A method and system for assessing vascular leakage based on changes in capillary bifurcation blood flow.

By identifying capillary bifurcation structures and calculating the degree of red blood cell flow imbalance, this method solves the problem that optical coherence tomography (OCT) cannot quantitatively assess capillary leakage, enabling non-invasive quantitative assessment of capillary leakage and improving early detection capability and reliability.

CN123058656APending Publication Date: 2026-11-10SHANDONG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611552218.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-09
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing optical coherence tomography angiography cannot directly characterize changes in vascular permeability, making it difficult to quantitatively assess the degree of capillary leakage. Furthermore, existing methods fail to incorporate the blood flow transmission relationship in capillary bifurcation structures for analysis, thus failing to meet the clinical need for non-invasive quantitative assessment of microvascular leakage.

Method used

By identifying capillary bifurcation structures, calculating the red blood cell flow rate of each vessel segment, and constructing leakage evaluation indicators based on the degree of imbalance in red blood cell flow before and after bifurcation, including main vessel leakage sub-items, first-order branch leakage sub-items, and comprehensive leakage index, a non-invasive quantitative assessment of the degree of capillary leakage can be achieved.

Benefits of technology

It enables non-invasive quantitative assessment of capillary leakage, improves the applicability and reliability of early detection, captures blood flow changes before leakage causes tissue edema, provides non-invasive and quantifiable assessment indicators, and expands the application scope of optical coherence tomography angiography in hemodynamic analysis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123058656A_ABST
    Figure CN123058656A_ABST
Patent Text Reader

Abstract

This invention pertains to fundus microvascular detection technology. It proposes a method and system for assessing vascular leakage based on changes in capillary bifurcation blood flow. The method involves acquiring optical coherence tomography (OCT) angiography images and corresponding dynamic blood flow images of the target region. Image preprocessing yields a capillary network image of the target region. Capillary bifurcation structures are identified, determining the main vessels, branches at all levels, and corresponding vessel segment boundaries. Red blood cell flow in each vessel segment is calculated. Based on the difference in red blood cell flow before and after bifurcation, a single-bifurcation comprehensive leakage index is calculated, resulting in a global fusion leakage index for the target region, thus characterizing the degree of vascular leakage. This invention eliminates the need for contrast agent injection and provides a non-invasive, quantitative assessment of microvascular leakage from a hemodynamic perspective, making it suitable for early assessment and dynamic follow-up of fundus microvascular lesions.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to fundus microvascular detection technology, specifically to a method and system for assessing vascular leakage based on changes in capillary bifurcation blood flow. Background Technology

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Capillary leakage is a common pathophysiological process in various fundus microvascular diseases, closely related to disease occurrence, development, and prognosis. Early assessment of this process has significant clinical value. With the development of fundus imaging technology, optical coherence tomography (OCT) angiography has become increasingly widespread. This technique can clearly display the microvascular network of the retina and choroid without the need for contrast agent injection, providing a technical basis for non-invasive observation of microcirculation. Currently, this technique is mainly used to analyze structural indicators such as vascular density, non-perfusion areas, and vascular morphology, reflecting morphological changes and perfusion status of microvessels, and providing effective support for the diagnosis and follow-up of fundus diseases.

[0004] Current optical coherence tomography (OCT) angiography still has limitations. It primarily presents vascular network morphology and blood flow signals, but cannot directly characterize changes in vessel wall permeability or quantitatively assess capillary leakage. Existing capillary blood flow measurement methods can only obtain blood flow velocity in a single vessel, without analyzing the blood flow transmission relationships at capillary bifurcation points. Therefore, they cannot reflect vascular leakage through changes in blood flow before and after bifurcation, failing to meet the clinical need for non-invasive quantitative assessment of microvascular leakage. Summary of the Invention

[0005] To address the limitations of existing fundus microvascular imaging techniques in quantitatively assessing capillary leakage and the lack of integration of blood flow measurements with the blood flow transmission relationship at bifurcation structures, this invention provides a method and system for assessing vascular leakage based on changes in capillary bifurcation blood flow. By identifying capillary bifurcation structures and calculating red blood cell flow in each segment, a leakage evaluation index is constructed based on the degree of imbalance in red blood cell flow before and after bifurcation. This achieves a non-invasive quantitative assessment of the degree of capillary leakage, improving the applicability and reliability of early detection of fundus microvascular leakage.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for assessing vascular leakage based on changes in capillary bifurcation blood flow.

[0007] A method for assessing vascular leakage based on changes in capillary bifurcation blood flow includes the following procedures: Acquire optical coherence tomography (OCT) angiography images and corresponding blood flow dynamic images of the target region. Preprocess the OCT angiography images to obtain capillary network images of the target region. Identify at least one capillary bifurcation structure in a capillary network image, and determine the main vessel, first-level branch vessels, and the corresponding subordinate vessels of each capillary bifurcation structure, as well as the centerline and vessel boundary of each vessel segment. Based on dynamic blood flow images, the red blood cell flow rate of each segment of each capillary bifurcation structure was calculated. The comprehensive leakage index of each capillary bifurcation structure is calculated based on the difference between the red blood cell flow rate of the main vessel and the sum of the red blood cell flow rates of all first-order branch vessels in each capillary bifurcation structure, and the difference between the red blood cell flow rate of each first-order branch vessel and the sum of the red blood cell flow rates of the corresponding lower-level vessels. Based on the comprehensive leakage index of all capillary bifurcation structures, the global fusion leakage index of the target area is calculated, and the global fusion leakage index is used to characterize the degree of vascular leakage in the target area.

[0008] In one implementation of the first aspect of the present invention, preprocessing of optical coherence tomography (OCT) angiography images includes: The optical coherence tomography (OCT) angiography images were sequentially processed with denoising, motion correction, image registration, vessel enhancement, and vessel segmentation to obtain capillary network images.

[0009] In one implementation of the first aspect of the present invention, identifying capillary bifurcation structures includes: Determine the bifurcation point location of each capillary bifurcation structure, and divide the vessel segments corresponding to the main vessel, first-order branch vessels, and lower-order vessels based on the bifurcation point location.

[0010] In one implementation of the first aspect of the present invention, calculating red blood cell flow based on dynamic blood flow images includes: Along the long axis of the corresponding blood vessel segment, the light and dark encoded image of the corresponding blood vessel segment as a function of time is extracted from the dynamic blood flow image, and the number of red blood cells passing through the corresponding blood vessel segment is counted based on the light and dark encoded image.

