Method and device for measuring central venous catheter implant length based on lung ct images
By using a region growing algorithm based on lung CT images and three-dimensional coordinate system reconstruction, the complexity and inaccuracy of central venous catheterization measurement are solved, enabling simple, objective, and high-precision measurement of catheter length.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- TIANJIN MEDICAL UNIVERSITY GENERAL HOSPITAL
- Filing Date
- 2026-05-21
- Publication Date
- 2026-06-19
AI Technical Summary
Existing central venous catheterization measurement methods are complex to operate and not very accurate. They require surgeons and nurses to have extensive experience in interpreting electrocardiograms and are easily affected by variations in cardiac anatomical location.
By acquiring CT images of the patient's lungs, a vascular model is reconstructed using a region growing algorithm and a three-dimensional coordinate system. The shortest distance between the proximal and distal ends is calculated to construct an accurate vascular model for measuring the catheter length.
The procedure was simplified, the objectivity and repeatability of the measurement were improved, the reliance on operator experience and the deviation of anatomical variations were reduced, and accurate measurement of the insertion length was achieved.
Smart Images
Figure CN122244043A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of medical image analysis technology, and in particular to a method and apparatus for measuring the length of a central venous catheter implanted based on lung CT images. Background Technology
[0002] Central venous catheterization is a medical technique that involves inserting a thin, flexible catheter into a large vein near the heart. The appropriate length of the catheter to be implanted in the blood vessel has always been a challenge for this type of surgery.
[0003] Current methods for measuring central venous catheterization typically employ the electrocardiogram (ECG) tip localization method. During port-a-catheter implantation, the surgeon and nurses work together on and off the operating table to connect the catheter to a dedicated ECG machine via a wire inside the catheter. The positional relationship between the catheter and the heart is roughly determined by observing changes in the ECG waveform. However, this method requires a certain training period, extensive ECG interpretation experience from both the surgeon and the nurse, and is relatively cumbersome. Furthermore, variations in cardiac anatomical position can lead to deviations in the results, posing a risk to the surgical outcome. Summary of the Invention
[0004] This application provides a method and apparatus for measuring the length of a central venous catheter implanted based on lung CT images, which at least to some extent overcomes the problems of complex operation and low accuracy in determining the length of the catheter implanted due to the limitations of related technologies.
[0005] Other features and advantages of this application will become apparent from the following detailed description, or may be learned in part from practice of this application.
[0006] According to one aspect of this application, a method for measuring the length of a central venous catheter implantation based on lung CT images is provided, comprising: Acquire a CT image of the patient's lungs, the CT image of the lungs including multiple tomographic images, the multiple tomographic images including a proximal tomographic image, a distal tomographic image and multiple intermediate tomographic images, the multiple tomographic images including a target vascular region, the intermediate tomographic images being tomographic images located between the proximal tomographic images and the distal tomographic images; Acquire proximal and distal tomographic images. Use the proximal and distal tomographic images as reference images respectively. Starting from the target blood vessel region in the reference images, and based on grayscale features and dynamic reference values, use a region growing algorithm to determine the target blood vessel region in each tomographic image layer by layer until the target blood vessel region in all tomographic images is determined. A three-dimensional coordinate system is constructed and all tomographic images are placed into the three-dimensional coordinate system. The coordinates of all pixels in the target blood vessel region of different tomographic layers in the three-dimensional coordinate system are extracted to obtain a blood vessel coordinate set. The blood vessel coordinate set includes an upward coordinate set with the proximal tomographic image as the reference layer image and a downward coordinate set with the distal tomographic image as the reference layer image. The intersection of the upward coordinate set and the downward coordinate set is taken to obtain the calibration coordinate set. Based on the calibration coordinate set, an accurate vascular model is constructed; based on the calibration coordinate set, the shortest distance between the target vascular region in the proximal end tomographic image and the target vascular region in the distal end tomographic image is calculated to obtain the central venous catheter implantation length.
[0007] In some embodiments, acquiring the proximal and distal tomographic images involves using the proximal and distal tomographic images as reference layer images, respectively. Starting from the target vessel region in the reference layer image, and using grayscale features and dynamic reference values as criteria, a region growing algorithm is employed to determine the target vessel region in each tomographic image layer by layer until the target vessel region in all tomographic images is determined. This includes: acquiring the proximal and distal tomographic images, using the proximal and distal tomographic images as reference layer images, acquiring the grayscale values of each pixel within the target vessel region in the reference layer image to obtain the grayscale value range of the reference layer; extracting the vessel contour grayscale and boundary grayscale in the reference layer image, wherein the vessel... The contour grayscale is the grayscale of pixels adjacent to the edge of the target blood vessel region within the target blood vessel region, and the boundary grayscale is the grayscale of pixels adjacent to the blood vessel contour but not belonging to the target blood vessel region. Dynamic reference values are calculated in the reference layer image, where the dynamic reference values are the blood vessel reference value and the boundary reference value. The blood vessel reference value is the maximum grayscale difference between adjacent pixels within the target blood vessel region, and the boundary reference value is the minimum grayscale difference between a pixel inside the target blood vessel region and its adjacent pixel outside the target blood vessel region along the boundary line of the target blood vessel region. The target blood vessel region in the reference layer image is mapped to adjacent tomographic images, and a region growing algorithm is used to determine the blood vessel regions in adjacent layers. This process is repeated layer by layer until the target blood vessel region in all tomographic images is determined.
[0008] In some embodiments, mapping the target blood vessel region in the reference layer image to adjacent tomographic images and using a region growing algorithm to determine the adjacent blood vessel region includes: mapping the target blood vessel region in the reference layer image to the same position in the adjacent tomographic images, denoted as a reference contour; obtaining pixels whose pixel values are within the grayscale value range of the reference layer in each pixel point of the reference contour, denoted as adjacent blood vessel regions; obtaining pixels adjacent to the outer side of the adjacent blood vessel region, denoted as points to be analyzed; extracting pixels adjacent to the points to be analyzed in the adjacent blood vessel region, denoted as adjacent pixels to be tested; calculating the difference between the grayscale value of the pixel to be tested and the grayscale value of its corresponding adjacent pixel to be tested, taking its absolute value, denoted as the difference to be analyzed; subtracting the difference to be analyzed from the blood vessel reference value and the boundary reference value respectively, taking the absolute value, to obtain a first difference and a second difference; if the second difference is less than the first difference, then the point to be analyzed is marked as a background point; if the first difference is less than or equal to the second difference, then the point to be analyzed is marked as an adjacent blood vessel region; repeating this process until no new adjacent blood vessel regions appear.
