of the catheter tip in the blood vessel, and the second position is the three-dimensional position of the catheter tip in the three-dimensional model of the blood vessel;

[0008] Step S4: in the process of moving the catheter, the center line of the current moving area is obtained, and the catheter is moved forward along the center line. Based on the internal image of the blood vessel and the DSA image, it is determined whether there is a special area in front of the moving direction of the catheter. If so, the moving direction of the catheter is changed.

[0009] As a preferred technical solution of the present invention, calculating the similarity between the internal image of the blood vessel and the corresponding simulated internal image comprises the following steps:

[0010] Step S31: Obtain the internal image of the blood vessel and one of the simulation internal images, obtain multiple first key regions of the internal image of the blood vessel, obtain multiple second key regions of the simulation internal image, and pair the first key regions and the second key regions based on the position information to generate key region pairs;

[0011] Step S32: For each of the first key regions, obtain the first value of each pixel point in the first key region, then obtain the second value of the pixel points adjacent to the pixel point, and calculate the first difference and the second difference between the first value and the multiple second values;

[0012] Step S33: Obtain the first average value of the multiple first differences, compare the size of the first average value and a preset first threshold. If the first average value is greater than the first threshold, set the first eigenvalue of the pixel point to a first preset value; otherwise, set the first eigenvalue of the pixel point to a second preset value. Combine the first eigenvalues of each pixel in the first key region to generate the first feature combination of the first key region;

[0013] Step S34: Obtain the second average value of the multiple second differences, compare the size of the second average value and a preset second threshold. If the first average value is greater than the second threshold, set the second eigenvalue of the pixel point to a first preset value; otherwise, set the second eigenvalue of the pixel point to a second preset value. Combine the second eigenvalues of each pixel in the first key region to generate the second feature combination of the first key region;

[0014] Step S35: Repeat Step S32 to Step S34 to obtain the first feature combination and the second feature combination of the second key region, and calculate the similarity of the key region pair based on the first feature combination and the second feature combination of the key region pair;

[0015] Step S36: Obtain the similarities of all the key region pairs, synthesize all the similarities, and obtain the overall similarity between the internal image of the blood vessel and the simulation internal image.

[0016] As a preferred technical solution of the present invention, determining the similarity of the key region pair based on the first feature combination and the second feature combination of the key region pair includes the following steps:

[0017] Step S351: Obtain the first quantity of the first preset value included in the first feature combination, and also obtain the second quantity of the first feature value included in the first feature combination. Divide the first quantity by the second quantity to obtain a first ratio, and also calculate the second ratio of the first preset value in the second feature combination.

[0018] Step S352: If both the first ratio and the second ratio are greater than a preset third threshold, classify the key area pair of the combination of the first key area and the corresponding second key area into the first category; otherwise, classify the key area pair into the second category.

[0019] Step S353: Obtain the two first feature combinations of the key area pair, initialize the first difference to zero, and compare each first feature value in the two first feature combinations in sequence. If the two first feature values are the same, the first difference remains unchanged; otherwise, increment the first difference by one until all the first feature values in the first feature combination are compared, obtain the final first difference, and divide the first difference by the second quantity to obtain a first similarity.

[0020] Step S354: Obtain the two second feature combinations of the key area pair, initialize the second difference to zero, and compare each second feature value in the two second feature combinations in sequence. If the two second feature values are the same, the second difference remains unchanged; otherwise, increment the second difference by one until all the second feature values in the second feature combination are compared, obtain the final second difference, and divide the second difference by the second quantity to obtain a second similarity.

[0021] Step S355: Obtain the average value of the first similarity and the second similarity, and use the average value as the similarity of the key area pair.

[0022] As a preferred technical solution of the present invention, all the similarities are synthesized, including the following steps:

[0023] Use a first formula to synthesize and calculate the overall similarity for all the similarities. The first formula is: , where S is the overall similarity, Xi is the similarity of the i-th key area pair, Wi is the weight value of the i-th key area pair, N is the total number of key area pairs. If the key area pair belongs to the first category, set the weight value of the key area pair to the first weight value; if the key area pair belongs to the second category, set the weight value of the key area pair to the second weight value.

[0024] As a preferred technical solution of the present invention, transmitting the DSA image, the internal blood vessel image, and the first position to the processing module includes the following steps:

[0025] Step S11: Obtain the first transmitted DSA image, the first internal blood vessel image, and the first first position, extract specific regions of the DSA image and the internal blood vessel image, encode the specific regions to obtain corresponding first encoded data, then generate corresponding first hash data based on the first encoded data, and obtain a corresponding check code based on the first position;

[0026] Step S12: Send the DSA image, the internal blood vessel region, and their respective corresponding first hash data to the processing module, and send the first position and the corresponding check code to the processing module;

[0027] Step S13: Receive the verification information from the processing module. If there is no error during the transmission, for a predetermined number of the DSA images, the internal blood vessel images, and the first positions to be transmitted next, do not generate the hash data and the check code. If there is an error during the transmission, check the transmission route and determine the cause of the error;

[0028] Step S14: Then repeat steps S11 to S12 for the DSA image, the internal blood vessel image, and the first position to be transmitted next until all the DSA images and all the internal blood vessel images are transmitted.