[0011] As a further limitation of the first aspect of the present invention, after counting the number of red blood cells passing through based on the light and dark encoded image, the method further includes: obtaining the scanning time of the corresponding blood vessel segment, and using the ratio of the number of red blood cells passing through to the scanning time as the red blood cell flow rate of the corresponding blood vessel segment. The comprehensive leakage index is calculated by: calculating the first difference between the red blood cell flow rate of the main vessel and the sum of the red blood cell flow rates of all primary branch vessels, and using the ratio of the first difference to the red blood cell flow rate of the main vessel as the leakage component of the main vessel. For each primary branch vessel, calculate the second difference between the red blood cell flow rate of the corresponding primary branch vessel and the sum of the red blood cell flow rates of all the subordinate vessels corresponding to the primary branch vessel. The ratio of the second difference to the red blood cell flow rate of the corresponding primary branch vessel is used as the primary branch leakage component of the corresponding primary branch vessel.

[0012] In one implementation of the first aspect of the present invention, calculating the comprehensive leakage index includes: The main vessel leakage component is assigned a first weight, and the average value of all first-level branch leakage components is assigned a second weight. The first weight is greater than the second weight. The product of the main vessel leakage component and the first weight is added to the product of the average value and the second weight to obtain the comprehensive leakage index of the corresponding capillary bifurcation structure.

[0013] In one implementation of the first aspect of the present invention, calculating the global fusion leakage index includes: The total number of capillary bifurcation structures within the target area is counted, and the ratio of the sum of the comprehensive leakage indices of all capillary bifurcation structures to the total number is used as the global fusion leakage index of the target area. The target area is any one of the following: the macular region of the retina, the region of retinal vascular disease, or the region of choroidal neovascularization.

[0014] Secondly, the present invention provides a vascular leakage assessment system based on changes in capillary bifurcation blood flow.

[0015] A vascular leakage assessment system based on changes in capillary bifurcation blood flow includes: The image acquisition unit is configured to: acquire optical coherence tomography angiography images and corresponding blood flow dynamic images of the target area, preprocess the optical coherence tomography angiography images, and obtain capillary network images of the target area. The bifurcation identification unit is configured to: identify at least one capillary bifurcation structure in a capillary network image, and determine the main vessel, first-level branch vessel, and the subordinate vessel corresponding to the first-level branch vessel in each capillary bifurcation structure, as well as the center line and vessel boundary of each vessel segment. The flow calculation unit is configured to calculate the red blood cell flow rate of each segment of each capillary bifurcation structure based on the blood flow dynamic image. The single bifurcation index unit is configured to calculate the comprehensive leakage index of each capillary bifurcation structure based on the difference between the red blood cell flow rate of the main vessel and the sum of the red blood cell flow rates of all first-level branch vessels in each capillary bifurcation structure, and the difference between the red blood cell flow rate of each first-level branch vessel and the sum of the red blood cell flow rates of the corresponding lower-level vessels. The global index unit is configured to calculate the global fusion leakage index of the target area based on the comprehensive leakage index of all capillary bifurcation structures, and use the global fusion leakage index to characterize the degree of vascular leakage in the target area.

[0016] Thirdly, the present invention provides a computer-readable storage medium storing a computer program adapted to be loaded by a processor and executed as described in the first aspect of the present invention: a method for assessing vascular leakage based on changes in capillary bifurcation blood flow.

[0017] Fourthly, the present invention provides a computer device, comprising: a processor and a computer-readable storage medium; A processor, adapted to execute computer programs; A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the vascular leakage assessment method based on capillary bifurcation blood flow changes as described in the first aspect of the present invention.

[0018] Compared with the prior art, the beneficial effects of the present invention are: This invention acquires optical coherence tomography (OCT) angiography images and corresponding dynamic blood flow images of a target region. After preprocessing, a capillary network image of the target region is obtained. Based on this, capillary bifurcation structures are identified, and the extent and boundaries of each level of vessel segment are defined. Combined with the dynamic blood flow images, the red blood cell flow rate of each vessel segment is calculated. The differences in red blood cell flow rate between the main vessel and first-order branches, and between first-order branches and lower-order vessels in the bifurcation structure, are transformed into calculable leakage components. Furthermore, a single-bifurcation comprehensive leakage index is obtained through weighted fusion, and a global fusion leakage index is calculated based on the multi-bifurcation results. Compared to existing methods that only observe vessel morphology and structure, this invention can reflect changes in vessel wall permeability from a hemodynamic perspective. It can capture blood flow changes caused by microvascular damage before structural changes such as tissue edema caused by leakage occur, improving the ability to detect leakage in its early stages. Simultaneously, it eliminates the need for contrast agent injection, making it suitable for multiple follow-ups and dynamic monitoring. It provides a non-invasive and quantifiable reference indicator for the assessment of fundus microvascular leakage-related diseases and expands the application scope of OCT angiography technology in hemodynamic analysis.

[0019] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0020] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0021] Figure 1 A flowchart illustrating a method for assessing vascular leakage based on changes in capillary bifurcation blood flow, provided as an exemplary embodiment of the present invention. Figure 2 A schematic diagram of a single-branching capillary structure is provided as an exemplary embodiment of the present invention; Figure 3 A colorimetric encoding diagram of the same capillary is provided as an exemplary embodiment of the present invention. Figure 4 A schematic diagram of a multi-branched capillary structure provided as an exemplary embodiment of the present invention; Figure 5 A schematic diagram of the principle of a vascular leakage assessment system based on capillary bifurcation blood flow changes, provided as an exemplary embodiment of the present invention; Figure 6 A schematic diagram of a computer device provided for an exemplary embodiment of the present invention. Detailed Implementation

[0022] Embodiments of the present invention will now be described in more detail with reference to the accompanying drawings. While some embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the invention. It should be understood that the drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the invention. In the description of the embodiments of the present invention, the terms "including" and similar expressions should be understood to have an open-ended meaning, i.e., including but not limited to.

[0023] The term "based on" should be understood as at least partially based on. The term "an embodiment" or "that embodiment" should be understood as at least one embodiment. In the description of this embodiment, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this embodiment, unless otherwise stated, "multiple" means two or more.

[0024] To facilitate understanding of the technical solutions of the embodiments of the present invention, the core technical terms involved in the specific embodiments of the present invention are explained and defined as follows: Optical Coherence Tomography Angiography (OCTA) is a non-invasive vascular imaging technique that uses optical coherence tomography to display blood flow signals and vascular morphology in the retinal and choroidal microvascular networks without the injection of contrast agents, providing basic image data for hemodynamic analysis at the capillary level.

[0025] Capillary bifurcation structure: refers to the hierarchical topology of a capillary network where a main vessel branches into multiple primary branches, and each primary branch further branches into lower-level vessels. This structure is the basic analytical unit for analyzing the conservation of erythrocyte flow before and after bifurcation and assessing the degree of local vascular leakage.

[0026] Main vessel (P): The segment of the capillary bifurcation structure that supplies blood to the first-order branches upstream and downstream of the bifurcation point. The main vessel carries the main blood flow within the bifurcation area, and changes in its red blood cell flow rate have the most direct effect on the local perfusion status and leakage level.