[0009] In some embodiments, constructing an accurate vascular model based on the calibration coordinate set includes: Based on the calibration coordinate set, a tomographic vascular unit is constructed; wherein, the tomographic vascular unit is a cube with the pixel point where each calibration coordinate in the calibration coordinate set is located as the base; adjacent tomographic vascular units are merged to form an overall vascular space; the overall vascular space is smoothed to obtain an accurate vascular model; the step of calculating the shortest distance between the target vascular region in the proximal tomographic image and the target vascular region in the distal tomographic image based on the calibration coordinate set to obtain the central venous catheter implantation length includes: obtaining the center point of the target vascular region in the proximal tomographic image and the distal tomographic image respectively to obtain the proximal point and the distal point; based on the calibration coordinate set, the target vascular region in each intermediate tomographic image is divided into multiple pixel units, the geometric center point of each pixel unit is obtained, and the proximal point, the geometric center point of a pixel unit in each intermediate tomographic image and the distal point are connected in sequence to obtain the proposed implantation line; traversing all pixel units, multiple proposed implantation lines are obtained; the length of the shortest proposed implantation line among the multiple proposed implantation lines is the central venous catheter implantation length.
[0010] In some embodiments, the pixel unit is a cube with dimensions of 1mm*1mm*Q, where Q is the tomographic thickness.
[0011] According to another aspect of this application, a device for measuring the implantation length of a central venous catheter based on lung CT images is also provided, comprising: an image acquisition module for acquiring a patient's lung CT image, the lung CT image including multiple tomographic images, the multiple tomographic images including a proximal tomographic image, a distal tomographic image and multiple intermediate tomographic images, the multiple tomographic images including a target vascular region, the intermediate tomographic images being tomographic images located between the proximal tomographic image and the distal tomographic image; and a vascular region determination module for acquiring the proximal tomographic image and the distal tomographic image from the image acquisition module, using the proximal tomographic image and the distal tomographic image as reference layers respectively, taking the target vascular region in the reference layers as the starting point, and using grayscale features and dynamic reference values as the determination criteria, employing a region growing algorithm. The target vascular region in each tomographic image is determined layer by layer until the target vascular region in all tomographic images is determined. A coordinate determination module is used to construct a three-dimensional coordinate system and place all tomographic images into the three-dimensional coordinate system. The coordinates of all pixels in the target vascular region of different tomographic images in the three-dimensional coordinate system are extracted to obtain a vascular coordinate set. This vascular coordinate set includes an upward coordinate set with the proximal tomographic image as the reference layer and a downward coordinate set with the distal tomographic image as the reference layer. The intersection of the upward and downward coordinate sets is taken to obtain a calibration coordinate set. A length extraction module is used to construct an accurate vascular model based on the calibration coordinate set. Based on the calibration coordinate set, the shortest distance between the target vascular region in the proximal tomographic image and the target vascular region in the distal tomographic image is calculated to obtain the central venous catheter implantation length.
[0012] In some embodiments, the blood vessel region determination module includes: a reference image acquisition module, used to acquire a proximal tomographic image and a distal tomographic image, respectively using the proximal tomographic image and the distal tomographic image as reference layer images, and acquiring the grayscale values of each pixel within the target blood vessel region in the reference layer image to obtain the grayscale value range of the reference layer; a grayscale extraction module, used to extract the grayscale of the blood vessel contour and the grayscale of the boundary in the reference layer image, wherein the grayscale of the blood vessel contour is the grayscale of the pixel adjacent to the edge of the target blood vessel region within the target blood vessel region, and the grayscale of the boundary is the grayscale of the pixel adjacent to the blood vessel contour but not belonging to the target blood vessel region; The reference value calculation module is used to calculate dynamic reference values in the reference layer image. The dynamic reference values are vascular reference values and boundary reference values. The vascular reference value is the maximum gray-level difference between adjacent pixels within the target vascular region, and the boundary reference value is the minimum gray-level difference between a pixel inside the target vascular region and its adjacent pixel outside the target vascular region along the boundary line of the target vascular region. The adjacent layer vascular determination module is used to map the target vascular region in the reference layer image to adjacent tomographic images and determine the adjacent layer vascular region using a region growing algorithm. The expansion module is used to expand layer by layer until the target vascular region in all tomographic images is determined.
[0013] In some embodiments, the extension module includes: a contrast module, configured to map a target vascular region in a reference layer image to the same position in an adjacent tomographic image, denoted as a reference contour; a vascular grayscale acquisition module, configured to acquire pixels whose pixel values are within the grayscale value range of the reference layer in each pixel point within the reference contour, denoted as adjacent vascular regions; an adjacent grayscale acquisition module, configured to acquire pixels adjacent to the outer side of the adjacent vascular regions, denoted as points to be analyzed; and a measured grayscale acquisition module, configured to extract pixels adjacent to the points to be analyzed in the adjacent vascular regions, denoted as adjacent pixels to be measured; The value calculation module is used to calculate the difference between the gray value of the pixel to be tested and the gray value of its corresponding neighboring pixel, and take the absolute value as the difference to be analyzed; the difference analysis module is used to subtract the difference to be analyzed from the blood vessel reference value and the boundary reference value respectively, and take the absolute value to obtain the first difference and the second difference. If the second difference is less than the first difference, the point to be analyzed is marked as a background point. If the first difference is less than or equal to the second difference, the point to be analyzed is marked as a neighboring blood vessel region; the loop execution module is used to execute cyclically until no new neighboring blood vessel regions appear.