[0029] As a preferred technical solution of the present invention, verifying the obtained DSA image and internal blood vessel image includes the following steps:

[0030] Step S21: Perform encoding processing on the DSA image and the internal blood vessel image for which the corresponding hash data is obtained to obtain second encoded data, and also generate corresponding second hash data based on the second encoded data. Compare the first hash data with the corresponding second hash data to determine whether they are the same. If they are the same, it indicates that there is no error during the transmission of the DSA image and the internal blood vessel image, and send a verification success message to the acquisition module. Otherwise, send a verification failure message to the acquisition module;

[0031] Step S22: Verify the corresponding first position based on the check code to determine whether the first position is in error. If it is in error, correct the first position based on the check code.

[0032] As a preferred technical solution of the present invention, determining whether there is a special area in front of the moving direction of the catheter based on the internal blood vessel image and the DSA image includes the following steps:

[0033] Step S41: Collect a number of the internal blood vessel images obtained historically, perform special marking on special regions of the internal blood vessel images as learning data, and generate an identification model using a machine learning algorithm based on the learning data. The identification model includes a first unit and a second unit;

[0034] Step S42: Segment the internal blood vessel images to generate a number of sub-images, and use the first unit to identify the key information of each of the sub-images;

[0035] Step S43: Use the second unit to classify the key information, classify different key information into different special categories, calculate an identification error based on the special categories and the special marking, and adjust the parameters of the first unit and the second unit based on the identification error;

[0036] Step S45: Obtain a number of the internal blood vessel images obtained historically as new learning data, train the identification model based on the learning data to obtain the updated identification model, obtain the identification error of the identification model. If the identification error is less than a preset error threshold, stop training; otherwise, repeat this step;

[0037] Step S45: After generating the final identification model, input the newly obtained internal blood vessel images into the identification model to determine whether there is a special region ahead.

[0038] The present invention further provides a catheter positioning system for vascular intervention, including the following modules:

[0039] An acquisition module, configured to acquire a number of DSA images. A micro camera is provided at the tip of the catheter. After the catheter is inserted into the blood vessel, as the tip of the catheter moves in the blood vessel, the micro camera takes pictures of the inside of the blood vessel at preset time intervals to obtain internal blood vessel images, and records the first position of the tip of the catheter in the blood vessel while taking the internal blood vessel images, and transmits the DSA images, the internal blood vessel images, and the first position to the processing module;

[0040] A processing module, configured to verify the DSA images, the internal blood vessel images, and the first position that need to be verified. After successful verification, establish a three-dimensional model of the blood vessel based on the acquired DSA images, the first position, and the corresponding internal blood vessel images, obtain the three-dimensional coordinate points corresponding to the first position in the three-dimensional model of the blood vessel, and simulate and obtain the simulated internal images taken at the position of the three-dimensional coordinate points in the three-dimensional model;

[0041] a positioning module, used for calculating the similarity between the internal image of the blood vessel and the corresponding simulated internal image, obtaining the simulated internal image having the greatest similarity with the internal image of the blood vessel, and obtaining the three-dimensional coordinate point corresponding to the simulated internal image, wherein the three-dimensional coordinate point is the second position of the catheter tip in the blood vessel, and the second position is the three-dimensional position of the catheter tip in the three-dimensional model of the blood vessel;

[0042] The control module is used to obtain the center line of the current moving area during the movement of the catheter, so that the catheter moves forward along the center line, and judge whether there is a special area in front of the moving direction of the catheter based on the internal image of the blood vessel and the DSA image. If so, change the moving direction of the catheter.

[0043] The present invention also provides a storage medium, wherein the storage medium stores program instructions, wherein when the program instructions are executed, the device where the storage medium is located is controlled to execute any one of the above-mentioned catheter positioning methods for vascular intervention.

[0044] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0045] In the present invention, by acquiring DSA images at different angles, a micro-shooting device is also arranged at the top of the catheter to acquire the internal image of the blood vessel, and the first position when the internal image of the blood vessel is captured is acquired at the same time, the DSA image, the internal image of the blood vessel and the first position are transmitted to the processing module, and verification data is added during the transmission process to ensure the correctness of the data transmission, and to ensure the accuracy of the subsequent establishment of the three-dimensional model, and a three-dimensional model of the blood vessel is established based on the DSA image and the internal image of the blood vessel, and a plurality of corresponding three-dimensional coordinate points in the three-dimensional model are acquired based on the first position, and simulated shooting is performed at the plurality of three-dimensional coordinate points in the three-dimensional model to acquire the corresponding simulated internal image, and the corresponding three-dimensional coordinate points are inferred from the simulated internal image with the greatest similarity to the real internal image of the blood vessel, so that the three-dimensional position of the top of the catheter can be acquired more accurately, thereby improving the accuracy and success rate of the interventional surgery, and in the process of moving the catheter, the catheter is moved along the center line of the blood vessel, and it is also determined whether there is a special area in front, and if so, the moving direction of the catheter is changed in advance to avoid the catheter from entering a non-target blood vessel by mistake or damaging the blood vessel wall. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A flowchart of the steps of the catheter positioning method for vascular intervention of the present invention;

[0047] Figure 2 It is a structural diagram of the catheter positioning system for vascular intervention of the present invention. DETAILED DESCRIPTION

[0048] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and 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.

