A method, system and device for calculating retinal vessel diameters from fundus photographs

Abstract

The application relates to a fundus photograph-based method, system and device for calculating the diameters of retinal blood vessels. The method comprises the following steps: acquiring a fundus photograph to be processed; segmenting the fundus photograph to be processed to obtain a fundus photograph with a tree structure and single-branch blood vessels; calculating the normal vector of each branch blood vessel segment pixel by pixel; and through nonlinear interpolation and smoothing processing, the average, maximum and minimum diameters of each branch blood vessel are obtained. The method aims to extract a target region based on an automatic segmentation algorithm, then traverse each branch blood vessel according to the tree structure to calculate the normal vector, and then calculate the parameters of each branch blood vessel through interpolation operation, so that the important clinical significance and potential application value of the method in the prediction and analysis of cardiovascular diseases and systemic diseases are discovered.

Description

Technical Field

[0001] This invention relates to the field of retinal fundus image analysis, and more specifically, to a method, device, system, computer-readable storage medium, and application thereof for calculating the diameter of retinal blood vessels in fundus imaging. Background Technology

[0002] In fundus image analysis, accurate localization of the macular center is crucial for computer-aided diagnostic systems of diabetic retinopathy. Furthermore, changes in retinal vascular morphology have significant clinical implications for cardiovascular and systemic diseases. For example, increased venule diameter is associated with the risk of diabetic retinopathy and stroke; hypertensive retinopathy can lead to arterial stenosis and a decreased arterial-venous ratio; and atherosclerosis can cause arterial stenosis.

[0003] Given that blood vessels in different regions have varying predictive significance for different diseases, clinicians currently rely entirely on experience to assess changes in arterial and venous morphology across different regions. Existing technologies only calculate the average vessel diameter and cannot automatically segment regions while simultaneously calculating vascular parameters within each region (e.g., the upper and lower sides of the nose). In recent years, the application of deep learning in vision has yielded significant results, with substantial work dedicated to developing image segmentation methods using deep learning models. Therefore, utilizing deep learning methods to automatically locate and segment target regions, while simultaneously combining this with computer vision techniques to calculate the diameter of individual blood vessels within those regions, holds significant importance and potential application value. Summary of the Invention

[0004] The method of this invention is based on an automatic segmentation algorithm to extract the tree structure and branch vessels of fundus images. Then, it traverses the single branch vessels according to the tree structure to calculate the normal vector of each branch pixel by pixel. Finally, it automatically calculates the diameter parameters of each branch vessel through interpolation and smoothing, thus exploring its computational power and potential application value.

[0005] This application discloses a method for calculating the diameter of retinal vessels in fundus photography, including:

[0006] Obtain fundus images to be processed;

[0007] The fundus image to be processed is segmented to obtain a fundus image with a tree-like structure and single-branched blood vessels;

[0008] For the fundus image with the tree structure and single-branch vessels, the normal vector of each branch is calculated by traversing the single-branch vessels according to the tree structure. Then, through interpolation and smoothing, the average, maximum, and minimum diameters of each vessel are obtained.

[0009] The segmentation process further includes automatically locating to the fovea of ​​the macula and automatically detecting and removing vascular pixels in the fundus image to be processed through a morphological refinement process to obtain a fundus image with a vascular centerline. Then, the fundus image with the vascular centerline is segmented by a segmentation algorithm, and the intersection points of the segmented skeleton lines and the endpoints of the vessels are calculated. After removing the skeleton line intersection points, each vessel segment is written into a tree structure to obtain a fundus image with a tree structure and single-branch vessels.

[0010] Specifically, preferably, the segmentation algorithm includes Unet+++, DeepLabv3+, or CSF-Net.

[0011] Furthermore, the calculation of the bifurcation point of the skeleton line is obtained by using the intersection operator to calculate each pixel and its surrounding pixels in the vascular tree with the vascular center line, and judging the calculated value against the set row; preferably, the calculation formula of the intersection operator is as follows:

[0012]

[0013] Wherein, j1-j5 are different convolutional kernels used to extract intersection point features, respectively extracting intersection point features of different shapes. When the extracted feature value is greater than the set intersection point threshold, it is marked as an intersection point; preferably, the set intersection point threshold is 4.

