Microvascular analysis method, system and device for eye fundus image and medium

Abstract

The invention relates to the technical field of image processing, and discloses a fundus image microvessel analysis method, system and device and a medium, and the method comprises the steps: obtaining an OCTA image, and carrying out the layering processing of the OCTA image; for any layered image, identifying an optic disc surrounding atrophy area in the layered image; determining an initial analysis area in the layered image, and extracting a corrected analysis area except the atrophy area around the optic disc in the initial analysis area; identifying an upper blood vessel bow sub-region and a lower blood vessel bow sub-region in the corrected analysis region; determining a first blood vessel density of the upper blood vessel bow sub-region based on the first blood flow signal of the upper blood vessel bow sub-region; determining a second blood vessel density of the lower blood vessel bow sub-region based on the second blood flow signal of the lower blood vessel bow sub-region; an asymmetry indicator is determined based on the first blood vessel density and the second blood vessel density. The method has the beneficial effect that more accurate and more specific quantitative evaluation of the fundus capillary state is realized.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method, system, device and medium for microvascular analysis of fundus images. Background Technology

[0002] Optical coherence tomography angiography (OCTA), a non-invasive, high-resolution retinal blood flow imaging technique, has become an important tool for visualizing and quantitatively assessing the microvascular system of the retina and choroid. Vascular density, as a primary parameter for assessing microvascular status, is widely used in the quantitative analysis of OCTA images. However, related techniques suffer from inherent limitations when processing retinal images containing peripapillary atrophy (PPA), including distorted vascular density measurements and an inability to avoid PPA interference that reduces the specificity of regional analyses. Summary of the Invention

[0003] This application provides a method, system, device, and medium for microvascular analysis of fundus images, which solves the technical problem of insufficient accuracy in related microvascular assessments and achieves a more accurate and specific quantitative assessment of the state of fundus microvessels.

[0004] To achieve the above objectives, the main technical solutions adopted in this application include: In a first aspect, embodiments of this application provide a method for microvascular analysis of fundus images. The method includes: acquiring an OCTA image of a target eye and performing layered processing on the OCTA image to obtain at least one layered image; for any layered image, identifying a peridiscal atrophy region in the layered image; determining an initial analysis region in the layered image and extracting a corrected analysis region from the initial analysis region excluding the peridiscal atrophy region; identifying an upper vascular arch region and an lower vascular arch region in the corrected analysis region; determining a first vascular density in the upper vascular arch region based on a first blood flow signal in the upper vascular arch region; determining a second vascular density in the lower vascular arch region based on a second blood flow signal in the lower vascular arch region; and determining an asymmetry index for quantifying the microvascular distribution between the upper and lower vascular arch regions based on the first and second vascular densities.

[0005] The microvascular analysis method for fundus images provided in this application, through layered processing of OCTA images, enables anatomical-level specificity in the analysis, allowing for independent evaluation of vascular networks at different depths. By identifying and excluding atrophic areas around the optic disc, the interference of avascular structures on the quantitative results is effectively eliminated, improving the purity of the data source and the accuracy of measurements. By precisely defining the superior and inferior vascular arch regions within the corrected area, the analysis focuses on the core functional areas with rich blood supply. By calculating and comparing the vascular density of the superior and inferior vascular arch regions, an asymmetric index for quantifying the uniformity of microvascular distribution is generated, achieving a more precise and specific quantitative assessment of the fundus microvascular status.

[0006] Optionally, determining the initial analysis region in the layered image includes: identifying the optic disc center and the macula center in the OCTA image; using the line segment connecting the optic disc center and the macula center as a baseline; using the midpoint of the baseline as the geometric center, generating a rectangular region with its long side parallel to the baseline according to a preset size, and determining the rectangular region as the initial analysis region.

[0007] Using the optic disc center and macula center—the two most stable and easily identifiable core anatomical landmarks in fundus images—as reference points, a stable coordinate system was created that is independent of individual axial length, refractive state, or optic disc morphology variations, ensuring the consistency of the analytical baseline. A standardized rectangular region of preset size was generated centered on the midpoint of the baseline, limiting the analysis to the central region with the highest image quality and greatest stability. This prevented low-quality data from being mixed into the quantitative analysis and improved the reliability of the results.