[0049] It can be understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish one element from another. For example, without departing from the scope of this application, the first xx script may be referred to as the second xx script, and similarly, the second xx script may be referred to as the first xx script.

[0050] The present invention provides a catheter positioning method for vascular intervention as Figure 1 shown, which is realized by performing the following steps:

[0051] Step S1: Obtain a plurality of DSA images. A micro camera is set at the tip of the catheter. After the catheter is inserted into the blood vessel, as the tip of the catheter moves in the blood vessel, the micro camera takes pictures of the inside of the blood vessel every preset time interval to obtain images of the inside of the blood vessel. While taking the images of the inside of the blood vessel, the first position of the tip of the catheter in the blood vessel is recorded, and the DSA images, the images of the inside of the blood vessel and the first position are transmitted to the processing module.

[0052] Specifically, in order to obtain an overall image of the blood vessel, a plurality of DSA images are obtained by taking pictures of the target object from different angles by an X-ray imaging device. The DSA images are two-dimensional perspective images that can clearly show the blood vessel. Based on the DSA images taken from different angles, a three-dimensional image of the blood vessel can be obtained using three-dimensional reconstruction technology. Images of the inside of the blood vessel are obtained by taking pictures of the inside of the blood vessel, and at the same time, the first position of the tip of the catheter in the blood vessel when the images of the inside of the blood vessel are taken is recorded. Since the DSA images are two-dimensional images, the first position is the two-dimensional position of the tip of the catheter in the blood vessel. If the catheter is only positioned and navigated based on the two-dimensional position, the difficulty and risk of vascular intervention surgery will increase. Therefore, a three-dimensional model of the blood vessel is generated based on the DSA images and the images of the inside of the blood vessel. After the DSA images and the images of the inside of the blood vessel are obtained, they need to be processed. The DSA images, the images of the inside of the blood vessel and the corresponding first position need to be transmitted to the processing module. Since the DSA images and the images of the inside of the blood vessel are related to the safety of the patient's life, in order to ensure that the data is not incorrect or lost during the transmission process, verification data such as hash data and check codes are added for synchronous transmission. The specific transmission method will be explained in detail later.

[0053] Step S2: Verify the DSA image, the internal blood vessel image, and the first position that need to be verified. After successful verification, establish a three-dimensional model of the blood vessel based on the DSA image, the first position, and the corresponding internal blood vessel image. Obtain the three-dimensional coordinate points corresponding to the first position in the three-dimensional model of the blood vessel, and simulate and obtain the simulated internal image taken at the position of the three-dimensional coordinate points in the three-dimensional model.

[0054] Specifically, during the transmission process, some verification data is added to the DSA image, the internal blood vessel image, and the first position. The verification data refers to the first hash data, the second hash data, and the check code mentioned later. Therefore, verify the DSA image, the internal blood vessel image, and the first position with added verification data. After successful verification, establish a three-dimensional model of the blood vessel based on the obtained DSA image, the first position, and the internal blood vessel image. The three-dimensional model can help us understand the overall situation and characteristics of the blood vessel more comprehensively. Obtain the three-dimensional coordinate points corresponding to the first position in the three-dimensional model. Since the first position is a two-dimensional coordinate, it will correspond to multiple three-dimensional coordinate points after being mapped into the three-dimensional model. In order to more accurately locate the catheter position, simulate and obtain the simulated internal image taken at the position of the three-dimensional coordinate points in the three-dimensional model.

[0055] In the case of unsuccessful verification, check the transmission line to troubleshoot the reasons for data transmission errors.

[0056] Step S3: Calculate the similarity between the internal blood vessel image and the corresponding simulated internal image, obtain the simulated internal image with the highest similarity to the internal blood vessel image, and obtain the three-dimensional coordinate points corresponding to the simulated internal image. The three-dimensional coordinate points are the second position of the catheter tip in the blood vessel. The second position is the three-dimensional position of the catheter tip in the three-dimensional model of the blood vessel.

[0057] Specifically, by calculating the similarity between the internal blood vessel image and the corresponding simulated internal image, the specific method for calculating the similarity will be explained in detail later. Obtain the simulated internal image with the highest similarity to the internal blood vessel image. The three-dimensional coordinate points of the taken simulated internal image are the second position of the catheter tip in the real blood vessel. Since the three-dimensional coordinate points are obtained from the three-dimensional model, the second position is the three-dimensional position in the real blood vessel. Through the above method, the exact position of the catheter tip in the blood vessel can be determined more accurately, thereby improving the accuracy and success rate of the operation.

[0058] Step S4: During the process of moving the catheter, obtain the center line of the current moving area, make the catheter move forward along the center line, and based on the internal blood vessel image and the DSA image, determine whether there is a special area in front of the moving direction of the catheter. If so, change the moving direction of the catheter.