[0027] First-order branch vessels ( ): The branch segments of a capillary bifurcation structure, which are the branches that directly branch off from the main vessel at the bifurcation point, where the subscripts are... This refers to the numbering of the primary branch vessels. Primary branch vessels receive blood from the main vessels and continue to transport blood to their subordinate vessels, serving as an intermediate link in analyzing the hierarchical blood flow transmission relationship.

[0028] Lower-level vessels ( ): The bifurcation structure of capillaries is composed of the first The branch segments formed by the further bifurcation of the primary branch vessels, where the subscript... This is the number of the subordinate vessels of this primary branch vessel. The subordinate vessels receive blood from the corresponding primary branch vessel and are the final link in the blood flow transmission chain.

[0029] Centerline and vessel boundary: The centerline refers to the skeletal line of the vessel segment extracted along its course, used to determine the long axis direction of the vessel and the scanning path; the vessel boundary refers to the contour line of the outer wall of the vessel lumen, used to determine the cross-sectional range of the vessel segment and the effective scanning area. Together, they form the basis for describing the geometric features of each vessel segment.

[0030] Red blood cell flow ( Red blood cell flow rate (RBC) is the number of red blood cells passing through a cross-section of a blood vessel segment per unit time, expressed as cells / second. RBC flow rate is the most direct quantitative indicator of microcirculatory blood flow within capillaries, reflecting the vessel segment's ability to transport red blood cells per unit time.

[0031] Bright-dark encoded images: Capillary imaging methods based on OCTA encoding acquire images of alternating bright and dark signals that change over time by repeatedly scanning the same capillary along its long axis. Since red blood cells typically pass through capillaries in a single file, the OCTA device generates a bright signal when scanning plasma and a dark signal when scanning red blood cells, thus forming a periodic bright-dark encoding sequence arranged along the time axis. Each bright-dark cycle corresponds to the signal of a red blood cell passing through the scan section; the number of red blood cells passing through can be obtained by counting the number of bright-dark cycles.

[0032] Main vessel leakage sub-item ( ): A normalized index that measures the degree of imbalance between the inflow of red blood cells into the main vessel and the sum of the outflow of red blood cells from each first-order branch vessel. Under ideal, leak-free conditions, the inflow of red blood cells into the main vessel should equal the sum of the outflow of red blood cells from each first-order branch vessel. When leakage occurs, the downstream red blood cell flow decreases, and the leakage component of the main vessel takes a positive value. The larger the value, the higher the degree of leakage at that bifurcation.

[0033] Primary branch leakage sub-item ( ): Measure the first A normalized index of the imbalance between the inflow of erythrocytes into primary branch vessels and the sum of the outflow of erythrocytes from all their subordinate vessels. Its physical meaning is similar to that of the main vessel leakage component, specifically used to evaluate the degree of blood flow loss between primary branch vessels and their subordinate vessels.

[0034] Comprehensive Leakage Index ( The comprehensive leakage index (CLEI) is obtained by weighted fusion of leakage components from the main vessel and each first-order branch, thus representing the degree of erythrocyte flow imbalance within a single capillary bifurcation structure. A higher CLEI indicates a greater degree of microvascular leakage, vessel wall damage, or perfusion abnormalities in that bifurcation region.

[0035] The overall integrated leakage index ( The global fusion leakage index (GFCI) is a regional quantitative indicator representing the overall capillary leakage or microvascular damage level in the entire detection area. It is obtained by averaging and fusing the comprehensive leakage indices of all independent capillary bifurcation structures within the detection area. The higher the value of the global fusion leakage index, the higher the overall microvascular leakage risk within the detection area.

[0036] This embodiment is based on the following technical background and the problems to be solved. Capillary leakage is a pathological process in which plasma components abnormally enter the interstitial space after the vascular endothelial barrier function is impaired. It is commonly seen in various fundus microvascular diseases such as diabetic retinopathy, diabetic macular edema, retinal vein occlusion, uveitis, and neovascularization. In the prior art, the assessment of fundus vascular leakage mainly relies on two methods: fluorescein fundus angiography and optical coherence tomography (OCT) structural imaging. Fluorescein fundus angiography can directly show the extent of vascular leakage, but it requires intravenous injection of contrast agents, which is an invasive examination and carries risks such as allergic reactions and renal burden, limiting its application in frequent follow-ups and special populations. OCT structural imaging can show structural changes such as macular edema, retinal interlaminar fluid accumulation, and retinal thickening that occur secondary to leakage, but it reflects the indirect results of leakage and cannot directly characterize the abnormal vascular permeability itself, especially in the early stages of leakage or when edema is not yet obvious, thus limiting its quantitative assessment capabilities. While OCTA can non-invasively visualize the microvascular network of the retina and choroid, current OCTA techniques are mainly used to observe morphological indicators such as vessel density, non-perfusion areas, vessel morphology, capillary occlusion, and neovascularization. They lack quantitative measurement and analysis of erythrocyte flow in each segment of capillary bifurcation structures, making it impossible to quantitatively assess the degree of capillary leakage from a hemodynamic perspective. Therefore, this invention proposes a method for assessing vascular leakage based on changes in blood flow at capillary bifurcations. Starting from the blood flow transmission relationship in capillary bifurcation structures, this method achieves a non-invasive and quantitative assessment of leakage degree by measuring the imbalance of erythrocyte flow before and after bifurcation.

[0037] To address the aforementioned technical problems, the vascular leakage assessment method provided by this invention unfolds according to the following overall approach: First, an OCTA system is used to scan the target area, acquiring OCTA vascular images and corresponding blood flow dynamic images. The OCTA images undergo preprocessing, including denoising, motion correction, image registration, vascular enhancement, and vascular segmentation, to obtain a capillary network image. Second, capillary bifurcation structures are identified in the capillary network image, determining the hierarchical relationships and geometric features of each vascular segment. Subsequently, based on the light and dark coding signals in the blood flow dynamic image, the red blood cell flow rate of each vascular segment is calculated by counting the number of light and dark change cycles. Furthermore, based on the conservation deviation of red blood cell flow rate before and after bifurcation, leakage components of the main vessel and first-level branches are calculated separately, and a single-bifurcation comprehensive leakage index is obtained through weighted fusion. Finally, the comprehensive leakage indices of all bifurcation structures within the detection area are averaged and fused to obtain a global fused leakage index, thereby characterizing the degree of vascular leakage in the target area. This method differs from existing technologies that rely on outcome-based assessments such as contrast agent extravasation or tissue edema. It directly addresses the hemodynamic mechanism by transforming the imbalance in red blood cell flow caused by capillary leakage into a calculable and comparable quantitative indicator.