[0008] Optionally, identifying the upper and lower vascular arch regions in the corrected analysis region includes: extracting a first rectangular sub-region located above the baseline in the corrected analysis region, and determining the extracted first rectangular sub-region as the upper vascular arch region; extracting a second rectangular sub-region located below the baseline in the corrected analysis region, and determining the extracted second rectangular sub-region as the lower vascular arch region; wherein the upper and lower vascular arch regions are symmetrical about the baseline.

[0009] Within the calibrated analysis area after eliminating PPA interference, rectangular sub-regions were delineated above and below the baseline to ensure that the analysis was completely focused on the most abundant and core retinal vascular arch distribution area. This eliminated potential signal dilution or interference from surrounding non-core areas, allowing the subsequently extracted microvascular parameters to highly specifically represent the physiological state of the core functional area. The symmetrical design of the upper and lower vascular arch sub-regions about the baseline ensured that the two sub-regions were completely identical in size, shape, and distance from the baseline, eliminating measurement biases that might be introduced by differences in the analysis window itself. This ensured that the final extracted asymmetry indicators accurately and sensitively reflected the spatial distribution characteristics of microvessels.

[0010] Optionally, determining a first vascular density in the upper vascular arch region based on a first blood flow signal in the upper vascular arch region; and determining a second vascular density in the lower vascular arch region based on a second blood flow signal in the lower vascular arch region, includes: calculating a first percentage between the pixel area of ​​the first blood flow signal and the pixel area of ​​the upper vascular arch region, and determining the first percentage as the first vascular density; calculating a second percentage between the pixel area of ​​the second blood flow signal and the pixel area of ​​the lower vascular arch region, and determining the second percentage as the second vascular density.

[0011] Blood vessel density is determined by calculating the percentage of blood flow signal pixel area to the sub-region area, transforming image data into an objective and standardized quantitative indicator. Since the purity of the upper and lower vascular arch sub-regions has been ensured through previous steps, the calculated first and second vessel densities can characterize the effective microvascular perfusion ratio of this sub-region. This provides a solid and reliable data foundation for subsequent quantification of its spatial differences (i.e., asymmetry).

[0012] Optionally, the asymmetry index includes an absolute value asymmetry index, which is determined as follows: the difference between the first vascular density and the second vascular density is calculated; the absolute value of the difference is taken, and the result of taking the absolute value is determined as the absolute value asymmetry index.

[0013] The absolute value asymmetry index is defined as the absolute value of the difference between the first and second vessel densities. This index directly quantifies the absolute separation of microvessel density between regions, and its calculation method ensures that it combines intuitive results, stable measurement, and unbiasedness in the direction of difference.

[0014] Optionally, the asymmetry index further includes a relative asymmetry index, which is determined as follows: calculating the arithmetic mean between the first vascular density and the second vascular density; calculating the ratio between the absolute asymmetry index and the arithmetic mean, and determining the ratio as the relative asymmetry index.

[0015] The relative asymmetry index is obtained by dividing the absolute asymmetry index by the arithmetic mean of the vascular density in the upper and lower vascular arch regions. This transforms the absolute difference into a standardized, unitless ratio. This process eliminates comparison bias caused by differences in overall perfusion levels among individuals, thus enabling fairer cross-individual comparisons.

[0016] Optionally, the at least one layered image includes: a retinal vascular layer image and / or a choroidal vascular layer image.

[0017] By layering OCTA images, it is possible to ensure that subsequent quantitative analysis corresponds to specific anatomical structures, thus making the analysis results anatomically and physiologically specific.

[0018] Secondly, embodiments of this application provide a microvascular analysis system for fundus images. The system includes: an image layering module for acquiring OCTA images of a target eye and performing layering processing on the OCTA images to obtain at least one layered image; a region correction module for identifying a peripapillary atrophy region in any layered image; determining an initial analysis region in the layered image and extracting a corrected analysis region from the initial analysis region excluding the peripapillary atrophy region; a sub-region identification module for identifying a superior vascular arch region and a inferior vascular arch region in the corrected analysis region; a density determination module for determining a first vascular density in the superior vascular arch region based on a first blood flow signal in the superior vascular arch region; and a second vascular density in the inferior vascular arch region based on a second blood flow signal in the inferior vascular arch region; and an index determination module for determining an index for quantifying the asymmetry of microvascular distribution between the superior and inferior vascular arch regions based on the first and second vascular densities.

[0019] Thirdly, embodiments of this application provide a computer device, including: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the above-described microvascular analysis method for fundus images.