[0059] Specifically, in order to avoid the catheter from entering a non-target blood vessel by mistake or damaging the blood vessel wall, and to reduce the surgical risks and complications, during the process of moving the catheter, the center line of the current moving area is obtained. The moving area refers to the front and rear area of the catheter tip in the blood vessel. The center line of the moving area is obtained, and the catheter is controlled to move forward along the center line. During the movement, it is also determined based on the internal blood vessel image and the DSA image whether there is a special area in the front of the moving direction of the catheter. The special areas include narrow areas, curved areas, bifurcation areas, etc. If there is, it is necessary to change the moving direction of the catheter in advance to prevent the catheter from entering a non-target blood vessel area or damaging the blood vessel wall.

[0060] The present invention obtains DSA images from different angles, and also sets a micro shooting device at the catheter tip to obtain the internal blood vessel image. At the same time, the first position when the internal blood vessel image is taken is obtained. The DSA image, the internal blood vessel image and the first position are transmitted to the processing module. Verification data is added during the transmission process to ensure the correctness of the data transmission and the accuracy of the subsequent establishment of the three-dimensional model. A three-dimensional model of the blood vessel is established based on the DSA image and the internal blood vessel image. Multiple three-dimensional coordinate points corresponding in the three-dimensional model are obtained based on the first position. Simulation shooting is performed at the multiple three-dimensional coordinate points in the three-dimensional model to obtain the corresponding simulated internal image. The three-dimensional coordinate points corresponding to the simulated internal image with the highest similarity to the real internal blood vessel image are deduced reversely, which can more accurately obtain the three-dimensional position of the catheter tip, thereby improving the accuracy and success rate of the interventional surgery. During the process of moving the catheter, the catheter is made to move along the center line of the blood vessel, and it is also determined whether there is a special area in the front. If there is, the moving direction of the catheter is changed in advance to avoid the catheter from entering a non-target blood vessel or damaging the blood vessel wall.

[0061] Furthermore, calculating the similarity between the internal blood vessel image and the corresponding simulated internal image includes the following steps:

[0062] Step S31: Obtain an internal blood vessel image and a simulated internal image, obtain multiple first key areas of the internal blood vessel image, obtain multiple second key areas of the simulated internal image, and pair the first key areas and the second key areas based on the position information to generate key area pairs.

[0063] Specifically, in order to accurately calculate the similarity between the simulated internal image and the internal blood vessel image, the key areas of the above two images are obtained first. For example, in the internal blood vessel area, the central area and the blood vessel wall area of the blood vessel generally have different characteristics, so the key areas can include the central area and the blood vessel wall area of the blood vessel. In order to more accurately divide the internal image, the internal image can also be divided into multiple key areas according to the spatial position. The key areas of the internal blood vessel image are collectively referred to as the first key areas, and the key areas of the simulated internal image are collectively referred to as the second key areas.

[0064] Step S32: For each first key region, obtain the first value of each pixel point in the first key region, then obtain the second values of the pixel points adjacent to the pixel point, and calculate the first difference and the second difference between the first value and the multiple second values.

[0065] Specifically, to calculate the similarity more accurately, obtain the first value of each pixel point in the first key region. The first value can be the brightness value of the pixel point. Also obtain the second values of the pixel points adjacent to the current pixel point. The second value refers to the brightness value of the adjacent pixel point. The first difference is the difference between the first value and the second value, and the second difference is the absolute value of the difference between the first value and the second value. The difference represents the difference in brightness level between the pixel point and the adjacent pixel points. The absolute value of the difference means not considering the positive or negative of the brightness difference, but only caring about the magnitude of the difference. By statistically analyzing the first difference and the second difference, it provides different perspectives to observe and analyze the image. Combining the two can more comprehensively understand the characteristics of the image and improve the accuracy of subsequent accuracy calculations.

[0066] Step S33: Obtain the first average value of the multiple first differences, compare the size of the first average value and the preset first threshold. If the first average value is greater than the first threshold, set the first feature value of the pixel point to the first preset value; otherwise, set the first feature value of the pixel point to the second preset value. Combine the first feature values of each pixel in the first key region to generate the first feature combination of the first key region.

[0067] Specifically, calculate the first average value of the first differences. Take the first average value as the average difference between the pixel point and the multiple adjacent pixel points. Preset the first threshold. Use the first threshold to judge whether the brightness difference is significant. If the difference is large, set the first feature value of the pixel point to the first preset value, such as setting it to 1; otherwise, set the first feature value of the pixel point to the second preset value, such as setting it to 0. Combine the first feature values of all pixel points in the first key region to generate the first feature combination. For example, the first feature combination of a first key region with 9 pixel points may be (1, 0, 1, 0, 1, 0, 1, 0, 1).

[0068] Step S34: Obtain the second average value of the multiple second differences, compare the size of the second average value and the preset second threshold. If the first average value is greater than the second threshold, set the second feature value of the pixel point to the first preset value; otherwise, set the second feature value of the pixel point to the second preset value. Combine the second feature values of each pixel in the first key region to generate the second feature combination of the first key region.

[0069] Specifically, obtain the second feature combination of the first key region through the above method. The second feature combination may be (1, 1, 1, 0, 1, 0, 1, 1, 1).

[0070] Step S35: Repeat steps S32 to S34 to obtain the first feature combination and the second feature combination of the second key region, and calculate the similarity of the key region pair based on the first feature combination and the second feature combination of the key region pair.