[0014] Furthermore, the calculation of the blood vessel endpoints is achieved by performing endpoint calculations on the edge pixels of each blood vessel segment, and then judging the calculated feature values ​​against a set endpoint operator threshold. Preferably, the calculation formula for the endpoint operator is as follows:

[0015]

[0016]

[0017] Wherein, k1-k8 are different convolution kernels used to extract endpoint features, respectively extracting endpoint features of different shapes. When the calculated feature value is greater than the set endpoint operator threshold, it is marked as an endpoint; preferably, the set endpoint operator threshold is 255.

[0018] Furthermore, the method for calculating the normal vector of the branch is as follows:

[0019] (1) First, calculate the vector V along the x-axis along the direction of the blood vessel. x The vector V along the y-axis y The specific calculation process is shown in the following formula;

[0020] V x =A x+3 -A x-3

[0021] V y =A y+3 -A y-3

[0022] Among them, A x A y For the origin of the coordinate system where the normal vector needs to be calculated, A x+3 For A x The third x-coordinate along the direction of the blood vessel, A x-3 Similarly; A y+3 For A y The third y-value along the direction of the blood vessel, A y-3 Similarly;

[0023] (2) Calculate the coordinates of the endpoint of the normal vector (C) x C y ), (D x D y As shown in the following formula, d is the preset normal vector length;

[0024]

[0025]

[0026]

[0027]

[0028] (3) Connect the coordinates of the endpoint of the normal vector and calculate the normal vector as shown in the following formula: τ x τ y As shown:

[0029] τ x =C x -D x

[0030] τ y =C y -D y

[0031] Where, τ x Let τ be the coordinate of the normal vector along the x-axis. y The coordinates of the normal vector along the y-axis.

[0032] Furthermore, the interpolation operation employs one or more of the following methods: nearest neighbor interpolation, bilinear interpolation, bicubic interpolation, interpolation, B-spline curve interpolation, nonlinear interpolation, and nearest neighbor interpolation. Preferably, the specific process of the interpolation operation is as follows: nonlinear interpolation is performed along the normal direction of the branch; non-integer contour points are found to generate pixel intensity contours; clustering is then performed along the normal direction of the branch; the pixel intensity contours are connected to form contour lines, and the edges of the contour lines are smoothed; the contour lines are arranged side by side and straightened to generate contours with each blood vessel; then nearest neighbor interpolation is used to process the contours with each blood vessel; the length of the contours with each blood vessel is calculated pixel by pixel along the center line to obtain the average, maximum, and minimum diameter of each blood vessel.

[0033] This application also discloses a system for calculating the diameter of retinal vessels in fundus photography, the system comprising:

[0034] The acquisition module is used to acquire fundus images to be processed;

[0035] The segmentation module is used to segment the fundus image to be processed to obtain a fundus image with a tree structure and single-branch vessels;

[0036] The calculation module is used to calculate the normal vector of the branch of the fundus image with tree structure and single branch vessels by traversing the single branch vessels according to the tree structure, and then obtain the average, maximum and minimum diameter of each vessel through interpolation and smoothing.

[0037] This application also discloses a computing device for calculating the diameter of retinal blood vessels in fundus photography, the device comprising: a memory and a processor;

[0038] The memory is used to store program instructions;

[0039] The processor is used to call program instructions, which, when executed, are used to implement the calculation method for the diameter of retinal blood vessels in fundus photography.

[0040] This application also discloses a computer-readable storage medium storing a computer program for calculating the diameter of retinal vessels in fundus photography. When the computer program is executed by a processor, it implements the method for calculating the diameter of retinal vessels in fundus photography.

[0041] The aforementioned devices or systems are used in the selection of analytical schemes to assist in the analysis of retinal vascular morphological changes in cardiovascular and systemic diseases; optionally, the analysis includes the correlation between increased venule diameter and the risk of diabetic retinopathy and stroke, hypertensive retinopathy leading to arterial stenosis and decreased arterial-venous ratio, and atherosclerosis leading to arterial stenosis analysis, which has a positive impact on the study of cardiovascular and systemic diseases.