[0020] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer instructions, which are used to cause a computer to execute the above-described microvascular analysis method for fundus images.

[0021] Fifthly, embodiments of this application provide a computer program product, including computer instructions, which are used to cause a computer to execute the above-described microvascular analysis method for fundus images. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0023] Figure 1 A flowchart of a microvascular analysis method for fundus images provided in this application embodiment; Figure 2 (A) A schematic diagram of a frontal projection image provided for an embodiment of this application; Figure 2 (B) A schematic diagram of layered images provided for an embodiment of this application; Figure 2 (C) is a schematic diagram of layered images provided in an embodiment of this application; Figure 2 (D) is a schematic diagram of layered images provided in an embodiment of this application; Figure 2 (E) is a schematic diagram of layered images provided in an embodiment of this application; Figure 2 (F) is a schematic diagram of layered images provided in an embodiment of this application; Figure 3 (A) is a schematic diagram of vascular density quantification provided in an embodiment of this application; Figure 3 (B) is a schematic diagram of vascular density quantification provided in an embodiment of this application; Figure 4 This is a schematic diagram of the corrected analysis region provided in an embodiment of this application; Figure 5 (A) is a schematic diagram of the upper and lower vascular density provided in an embodiment of this application; Figure 5 (B) is a schematic diagram of the upper and lower vascular arch regions provided in the embodiments of this application; Figure 6 This is a schematic diagram of a microvascular analysis system provided in an embodiment of this application; Figure 7 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0025] Optical coherence tomography angiography (OCTA), a non-invasive, high-resolution retinal blood flow imaging technique, has become an important tool for visualizing and quantitatively assessing the microvascular system of the retina and choroid. By detecting signal changes caused by blood flow, this technique can generate three-dimensional images of the layered capillary networks of the fundus without the need for contrast agents, providing a rich foundation of imaging data for the objective assessment of ocular circulation.

[0026] In quantitative analysis of microvessels in OCTA images, especially in the peripheral region of the optic disc surrounding the optic nerve head, vessel density is one of the most commonly used quantitative parameters. Common quantitative methods typically rely on pre-set, fixed analysis templates. For example, a standard ring-shaped region or an equal-area grid partition is generated based on the geometric center of the optic disc, and the proportion of pixels occupied by blood flow signal within each partition is calculated as the vessel density value for that region. However, this standardized analysis process based on fixed templates reveals significant technical limitations when dealing with fundus conditions exhibiting related structural changes. Specifically, peripapillary atrophy (PPA) is a common morphological change in the fundus. Histologically, PPA regions show atrophy of the retinal pigment epithelium and choroidal capillary layer; on OCTA images, they appear as avascular areas with absent or significantly reduced blood flow signals. When calculating overall vascular density using a fixed analysis region containing PPA (Polymeric Propagated Avascular Area), the avascular region is included in the denominator of the total area, systematically "diluting" the proportion of blood flow signal in the whole or a local area. This results in a calculated vascular density value that is significantly lower than the actual microvascular density of perfused retinal tissue. This measurement bias introduced by PPA makes comparisons of vascular density between different individuals or between follow-ups of the same individual inaccurate. Due to the high individual variability in the extent, morphology, and location of PPA, fixed zones are often infiltrated to varying degrees by PPA regions of different sizes. This results in each zone containing both vascularized retinal tissue and avascular atrophic areas, with inconsistent degrees of PPA "contamination" across zones. This contamination not only severely interferes with the measurement of absolute vascular density within a zone but also significantly reduces the reliability of zone comparisons (such as vertical asymmetry analysis) aimed at revealing regional blood flow differences, because the observed differences may primarily stem from uneven PPA distribution rather than genuine physiological changes in microvascular status. Currently, while related technologies recognize that PPA (proximal retinal endothelial lesions) is a significant confounding factor in OCTA (Optical Cord Apparatus Therapy) quantitative analysis, a systematic technical solution that can accurately adapt to individual fundus anatomical variations is lacking to effectively identify and eliminate interference from PPA regions before quantification. Most related methods either completely ignore this influence or only perform coarse, rule-based approximations, failing to achieve precise matching with the actual complex morphology of PPA. Therefore, they cannot fundamentally solve the measurement bias problem and are unable to perform highly specific and sensitive microvascular regionalization analysis within the "pure" functional retinal region after eliminating PPA interference.