[0038] Figure 1 This paper illustrates the overall technical flowchart of the method for assessing vascular leakage based on capillary bifurcation blood flow changes provided in this invention. The assessment system of this invention operates in a detection environment comprised of an OCTA acquisition device and a computing device. The OCTA acquisition device performs non-invasive tomographic scanning on the target area, acquiring raw OCTA vascular images and corresponding dynamic blood flow images. The target area can be the macular region of the retina, retinal vascular lesion areas, choroidal neovascularization areas, or other microvascular areas that can be imaged by OCTA. The computing device is equipped with a processor and a computer-readable storage medium. The processor executes the assessment program stored in the storage medium, sequentially completing all calculation processes including image preprocessing, bifurcation structure recognition, red blood cell flow calculation, leakage component calculation, comprehensive leakage index calculation, and full-domain fusion leakage index output. The leakage assessment results are then output to a display device or data management system. The system can also be expanded into a networked detection platform, supporting batch testing of multiple patients, and is suitable for clinical auxiliary assessment and follow-up monitoring scenarios of fundus microvascular lesions. Data acquisition and processing in this invention strictly comply with relevant laws and regulations and are used only for vascular leakage assessment, not for any other unauthorized purposes. It should be understood that the above system environment is described for illustrative purposes only and does not imply any limitation on the scope of protection of the present invention.

[0039] In one implementation, such as Figure 1 As shown, the method for assessing vascular leakage based on changes in capillary bifurcation blood flow provided by the present invention includes the following steps, and the steps are connected by data transfer to form a complete assessment process from image acquisition to leakage quantification.

[0040] Step S101 is the image acquisition and preprocessing step, which is executed collaboratively by the OCTA acquisition device and the computing device. It is responsible for acquiring the OCTA vascular image and the corresponding blood flow dynamic image of the target area, and preprocessing the OCTA vascular image to obtain the capillary network image of the target area.

[0041] In step S101, the OCTA acquisition device scans the target area and simultaneously outputs two types of data: one is an OCTA vascular image, reflecting the blood flow signal distribution of the microvascular network within the target area; the other is a dynamic blood flow image, recording the light and dark encoded signal sequence of the same capillary region over time, for subsequent red blood cell flow calculation. After receiving the above two types of images, the computing device sequentially performs denoising processing, motion correction processing, image registration processing, vessel enhancement processing, and vessel segmentation processing on the OCTA vascular image to obtain the capillary network image of the target area.

[0042] Denoising is used to eliminate speckle noise and background interference generated during OCTA acquisition, improving the image signal-to-noise ratio. Common methods include denoising algorithms based on nonlocal mean filtering or wavelet transform, which can effectively suppress random noise while preserving the edge information of small blood vessels. Motion correction is used to compensate for image displacement caused by the patient's eye movements, ensuring spatial consistency between different frames. By extracting rigid body transformation parameters between adjacent frames and performing inverse transformation alignment, inter-frame misalignment caused by micro-movements of the eye is eliminated. Image registration is used to align multiple frames to a unified coordinate system, ensuring spatial accuracy for subsequent analysis. Typically, methods based on specific coordinate systems are employed. The registration algorithm, based on feature points or mutual information, maps each frame of the image to the coordinate system of the reference frame. Vessel enhancement processing improves the contrast and continuity of capillaries, enhancing the visibility of small vessels. Common methods include multi-scale vessel enhancement algorithms such as the Frangi filter, which can selectively enhance tubular structures based on local curvature features of the image. Vessel segmentation processing extracts the binary skeleton of the capillary network from the enhanced image, forming a capillary network image that can be used for subsequent bifurcation identification. Threshold-based segmentation and deep learning semantic segmentation methods can be used to separate vessel pixels from background pixels, generating a binary mask for the capillary network. The output of step S101 is the capillary network image of the target region and the corresponding blood flow dynamic image, both of which will serve as input data for steps S102 and S103. After the above five preprocessing steps, noise interference, motion artifacts, and vessel blurring in the original OCTA image are systematically eliminated, providing a high-quality image foundation for subsequent bifurcation structure identification and red blood cell flow calculation.

[0043] Step S102 is the capillary bifurcation structure identification step, which is executed by the computing device. It is responsible for identifying at least one capillary bifurcation structure in the capillary network image and determining the main blood vessel, first-level branch blood vessel, lower-level blood vessel, center line and blood vessel boundary of each blood vessel segment in each capillary bifurcation structure.

[0044] In step S102, the computing device performs topological analysis on the capillary network image to identify bifurcation nodes in the vascular network and determine the bifurcation point location of each capillary bifurcation structure. Specifically, the computing device refines the binarized skeleton of the capillary network image and extracts the vascular centerline skeleton; then, it performs connected component analysis on the skeleton image, counts the number of connected neighbors for each skeleton pixel, and identifies skeleton pixels with a connected neighbor number greater than or equal to 3 as bifurcation nodes; using each bifurcation node as the center, it traces the corresponding vascular segment along each connected direction until it reaches the next bifurcation node or the end of the vascular segment, thus completing the boundary delineation of each vascular segment. Figure 2 As shown, for each identified capillary bifurcation structure, the hierarchical classification of each vessel segment is determined based on the bifurcation point location: the vessel segment upstream of the bifurcation point is identified as the main vessel, denoted as... The downstream segments that branch directly from the bifurcation point are identified as first-order branch vessels, and the second-order branch vessels are identified as third-order branch vessels. First-order branch vessels of the root are denoted as The vascular segments formed by further bifurcation of each primary branch vessel are identified as lower-level vessels. The first branch of the root vessel The lower-level vessels are denoted as During the hierarchical segmentation process, the relationship between main vessels and branch vessels is determined by analyzing the diameter and blood flow direction of each vessel segment. Generally, vessel segments with larger diameters and located upstream of the bifurcation are identified as main vessels, while those with smaller diameters and located downstream are identified as primary branch vessels. After hierarchical segmentation, the centerline and vessel boundary are extracted for each vessel segment: the centerline is extracted along the vessel's direction of travel using a distance-transform-based skeleton extraction algorithm to generate a uniformly distributed sequence of centerline points along the vessel's long axis, used to determine the vessel's long axis direction and subsequent scanning path; the vessel boundary is extracted along the outer wall contour of the lumen using a boundary detection algorithm based on an active contour model or morphological dilation to determine the cross-sectional range of the effective scanning area. The output of step S102 is the hierarchical relationship, centerline, and vessel boundary of each vessel segment in each capillary bifurcation structure. This information will serve as the geometric input for calculating red blood cell flow in step S103. Through the above topology analysis and hierarchical division, the computing device can systematically extract all analyzable bifurcation structures from complex capillary network images, laying a structured data foundation for subsequent batch calculation of red blood cell flow and assessment of leakage index.

[0045] Step S103 is the red blood cell flow calculation step, executed by the computing device. It is responsible for calculating the red blood cell flow in each segment of each capillary bifurcation structure based on the dynamic blood flow image, including the red blood cell flow in the main vessel. , No. erythrocyte flow in primary branch vessels and the The first branch of the root vessel Root-level vascular erythrocyte flow .