[0042] The above-mentioned equipment or systems are used in the intelligent calculation of various blood vessel diameter parameters;

[0043] The above-mentioned equipment or system is used in the automatic segmentation of the target area of ​​fundus photography; optionally, the automatic segmentation includes extracting the tree structure and single-branch vessels of the fundus photography to obtain fundus photography of areas such as the optic cup and optic disc;

[0044] The aforementioned equipment or system is used in predicting the occurrence and development of eye diseases. Optionally, the prediction includes the predictive analysis of various vascular parameters in the segmented fundus images. These various vascular parameters include the diameters of the infratemporal and superior temporal arteries, the diameter of the macular vessels, the diameter of the infratemporal artery, and the diameter of the superior temporal artery, etc., which have a positive impact on and promote the research of optic disc glaucoma, peripheral glaucoma, and macular diseases.

[0045] This invention utilizes deep learning to train a model for calculating the macular center coordinates and a model for segmenting the optic cup and optic disc, based on a dataset containing the coordinates of the macula and the optic cup and optic disc. It automatically locates the fovea centralis and segments the target region. Then, through a series of calculations including the normal vectors of each branch, interpolation, and smoothing, it obtains various vascular parameters. This overcomes the limitations of methods that rely solely on experience to calculate arteriovenous morphological changes and methods that only calculate the average diameter of blood vessels. It has significant clinical implications and potential application value in the predictive analysis of cardiovascular and systemic diseases, demonstrating strong innovation and advancing the analysis and research of fundus imaging data.

[0046] Advantages of this application:

[0047] 1. This application innovatively discloses a method for calculating the diameter of retinal vessels in fundus photography. The method is based on an automatic segmentation algorithm to extract the target region in the fundus photography, and then calculates the parameters of each vessel based on interpolation and smoothing of each branch vessel in the target region, which objectively improves the accuracy and depth of data analysis.

[0048] 2. This application innovatively detects and removes vascular pixels in the fundus image to be processed through a morphological refinement process to obtain a vascular tree with a vascular centerline. Then, the vascular tree with a vascular centerline is automatically divided into vascular segments by a segmentation algorithm. At the same time, the intersection point operator is used to calculate the bifurcation point of the skeleton line and the endpoint operator is used to calculate the endpoint of the vascular segment, resulting in a fundus image with a tree structure and single-branched vessels, which significantly improves efficiency.

[0049] 3. This application creatively discloses a method based on fundus images with a tree structure and single-branch vessels. The method traverses the single-branch vessels according to the tree structure to obtain the normal vector. Through a series of calculations such as interpolation and smoothing, various vascular parameters, including the diameter of the inferior temporal artery, the superior temporal artery, and the diameter of the macular vessels, are obtained. This method better reflects the characteristics of each vascular parameter. Given that the vessels in different regions have different predictive significance for different diseases, this application can be more accurately applied to the auxiliary analysis of the occurrence and development of diseases related to fundus image data. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a schematic flowchart of the method for calculating the diameter of retinal vessels in fundus photography provided in an embodiment of the present invention;

[0052] Figure 2 This is a fundus image with a vascular centerline provided in an embodiment of the present invention;

[0053] Figure 3 This is an example of obtaining a pipe diameter based on a target region, provided by an embodiment of the present invention.

[0054] Figure 4 This is an analysis diagram of the effect of interpolation based on branch normal vectors provided in an embodiment of the present invention;

[0055] Figure 5 This is a schematic diagram of a device for calculating the diameter of retinal blood vessels in fundus photography, provided in an embodiment of the present invention. Detailed Implementation

[0056] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0057] In some of the processes described in the specification, claims, and accompanying drawings of this invention, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be performed in the order they appear herein, or may be performed in parallel. The operation numbers, such as S101, S102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be performed sequentially or in parallel.

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] Figure 1 This is a schematic flowchart illustrating the method for calculating the diameter of retinal vessels in fundus photography according to an embodiment of the present invention. Specifically, the method includes the following steps:

[0060] S101: Obtain the fundus image to be processed.

[0061] In one embodiment, the fundus image to be processed contains data on the coordinates of the macula's center point and the optic cup and optic disc. The data containing the coordinates of the macula's center point and the optic cup and optic disc are trained using deep learning methods to obtain a macula's center point coordinate calculation model and an optic cup and optic disc segmentation model, respectively.

[0062] In one specific embodiment, the data used in the model includes 40 fundus images from the public dataset Drive as the training and testing set for vascular segmentation, and 100 images from the drishtiGS dataset as the training and testing set for optic cup, optic disc, and fovea localization.