[0014] The technical solutions provided in the embodiments of this application include at least the following beneficial effects: By using bidirectional vascular tracking and regional growth analysis of CT tomographic images, combined with the intersection calibration of ascending and descending coordinates, the precise three-dimensional model of the target blood vessel can be automatically reconstructed and the catheter length extracted. This avoids the dependence of the electrocardiogram tip positioning method on the operator's experience and the deviation caused by cardiac anatomical variations. It has the advantages of simple operation, objective measurement, and good repeatability.
[0015] Furthermore, by extracting the grayscale features of the target blood vessels in the reference layer and calculating the blood vessel reference value and boundary reference value, efficient mapping and regional growth expansion of the blood vessel region between adjacent sections were achieved, reducing manual intervention.
[0016] Furthermore, by performing ascending and descending analyses from the distal and proximal ends respectively and taking the intersection of their vascular coordinates as calibration coordinates, the interference of ascending and descending branches of the blood vessels is effectively eliminated, improving the accuracy of main vessel tracking.
[0017] Furthermore, by expanding the pixels corresponding to the calibration coordinates into tomographic vascular units and merging the common contour surfaces of adjacent units, and then constructing an accurate model through smoothing, continuous and smooth three-dimensional vascular morphology can be quickly generated based on individualized CT images. Then, the precise catheter length can be obtained by extracting the centerline, which significantly improves the quantifiability and clinical applicability of preoperative planning for central venous catheterization. Attached Figure Description
[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.
[0019] Figure 1 This document illustrates a flowchart of a method for measuring the length of a central venous catheter implanted based on lung CT images, as described in an embodiment of this application.
[0020] Figure 2 An image of the base layer is shown in an embodiment of this application.
[0021] Figure 3 An image of the target blood vessel region is shown in an embodiment of this application.
[0022] Figure 4 The images of adjacent faults in an embodiment of this application are shown.
[0023] Figure 5 This illustration shows a schematic diagram of mapping the target vascular region of the reference layer onto adjacent tomographic images in an embodiment of this application.
[0024] Figure 6 A schematic diagram of the complete adjacent layer vascular region is shown in an embodiment of this application.
[0025] Figure 7 This diagram illustrates the three-dimensional coordinate system and the location of the reference layer in an embodiment of this application.
[0026] Figure 8 This diagram illustrates the tomographic images of each part in the embodiments of this application in a three-dimensional coordinate system.
[0027] Figure 9 This diagram illustrates the positional relationship of three tomographic vascular units in an embodiment of this application.
[0028] Figure 10 A schematic diagram of the overall vascular space in an embodiment of this application is shown.
[0029] Figure 11 This diagram illustrates the smoothing of the overall vascular space model outline in an embodiment of this application.
[0030] Figure 12 This illustration shows a schematic diagram of a device for measuring the length of a central venous catheter implanted based on lung CT images, according to one embodiment of this application. Detailed Implementation
[0031] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided to make this application more comprehensive and complete, and to fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
[0032] Furthermore, the accompanying drawings are merely illustrative of this application and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.
[0033] It should be noted that the acquisition, storage, use, and processing of data in the technical solution of this application all comply with the relevant provisions of national laws and regulations. The various types of data, such as personal identity data, operational data, and behavioral data related to individuals, customers, and groups, obtained in the embodiments of this application have all been authorized.
[0034] The specific implementation methods of the embodiments of this application will be described in detail below with reference to the accompanying drawings.
[0035] Figure 1 This application illustrates a flowchart of a method for measuring the implantation length of a central venous catheter based on lung CT images, as shown in the embodiment of this application. Figure 1 As shown, the method for measuring the implantation length of a central venous catheter based on lung CT images provided in this application embodiment includes the following steps S101 to S104: S101. Acquire CT images of the patient's lungs. The CT images of the lungs include multiple tomographic images, which include proximal tomographic images, distal tomographic images, and multiple intermediate tomographic images. The multiple tomographic images include the target vascular region. The intermediate tomographic images are tomographic images located between the proximal and distal tomographic images.
[0036] In some embodiments, CT scans can obtain different tomographic images of the lungs, which are labeled in order from top to bottom. ,in, n are non-zero natural numbers and n for TI The number of tomographic images is not limited in this application. For example, a CT scan may yield 60 tomographic images, where 1 ≤ n ≤ 60. The tomographic images are automatically sorted from top to bottom.
[0037] In some embodiments, medical personnel select proximal and distal tomographic images from different tomographic images and mark the target vascular region for which central venous catheterization is required in the proximal and distal tomographic images.
[0038] The near-central tomographic image is recorded as being in ,in, max () is the maximum value operator; the distal end tomographic image is denoted as... For example, medical staff Mark the distal end, in The proximal end is marked, which is actually the cardiac entry point that the catheter needs to reach during central venous catheterization, while the distal end is the point where the catheter is punctured into the vein. The length of the vessel between the proximal and distal ends is actually the catheter length that needs to be measured. The selection of the proximal and distal ends needs to be adjusted according to the specific surgical requirements and the patient's condition. It cannot be confirmed by a machine, so it must be done by medical personnel and cannot be selected by intelligent systems.
[0039] It should be noted that the patient's lung CT images were obtained by scanning the lungs while the patient was holding their breath, as holding one's breath reduces the movement of the lungs.
[0040] S102. Obtain the proximal end tomographic image and the distal end tomographic image. Use the proximal end tomographic image and the distal end tomographic image as the reference layer images respectively. Starting from the target blood vessel region in the reference layer image, and using grayscale features and dynamic reference values as the judgment criteria, use the region growing algorithm to determine the target blood vessel region in each tomographic image layer by layer until the target blood vessel region in all tomographic images is determined.
[0041] In some embodiments, step S102 includes S1021 to S1025: S1021. Obtain the proximal end tomographic image and the distal end tomographic image. Using the proximal end tomographic image and the distal end tomographic image as reference layer images, obtain the gray value of each pixel in the target blood vessel region in the reference layer image.