[0046] In step S103, for each blood vessel segment to be tested, the computing device extracts a light-dark encoded image of the segment along its long axis over time from the blood flow dynamic image, based on the centerline direction of the segment determined in step S102. Specifically, a scan line is set along the centerline direction of the segment, and the scan line is used as the sampling path to extract the scan signal intensity values ​​of the segment at different times from the blood flow dynamic image. The signal intensity values ​​at each time are arranged in chronological order to form a one-dimensional signal sequence unfolded along the time axis, i.e., the light-dark encoded image. Figure 3As shown, since red blood cells typically pass through capillaries in a single file, the OCTA device generates a bright signal when scanning plasma and a dark signal when scanning red blood cells. This creates a periodic alternating bright and dark coding sequence along the time axis, with each bright-dark cycle corresponding to the signal of a red blood cell passing through the scan section. The computing device performs peak and valley detection on the bright-dark coded image, identifying local maxima (corresponding to plasma passage times) and local minima (corresponding to red blood cell passage times) of the signal intensity, and counts the number of complete bright-dark cycles to obtain the signal intensity over the scan duration. The number of red blood cells passing through the cross-section of this blood vessel segment Then, the red blood cell flow rate of this blood vessel segment is calculated according to the following formula. : (1); In the formula, Red blood cell flow rate, measured in cells / second; Scan duration The number of red blood cells in the cross-section of the blood vessel segment scan; The scan duration is denoted as . This formula converts the period count results in the coded image into the number of red blood cells passing through per unit time, directly reflecting the microcirculatory blood flow within the capillaries. It is the core input for subsequent leakage index calculations. Red blood cell flow, as a direct quantitative indicator of microcirculatory blood flow, has a more direct physiological significance than blood flow velocity: blood flow velocity reflects the speed of red blood cell movement within the blood vessels, while red blood cell flow directly reflects the number of red blood cells transported per unit time, and is more closely related to the local tissue oxygen supply capacity. In cases of low signal quality, autocorrelation analysis or spectral analysis methods can be used to filter and enhance the coded image to improve the accuracy of period counting. By performing the above coded image extraction and period counting process separately on each blood vessel segment, the red blood cell flow of the main blood vessel is obtained sequentially. Red blood cell flow in each primary branch vessel and red blood cell flow in each lower-level blood vessel The flow data is then passed to step S104. The output of step S103 is the red blood cell flow value for all segments of each capillary bifurcation structure. This flow data constitutes the core input for subsequent leakage sub-item calculation and comprehensive leakage index assessment, ensuring that the leakage assessment index is based on directly measurable hemodynamic parameters, rather than indirectly inferred structural changes.

[0047] Step S104 is the comprehensive leakage index calculation step, which is executed by the computing device. It is responsible for calculating the leakage sub-item of the main vessel and the leakage sub-item of the first-level branch vessels based on the difference between the red blood cell flow of the main vessel and the sum of the red blood cell flow of all first-level branch vessels in each capillary bifurcation structure, and the difference between the red blood cell flow of each first-level branch vessel and the sum of the red blood cell flow of its subordinate vessels. The comprehensive leakage index of each capillary bifurcation structure is calculated by weighted fusion.

[0048] In step S104, the computing device first analyzes the blood flow conservation relationship between the main vessel and the first-order branch vessels. Under ideal, leak-free conditions, the red blood cell inflow into the main vessel should be equal to the sum of the red blood cell outflows from each first-order branch vessel, i.e., satisfying the following conservation equation: (2); In the formula, Main blood vessel erythrocyte flow; For the first Red blood cell flow in primary branch vessels. When capillary leakage, vessel wall damage, or plasma extravasation occurs, the increased local blood viscosity and resistance to red blood cell passage reduce the number of downstream red blood cells passing through per unit time, causing a deviation from the above conservation relationship. Based on this, a main vessel leakage component is defined. for: (3); In the formula, The main vascular leakage category; Main blood vessel erythrocyte flow; For the first Red blood cell flow in primary branch vessels. The main vessel leakage component is obtained by normalizing the difference between the main vessel inflow and the sum of the primary branch outflow to the main vessel flow, resulting in a dimensionless ratio index. The larger the value, the greater the imbalance between the inflow of red blood cells into the main blood vessel and the outflow of red blood cells from the first-order branches, suggesting that there may be more significant capillary leakage or microvascular damage at the bifurcation point.

[0049] The computing device then analyzes the blood flow conservation relationships between each first-order branch vessel and its subordinate vessels. For the first... For a primary branch vessel, under ideal, leak-free conditions, the erythrocyte inflow rate of that primary branch vessel should be equal to the sum of the erythrocyte inflow rates of all its subordinate vessels, satisfying the following conservation equation: (4); In the formula, For the first erythrocyte flow in the first-order branch vessels of the root; For the first The first branch of the root vessel Red blood cell flow in the primary branch vessels. When leakage occurs, red blood cell flow downstream of the primary branch vessels also decreases, causing a deviation from the above conservation relationship. Based on this, the first... Leakage sub-items of primary branch vessels for: (5); In the formula, For the first Leakage items of primary branch vessels; For the first erythrocyte flow in the first-order branch vessels of the root; For the first The first branch of the root vessel Red blood cell flow in the root subvascular system. The larger the value, the higher the value. The greater the imbalance in erythrocyte flow between the primary branch vessels and their subordinate vessels, the more likely there is leakage or loss of blood flow near that primary branch vessel segment.

[0050] After obtaining the leakage components of the main vessel and each first-order branch, the computing device performs a weighted fusion of the two types of leakage components to calculate the comprehensive leakage index of the capillary bifurcation structure. Assume that the bifurcation structure contains... Root primary branch vessels, comprehensive leakage index The calculation formula is: (6); In the formula, The single-branch comprehensive leakage index; It is the main vessel weighting coefficient, and ; This refers to the number of primary branch vessels. For the first Leakage component of primary branch vessels. In the calculation of the comprehensive leakage index, the main vessel weighting coefficient... A value greater than 0.5 is significant because the main vessel carries the majority of blood flow within the bifurcation region, and changes in its erythrocyte flow have a more direct impact on local perfusion status and leakage levels. Giving the main vessel a higher weight enhances the sensitivity and stability of the comprehensive leakage index in responding to bifurcation leakage, vessel wall damage, and perfusion abnormalities. The leakage component of the first-level branches uses the arithmetic mean of the leakage components of each first-level branch, reflecting the average leakage level at each first-level branch level within the bifurcation structure. Used to characterize the degree of erythrocyte flow imbalance within a single capillary bifurcation structure. The higher the value, the greater the degree of microvascular leakage, vessel wall damage, or perfusion abnormality in the bifurcation region. The output of step S104 is the comprehensive leakage index corresponding to each capillary bifurcation structure, and these values ​​will be passed to step S105 for global fusion calculation.