[0063] S102: Segment the fundus image to be processed to obtain a fundus image with a tree structure and single-branch vessels.

[0064] In one embodiment, the segmentation of the fundus image to be processed includes automatically detecting and removing vascular pixels in the fundus image through a morphological refinement process, resulting in, for example, Figure 2 The fundus image shown has the vascular centerline, and then it is segmented using a segmentation algorithm. Figure 2 Then, the intersection points of the obtained skeleton lines and the endpoints of the blood vessels are calculated. After removing the intersection points, each blood vessel segment is written into the tree structure to obtain a fundus image with a tree structure and single-branch blood vessels.

[0065] Specifically, the segmentation algorithm adopts any one or more of the following algorithms: FCN, U-Net, Unet++, Unet+++, DeepLabv3+, SegNet, RefineNet, CSF-Net; preferably, the segmentation algorithm adopts Unet+++, DeepLabv3+, or CSF-Net.

[0066] Each decoder layer in Unet+++ fuses small-scale and same-scale feature maps from the encoder, as well as large-scale feature maps from the decoder, which capture both fine-grained and coarse-grained semantics across the entire scale.

[0067] Deeplabv3+ is a network architecture in the Deeplab series. It is a culmination of dilated convolution, encoding / decoding, and multi-scale models, and is also one of the most commonly used semantic segmentation network models.

[0068] The RefineNet model downsamples using ResNet and achieves deeper fusion of multi-scale features through a series of convolutions, fusions, and pooling. It employs chained residual pooling, uses pooling with windows of different sizes, and fuses features using residual connections and learnable weights.

[0069] CSF-Net is a color spectral fusion network for semantic labeling of airborne laser scanning point clouds. It is trained in an end-to-end manner using weighted cross-entropy loss to build a multimodal fusion network.

[0070] Furthermore, the bifurcation points of the skeleton lines are calculated by using the intersection operator on each pixel and its surrounding pixels in the vascular tree with the vascular centerline, and then comparing the calculated value with a set threshold. Preferably, the intersection threshold is set to 4. Preferably, the calculation formula for the intersection operator is as follows:

[0071]

[0072] Wherein, j1-j5 are different convolutional kernels used to extract intersection point features, respectively extracting intersection point features of different shapes. When the extracted feature value is greater than the set threshold, it is marked as an intersection point; preferably, the set threshold is 4.

[0073] Furthermore, endpoint calculations are performed on the edge pixels of each blood vessel segment. The calculated feature values ​​and a set endpoint operator threshold are used to determine the blood vessel endpoints. Preferably, the formula for the endpoint operator is as follows:

[0074]

[0075]

[0076] Wherein, k1-k8 are different convolution kernels used to extract endpoint features, respectively extracting endpoint features of different shapes. When the extracted feature value is greater than the set endpoint operator threshold, it is marked as an endpoint; preferably, the set endpoint operator threshold is 255.

[0077] S103: Calculate the normal vector of each pixel of a single branch vessel segment, and then obtain the average, maximum, and minimum diameters of each branch vessel through nonlinear interpolation and smoothing.

[0078] In one example, the fundus image with a tree structure and single-branch vessels obtained in step S102 is used to traverse the single-branch vessels according to the tree structure to calculate the normal vector of the branch, and then interpolation and smoothing are performed to obtain the following result: Figure 3 The diagram shown is based on the pipe diameter of the target area; preferably, the normal vector calculation method is as follows:

[0079] (1) Calculate the vector V along the x-axis along the direction of the blood vessel. x The vector V along the y-axis y As shown in the formula below;

[0080] V x =A x+3 -A x-3

[0081] V y =A y+3 -A y-3

[0082] Among them, A x A y For the origin of the coordinate system where the normal vector needs to be calculated, A x+3 For A x The third x-coordinate along the direction of the blood vessel, A x-3 Similarly; A y+3 For A y The third y-value along the direction of the blood vessel, A y-3 Similarly;

[0083] (2) Formula for calculating the coordinates of the endpoint of the normal vector (C) x C y ), (D x D y As shown in the following formula, d is the preset normal vector length;

[0084]

[0085]

[0086]

[0087]

[0088] (3) Connect the coordinates of the endpoint of the normal vector and calculate the normal vector as shown in the following formula: τ x τ y As shown.