[0042] In some embodiments, the reference layer image is the proximal or distal tomographic image marked by medical personnel in step S101. The grayscale values of all pixels within the target blood vessel region in the reference layer image are obtained to determine the grayscale value range of the reference layer. For example, the grayscale value range of the reference layer is [241, 255]. It should be noted that, in the embodiments of this application, analysis is performed using both proximal tomographic images and distal tomographic images as reference layers, referred to as uplink analysis and downlink analysis, respectively. Downlink analysis is based on... Perform step analysis for the baseline layer. Each time the adjacent layer vessel points are analyzed, if the current slice is... Then the adjacent fault is taken That is, the analysis proceeds layer by layer from the distal end towards the heart; ascending analysis is based on... Perform step analysis for the baseline layer. Each time the adjacent layer vessel points are analyzed, if the current slice is... Then the adjacent fault is taken That is, the analysis proceeds layer by layer from the proximal end to the distal end.
[0043] S1022. Extract the grayscale of blood vessel contours and boundaries from the reference layer image.
[0044] In some embodiments, the blood vessel contour is the pixel point adjacent to the edge of the target blood vessel region within the target blood vessel region; the boundary is the pixel point adjacent to the blood vessel contour but not belonging to the target blood vessel region. The gray value of each pixel point in the blood vessel contour is extracted and recorded as the blood vessel contour gray value, and the gray value of each pixel point in the boundary is extracted and recorded as the boundary gray value.
[0045] It should be noted that in this application, adjacent pixels are defined as pixels that share only one common edge.
[0046] S1023. Calculate the dynamic reference value in the reference layer image, where the dynamic reference value is the blood vessel reference value and the boundary reference value.
[0047] In some embodiments, the blood vessel reference value is the maximum grayscale difference between adjacent pixels within the target blood vessel region. The boundary reference value is the minimum grayscale difference between a pixel inside the target blood vessel region (blood vessel contour grayscale) and its adjacent pixel outside the target blood vessel region (boundary grayscale) along the boundary line of the target blood vessel region.
[0048] Because of or The analysis process for the reference layer is the same, with the only difference being the selection of adjacent faults. Therefore, this embodiment only uses the reference layer as the reference layer. For the baseline layer (downlink analysis), the analysis process of steps S1021 to S1023 is illustrated with an example: Figure 2 An image of the base layer is shown in an embodiment of this application. Figure 3 The image shows a target blood vessel region in an embodiment of this application, where the dashed line represents the boundary line of the target blood vessel region, and the pixels inside the target blood vessel region are the target blood vessel regions selected by medical personnel. Figure 3 As shown, the gray value of the pixel containing the number 0 is the boundary gray value, and the gray value of the pixel containing the number 1 is the blood vessel contour gray value. If there is an overlap between two pixels, they are considered adjacent. For example, if the gray value of a blood vessel contour is 253, there are two adjacent pixels with boundary gray values of 183 and 188, respectively. The boundary differences can be calculated as 70 and 65, respectively. All boundary differences are calculated. At the same time, the blood vessel contour gray value belongs to the blood vessel gray value. The pixel with the blood vessel contour gray value of 253 mentioned above also has 3 adjacent blood vessel gray values, namely 255, 253, and 255, respectively. The blood vessel differences are calculated as 2, 0, and 2, respectively. The blood vessel differences between all adjacent blood vessel gray values are calculated, and finally, the boundary reference value = 50 and the blood vessel reference value = 14 are extracted.
[0049] S1024. Map the target blood vessel region in the reference layer image to the adjacent tomographic image, and use the region growing algorithm to determine the blood vessel region in the adjacent layer.
[0050] In some embodiments, the target vascular region in the reference layer image is mapped to the same position in the adjacent tomographic image, denoted as the reference contour; pixels with the same gray level as the vascular region within the reference contour are obtained, denoted as the adjacent vascular region. Pixels adjacent to the outer side of the adjacent vascular region are obtained, denoted as the point to be analyzed; pixels adjacent to the point to be analyzed are extracted from the adjacent vascular region, denoted as the adjacent pixel to be tested; the difference between the gray value of the pixel to be tested and the gray value of its corresponding adjacent pixel to be tested is calculated, and its absolute value is denoted as the difference to be analyzed; the difference to be analyzed is subtracted from the vascular reference value and the boundary reference value respectively, and the absolute value is obtained to obtain the first difference and the second difference. If the second difference is less than the first difference, the point to be analyzed is marked as a background point; if the first difference is less than or equal to the second difference, the point to be analyzed is marked as the adjacent vascular region. This process is repeated until no new adjacent vascular regions appear.
[0051] S1025. Expand layer by layer until the target vascular region is identified in all tomographic images.
[0052] In some embodiments, adjacent slices are used as new reference slices, and adjacent vascular regions are used as new target vascular regions. New adjacent slice images are searched again, and steps S1021 to S1024 are repeated. The analysis is carried out in a loop until adjacent vascular regions in all slice images are obtained. The adjacent vascular regions are the target vascular regions distributed in different slice images.
[0053] For example, the base layer is, n =1, will As an adjacent fault, i.e. These are adjacent faults. Figure 4 The images of adjacent faults in an embodiment of this application are shown. Figure 5 This illustration shows a schematic diagram of mapping the target vascular region of the reference layer onto adjacent tomographic images in an embodiment of this application. The dashed lines represent reference contours, for example, in... The statistical analysis yielded a grayscale range of [241, 255] for blood vessels. Therefore, [the following text is incomplete and requires further context: "the range of grayscale values for blood vessels is [241, 255], then..."] Figure 5Pixels within the reference contour and with grayscale values in the range [241, 255] are named neighboring vessel points. Pixels adjacent to these neighboring vessel points are then identified as the points to be analyzed. For example, if a neighboring vessel point has a grayscale value of 245 and the point to be analyzed has a grayscale value of 255, the calculated difference is 10. The boundary reference value and vessel reference value are 50 and 14, respectively. It can be seen that the difference is closer to the vessel reference value because 10 differs from 50 by 40 (the second difference), and 10 differs from 14 by only 4 (the first difference). This indicates that the current point to be analyzed also belongs to the target vessel region and has the same properties as the neighboring vessel points. Therefore, it is also included in the neighboring vessel points. The analysis is performed on each neighboring vessel point and its adjacent points, repeating this process until no new neighboring vessel points appear. Finally, a complete set of neighboring vessel points is obtained, and the region formed by these complete neighboring vessel points is the neighboring vessel region. Figure 6 This diagram illustrates a complete adjacent vascular region in an embodiment of this application. The pixels within the closed area enclosed by the dashed line represent the adjacent vascular region, and the dashed line also represents the boundary line of the new adjacent vascular region. As a new baseline layer, and at the same time As a new adjacent fault, iterative analysis is performed until the analysis yields a result. to Until all target vascular regions are reached.