[0027] In summary, existing OCTA microvascular quantitative analysis methods, when applied to fundus images containing PPA (Potentially Painful Vessel) areas, suffer from inherent technical limitations. These include distortion of vascular density measurements due to the inclusion of avascular atrophy regions in the calculation, and a decrease in regional analysis specificity due to the inability of fixed partitioning patterns to avoid PPA interference. This restricts the ability to accurately and reliably assess the fundus microcirculation status. Therefore, there is an urgent need to develop a method capable of identifying PPA regions and performing precise microvascular quantification and analysis after excluding these regions, in order to improve the accuracy of OCTA image quantitative analysis.

[0028] This application provides a method for microvascular analysis of fundus images. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0029] Please refer to Figure 1 , Figure 1 A flowchart of a microvascular analysis method for fundus images provided in this application embodiment is shown below. Figure 1 As shown, the process includes the following steps: Step S1: Obtain an OCTA image of the target eye and perform layer processing on the OCTA image to obtain at least one layered image.

[0030] The target eye can be a human or an animal eye. In practical applications, OCTA images refer to wide-angle OCTA images of the posterior pole acquired using scanned-frequency optical coherence tomography (OCT). Layered processing refers to the process of segmenting three-dimensional OCTA image data along the depth direction into two-dimensional frontal projection images corresponding to different anatomical layers, based on the reflection signal characteristics of different retinal and choroidal structures, to accurately analyze the microvascular networks of different anatomical layers. This provides the possibility for subsequent precise exclusion of interference areas and quantification of microvascular features on specific vascular layers. In practical applications, layered processing typically covers the main vascular layers of the retina and choroid, including the superficial retinal capillary plexus, deep capillary plexus, full-thickness retina, and choroidal capillary layer.

[0031] Step S3: For any layered image, identify the peridiscal atrophy region in the layered image; determine the initial analysis region in the layered image, and extract the corrected analysis region from the initial analysis region excluding the peridiscal atrophy region.

[0032] In quantitative microvascular analysis, by clearly separating and defining the core structural confounding factor, PPA, from layered images, a standardized and image-quality-stable corrected analysis region is obtained. This ensures that the final obtained microvascular parameters accurately reflect the state of perfused retinal tissue, rather than being misled by avascular atrophy areas, thus improving the accuracy and specificity of the analysis.

[0033] Step S5: Identify the upper vascular arch region and the lower vascular arch region in the corrected analysis region.

[0034] By further selecting the most retinal blood supply and core functional area (vascular arch region) within the calibrated analysis area for measurement, the final microvascular measurement parameters specifically represent the functional state of the retinal core circulation, thus improving the physiological relevance of the parameters.

[0035] Step S7: Based on the first blood flow signal of the upper vascular arch region, determine the first vascular density of the upper vascular arch region; based on the second blood flow signal of the lower vascular arch region, determine the second vascular density of the lower vascular arch region.

[0036] The first and second vascular densities are quantitative indicators of the microvascular richness within the superior and inferior vascular arch regions, respectively. By completely eliminating PPA interference, the calculated vascular densities of the superior and inferior vascular arch regions can highly specifically characterize the true microvascular perfusion level of the corresponding regions, providing a high-quality data foundation for subsequent accurate quantification of their asymmetry.

[0037] Step S9: Based on the first blood vessel density and the second blood vessel density, determine an asymmetry index for quantifying the microvascular distribution between the upper vascular arch region and the lower vascular arch region.

[0038] By using the first and second vascular densities, the asymmetry of microvascular distribution in the upper and lower vascular arch regions is quantified, transforming complex morphological information into a single, objective quantitative indicator, thus promoting the standardization and automation of the analysis.

[0039] The microvascular analysis method for fundus images provided in this application, through layered processing of OCTA images, enables anatomical-level specificity in the analysis, allowing for independent evaluation of vascular networks at different depths. By identifying and excluding atrophic areas around the optic disc, the interference of avascular structures on the quantitative results is effectively eliminated, improving the purity of the data source and the accuracy of measurements. By precisely defining the superior and inferior vascular arch regions within the corrected area, the analysis focuses on the core functional areas with rich blood supply. By calculating and comparing the vascular density of the superior and inferior vascular arch regions, an asymmetric index for quantifying the uniformity of microvascular distribution is generated, achieving a more precise and specific quantitative assessment of the fundus microvascular status.