[0089] τ x =C x -D x

[0090] τy =C y -D y

[0091] Where, τ x Let τ be the coordinate of the normal vector along the x-axis. y The coordinates of the normal vector along the y-axis.

[0092] In one specific embodiment, the interpolation operation employs one or more of the following methods: nearest neighbor interpolation, bilinear interpolation, bicubic interpolation, interpolation, B-spline curve interpolation, nonlinear interpolation, and nearest neighbor interpolation. Preferably, the interpolation operation process is as follows: Figure 4 Shown: Figure 4 The process of drawing normal vectors (A) is based on non-linear interpolation along the normal vector direction of the branch, then generating pixel intensity contours by finding non-integer contour points, and finally clustering along the normal direction of the branch to obtain... Figure 4 The clustering contour shown in B is then connected to the pixel intensity contours to form a contour line, and the edges of the contour line are smoothed to obtain the following... Figure 4 The smoothing filter diagram shown in D and Figure 4 The connecting contour lines shown in E are arranged side-by-side and straightened to generate contours containing each blood vessel. Then, the following steps are taken: Figure 4 The nearest neighbor interpolation process shown in C has the outline of each blood vessel. The length of each blood vessel is calculated pixel by pixel along the center line to obtain the average, maximum and minimum diameter of each blood vessel.

[0093] Specifically, the diameters of each blood vessel include the diameter of the optic disc area, the diameter of the macular area, the temporal side of the macular area, the superior temporal side, the inferior temporal side, the superior nasal side, and the inferior nasal side; preferably, the parameters of each blood vessel diameter include the average diameter, the maximum diameter, and the minimum diameter.

[0094] further, Figure 3 The target area shown is the fundus image area segmentation steps corresponding to the obtained blood vessel contours, as shown in a, b, c, d, and e:

[0095] a. Connect the coordinates of the fovea center and the center of the test disc, and draw a straight line to divide the fundus imaging area into the upper and lower sides;

[0096] b. On the original image, with the optic disc and the fovea of ​​the macula as the center, and the diameter of the optic disc as the diameter, draw circles at the optic disc and the fovea of ​​the macula respectively. The circular area at the macula is the macular area, and the circular area at the optic disc is the optic disc area.

[0097] c. At the highest and lowest points of the viewing disk boundary, draw lines parallel to the straight line drawn in step a;

[0098] d. Draw a perpendicular line from the straight line in step a at the center of the macula to divide the fundus imaging area into the temporal and nasal sides;

[0099] e. Within the boundaries of the upper and lower parallel lines drawn in step c, draw the temporal side outside the circular boundary of the macular region drawn in step b, which is the temporal side of the macular region.

[0100] The above method is used to calculate the diameter of retinal vessels in fundus photography. Based on the fundus photography to be processed, the target region is segmented and the diameter parameters of each vessel in the target region are calculated based on interpolation and normal vectors. Figure 3 These are renderings provided by embodiments of the present invention, showcasing the diameter of retinal vessels based on automatically segmented target regions. This demonstrates that the proposed method for calculating retinal vessel diameters in fundus photography possesses strong discrimination, automatic analysis, and calculation capabilities. Therefore, by interpolating and obtaining various vascular parameters, including the diameters of the inferotemporal artery, superior temporal artery, and macular vessels, the characteristics of each vascular parameter can be better reflected. Given the different disease predictive significance of vessels in different regions, this application is more advantageous in the auxiliary analysis of disease development related to fundus photography data.

[0101] This invention provides a system for calculating the diameter of retinal vessels in fundus photography. The system includes:

[0102] The acquisition module is used to acquire fundus images to be processed;

[0103] The segmentation module is used to segment the fundus image to be processed, resulting in a fundus image with a tree structure and single-branch vessels;

[0104] The calculation module is used to traverse the single-branch blood vessels pixel by pixel according to the tree structure of the fundus image with tree structure and single-branch blood vessels to calculate the normal vector of the branch. Then, through interpolation and smoothing, the average, maximum and minimum diameter of each blood vessel is obtained.

[0105] Figure 5 This invention provides a calculation device for calculating the diameter of retinal blood vessels in fundus photography, comprising: a memory and a processor; the device may further include: an input device and an output device.