[0054] S103. Construct a three-dimensional coordinate system, determine the coordinates of pixels within the target vascular region in different slices, and obtain the vascular coordinate set; wherein, the vascular coordinate set includes an upward coordinate set with the proximal slice as the reference layer and a downward coordinate set with the distal slice as the reference layer; take the intersection of the upward coordinate set and the downward coordinate set to obtain the calibration coordinate set.
[0055] In some embodiments, step S103 includes S1031 to S1034: S1031. Establish a three-dimensional coordinate system.
[0056] In some embodiments, a three-dimensional coordinate system is established with the lower left vertex of the reference layer image as the origin, the lower boundary as the X-axis, the left boundary as the Y-axis, and the Z-axis constructed perpendicular to the reference layer image.
[0057] For example, Figure 7 This diagram illustrates the three-dimensional coordinate system and the position of the reference layer in an embodiment of this application. The origin is the lower left vertex of the reference layer, the lower boundary is the X-axis, the left boundary is the Y-axis, and the reference layer is perpendicular to the reference layer. Construct the Z-axis and establish a three-dimensional coordinate system.
[0058] S1032. Place the images of each fault into the three-dimensional coordinate system.
[0059] In some embodiments, the slice thickness of the tomographic image is taken and named the fault slice thickness, denoted by the symbol Q. The numerical value of the fault slice thickness is not limited in this application; for example, Q = 2 mm.
[0060] Figure 8 This diagram illustrates the tomographic images in a three-dimensional coordinate system according to embodiments of this application. The reference tomographic image is placed parallel to the plane containing the X and Y axes, at position Z = Q / 2. Simultaneously, while ensuring that adjacent tomographic images are parallel to the plane containing the X and Y axes, adjacent tomographic images are placed at position Z = Q / 2 + Q × (n-1), with the lower left vertex of each adjacent tomographic image on the Z-axis, the lower boundary of each adjacent tomographic image parallel to the X-axis, and the left boundary of each adjacent tomographic image parallel to the Y-axis.
[0061] For example, The plane parallel to the X and Y axes will As for the point Z = Q / 2, while ensuring Under the premise of being parallel to the plane containing the X and Y axes, Placed at Z = Q / 2 + Q × (n-1), and The bottom left vertex must be on the Z-axis. The lower boundary of is parallel to the X-axis, and the left boundary of must be parallel to the Y-axis.
[0062] S1033. Extract the coordinates of all pixels in the target blood vessel region of different layers in the three-dimensional coordinate system to obtain the blood vessel coordinate set.
[0063] In some embodiments, the geometric center coordinates of each pixel in the target vascular region are used as the coordinates of that pixel. The coordinates of all pixels in the target vascular region at different slices in the three-dimensional coordinate system are extracted to obtain a vascular coordinate set. The vascular coordinate set includes an upward coordinate set with the proximal slice image as the reference slice image and a downward coordinate set with the distal slice image as the reference slice image.
[0064] S1034. Take the intersection of the up-row coordinate set and the down-row coordinate set as the calibration coordinate set.
[0065] It should be noted that during the upward analysis, in addition to the target vessel itself, the upward branches of the target vessel will also be analyzed. Similarly, the downward analysis will also obtain the downward branches of the target vessel. By comparing the intersection of the upward coordinate set and the downward coordinate set, the upward and downward branches can be eliminated, thereby obtaining the main body of the target vessel, i.e., the calibration coordinate set.
[0066] S104. Construct an accurate vascular model based on the calibration coordinate set; calculate the shortest distance between the target vascular region in the proximal tomographic image and the target vascular region in the distal tomographic image based on the calibration coordinate set, and obtain the central venous catheter implantation length.
[0067] In some embodiments, step S104 specifically includes S1041~S1044: S1041. Construct a tomographic vascular unit based on the calibration coordinate set.
[0068] In some embodiments, the pixel at each calibration coordinate in the calibration coordinate set is named a vascular calibration point. The vascular calibration point is then vertically moved upwards and downwards by the tomographic slice thickness Q, respectively, to obtain the unit upper base and unit lower base. That is, each vascular calibration point is expanded into a cube with a height of 2Q, where Q is the slice thickness of the tomographic image. The corresponding vertices of the unit upper base and unit lower base are then connected to form a cube, named a tomographic vascular unit.
[0069] Repeat the above operation for each calibration coordinate to obtain all tomographic vascular units. Name the contour line of each tomographic vascular unit the unit contour.
[0070] S1042. Merge adjacent vascular units to form an overall vascular space.
[0071] In some embodiments, the enclosed space within each tomographic vascular unit is named the unit vascular space. For adjacent unit vascular spaces, if they share a common surface, i.e., their adjacent unit outlines overlap, the unit outlines of that common surface are removed, thus merging the two spaces. In other words, after merging, all unit vascular spaces together form the overall vascular space, and the remaining unit outlines together form the model outline of the overall vascular space. This step fuses discrete tomographic vascular units into a continuous vascular model, eliminating internal overlapping surfaces.
[0072] S1043. Smooth the overall vascular space to obtain an accurate vascular model.
[0073] In this step, the overall vascular space model contour is smoothed. The smoothed model transforms from sharp edges into a smooth curved surface, making the surface more closely resemble the shape of a real blood vessel, ultimately resulting in an accurate vascular model of the target vessel. The smoothing method used is not limited in this application; algorithms such as Laplacian smoothing, surface subdivision, or spline fitting can be used.