[0040] In some specific embodiments, determining the initial analysis region in the layered image includes: identifying the optic disc center and the macula center in the OCTA image; using the line segment connecting the optic disc center and the macula center as a baseline; using the midpoint of the baseline as the geometric center, generating a rectangular region with its long side parallel to the baseline according to a preset size, and determining the rectangular region as the initial analysis region.

[0041] Using the optic disc center and macula center—the two most stable and easily identifiable core anatomical landmarks in fundus images—as reference points, a stable coordinate system was created that is independent of individual axial length, refractive state, or optic disc morphology variations, ensuring the consistency of the analytical baseline. A standardized rectangular region of preset size was generated centered on the midpoint of the baseline, limiting the analysis to the central region with the highest image quality and greatest stability. This prevented low-quality data from being mixed into the quantitative analysis, improving the reliability of the analysis results.

[0042] In some specific embodiments, identifying the upper and lower vascular arch regions in the corrected analysis region includes: extracting a first rectangular sub-region located above the baseline in the corrected analysis region, and determining the extracted first rectangular sub-region as the upper vascular arch region; extracting a second rectangular sub-region located below the baseline in the corrected analysis region, and determining the extracted second rectangular sub-region as the lower vascular arch region; wherein the upper and lower vascular arch regions are symmetrical about the baseline.

[0043] Within the calibrated analysis area after eliminating PPA interference, rectangular sub-regions were delineated above and below the baseline to ensure that the analysis was fully focused on the most abundant and core retinal vascular arch distribution area. This eliminated potential signal dilution or interference from surrounding non-core areas, allowing the subsequently extracted microvascular parameters to highly specifically represent the physiological state of the core functional area. The symmetrical design of the upper and lower vascular arch sub-regions about the baseline ensured that the two sub-regions were completely identical in size, shape, and distance from the baseline, eliminating measurement biases that might be introduced by differences in the analysis window itself. This ensured that the final calculated asymmetry index accurately and sensitively reflected the spatial distribution characteristics of microvessels.

[0044] In some specific embodiments, determining a first vascular density of the upper vascular arch region based on a first blood flow signal of the upper vascular arch region; and determining a second vascular density of the lower vascular arch region based on a second blood flow signal of the lower vascular arch region, includes: calculating a first percentage between the pixel area of ​​the first blood flow signal and the pixel area of ​​the upper vascular arch region, and determining the first percentage as the first vascular density; calculating a second percentage between the pixel area of ​​the second blood flow signal and the pixel area of ​​the lower vascular arch region, and determining the second percentage as the second vascular density.

[0045] Blood vessel density is determined by calculating the percentage of blood flow signal pixel area to the sub-region area, transforming image data into an objective and standardized quantitative indicator. Since the purity of the upper and lower vascular arch sub-regions has been ensured through previous steps, the calculated first and second vessel densities can characterize the effective microvascular perfusion ratio of this sub-region. This provides a solid and reliable data foundation for subsequent quantification of its spatial differences (i.e., asymmetry).

[0046] In some specific embodiments, the asymmetry index includes an absolute value asymmetry index, which is determined as follows: the difference between the first vascular density and the second vascular density is calculated; the absolute value of the difference is taken, and the result of taking the absolute value is determined as the absolute value asymmetry index.

[0047] The absolute value asymmetry index is defined as the absolute value of the difference between the first and second vessel densities. This index directly quantifies the absolute separation of microvessel density between regions, and its calculation method ensures that it combines intuitive results, stable measurement, and unbiasedness in the direction of difference.

[0048] In some specific embodiments, the asymmetry index further includes a relative asymmetry index, which is determined as follows: calculating the arithmetic mean between the first vascular density and the second vascular density; calculating the ratio between the absolute asymmetry index and the arithmetic mean, and determining the ratio as the relative asymmetry index.

[0049] The relative asymmetry index is obtained by dividing the absolute asymmetry index by the arithmetic mean of the vascular density in the upper and lower vascular arch regions. This transforms the absolute difference into a standardized, unitless ratio. This process eliminates comparison bias caused by differences in overall perfusion levels among individuals, thus enabling fairer cross-individual comparisons.

[0050] In some specific embodiments, the at least one layered image includes: a retinal vascular layer image and / or a choroidal vascular layer image.