[0071] Specifically, repeat the above steps S32 to S34 to obtain the first feature combination and the second feature combination corresponding to the second key region, and also calculate the similarity of the key region pair based on the first feature combination and the second feature combination of the key region pair. The specific method for calculating the similarity will be explained in detail later.

[0072] Step S36: Obtain the similarities of all key region pairs, and synthesize all the similarities to obtain the overall similarity between the internal blood vessel image and the internal simulation image.

[0073] Specifically, since there are multiple groups of key region pairs for a group of internal blood vessel images and internal simulation images, after obtaining the similarities of the key region pairs, it is necessary to synthesize the similarities of the key region pairs to obtain the overall similarity between the internal blood vessel image and the internal simulation image. The specific method for synthesizing the similarities to obtain the overall similarity will be explained in detail later.

[0074] Furthermore, determining the similarity of the key region pair based on the first feature combination and the second feature combination of the key region pair includes the following steps:

[0075] Step S351: Obtain the first quantity of the first preset value included in the first feature combination, and also obtain the second quantity of the first feature value included in the first feature combination. Divide the first quantity by the second quantity to obtain the first ratio, and the second ratio of the first preset value in the second feature combination.

[0076] Specifically, the first quantity represents the number of pixel points with a large difference in luminance values in the first key region, and the second quantity represents the number of all pixel points in the first key region. Use the above method to calculate the first ratio of the number of pixel points with a large difference in luminance values to the total number of pixel points, and also use the above method to calculate the second ratio of the pixel points with a large absolute value of luminance difference to the total number of pixel points, which serves as the data basis for calculating the similarity later.

[0077] Step S352: If both the first ratio and the second ratio are greater than a preset third threshold, classify the key region pair of the combination of the first key region and the corresponding second key region into the first category; otherwise, classify the key region pair into the second category.

[0078] Specifically, if both the first ratio and the second ratio are greater than a preset third threshold, and the third threshold is a standard value set based on historical experimental data, it indicates that both the brightness value difference and the absolute value of the brightness value difference exceed the standard value. The corresponding key region pair is classified into the first category; otherwise, it is classified into the second category. The first category represents that the corresponding key region pair plays a greater role in determining whether two images are similar, and the second category represents that the key region pair plays a less significant role in determining whether two images are similar compared to the first category.

[0079] Step S353: Obtain two first feature combinations of the key region pair, initialize the first difference to zero, and compare each first feature value in the two first feature combinations in sequence. If the two first feature values are the same, the first difference remains unchanged; otherwise, increment the first difference by one until all first feature values in the first feature combination have been compared, obtaining the final first difference. Divide the first difference by the second quantity to obtain the first similarity.

[0080] Specifically, obtain the first difference of the two first feature combinations of the key region pair through the above method. The smaller the first difference, the higher their similarity. Divide the first difference by the total number of pixel points to obtain the first similarity calculated based on the first feature combination. The first similarity is calculated based on the first feature combination, and the first feature combination is calculated based on the brightness value difference. Therefore, the first similarity is the similarity calculated considering the difference in brightness values.

[0081] Step S354: Obtain two second feature combinations of the key region pair, initialize the second difference to zero, and compare each second feature value in the two second feature combinations in sequence. If the two second feature values are the same, the second difference remains unchanged; otherwise, increment the second difference by one until all second feature values in the second feature combination have been compared, obtaining the final second difference. Divide the second difference by the second quantity to obtain the second similarity.

[0082] Specifically, obtain the second difference of the two second feature combinations corresponding to the key region pair through the above method. The larger the second difference, the smaller the similarity between the two. Divide the second difference by the total number of pixel points to obtain the second similarity calculated based on the first feature combination. The second similarity is calculated based on the second feature combination, and the second feature combination is calculated based on the absolute value of the brightness value difference. Therefore, the second similarity is the similarity calculated considering the absolute value of the brightness value difference.

[0083] Step S355: Calculate the average value of the first similarity and the second similarity, and use the average value as the similarity of the key region pair.

[0084] Specifically, obtain the average value of the first similarity and the second similarity, and use the average value as the similarity of the key region pair. The similarity calculated from two different perspectives in the above method is averaged to obtain the final similarity. By comprehensively considering the similarity calculated from two perspectives, the calculated similarity can be made more accurate.

[0085] Furthermore, comprehensively combine all similarities, including the following steps:

[0086] Use the first formula to comprehensively calculate the overall similarity for all similarities. The first formula is: , where S is the overall similarity, Xi is the similarity of the i-th key region pair, Wi is the weight value of the i-th key region pair, N is the total number of key region pairs. If the key region pair belongs to the first category, set the weight value of the key region pair to the first weight value. If the key region pair belongs to the second category, set the weight value of the key region pair to the second weight value.

[0087] Specifically, since the internal blood vessel image and the internal simulation image are composed of multiple pairs of key region pairs, and the key region pairs have been classified before. The first category represents that the corresponding key region pair plays a greater role in judging whether two images are similar. Therefore, set the weight value of the key similar pair belonging to the first category to the first weight value, such as 0.6. The second category represents that the key region pair plays a relatively smaller role in judging whether two images are similar. Therefore, set the weight value of the key similar pair belonging to the second category to the second weight value, such as 0.4. Perform weighted average on all key similar pairs to obtain the overall similarity.