[0106] Memory, processor, input devices, and output devices can be connected via a bus or other means. Figure 5 The example shown is a bus connection.

[0107] Memory is used to store program instructions;

[0108] The processor is used to call program instructions, which, when executed, are used to perform the above-mentioned calculation method for the diameter of retinal vessels in fundus photography.

[0109] The present invention provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described method for calculating the diameter of retinal vessels in fundus photography.

[0110] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0111] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection between apparatuses or modules, and may be electrical, mechanical, or other forms.

[0112] The modules described as separate components may or may not be physically separate. Similarly, the components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Specifically, some or all of the modules can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0113] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The aforementioned integrated modules can be implemented in hardware or as software functional modules.

[0114] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc.

[0115] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0116] The computer device provided by the present invention has been described in detail above. For those skilled in the art, there will be changes in the specific implementation and application scope based on the ideas of the embodiments of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for calculating the diameter of retinal vessels in fundus photography, comprising: Obtain fundus images to be processed; The fundus image to be processed is segmented to obtain a fundus image with a tree-like structure and single-branched blood vessels; For the fundus image with tree structure and single-branch vessels, the normal vector of the branch is calculated by traversing the single-branch vessels according to the tree structure. Then, the average, maximum and minimum diameter of each vessel is obtained through interpolation and smoothing. The segmentation process includes automatically locating to the fovea of ​​the macula and automatically detecting and removing vascular pixels in the fundus image to be processed through a morphological refinement process to obtain a fundus image with a vascular centerline. Then, the fundus image with the vascular centerline is segmented by a segmentation algorithm, and the intersection points of the segmented skeleton lines and the endpoints of the vessels are calculated. After removing the intersection points of the skeleton lines, each vessel segment is written into a tree structure to obtain a fundus image with a tree structure and single-branch vessels. The specific process of the interpolation operation is to perform nonlinear interpolation along the normal vector direction of each branch, generate pixel intensity contours by finding non-integer contour points, then perform clustering along the normal direction of the branches, connect the pixel intensity contours to form contour lines and smooth the edges of the contour lines, arrange the contour lines side by side and straighten them to generate the contours of each blood vessel, and then use nearest neighbor interpolation to process the contours of each blood vessel, calculate the length of the contours of each blood vessel pixel by pixel along the center line, and obtain the average, maximum and minimum diameter of each blood vessel. The method for calculating the normal vector of the branch is as follows: (1) Calculate the vector V along the x-axis of the single-branch vessel. x The vector V along the y-axis y As shown in the following formula: Among them, A x A y Let A be the origin of the coordinate system for the normal vector to be calculated. x+3 For A x The third x-coordinate along the direction of the blood vessel, A x-3 Similarly; A y+3 For A y The third y-value along the direction of the blood vessel, A y-3 Similarly; (2) Calculate the coordinates of the endpoint of the normal vector (C) x C y ), (D x D y As shown in the following formula, d is the preset normal vector length; (3) Connect the coordinates of the endpoint of the normal vector, and calculate the normal vector of the branch as shown in the following formula. , As shown: in, Let be the coordinates of the normal vector along the x-axis. The coordinates of the normal vector along the y-axis; The vessels mentioned include the inferotemporal artery, superior temporal artery, superior nasal artery, inferior nasal artery, optic disc region vessels, and macular region vessels.

2. The method for calculating the diameter of retinal vessels in fundus photography according to claim 1, characterized in that, The intersection points of the skeleton lines are obtained by calculating the intersection point operator for each pixel in the vascular tree with the vascular center line and its surrounding pixels, and judging the calculated value against the set intersection point threshold.

3. The method for calculating the diameter of retinal vessels in fundus photography according to claim 2, characterized in that, The calculation formula for the intersection operator is as follows: Here, j1-j5 are different convolutional kernels used to extract intersection point features. j1-j5 are used to extract intersection point features of different shapes. When the extracted feature value is greater than the set intersection point threshold, it is marked as an intersection point.

4. The method for calculating the diameter of retinal vessels in fundus photography according to claim 3, characterized in that, The set threshold for the intersection point is 4.

5. The method for calculating the diameter of retinal vessels in fundus photography according to claim 2, characterized in that, The blood vessel endpoints are obtained by performing endpoint calculations on the edge pixels of each blood vessel segment, and then judging the calculated feature values ​​and the set endpoint operator thresholds.