[0051] The retinal vascular layer images include, but are not limited to: images of the superficial retinal capillary plexus, images of the deep retinal capillary plexus, and full-thickness retinal blood flow images obtained by integrating blood flow signals from all layers of the retina. The choroidal vascular layer images include, but are not limited to: images of the choroidal capillary layer and images of the large vessels in the choroid. By processing OCTA images in layers, it is possible to ensure that subsequent quantitative analysis corresponds to specific anatomical structures, thus giving the analysis results anatomical and physiological specificity.

[0052] In some specific embodiments, this application provides a method for microvascular analysis of fundus images, the method specifically including the following steps: 1) Acquiring wide-angle OCTA images of the posterior pole and performing layered processing. Specifically, a swept-source OCTA device is used to scan the target eye. The scanning center is set to the midpoint of the line connecting the center of the macula and the center of the optic disc, and the scanning range is, for example, 15 mm × 15 mm, to obtain wide-angle blood flow imaging data covering key structures in the posterior pole (optic disc, macula, and major vascular arches). Please refer to [reference needed]. Figure 2 (A)-(F), Figure 2 (A) is a schematic diagram of the original blood flow projection image provided in the embodiment of this application, wherein the red arrow represents atrophy around the optic disc. Figure 2 (B)-(F) Schematic diagrams of layered images provided in embodiments of this application. The system automatically generates different layers of the retina, including: the superficial capillary layer of the retina (e.g., the retinal capillary layer). Figure 2 (As shown in (B)), the deep capillary layer of the retina (such as...) Figure 2 (C) shown), full-thickness retina (as shown) Figure 2 (D) shown), choroidal capillary layer (as shown) Figure 2 (E) shown) and the large blood vessel layer in the choroid (as shown) Figure 2 (F) shows the frontal projection (en face) blood flow image. Only images whose image quality scores meet the preset standards are selected for subsequent analysis.

[0053] 2) Define the initial analysis region and identify the PPA region. Please refer to [reference needed]. Figure 3 (A), Figure 3 (A) is a schematic diagram of vascular density quantification provided in an embodiment of this application. On the layered image, firstly, an initial analysis region is defined, and the optic disc center point in the layered image is determined (e.g., ...). Figure 3 (As shown by the blue cross in (A)) and the center point of the macula (as shown in (A)). Figure 3 (As shown by the red cross in (A)), connecting the two points yields a baseline. The system centers on the midpoint of this baseline (e.g., Figure 3 (As shown by the green cross in (A)), a rectangular area (e.g., 12mm × 11mm) with its long side parallel to the baseline is automatically generated. To facilitate partitioning and quantization, this rectangular area is uniformly divided into an M-row × N-column grid array, for example, 8 rows × 8 columns (64 small cells, e.g., ...). Figure 3 (As shown by the white squares in (A)). The area covered by this grid is the initial analysis region. Next, identify the PPA region. Please refer to... Figure 3 (B) Figure 3 (B) is a schematic diagram of vascular density quantification provided in an embodiment of this application. Figure 3 As shown in (B), the boundaries of the atrophic region around the optic disc are identified and delineated on the OCTA image or its registered fundus image (a fundus photograph image that is spatially aligned with the OCTA image at the pixel level). Figure 3 (B) The yellow dashed area indicates the identified PPA area. Figure 3 (A) and Figure 3 (B) The comparison shows the changes in the vascular density values ​​of each small grid in the choroidal capillary layer before and after excluding the PPA region.

[0054] 3) Generate the corrected analysis region. To eliminate the systematic interference of the avascular or low-vascularity PPA structure on the calculation of microvascular parameters, the identified PPA regions are excluded from the initial analysis region. Specifically, in subsequent calculations, all portions (pixels or entire small grids) of the initial analysis region's grid that fall within the PPA region are ignored or masked. The effective analysis region obtained after this processing, which does not contain any PPA pixels, is the "corrected analysis region." Please refer to [reference needed]. Figure 4 , Figure 4 This is a schematic diagram of the corrected analysis area provided in an embodiment of this application. To reduce the influence of surrounding avascular areas within the scanning range, this embodiment of the application selects... Figure 4The central 6×6 grid in the 8×8 grid shown is used as the research and analysis area for each layer of the image.

[0055] 4) Within the corrected analysis area, delineate the superior and inferior vascular arch regions. Define the baseline based on the optic disc-macula line. Please refer to [reference needed]. Figure 5 (A), Figure 5 (A) is a schematic diagram of the upper and lower vascular density provided in an embodiment of this application. Figure 5 As shown in (A), the 3×6 grid area above the baseline, i.e., the area within the blue box, has a vessel density defined as the upper hemifield vessel density (SHVD). The 3×6 grid area below the baseline, i.e., the area within the yellow box, has a vessel density defined as the lower hemifield vessel density (IHVD).