[0088] Furthermore, transmit the DSA image, the internal blood vessel image, and the first position to the processing module, including the following steps:

[0089] Step S11: Obtain the first transmitted DSA image, the first internal blood vessel image, and the first first position, extract the specific regions of the DSA image and the internal blood vessel image, encode the specific regions to obtain the corresponding first encoded data, then generate the corresponding first hash data based on the first encoded data, and obtain the corresponding check code based on the first position;

[0090] Step S12: Send the DSA image, the internal blood vessel region, and their respective first hash data to the processing module, and send the first position and the corresponding check code to the processing module;

[0091] Step S13: Receive the verification information from the processing module. If there is no error during the transmission, for the next predetermined number of transmitted DSA images, internal blood vessel images, and first positions, do not generate hash data and check codes. If there is an error during the transmission, check the transmission route and judge the cause of the error;

[0092] Step S14: Subsequently, steps S11 to S12 are repeatedly executed for the next DSA image, internal blood vessel image, and first position to be transmitted until all DSA images and all internal blood vessel images are transmitted completely.

[0093] Specifically, since frequent verification of the transmitted images will increase the transmission time, but in order to ensure the correctness of the images during transmission, it is considered that after sending a predetermined number of DSA images, internal blood vessel images, and first positions, verification data is added to the above data to be sent next time. The verification data includes first hash data, second hash data, and a check code. In the case of no error in verification, after sending a predetermined number of image data, verification data is added to the newly sent images again. Through the above method, the accuracy of the image data during transmission is ensured, and the transmission time will not be delayed too long.

[0094] Furthermore, verifying the obtained DSA images and internal blood vessel images includes the following steps:

[0095] Step S21: For the DSA images and internal blood vessel images for which corresponding hash data is obtained, perform encoding processing to obtain second encoded data, and also generate corresponding second hash data based on the second encoded data. Compare the first hash data and the corresponding second hash data to determine whether they are the same. If they are the same, it indicates that there is no error in the transmission of the DSA images and internal blood vessel images, and send a verification success message to the acquisition module; otherwise, send a verification failure message to the acquisition module.

[0096] Step S22: Verify the corresponding first position based on the check code to determine whether the first position is in error. If it is in error, correct the first position based on the check code.

[0097] Specifically, since verification data including first hash data, second hash data, and a check code is added when sending certain images, at the receiving end, that is, the processing module, it is necessary to perform verification processing on the image data. Since DSA images and internal blood vessel images are image data, if correction data is used for correction, it will take a lot of time. Therefore, in the case of verification failure, it is only necessary to check and correct the transmission circuit and then re-transmit. The position information is relatively small, and in the case of an error, the check code can be used for correction. The check code can use a check code with a correction function such as Hamming code. Through the above method, the internal blood vessel images and DSA images can be verified quickly and accurately.

[0098] Furthermore, determining whether there is a special area in front of the catheter movement direction based on the internal blood vessel image includes the following steps:

[0099] Step S41: Collect a number of previously acquired internal vascular images, specially mark special regions of the internal vascular images as learning data, and use a machine learning algorithm based on the learning data to generate an identification model. The identification model includes a first unit and a second unit;

[0100] Step S42: Segment the internal vascular image to generate a number of sub-images, and use the first unit to identify each sub-image to obtain key information of each sub-image;

[0101] Step S43: Use the second unit to classify the key information, classify different key information into different special categories, calculate the identification error based on the special category and the special mark. Based on the identification error, adjust the parameters of the first unit and the second unit;

[0102] Step S45: Obtain a number of newly acquired internal vascular images as new learning data, train the identification model based on the learning data to obtain an updated identification model, obtain the identification error of the identification model. If the identification error is less than a preset error threshold, stop training; otherwise, repeat this step;

[0103] Step S45: After generating the final identification model, input the newly acquired internal vascular image into the identification model to determine whether there is a special region ahead.

[0104] Specifically, the first unit is a feature extraction unit, and the second unit is a classification unit. Through the above method, a model that can effectively identify special regions of blood vessels can be quickly trained. Input the newly acquired internal vascular image into the identification model, and the identification model identifies whether there is a special region in the internal vascular image.

[0105] According to another aspect of the embodiments of the present invention, as shown in Figure 2 also provided is a catheter positioning system for vascular intervention, including an acquisition module, a processing module, a positioning module, and a control module, used to implement the catheter positioning method for vascular intervention described above. The specific functions of each module are as follows:

[0106] The acquisition module is used to acquire a number of DSA images. The DSA images are two-dimensional images and can clearly show blood vessels. A micro camera is set at the tip of the catheter. After the catheter is inserted into the blood vessel, as the tip of the catheter moves in the blood vessel, the micro camera takes pictures of the inside of the blood vessel at preset time intervals to obtain internal vascular images. While taking pictures of the internal vascular images, record the first position of the tip of the catheter in the blood vessel, and transmit the DSA images, internal vascular images, and the first position to the processing module;

[0107] A processing module for verifying a DSA image, an internal blood vessel image, and a first position that need to be verified. After successful verification, a three-dimensional model of the blood vessel is established based on the obtained DSA image, the first position, and the corresponding internal blood vessel image. The three-dimensional coordinate point corresponding to the first position is obtained in the three-dimensional model of the blood vessel, and a simulated internal image taken at the position of the three-dimensional coordinate point is obtained in the three-dimensional model.