[0074] For example, constructing a tomographic vascular unit involves building a 1mm cube upwards from the vascular calibration point as the lower base, and simultaneously building a 1mm cube downwards from the vascular calibration point as the upper base. This ultimately expands the pixels vertically into a 2mm high cube model, which is the tomographic vascular unit. Since the tomographic vascular unit is a cube, it has an internal enclosed space, which is the unit's vascular space, and the cube's outline is the unit's outline. For example, there might be a positional relationship between three tomographic vascular units. Figure 9This diagram illustrates the positional relationship of three tomographic vascular units in an embodiment of this application. Figure 10 This diagram illustrates the overall vascular space in an embodiment of this application. Figure 9 As shown, they are adjacent to each other. If two unit outlines overlap, the overlapping parts of the unit outlines are completely removed, and the overall vascular space is finally obtained as shown. Figure 10 As shown, Figure 10 The remaining line segments are the model outline, and smoothing can make the connection parts of the model outline smoother. Figure 11 This diagram illustrates the smoothing process applied to the overall vascular space model contour in an embodiment of this application. For example... Figure 11 As shown, the model surface is changed from sharp edges to curved surfaces, which better fits the shape of blood vessels. The smoothing technology is an existing processing technology, and its principle will not be explained in this embodiment. Finally, the model outline is smoothed to obtain an accurate model of the target blood vessel. The accurate model reveals the direction and length of the target blood vessel in the human body.
[0075] S1044. Obtain the center point of the target blood vessel region in the proximal and distal tomographic images respectively to obtain the proximal point and the distal point; based on the calibration coordinate set, divide the target blood vessel region in each intermediate tomographic image into multiple pixel units, obtain the geometric center point of each pixel unit, and connect the proximal point, the geometric center point of a pixel unit in each intermediate tomographic image and the distal point in sequence to obtain the proposed implantation line; traverse all pixel units to obtain multiple proposed implantation lines; the length of the shortest proposed implantation line among the multiple proposed implantation lines is the length of the central venous catheter implantation.
[0076] In some embodiments, the pixel unit is a cube with dimensions of 1mm * 1mm * Q, where Q is the slice thickness. For example, the pixel unit is 1mm * 1mm * 2mm. The purpose of this section is to find the shortest distance within the vascular region in the proximal and distal slices, which is the central venous catheter implantation length.
[0077] For example, in this embodiment, the extracted tube length is 13.9 cm.
[0078] In some embodiments, the method for calculating the shortest distance between the target vascular region in the proximal tomographic image and the target vascular region in the distal tomographic image can also be: extracting the centerline of the precise vascular model and obtaining the total length of the centerline, which is the length of the central venous catheter implantation. This application does not limit the method for extracting the centerline. For example, importing the precise model into Mimics software and using the centerline extraction function in Mimics to automatically extract the centerline of the precise model can also obtain the length of the centerline.
[0079] Based on the same inventive concept, this application also provides a device for measuring the length of a central venous catheter implanted based on lung CT images, as shown in the following embodiment. Since the principle by which this device solves the problem is similar to that of the method embodiment described above, the implementation of this device embodiment can refer to the implementation of the method embodiment described above, and repeated details will not be elaborated further.
[0080] Figure 12 This illustration shows a schematic diagram of a device for measuring the length of a central venous catheter implanted based on lung CT images, as described in an embodiment of this application. Figure 12 As shown, the device includes: The image acquisition module 121 is used to acquire CT images of the patient's lungs. The CT images of the lungs include multiple tomographic images, which include a proximal tomographic image, a distal tomographic image, and multiple intermediate tomographic images. The multiple tomographic images include the target vascular region, and the intermediate tomographic images are tomographic images located between the proximal and distal tomographic images.
[0081] The blood vessel region determination module 122 is used to acquire the proximal tomographic image and the distal tomographic image in the image acquisition module 121. The proximal tomographic image and the distal tomographic image are used as reference layer images respectively. The target blood vessel region in the reference layer image is used as the starting point. The gray-scale features and dynamic reference values are used as the judgment criteria. The region growing algorithm is used to determine the target blood vessel region in each tomographic image layer by layer until the target blood vessel region in all tomographic images is determined. The coordinate determination module 123 is used to construct a three-dimensional coordinate system and place all tomographic images into the three-dimensional coordinate system, extract the coordinates of all pixel points in the target blood vessel region of different tomographic layers in the three-dimensional coordinate system, and obtain a blood vessel coordinate set. The blood vessel coordinate set includes an upward coordinate set with the proximal tomographic image as the reference layer image and a downward coordinate set with the distal tomographic image as the reference layer image. The intersection of the upward coordinate set and the downward coordinate set is taken to obtain a calibration coordinate set. The length extraction module 124 is used to construct an accurate vascular model based on the calibration coordinate set; based on the calibration coordinate set, it calculates the shortest distance between the target vascular region in the proximal end tomographic image and the target vascular region in the distal end tomographic image to obtain the central venous catheter implantation length.
[0082] In some embodiments, the blood vessel region determination module 122 includes: The reference image acquisition module is used to acquire proximal tomographic images and distal tomographic images. The proximal tomographic images and distal tomographic images are used as reference layer images respectively. The gray values of each pixel in the target blood vessel region in the reference layer image are acquired to obtain the gray value range of the reference layer. The grayscale extraction module is used to extract the grayscale of blood vessel contours and the grayscale of boundaries in the reference layer image. The grayscale of blood vessel contours is the grayscale of pixels adjacent to the edge of the target blood vessel region within the target blood vessel region, and the grayscale of boundaries is the grayscale of pixels adjacent to the blood vessel contours but not belonging to the target blood vessel region. The reference value calculation module is used to calculate the dynamic reference value in the reference layer image. The dynamic reference value is the blood vessel reference value and the boundary reference value. The blood vessel reference value is the maximum gray level difference between adjacent pixels in the target blood vessel region. The boundary reference value is the minimum gray level difference between the inner pixel of the target blood vessel region and its adjacent outer pixel along the boundary line of the target blood vessel region. The adjacent layer vessel determination module is used to map the target vessel region in the reference layer image to the adjacent tomographic image and uses a region growing algorithm to determine the adjacent layer vessel region. An extension module is used to extend layer by layer until the target vascular region is identified in all tomographic images.