[0056] To avoid interference from the edges of the PPA and focus on the core functional area rich in blood vessels, a continuous grid block completely outside the PPA region and covering the course of the major vascular arches above the retina was selected above the baseline as the superior vascular arch region; a continuous grid block completely outside the PPA region and covering the course of the major vascular arches below the retina was selected below the baseline as the inferior vascular arch region. Please refer to [reference needed]. Figure 5 (B) Figure 5 (B) is a schematic diagram of the upper and lower vascular arch regions provided in an embodiment of this application. Figure 5 As shown in (B), excluding the PPA region, the area above the baseline containing the upper vascular arch (the area within the purple box) is defined as the upper vascular arch region, and its vascular density is defined as the upper half vascular density (aSHVD, Superior Hemifield Vessel Density of the Vascular Arcade). Excluding the PPA region, the area below the baseline containing the lower vascular arch (the area within the green box) is defined as the lower half vascular density (aIHVD, Inferior Hemifield Vessel Density of the Vascular Arcade).

[0057] 5) Calculate the vascular density of the vascular arch regions. Calculate the vascular density of the upper and lower vascular arch regions separately. Vascular density is defined as the percentage of pixel area occupied by OCTA blood flow signals within a sub-region relative to the total pixel area of ​​the sub-region. This calculation process can be performed separately on multiple layers of the retina, such as the surface, deep, and full layers, to obtain layered microvascular information.

[0058] 6) Calculate asymmetry indices characterizing microvascular distribution. Based on the obtained vascular density values ​​of the superior and inferior vascular arch regions, calculate asymmetry parameters quantifying their differences, mainly including: absolute asymmetry index and relative asymmetry index. These parameters objectively reflect the uniformity of blood flow distribution in the superior and inferior core vascular network on the temporal side of the retina. The absolute asymmetry index reflects the absolute difference in vascular density between the superior and inferior vascular arch regions, while the relative asymmetry index reflects the relative magnitude of this difference with respect to the average vascular density.

[0059] The microvascular analysis method for fundus images provided in this application avoids the "dilution" effect of PPA (porphyria perforata) on microvascular density measurements by actively identifying and excluding PPA regions. It focuses on analyzing the functional core area rich in blood flow by precisely delineating the anatomically significant superior and inferior vascular arches. The resulting vascular density and asymmetry indices are a purer, more stable, and repeatable set of quantitative parameters, providing numerical basis for accurate assessment of fundus microvascular status.

[0060] Accordingly, please refer to Figure 6 , Figure 6 This application provides a schematic diagram of a microvascular analysis system for fundus images, as shown in the embodiments. Figure 6 As shown, the system includes: an image acquisition module for acquiring OCTA images of the target eye; an image layering module for performing layering processing on the OCTA images to obtain at least one layered image; a region recognition module for identifying the peripapillary atrophy region in any layered image; a region correction module for determining an initial analysis region in the layered image and extracting a corrected analysis region from the initial analysis region excluding the peripapillary atrophy region; a sub-region recognition module for identifying the superior vascular arch region and the inferior vascular arch region in the corrected analysis region; a density determination module for determining a first vascular density of the superior vascular arch region based on a first blood flow signal of the superior vascular arch region; and a second vascular density of the inferior vascular arch region based on a second blood flow signal of the inferior vascular arch region; and an index determination module for determining an index for quantifying the asymmetry of microvascular distribution between the superior and inferior vascular arch regions based on the first and second vascular densities.

[0061] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0062] The microvascular analysis system in this embodiment is presented in the form of functional units. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0063] Please see Figure 7 , Figure 7 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application, such as... Figure 7 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 7 Take a processor 10 as an example.

[0064] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0065] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0066] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0067] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0068] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0069] This application also provides a computer-readable storage medium. The methods described in this application can be implemented in hardware or firmware, or implemented as recordable on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and subsequently stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code when the software or computer code is accessed by the computer, processor, or hardware. This application provides a computer program product including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the method of any embodiment of this application.