[0108] A positioning module for calculating the similarity between the internal blood vessel image and the corresponding simulated internal image, obtaining the simulated internal image with the highest similarity to the internal blood vessel image, and obtaining the three-dimensional coordinate point corresponding to the simulated internal image. The three-dimensional coordinate point is the second position of the catheter tip in the blood vessel, and the second position is the three-dimensional position of the catheter tip in the three-dimensional model of the blood vessel.

[0109] A control module for obtaining the center line of the current moving area during the process of moving the catheter, moving the catheter forward along the center line, and determining whether there is a special area in front of the moving direction of the catheter based on the internal blood vessel image and the DSA image. If so, the moving direction of the catheter is changed.

[0110] According to another aspect of the embodiments of the present invention, a storage medium is further provided. The storage medium stores program instructions, and when the program instructions run, the device where the storage medium is located is controlled to execute any one of the above catheter positioning methods for vascular intervention.

[0111] In summary, the catheter positioning method for vascular intervention of the present invention includes obtaining a plurality of DSA images, taking internal blood vessel images at preset time intervals, and also obtaining the corresponding first position at the time of shooting. The DSA images, internal blood vessel images, and the first position are transmitted to the processing module to establish a three-dimensional model of the blood vessel. The three-dimensional coordinate point corresponding to the first position is obtained in the three-dimensional model of the blood vessel, and a simulated internal image taken at the position of the three-dimensional coordinate point is obtained in the three-dimensional model. The similarity between the internal blood vessel image and the corresponding simulated internal image is calculated, the simulated internal image with the highest similarity to the internal blood vessel image is obtained, the three-dimensional coordinate point corresponding to the simulated internal image is obtained, the center line of the current moving area is obtained, the catheter is moved forward along the center line, and if a special area is encountered, the moving direction of the catheter is changed in advance. The present invention can improve the accuracy and success rate of vascular intervention surgery.

[0112] It should be understood that although the steps in the flowcharts of the embodiments of the present invention are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in each embodiment may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or sub-steps or stages of other steps.

[0113] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The above program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0114] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0115] The above-mentioned embodiments only represent several implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent for the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several deformations and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent for the present invention shall be subject to the appended claims.

[0116] The above is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A catheter positioning system for vascular intervention, characterized in that: include: an acquisition module, for acquiring a plurality of DSA images, wherein a micro camera is arranged at the top of the catheter, and after the catheter is inserted into the blood vessel, as the top of the catheter moves in the blood vessel, the micro camera takes pictures of the inside of the blood vessel at preset time intervals to acquire an image of the inside of the blood vessel, and simultaneously records a first position of the top of the catheter in the blood vessel while taking the image of the inside of the blood vessel, and transmits the DSA image, the image of the inside of the blood vessel and the first position to the processing module; a processing module, configured to verify the DSA image, the internal image of the blood vessel and the first position to be verified, and after successful verification, establish a three-dimensional model of the blood vessel based on the DSA image, the first position and the corresponding internal image of the blood vessel, obtain a three-dimensional coordinate point corresponding to the first position in the three-dimensional model of the blood vessel, and simulate in the three-dimensional model to obtain a simulated internal image taken at the position of the three-dimensional coordinate point; a positioning module, used for calculating the similarity between the internal image of the blood vessel and the corresponding simulated internal image, obtaining the simulated internal image having the greatest similarity with the internal image of the blood vessel, and obtaining the three-dimensional coordinate point corresponding to the simulated internal image, wherein the three-dimensional coordinate point is the second position of the catheter tip in the blood vessel, and the second position is the three-dimensional position of the catheter tip in the three-dimensional model of the blood vessel; A control module, used for obtaining the center line of the current moving area during the process of moving the catheter, moving the catheter forward along the center line, and judging whether there is a special area in front of the moving direction of the catheter based on the internal image of the blood vessel and the DSA image, and if so, changing the moving direction of the catheter; The calculating the similarity between the internal image of the blood vessel and the corresponding simulated internal image comprises: Acquire the blood vessel internal image and one of the simulated internal images, acquire multiple first key areas of the blood vessel internal image, acquire multiple second key areas of the simulated internal image, and pair the first key areas with the second key areas based on position information to generate key area pairs; For each of the first key areas, obtaining a first value of each pixel in the first key area, then obtaining a second value of a pixel adjacent to the pixel, and calculating a first difference and a second difference between the first value and a plurality of the second values; Obtaining a first average value of a plurality of the first difference values, comparing the first average value with a preset first threshold, and if the first average value is greater than the first threshold, setting the first characteristic value of the pixel point to a first preset value; otherwise, setting the first characteristic value of the pixel point to a second preset value, and combining the first characteristic values ​​of each pixel in the first key area to generate a first characteristic combination of the first key area; Obtain a second average value of the plurality of second difference values, compare the second average value with a preset second threshold, if the first average value is greater than the second threshold, set the second characteristic value of the pixel point to a first preset value, otherwise set the second characteristic value of the pixel point to the second preset value, and combine the second characteristic value of each pixel in the first key area to generate a second characteristic combination of the first key area; Acquire the first feature combination and the second feature combination of the second key area, and calculate the similarity of the key area pair based on the first feature combination and the second feature combination of the key area pair; The similarities of all the key area pairs are obtained, and all the similarities are integrated to obtain the overall similarity between the internal image of the blood vessel and the simulated internal image.