[0051] In some implementations, the calculation process for the main vessel leakage component in step S104 specifically includes: calculating the main vessel erythrocyte flow rate. Red blood cell flow in all primary branch vessels The difference between the sums is used to determine the main vessel red blood cell flow rate. The ratio is used as the leakage component of the main vessel; for each primary branch vessel, the corresponding erythrocyte flow rate of the primary branch vessel is calculated. Red blood cell flow in all subordinate vessels below the corresponding first-order branch vessel The difference between the sums is used to determine the erythrocyte flow rate of the corresponding first-order branch vessels. The ratio of the values ​​is used as the leakage component of the primary branch vessel. The above normalization process ensures that each leakage component is a dimensionless ratio, eliminating the influence of differences in absolute flow rates across different vessel segments on the comparison of leakage levels, thus making the leakage index comparable across vessel segments and patients. It should be noted that under normal physiological conditions, due to measurement errors, blood flow fluctuations, and local hydrodynamic effects at vessel bifurcation points, there may be slight deviations between the sum of erythrocyte flow rates in the main vessel and those in the primary branch vessels. Therefore, the interpretation of leakage components should be based on a comprehensive analysis of the clinical context and statistical reference range. When a leakage component exceeds the mean plus two standard deviations of the normal control group, a statistically significant leakage abnormality can be considered. In some implementations, longitudinal comparisons of leakage components at different time points for the same patient can be performed. The dynamic trends in leakage components can be used to assess disease progression or treatment effectiveness, providing quantitative evidence for clinical follow-up monitoring.

[0052] Step S105 is the global fusion leakage index calculation step, which is executed by the computing device. It is responsible for calculating the global fusion leakage index of the target area based on the comprehensive leakage index of all capillary bifurcation structures, and using the global fusion leakage index to characterize the degree of vascular leakage in the target area.

[0053] In step S105, as Figure 4 As shown, if a total of [number] targets are identified within the target area... Each independent capillary bifurcation structure is analyzed in step S104, and the comprehensive leakage index corresponding to each bifurcation structure is calculated. , ... The overall fusion leakage index of the target area is then... The calculation formula is: (7); In the formula, The overall integrated leakage index, namely ; The total number of independent capillary bifurcation structures within the target area; For the first The comprehensive leakage index of capillary bifurcation structures. The global fusion leakage index integrates leakage information at the local bifurcation level into a regional overall assessment index by arithmetically averaging the comprehensive leakage indices of all independent capillary bifurcation structures within the detection area. This eliminates the impact of measurement errors of individual bifurcation structures on the overall assessment results and improves the stability and representativeness of the assessment. The higher the value, the higher the overall risk of microvascular leakage within the target area; A value close to zero indicates that the erythrocyte flow transmission relationship in the capillary bifurcation structure within the target area tends to be conserved, and the vascular barrier function is basically intact.

[0054] In some embodiments, the statistical range of the target area in step S105 can be any one of the retinal macular region, retinal vascular disease region, or choroidal neovascularization region. For the retinal macular region, the global fusion leakage index reflects the overall leakage degree of the capillary network in the macular region and can be used for early quantitative assessment of diseases such as diabetic macular edema. For the retinal vascular disease region, the global fusion leakage index reflects the overall level of microvascular leakage within the disease area and can be used to quantify the leakage degree of diseases such as retinal vein occlusion. For the choroidal neovascularization region, the global fusion leakage index reflects the active leakage status of the neovascularization region and can be used to determine the activity of neovascular diseases. In some embodiments, the target area can also be divided into several sub-regions, and the global fusion leakage index of each sub-region can be calculated to form a spatial distribution map of the leakage degree. This helps clinicians locate the most severely leaking local area and provides a more spatially resolved quantitative basis for target selection in laser photocoagulation therapy or efficacy evaluation of anti-vascular endothelial growth factor injection therapy. The calculation results of the global fusion leakage index can be output to the electronic medical record system in numerical form, or overlaid on the OCTA vascular image in the form of a pseudo-color heat map, intuitively presenting the spatial distribution of leakage degree in the target area, improving the convenience and interpretability of clinical use.

[0055] In some embodiments, the present invention also provides a system for assessing vascular leakage based on changes in capillary bifurcation blood flow, such as... Figure 5 As shown, the system includes the following functional units, which exchange parameters through a data interface to form a complete evaluation process, corresponding one-to-one with each step of the above method embodiment.

[0056] The image acquisition unit 501 is configured to perform the function of step S101, specifically responsible for acquiring the OCTA vascular image and corresponding blood flow dynamic image of the target area, and sequentially performing denoising processing, motion correction processing, image registration processing, vascular enhancement processing, and vascular segmentation processing on the OCTA vascular image to obtain the capillary network image of the target area. The capillary network image and blood flow dynamic image are then transmitted to the recognition unit and the flow calculation unit. The image acquisition unit can be further divided into an image receiving subunit and an image preprocessing subunit: the image receiving subunit is responsible for establishing a data communication interface with the OCTA acquisition device, receiving raw OCTA vascular image frames and blood flow dynamic image frames in real time, and performing format conversion and storage on the received images; the image preprocessing subunit is responsible for sequentially executing each preprocessing algorithm in a fixed order of denoising, motion correction, registration, vascular enhancement, and vascular segmentation, outputting a standardized capillary network image to ensure that the image quality for subsequent analysis meets the requirements of bifurcation recognition and flow calculation.

[0057] The bifurcation identification unit 502 is configured to perform the function of step S102, specifically responsible for identifying at least one capillary bifurcation structure in the capillary network image and determining the main vessel in each capillary bifurcation structure. First-order branch vessels Lower-level blood vessels The system transmits the bifurcation structure information, including the centerlines and boundaries of each vessel segment, to the flow calculation unit. The identification unit employs a bifurcation detection algorithm based on image topology analysis to perform connected component analysis on the capillary network skeleton image. This automatically locates bifurcation nodes and classifies the hierarchical affiliation of each vessel segment. Simultaneously, it extracts the centerline point sequence and boundary contours of each vessel segment, forming structured bifurcation topology description data, providing accurate geometric location information for subsequent flow calculation.

[0058] The flow calculation unit 503 is configured to perform the function of step S103, specifically responsible for extracting the light and dark encoded images along the centerline direction of each blood vessel segment based on the blood flow dynamic image, and obtaining the number of red blood cells by counting the number of light and dark change cycles. Combined with scan duration According to the formula Calculate the erythrocyte flow rate of each segment in each capillary bifurcation structure, including the erythrocyte flow rate of the main vessel. Red blood cell flow in each primary branch vessel and red blood cell flow in each lower-level blood vessel The flow data is then transmitted to the bifurcation index unit. The flow calculation unit independently performs the light and dark encoded image extraction and cycle counting process for each vessel segment, supporting parallel calculation of multiple vessel segments to improve overall calculation efficiency. At the same time, the calculated red blood cell flow data is validated for reasonableness. If the red blood cell flow of a certain vessel segment is abnormally high or low, it is marked as an outlier and processed accordingly in subsequent leakage index calculations to ensure the reliability of the evaluation results.