[0083] In some embodiments, the extension module includes: The contrast module is used to map the target vascular region in the reference layer image to the same location in the adjacent tomographic image, which is denoted as the reference contour. The blood vessel grayscale acquisition module is used to acquire pixels whose pixel values are within the grayscale value range of the reference layer in each pixel point within the reference contour, and these are denoted as adjacent blood vessel regions. The adjacent grayscale acquisition module is used to acquire pixels that are adjacent to the outer side of the adjacent blood vessel region, and these pixels are recorded as the points to be analyzed. The grayscale acquisition module is used to extract pixels adjacent to the point to be analyzed in the adjacent blood vessel region, and these pixels are recorded as adjacent pixels to be analyzed. The difference calculation module is used to calculate the difference between the gray value of the pixel to be tested and the gray value of its corresponding neighboring pixel to be tested, and take its absolute value as the difference to be analyzed. The difference analysis module is used to calculate the absolute value of the difference to be analyzed by subtracting the vascular reference value and the boundary reference value respectively, and obtain the first difference and the second difference. If the second difference is less than the first difference, the point to be analyzed is marked as a background point. If the first difference is less than or equal to the second difference, the point to be analyzed is marked as an adjacent vascular region. The loop execution module is used to execute the operation repeatedly until no new adjacent vascular regions appear.
[0084] It should be noted that the image acquisition module 121, blood vessel region determination module 122, coordinate determination module 123, and length extraction module 124 mentioned above correspond to S101 to S104 in the method embodiment. The examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the content disclosed in the above method embodiment. It should be noted that the above modules, as part of the device, can be executed in a computer system such as a set of computer-executable instructions.
[0085] Those skilled in the art will understand that various aspects of this application can be implemented as a system, method, or program product. Therefore, various aspects of this application can be specifically implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or a combination of hardware and software implementations, collectively referred to herein as a "circuit," "module," or "system."
[0086] Furthermore, although the steps of the method in this application are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or a step may be broken down into multiple steps.
[0087] Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, mobile terminal, or network device, etc.) to execute the methods according to the embodiments of this application.
[0088] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the appended claims.
Claims
1. A method for measuring the length of a central venous catheter implanted based on lung CT images, characterized in that, include: Acquire a CT image of the patient's lungs, the CT image of the lungs including multiple tomographic images, the multiple tomographic images including a proximal tomographic image, a distal tomographic image and multiple intermediate tomographic images, the multiple tomographic images including a target vascular region, the intermediate tomographic images being tomographic images located between the proximal tomographic images and the distal tomographic images; The proximal end tomographic image and the distal end tomographic image are acquired. The proximal end tomographic image and the distal end tomographic image are used as reference layer images respectively. The target blood vessel region in the reference layer image is taken as the starting point. Based on grayscale features and dynamic reference values, the target blood vessel region in each tomographic image is determined layer by layer using a region growing algorithm until the target blood vessel region in all tomographic images is determined. A three-dimensional coordinate system is constructed and all tomographic images are placed into the three-dimensional coordinate system. The coordinates of all pixels in the target blood vessel region of different tomographic layers in the three-dimensional coordinate system are extracted to obtain a blood vessel coordinate set. The blood vessel coordinate set includes an upward coordinate set with the proximal tomographic image as the reference layer image and a downward coordinate set with the distal tomographic image as the reference layer image. The intersection of the upward coordinate set and the downward coordinate set is taken to obtain the calibration coordinate set. Based on the calibration coordinate set, an accurate vascular model is constructed; based on the calibration coordinate set, the shortest distance between the target vascular region in the proximal end tomographic image and the target vascular region in the distal end tomographic image is calculated to obtain the central venous catheter implantation length.
2. The method for measuring the length of a central venous catheter implanted based on lung CT images according to claim 1, characterized in that, The acquisition of the proximal tomographic image and the distal tomographic image involves using the proximal tomographic image and the distal tomographic image as reference layers, respectively. Starting from the target vessel region in the reference layer image, and based on grayscale features and dynamic reference values, a region growing algorithm is used to determine the target vessel region in each tomographic image layer by layer until the target vessel region in all tomographic images is determined. This includes: The proximal tomographic image and the distal tomographic image are acquired. The proximal tomographic image and the distal tomographic image are used as reference layer images respectively. The gray values of each pixel in the target blood vessel region in the reference layer image are acquired to obtain the gray value range of the reference layer. Extract the grayscale values of the blood vessel contour and the boundary grayscale values from the reference layer image. The grayscale values of the blood vessel contour are the grayscale values of the pixels adjacent to the edge of the target blood vessel region within the target blood vessel region, and the boundary grayscale values are the grayscale values of the pixels adjacent to the blood vessel contour but not belonging to the target blood vessel region. Calculate dynamic reference values in the reference layer image, wherein the dynamic reference values are blood vessel reference values and boundary reference values. The blood vessel reference value is the maximum gray level difference between adjacent pixels within the target blood vessel region, and the boundary reference value is the minimum gray level difference between a pixel inside the target blood vessel region and its adjacent pixel outside the target blood vessel region along the boundary line of the target blood vessel region. The target blood vessel region in the baseline image is mapped to the adjacent tomographic image, and the adjacent blood vessel region is determined by the region growing algorithm. Expand layer by layer until the target vascular region is identified in all tomographic images.
3. The method for measuring the length of a central venous catheter implanted based on lung CT images according to claim 2, characterized in that, The process of mapping the target vascular region in the reference layer image to adjacent tomographic images, and using a region growing algorithm to determine the vascular region in adjacent layers, includes: The target vascular region in the baseline image is mapped to the same location in the adjacent tomographic image, and this is denoted as the reference contour. The pixels whose pixel values are within the gray value range of the reference layer are obtained from each pixel point in the reference contour and are recorded as the adjacent blood vessel region. Obtain the pixels adjacent to the outer side of the adjacent blood vessel region and record them as the points to be analyzed; In the adjacent blood vessel region, extract the pixels adjacent to the point to be analyzed, and record them as the adjacent pixels to be tested; Calculate the difference between the gray value of the pixel to be tested and the gray value of its corresponding neighboring pixel to be tested, and take its absolute value as the difference to be analyzed. The difference to be analyzed is subtracted from the vascular reference value and the boundary reference value respectively, and the absolute value is taken to obtain the first difference and the second difference. If the second difference is less than the first difference, the point to be analyzed is marked as a background point. If the first difference is less than or equal to the second difference, the point to be analyzed is marked as an adjacent vascular region. Repeat this process until no new adjacent vascular regions appear.