2. The system according to claim 1, characterized in that Determining the similarity of the key area pair based on the first feature combination and the second feature combination of the key area pair includes: Obtaining a first quantity of the first preset values ​​included in the first feature combination, further obtaining a second quantity of the first feature values ​​included in the first feature combination, dividing the first quantity by the second quantity to obtain a first ratio, and further calculating a second ratio of the first preset values ​​in the second feature combination; If both the first ratio and the second ratio are greater than a preset third threshold, the key area pair consisting of the first key area and the corresponding second key area is classified into the first category; otherwise, the key area pair is classified into the second category; Obtain two first feature combinations of the key area pair, initialize the first difference to zero, and compare each first feature value in the two first feature combinations in turn. If the two first feature values ​​are the same, the first difference remains unchanged; otherwise, the first difference is increased by one until all the first feature values ​​in the first feature combinations are compared to obtain a final first difference, and the first difference is divided by the second number to obtain a first similarity; Obtain two second feature combinations of the key area pair, initialize the second difference to zero, and compare each second feature value in the two second feature combinations in turn. If the two second feature values ​​are the same, the second difference remains unchanged; otherwise, the second difference is increased by one until all the second feature values ​​in the second feature combinations are compared to obtain a final second difference, and the second difference is divided by the second number to obtain a second similarity; An average value of the first similarity and the second similarity is obtained, and the average value is used as the similarity of the key region pair.

3. The system according to claim 1, characterized in that All the similarities are combined, including: All the similarities are combined to calculate the overall similarity using a first formula, where the first formula is: , where S is the overall similarity, Xi is the similarity of the i-th key area pair, Wi is the weight value of the i-th key area pair, and N is the total number of key area pairs. If the key area pair belongs to the first category, the weight value of the key area pair is set to the first weight value. If the key area pair belongs to the second category, the weight value of the key area pair is set to the second weight value.

4. The system according to claim 1, characterized in that Transmitting the DSA image, the intravascular image and the first position to a processing module includes: Obtaining the transmitted first DSA image, the first intravascular image, and the first first position, extracting specific areas of the DSA image and the intravascular image, encoding the specific areas to obtain corresponding first coded data, generating corresponding first hash data based on the first coded data, and obtaining a corresponding check code based on the first position; Sending the DSA image and the intravascular image and the first hash data corresponding to each to the processing module, and sending the first position and the corresponding verification code to the processing module; receiving the verification information of the processing module, and if no error occurs during the transmission process, not generating the hash data and the check code for the predetermined number of the DSA images, the intravascular images and the first position to be transmitted next, and if an error occurs during the transmission process, checking the transmission route and determining the cause of the error; Afterwards, the steps of acquiring the first DSA image, the first intravascular image and the first first position to be transmitted are repeated for the DSA image, the intravascular image and the first position to be transmitted next, extracting specific areas of the DSA image and the intravascular image, encoding the specific areas to obtain corresponding first encoded data, generating corresponding first hash data based on the first encoded data, and acquiring a corresponding verification code based on the first position, to sending the DSA image, the intravascular area and the corresponding first hash data to the processing module, and sending the first position and the corresponding verification code to the processing module, until the transmission of all the DSA images and all the intravascular images is completed.

5. The system according to claim 1, characterized in that Verifying the acquired DSA image and the intravascular image includes: Perform encoding processing on the DSA image and the intravascular image that have obtained corresponding hash data to obtain second encoded data, further generate corresponding second hash data based on the second encoded data, compare the first hash data and the corresponding second hash data to determine whether they are the same, if they are the same, it means that there is no error in the transmission process of the DSA image and the intravascular image, and send a verification success message to the acquisition module, otherwise send a verification failure message to the acquisition module; The corresponding first position is verified based on the check code to determine whether the first position is wrong. If there is an error, the first position is corrected based on the check code.

6. The system according to claim 1, characterized in that Judging whether there is a special area ahead of the moving direction of the catheter based on the internal image of the blood vessel and the DSA image includes: Collecting a number of the intravascular images acquired historically, specially marking special areas of the intravascular images as learning data, and generating a recognition model based on the learning data using a machine learning algorithm, wherein the recognition model includes a first unit and a second unit; Segmenting the blood vessel internal image to generate a plurality of sub-images, and using the first unit to identify the plurality of sub-images to obtain key information of each of the sub-images; Using the second unit to classify the key information, classify different key information into different special categories, calculate recognition errors based on the special categories and the special marks, and adjust parameters of the first unit and the second unit based on the recognition errors; Acquire a number of new historically acquired internal images of the blood vessels as new learning data, train the recognition model based on the learning data to obtain an updated recognition model, obtain a recognition error of the recognition model, and if the recognition error is less than a preset error threshold, stop training, otherwise, repeat this step; After the final recognition model is generated, the newly acquired internal image of the blood vessel is input into the recognition model to determine whether there is a special area in front.

Images