[0059] Bifurcation index unit 504 is configured to perform the function of step S104, specifically responsible for calculating the main vessel leakage component based on the difference between the main vessel erythrocyte flow rate and the sum of erythrocyte flow rates of all first-order branch vessels in each capillary bifurcation structure and its normalized value. The leakage component of the first-level branch is calculated based on the difference between the erythrocyte flow rate of each first-level branch vessel and the sum of the erythrocyte flow rates of the corresponding lower-level vessels, and its normalized value. The comprehensive leakage index of each capillary bifurcation structure was calculated using a weighted fusion formula. The comprehensive leakage index is transmitted to the global index unit. The bifurcation index unit contains the main vessel weight coefficient. The configuration interface allows clinical users to adjust the weighting coefficients according to specific application scenarios to adapt to the differences in the contribution ratio of leakage from main vessels and branch vessels in different disease types. It also supports batch comprehensive leakage index calculation for multiple capillary bifurcation structures and outputs the leakage index results of each bifurcation structure in a structured data format for fusion and summarization by the global index unit.

[0060] Global index unit 505 is configured to perform the function of step S105, specifically responsible for counting the total number of capillary bifurcation structures within the target area. The sum of the comprehensive leakage indices of all capillary bifurcation structures and the total number The ratio of the two values ​​is used as the overall fusion leakage index of the target area. The results are then output to a display device or data management system to characterize the degree of vascular leakage in the target area using a global fusion leakage index.

[0061] This invention also provides a computer-readable storage medium storing a computer program adapted to be loaded and executed by a processor to implement all steps of the aforementioned method for assessing vascular leakage based on capillary bifurcation blood flow changes, including image acquisition and preprocessing step S101, capillary bifurcation structure identification step S102, red blood cell flow calculation step S103, comprehensive leakage index calculation step S104, and global fusion leakage index calculation step S105. When the evaluation program stored in the computer-readable storage medium is executed by the processor, it can sequentially complete the entire calculation process of OCTA image preprocessing, bifurcation structure topology identification, light and dark encoded image extraction, red blood cell flow calculation, leakage component calculation, comprehensive leakage index weighted fusion, and global fusion leakage index output. The computer-readable storage medium can be a read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, hard disk, optical disk, or other types of tangible non-volatile storage media to ensure that the program and data are not lost after power failure and to support persistent storage and retrieval on computing devices.

[0062] This invention also provides a computer device, such as... Figure 6 As shown, the device includes a processor and a computer-readable storage medium. The processor is adapted to execute a computer program; the computer-readable storage medium stores the computer program, which, when executed by the processor, implements all the steps of the aforementioned method for assessing vascular leakage based on changes in capillary bifurcation blood flow. The computer device can be a clinical ophthalmology testing workstation, a server, or an embedded computing terminal, whose processor has sufficient computing power to quickly complete OCTA image preprocessing, bifurcation structure recognition, red blood cell flow calculation, and leakage index output, and transmit the assessment results to a display device or electronic medical record system interface.

[0063] The memory may include one or more non-volatile memories and one or more volatile memories. Examples of non-volatile memories include, but are not limited to, at least one of the following: read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, hard disk, optical disc (CD), digital video disc (DVD), or other magnetic and optical storage. Examples of volatile memories include, but are not limited to, at least one of the following: random access memory (RAM) or other volatile memories that do not persist during the duration of a power outage. In some embodiments, a program may be tangibly contained in a computer-readable medium, which may include in a device or other storage device accessible by the device, and the program may be loaded from the computer-readable medium into RAM for execution. The computer-readable medium may include any type of tangible non-volatile memory storing a computer program that, when executed by a processor, implements the above-described method for assessing vascular leakage based on changes in capillary bifurcation blood flow.

[0064] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented entirely or partially in the form of a computer program product. A computer program product includes one or more computer instructions that, when loaded and executed on a computer device, generate all or part of the flow or function according to the embodiments of the present invention. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. The computer-readable storage medium can be any available medium accessible by a computer device or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium, an optical medium, or a semiconductor medium.

[0065] To further illustrate the application process and evaluation effect of the technical solution of the present invention in actual clinical scenarios, the following detailed description is provided using specific application scenarios.

[0066] In the scenario of assessing vascular leakage in the macular region of diabetic retinopathy patients, clinicians use the method of this invention to quantitatively assess vascular leakage in the macular region of diabetic retinopathy patients. An OCTA acquisition device performs a non-invasive scan of the patient's macular region, simultaneously acquiring OCTA vascular images and dynamic blood flow images of the macular region. This eliminates the need for intravenous injection of fluorescein contrast agents, avoiding the risks of contrast agent-related allergic reactions and renal burden, making it suitable for diabetic patients requiring frequent follow-up. After preprocessing the OCTA vascular images, the computing device identifies multiple capillary bifurcation structures in the macular capillary network image. For each bifurcation structure, the main vessel, each first-order branch vessel, and each lower-order vessel are extracted with coded images of light and dark. Red blood cell flow in each vessel segment is calculated, and then the leakage components of the main vessel, each first-order branch, and the comprehensive leakage index of each bifurcation structure are calculated. Finally, a fused leakage index for the entire macular region is obtained. This index can quantitatively reflect abnormal vascular permeability by measuring the degree of imbalance in red blood cell flow at capillary bifurcation in the early stages when macular edema is not yet obvious. Compared with simply relying on OCT structural imaging to observe fluid accumulation, it has the potential for earlier leakage detection and can be used for early auxiliary diagnosis and follow-up monitoring of diabetic macular edema.

[0067] In the scenario of monitoring leakage in the retinal vascular lesion area of ​​patients with retinal vein occlusion, clinicians use the method of this invention to perform multiple non-invasive follow-up examinations at different treatment time points. During each follow-up, the OCTA acquisition device scans the retinal vascular lesion area, and the calculation device calculates the global fusion leakage index of the lesion area according to the complete process from steps S101 to S105. By comparing multiple time points before and after treatment. By tracking the changing trends, clinicians can quantitatively monitor the progression and regression of retinal vein occlusion-related leakage, and objectively evaluate the effects of anti-vascular endothelial growth factor therapy or laser photocoagulation on improving vascular leakage. This method eliminates the need for repeated injections of contrast agents and can be safely performed multiple times in a short period, providing a non-invasive and quantitative assessment tool for dynamic monitoring of treatment efficacy.