4. The method for measuring the length of a central venous catheter implanted based on lung CT images according to claim 1, characterized in that, The construction of an accurate vascular model based on the calibration coordinate set includes: Based on the calibration coordinate set, a tomographic vascular unit is constructed; wherein, the tomographic vascular unit is a cube with the pixel point where each calibration coordinate in the calibration coordinate set is located as the base; Adjacent sectional vascular units are merged to form an overall vascular space; The entire vascular space is smoothed to obtain an accurate vascular model; The process of calculating the shortest distance between the target vessel region in the proximal tomographic image and the target vessel region in the distal tomographic image based on the calibration coordinate set, to obtain the central venous catheter implantation length, includes: The center points of the target blood vessel region in the proximal and distal tomographic images are obtained respectively to obtain the proximal and distal points. Based on the calibration coordinate set, the target blood vessel region in each intermediate tomographic image is divided into multiple pixel units, and the geometric center point of each pixel unit is obtained. The proximal point, the geometric center point of a pixel unit in each intermediate tomographic image, and the distal point are connected in sequence to obtain the proposed implantation line. All pixel units are traversed to obtain multiple proposed implantation lines. The length of the shortest proposed implantation line among the multiple proposed implantation lines is the length of the central venous catheter implantation.
5. The method for measuring the length of a central venous catheter implanted based on lung CT images according to claim 4, characterized in that, The pixel unit is a cube with dimensions of 1mm*1mm*Q, where Q is the thickness of the tomographic layer.
6. A device for measuring the length of a central venous catheter implanted based on lung CT images, characterized in that, The device includes: The image acquisition module is used to acquire CT images of the patient's lungs. The CT images of the lungs include multiple tomographic images, including a proximal tomographic image, a distal tomographic image, and multiple intermediate tomographic images. The multiple tomographic images include a target vascular region. The intermediate tomographic images are tomographic images located between the proximal tomographic images and the distal tomographic images. The blood vessel region determination module is used to acquire the proximal tomographic image and the distal tomographic image from the image acquisition module. The proximal tomographic image and the distal tomographic image are used as reference layer images respectively. The target blood vessel region in the reference layer image is used as the starting point. Based on grayscale features and dynamic reference values, the target blood vessel region in each tomographic image is determined layer by layer using a region growing algorithm until the target blood vessel region in all tomographic images is determined. The coordinate determination module is used to construct a three-dimensional coordinate system and place all tomographic images into the three-dimensional coordinate system. It extracts the coordinates of all pixels in the target blood vessel region of different tomographic layers in the three-dimensional coordinate system to obtain a blood vessel coordinate set. The blood vessel coordinate set includes an upward coordinate set with the proximal tomographic image as the reference layer image and a downward coordinate set with the distal tomographic image as the reference layer image. The intersection of the upward coordinate set and the downward coordinate set is taken to obtain the calibration coordinate set. The length extraction module is used to construct an accurate vascular model based on the calibration coordinate set; and to calculate the shortest distance between the target vascular region in the proximal tomographic image and the target vascular region in the distal tomographic image based on the calibration coordinate set, thereby obtaining the central venous catheter implantation length.
7. The device for measuring the length of a central venous catheter implantation based on lung CT images according to claim 6, characterized in that, The vascular region determination module includes: The reference image acquisition module is used to acquire proximal tomographic images and distal tomographic images. The proximal tomographic images and distal tomographic images are used as reference layer images respectively. The gray values of each pixel in the target blood vessel region in the reference layer image are acquired to obtain the gray value range of the reference layer. The grayscale extraction module is used to extract the grayscale of blood vessel contours and the grayscale of boundaries in the reference layer image. The grayscale of blood vessel contours is the grayscale of pixels adjacent to the edge of the target blood vessel region within the target blood vessel region, and the grayscale of boundaries is the grayscale of pixels adjacent to the blood vessel contours but not belonging to the target blood vessel region. The reference value calculation module is used to calculate the dynamic reference value in the reference layer image. The dynamic reference value is the blood vessel reference value and the boundary reference value. The blood vessel reference value is the maximum gray level difference between adjacent pixels in the target blood vessel region. The boundary reference value is the minimum gray level difference between a pixel inside the target blood vessel region and its adjacent pixel outside the target blood vessel region along the boundary line of the target blood vessel region. The adjacent layer vessel determination module is used to map the target vessel region in the reference layer image to the adjacent tomographic image and uses a region growing algorithm to determine the adjacent layer vessel region. An extension module is used to extend layer by layer until the target vascular region is identified in all tomographic images.
8. The device for measuring the length of a central venous catheter implanted based on lung CT images according to claim 7, characterized in that, The extension module includes: The contrast module is used to map the target vascular region in the reference layer image to the same location in the adjacent tomographic image, which is denoted as the reference contour. The blood vessel grayscale acquisition module is used to acquire pixels whose pixel values are located within the grayscale value range of the reference layer in each pixel point of the reference contour, and denoted as the adjacent layer blood vessel region. The adjacent grayscale acquisition module is used to acquire pixels that are adjacent to the outer side of the adjacent blood vessel region, and these pixels are recorded as the points to be analyzed. The grayscale acquisition module is used to extract pixels adjacent to the point to be analyzed in the adjacent blood vessel region, and these pixels are recorded as adjacent pixels to be analyzed. The difference calculation module is used to calculate the difference between the gray value of the pixel to be tested and the gray value of its corresponding neighboring pixel to be tested, and take its absolute value as the difference to be analyzed. The difference analysis module is used to subtract the difference to be analyzed from the vascular reference value and the boundary reference value respectively and take the absolute value to obtain the first difference and the second difference. If the second difference is less than the first difference, the point to be analyzed is marked as a background point. If the first difference is less than or equal to the second difference, the point to be analyzed is marked as an adjacent vascular region. The loop execution module is used to execute the operation repeatedly until no new adjacent vascular regions appear.