6. The method for calculating the diameter of retinal vessels in fundus photography according to claim 5, characterized in that, The calculation formula for the endpoint operator is as follows: Among them, k1-k8 are different convolution kernels used to extract endpoint features. k1-k8 extract endpoint features of different shapes respectively. When the calculated feature value is greater than the set endpoint operator threshold, it is marked as an endpoint.

7. The method for calculating the diameter of retinal vessels in fundus photography according to claim 6, characterized in that, The set endpoint operator threshold is 255.

8. The method for calculating the diameter of retinal vessels in fundus photography according to claim 2, characterized in that, The segmentation algorithm adopts any one or more of the following algorithms: FCN, U-Net, Unet++, Unet+++, DeepLab v3+, SegNet, RefineNet, CSF-Net.

9. The method for calculating the diameter of retinal vessels in fundus photography according to claim 1, characterized in that, The interpolation operation employs one or more of the following interpolation methods: nearest neighbor interpolation, bilinear interpolation, bicubic interpolation, interpolation, B-spline curve interpolation, nonlinear interpolation, and nearest neighbor interpolation.

10. A system for calculating the diameter of retinal vessels in fundus photography, characterized in that, The system includes: The acquisition module is used to acquire fundus images to be processed; The segmentation module is used to segment the fundus image to be processed to obtain a fundus image with a tree structure and single-branch vessels; The calculation module is used to calculate the normal vector of the branch of the fundus image with tree structure and single branch vessels by traversing the single branch vessels according to the tree structure, and then obtain the average, maximum and minimum diameter of each vessel through interpolation and smoothing. The segmentation process includes automatically locating to the fovea of ​​the macula and automatically detecting and removing vascular pixels in the fundus image to be processed through a morphological refinement process to obtain a fundus image with a vascular centerline. Then, the fundus image with the vascular centerline is segmented by a segmentation algorithm, and the intersection points of the segmented skeleton lines and the endpoints of the vessels are calculated. After removing the intersection points of the skeleton lines, each vessel segment is written into a tree structure to obtain a fundus image with a tree structure and single-branch vessels. The specific process of the interpolation operation is to perform nonlinear interpolation along the normal vector direction of each branch, generate pixel intensity contours by finding non-integer contour points, then perform clustering along the normal direction of the branches, connect the pixel intensity contours to form contour lines and smooth the edges of the contour lines, arrange the contour lines side by side and straighten them to generate the contours of each blood vessel, and then use nearest neighbor interpolation to process the contours of each blood vessel, calculate the length of the contours of each blood vessel pixel by pixel along the center line, and obtain the average, maximum and minimum diameter of each blood vessel. The method for calculating the normal vector of the branch is as follows: (1) Calculate the vector V along the x-axis of the single-branch vessel. x The vector V along the y-axis y As shown in the following formula: Among them, A x A y Let A be the origin of the coordinate system for the normal vector to be calculated. x+3 For A x The third x-coordinate along the direction of the blood vessel, A x-3 Similarly; A y+3 For A y The third y-value along the direction of the blood vessel, A y-3 Similarly; (2) Calculate the coordinates of the endpoint of the normal vector (C) x C y ), (D x D y As shown in the following formula, d is the preset normal vector length; (3) Connect the coordinates of the endpoint of the normal vector, and calculate the normal vector of the branch as shown in the following formula. , As shown: in, Let be the coordinates of the normal vector along the x-axis. The coordinates of the normal vector along the y-axis; The vessels mentioned include the inferotemporal artery, superior temporal artery, superior nasal artery, inferior nasal artery, optic disc region vessels, and macular region vessels.

11. A device for calculating the diameter of retinal blood vessels in fundus photography, characterized in that, The device includes: a memory and a processor; the memory is used to store program instructions; the processor is used to call the program instructions, and when the program instructions are executed, they are used to execute the method for calculating the diameter of retinal vessels in fundus imaging as described in any one of claims 1-9.

12. A computer-readable storage medium storing thereon a computer program for calculating the diameter of retinal vessels in fundus photography, characterized in that, When the computer program is executed by the processor, it implements the method for calculating the diameter of retinal blood vessels in fundus photography as described in any one of claims 1-9.

Images