[0068] In the context of choroidal neovascularization leakage assessment for neovascular diseases, clinicians employ the method of this invention to assess active leakage in the choroidal neovascularization region of patients with neovascular diseases such as wet age-related macular degeneration. An OCTA acquisition device scans the choroidal neovascularization region, and the calculation device identifies capillary bifurcation structures within this region, calculates the comprehensive leakage index of each bifurcation structure, and fuses them to obtain a global fused leakage index. When neovascularization is active, the vessel wall structure is immature, the endothelial barrier function is incomplete, the imbalance of erythrocyte flow at capillary bifurcation is high, and the global fusion leakage index is large. When neovascularization tends to quiescent after treatment, the vessel wall structure gradually stabilizes, the conservation of erythrocyte flow improves, and the global fusion leakage index decreases accordingly. Dynamic monitoring is used to monitor this process. The changes can help clinicians assess the activity of choroidal neovascularization, providing quantitative reference for the timing and course management of anti-vascular endothelial growth factor therapy.

[0069] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention. The above are merely exemplary embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for assessing vascular leakage based on changes in capillary bifurcation blood flow, characterized in that, The process includes the following: Acquire optical coherence tomography (OCT) angiography images and corresponding blood flow dynamic images of the target region. Preprocess the OCT angiography images to obtain capillary network images of the target region. Identify at least one capillary bifurcation structure in a capillary network image, and determine the main vessel, first-level branch vessels, and the corresponding subordinate vessels of each capillary bifurcation structure, as well as the centerline and vessel boundary of each vessel segment. Based on dynamic blood flow images, the red blood cell flow rate of each segment of each capillary bifurcation structure was calculated. The comprehensive leakage index of each capillary bifurcation structure is calculated based on the difference between the red blood cell flow rate of the main vessel and the sum of the red blood cell flow rates of all first-order branch vessels in each capillary bifurcation structure, and the difference between the red blood cell flow rate of each first-order branch vessel and the sum of the red blood cell flow rates of the corresponding lower-level vessels. Based on the comprehensive leakage index of all capillary bifurcation structures, the global fusion leakage index of the target area is calculated, and the global fusion leakage index is used to characterize the degree of vascular leakage in the target area.

2. The method for assessing vascular leakage based on changes in capillary bifurcation blood flow as described in claim 1, characterized in that, Preprocessing of optical coherence tomography (OCT) angiography images includes: The optical coherence tomography (OCT) angiography images were sequentially processed with denoising, motion correction, image registration, vessel enhancement, and vessel segmentation to obtain capillary network images.

3. The method for assessing vascular leakage based on changes in capillary bifurcation blood flow as described in claim 1, characterized in that, Identifying capillary bifurcation structures, including: Determine the bifurcation point location of each capillary bifurcation structure, and divide the vessel segments corresponding to the main vessel, first-order branch vessels, and lower-order vessels based on the bifurcation point location.

4. The method for assessing vascular leakage based on changes in capillary bifurcation blood flow as described in claim 1, characterized in that, Calculating red blood cell flow based on dynamic blood flow images includes: Along the long axis of the corresponding blood vessel segment, the light and dark encoded image of the corresponding blood vessel segment as a function of time is extracted from the dynamic blood flow image, and the number of red blood cells passing through the corresponding blood vessel segment is counted based on the light and dark encoded image.

5. The method for assessing vascular leakage based on changes in capillary bifurcation blood flow as described in claim 4, characterized in that, After counting the number of red blood cells passing through based on the light and dark encoded image, the method also includes: obtaining the scanning time of the corresponding blood vessel segment, and using the ratio of the number of red blood cells passing through to the scanning time as the red blood cell flow rate of the corresponding blood vessel segment; The comprehensive leakage index is calculated by: calculating the first difference between the red blood cell flow rate of the main vessel and the sum of the red blood cell flow rates of all primary branch vessels, and using the ratio of the first difference to the red blood cell flow rate of the main vessel as the leakage component of the main vessel. For each primary branch vessel, calculate the second difference between the red blood cell flow rate of the corresponding primary branch vessel and the sum of the red blood cell flow rates of all the subordinate vessels corresponding to the primary branch vessel. The ratio of the second difference to the red blood cell flow rate of the corresponding primary branch vessel is used as the primary branch leakage component of the corresponding primary branch vessel.

6. The method for assessing vascular leakage based on changes in capillary bifurcation blood flow as described in claim 1, characterized in that, The comprehensive leakage index is calculated by including: The main vessel leakage component is assigned a first weight, and the average value of all first-level branch leakage components is assigned a second weight. The first weight is greater than the second weight. The product of the main vessel leakage component and the first weight is added to the product of the average value and the second weight to obtain the comprehensive leakage index of the corresponding capillary bifurcation structure.

7. The method for assessing vascular leakage based on changes in capillary bifurcation blood flow as described in claim 1, characterized in that, Calculating the global fusion leakage index includes: The total number of capillary bifurcation structures within the target area is counted, and the ratio of the sum of the comprehensive leakage indices of all capillary bifurcation structures to the total number is used as the global fusion leakage index of the target area. The target area is any one of the following: the macular region of the retina, the region of retinal vascular disease, or the region of choroidal neovascularization.

8. A vascular leakage assessment system based on changes in capillary bifurcation blood flow, characterized in that, include: The image acquisition unit is configured to: acquire optical coherence tomography angiography images and corresponding blood flow dynamic images of the target area, preprocess the optical coherence tomography angiography images, and obtain capillary network images of the target area. The bifurcation identification unit is configured to: identify at least one capillary bifurcation structure in a capillary network image, and determine the main vessel, first-level branch vessel, and the subordinate vessel corresponding to the first-level branch vessel in each capillary bifurcation structure, as well as the center line and vessel boundary of each vessel segment. The flow calculation unit is configured to calculate the red blood cell flow rate of each segment of each capillary bifurcation structure based on the blood flow dynamic image. The single bifurcation index unit is configured to calculate the comprehensive leakage index of each capillary bifurcation structure based on the difference between the red blood cell flow rate of the main vessel and the sum of the red blood cell flow rates of all first-level branch vessels in each capillary bifurcation structure, and the difference between the red blood cell flow rate of each first-level branch vessel and the sum of the red blood cell flow rates of the corresponding lower-level vessels. The global index unit is configured to calculate the global fusion leakage index of the target area based on the comprehensive leakage index of all capillary bifurcation structures, and use the global fusion leakage index to characterize the degree of vascular leakage in the target area.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program adapted to be loaded by a processor and executed as described in any one of claims 1 to 7 for assessing vascular leakage based on changes in capillary bifurcation blood flow.

10. A computer device, characterized in that, include: Processor and computer-readable storage media; A processor, adapted to execute computer programs; A computer-readable storage medium storing a computer program, which, when executed by the processor, implements the vascular leakage assessment method based on capillary bifurcation blood flow changes as described in any one of claims 1 to 7.