[0070] The systems and modules described in the above embodiments can be implemented by computer chips or physical entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0071] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0072] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0073] This application is described with reference to flowchart illustrations and / or block diagrams of methods, systems, and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0074] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0075] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0076] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0077] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on its differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0078] The above description is merely an embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of this application should be included within the scope of the claims of this application.

[0079] Although embodiments of this application have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of this application, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method of microvessel analysis of an ocular fundus image, characterized by, The method includes: Acquire an OCTA image of the target eye and perform layer processing on the OCTA image to obtain at least one layered image; For any layered image, identify the peripapillary atrophy region in the layered image; determine the initial analysis region in the layered image, and extract the corrected analysis region from the initial analysis region excluding the peripapillary atrophy region; Identify the upper and lower vascular arch regions within the corrected analysis area; Based on the first blood flow signal of the upper vascular arch region, the first vascular density of the upper vascular arch region is determined; based on the second blood flow signal of the lower vascular arch region, the second vascular density of the lower vascular arch region is determined. Based on the first vascular density and the second vascular density, an index is determined to quantify the asymmetry of microvascular distribution between the upper vascular arch region and the lower vascular arch region.

2. The method of claim 1, wherein, Determining the initial analysis region in the layered image includes: Identify the optic disc center and macular center in the OCTA image; The line segment connecting the center of the optic disc and the center of the macula is used as the baseline; Using the midpoint of the baseline as the geometric center, a rectangular region with its long side parallel to the baseline is generated according to a preset size, and the rectangular region is determined as the initial analysis region.

3. The method according to claim 2, characterized in that, Identifying the upper and lower vascular arch regions within the corrected analysis region includes: Extract the first rectangular sub-region located above the baseline in the corrected analysis region, and determine the extracted first rectangular sub-region as the upper vascular arch sub-region; Extract the second rectangular sub-region located below the baseline in the corrected analysis region, and determine the extracted second rectangular sub-region as the lower vascular arch sub-region; The upper vascular arch region and the lower vascular arch region are symmetrical about the baseline.

4. The method according to claim 1, characterized in that, Based on the first blood flow signal in the upper vascular arch region, a first vascular density in the upper vascular arch region is determined; based on the second blood flow signal in the lower vascular arch region, a second vascular density in the lower vascular arch region is determined, including: Calculate a first percentage between the pixel area of ​​the first blood flow signal and the pixel area of ​​the upper vascular arch region, and determine the first percentage as the first vascular density; Calculate a second percentage between the pixel area of ​​the second blood flow signal and the pixel area of ​​the lower vascular arch region, and determine the second percentage as the second vascular density.

5. The method according to claim 1, characterized in that, The asymmetry index includes an absolute value asymmetry index, which is determined as follows: Calculate the difference between the first blood vessel density and the second blood vessel density; The difference is processed by taking the absolute value, and the result of taking the absolute value is determined as the absolute value asymmetry index.

6. The method according to claim 5, characterized in that, The asymmetry index also includes a relative asymmetry index, which is determined as follows: Calculate the arithmetic mean between the first blood vessel density and the second blood vessel density; Calculate the ratio between the absolute value asymmetry index and the arithmetic mean, and determine the ratio as the relative asymmetry index.

7. The method according to claim 1, characterized in that, The at least one layered image includes: a retinal vascular layer image and / or a choroidal vascular layer image.

8. A microvascular analysis system for fundus images, characterized in that, The system includes: An image layering module is used to acquire an OCTA image of the target eye and perform layering processing on the OCTA image to obtain at least one layered image; The region correction module is used to identify the peripapillary atrophy region in any layered image; determine the initial analysis region in the layered image; and extract the corrected analysis region from the initial analysis region excluding the peripapillary atrophy region. The sub-region identification module is used to identify the upper vascular arch sub-region and the lower vascular arch sub-region in the corrected analysis region; The density determination module is used to determine the first vascular density of the upper vascular arch region based on the first blood flow signal of the upper vascular arch region; and to determine the second vascular density of the lower vascular arch region based on the second blood flow signal of the lower vascular arch region. The index determination module is used to determine an index for quantifying the asymmetry of microvascular distribution between the upper vascular arch region and the lower vascular arch region based on the first vascular density and the second vascular density.

9. A computer device, characterized in that, include: A memory and a processor are communicatively connected, the memory storing computer instructions, and the processor executing the computer instructions to perform the microvascular analysis method of fundus images according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the microvascular analysis method of the fundus image according to any one of claims 1 to 7.

Images