Method and apparatus for determining cataracts in animal

Abstract

According to one embodiment of the present disclosure, a method for determining the progression of cataracts in an animal may be provided, comprising the steps of: receiving an eye image of the animal from a user device; extracting a lens region from the eye image; detecting a Y-suture structure of the lens within the lens region; normalizing the image characteristics of the detected Y-suture structure with a reference value of the lens region to calculate one or more quantitative indicators representing the relative visibility of the Y-suture; determining the progression of the cataract based on the quantitative indicators; and transmitting the determined progression of the cataract to the user device.

Description

Technology Field

[0001] The present invention relates to a method and apparatus for determining the degree of cataract progression in an animal, and more specifically, to a method and apparatus for determining the degree of cataract progression in an animal based on a Y suture structure detected from an image of the animal's eye. Background Technology

[0002] Cataracts are a condition in which the lens becomes cloudy, and they are a common ophthalmic disease in companion animals, including dogs and cats. As cataracts progress, they can lead to vision loss and blindness, and cause complications such as lens-induced uveitis due to the leakage of proteins from the clouded lens, as well as secondary glaucoma. Therefore, early detection and accurate assessment of the stage of progression are crucial for determining the timing of medication or surgery and for preventing complications.

[0003] Traditionally, the diagnosis of cataracts in companion animals relied primarily on veterinarians directly observing the lens using specialized equipment, such as slit-lamp microscopes, to determine the stage of progression. However, this method had limitations as it required specialized equipment and skilled personnel, and examinations could only be performed by visiting an animal hospital, making it difficult for owners to frequently check the onset and progression of cataracts at home.

[0004] To address this, attempts have been made to automatically estimate the presence or severity of cataracts from eye images. For instance, methods utilizing learning models such as convolutional neural networks to directly estimate the degree of lens opacity from eye images are known. However, methods that directly measure the absolute brightness or degree of opacity are sensitive to differences in lighting, exposure, white balance, imaging equipment, and shooting angle during capture; consequently, there is a problem where measurement values ​​vary significantly depending on the shooting conditions, even for the same eye. In particular, since the quality and lighting conditions of videos directly captured by guardians using devices like smartphones are inconsistent, it is difficult to ensure the reliability of judgments based on absolute values.

[0005] Furthermore, if the learning model fails to provide the basis for its judgment, the results are difficult for humans to interpret and trust, leading to low clinical acceptance. Moreover, nuclear sclerosis, in which the lens nucleus hardens with aging, is difficult to distinguish from true cataracts due to their similar appearance; additionally, animals do not cooperate with examinations, and the appearance of normal lenses varies by species or breed, all of which contribute to lowering the accuracy of automated diagnosis.

[0006] Therefore, a technology is required that is robust to variations in the imaging environment, allows for the interpretation of the basis for judgment, and can consistently determine the stage of cataract progression. Prior art literature

[65535] Republic of Korea Published Patent Application No. 10-2026-0023306 (February 20, 2026) The problem to be solved

[0007] The embodiments of the present invention aim to provide a method and apparatus for determining the progression of cataracts in an animal without visiting an animal hospital, using images captured by a user with a smartphone or the like.

[0008] The embodiments of the present invention aim to provide a method and apparatus for determining the progression of cataracts in animals, which can distinguish conditions that cannot be distinguished by visibility alone, by considering together a quantitative indicator representing the visibility of the Y suture and an indicator of opacity in the lens region.

[0009] The embodiments of the present invention aim to provide a method and apparatus for determining the degree of cataract progression in an animal, which can predict future progression trends by calculating the rate of cataract progression based on changes in quantitative indicators derived from images taken at different times for the same individual.

[0010] The embodiments of the present invention aim to provide a method and apparatus for determining the progression of cataracts in animals, which can differentiate between nuclear sclerosis and true cataracts based on the spatial distribution of opacity, thereby preventing misdiagnosis of age-related changes as cataracts. means of solving the problem

[0011] A method for determining the degree of cataract progression in an animal according to an embodiment of the present invention may include the steps of receiving an eye image of the animal from a user device, extracting a lens region from the eye image, detecting a Y-suture structure of the lens within the lens region, normalizing the image characteristics of the detected Y-suture structure with a reference value of the lens region to calculate one or more quantitative indicators representing the relative visibility of the Y-suture, determining the degree of cataract progression based on the quantitative indicators, and transmitting the determined degree of cataract progression to the user device.

[0012] According to one embodiment of the present invention, the quantitative indicator may include one or more of Y suture length, Y suture contrast, Y suture edge clarity, Y suture branch count, or Y suture visibility index.

[0013] According to one embodiment of the present invention, the Y suture line visibility index can be calculated based on the value obtained by dividing the average brightness or average contrast value of the Y suture line region by the overall average brightness of the lens region.

[0014] The step of calculating the quantitative index according to one embodiment of the present invention may include: setting the branch point of the Y suture as the origin and setting a suture-based coordinate system with a plurality of branches of the Y suture as reference axes; and mapping the opacity of the lens region on the coordinate system as a function of angle and radius to calculate a normalized index for the difference in eye size or imaging magnification.

[0015] The step of calculating the quantitative indicator according to one embodiment of the present invention includes calculating a symmetry score by comparing the shape of the Y suture with a reference shape or calculating the degree of asymmetry for each of the plurality of branches of the Y suture, and the step of determining the degree of cataract progression can detect a localized or early cataract lesion based on the symmetry score or the degree of asymmetry for each branch.

[0016] According to one embodiment of the present invention, the step of calculating the quantitative index includes: quantifying radial opacity within a fan-shaped area set along the branch of the Y suture line, and the step of determining the degree of cataract progression may include: localizing the location of progression of the opacity based on the quantitative value of the radial opacity.

[0017] According to one embodiment of the present invention, the degree of cataract progression may be classified into one or more of normal, early cataract, immature cataract, mature cataract, or hypermature cataract.

[0018] According to one embodiment of the present invention, the step of calculating a turbidity index indicating the turbidity of the lens region is further included, and the step of determining the degree of cataract progression may include: a step of distinguishing between normal or mature cataracts based on the quantitative index and the turbidity index.

[0019] The step of determining the degree of cataract progression according to one embodiment of the present invention may include: a step of differentiating nuclear sclerosis and cataract based on the spatial distribution of opacity indicated by the quantitative indicator.

[0020] According to one embodiment of the present invention, based on the degree of cataract progression and the quantitative indicator, the method may further include the step of outputting one or more of the indication for surgery, the risk of lens-induced uveitis, and the fragility of the lens capsule as auxiliary indicators.

[0021] According to one embodiment of the present invention, the method may further include a step of calculating the cataract progression rate based on the time-interval change of the quantitative indicator calculated from each of a plurality of ocular images taken at different times for the same object.

[0022] According to one embodiment of the present invention, when the eye image is a video, the method may further include a step of selecting a frame to be analyzed based on the visibility or clarity of the Y suture line among a plurality of frames of the video.

[0023] The step of detecting the Y suture structure according to one embodiment of the present invention includes the step of calculating the detection reliability of the Y suture structure, and may further include the step of generating uncertainty information in response to the detection reliability being less than a reference value.

[0024] The step of determining the degree of cataract progression according to one embodiment of the present invention may include: applying a correction parameter according to the breed of the animal.

[0025] The step of detecting the Y suture structure according to one embodiment of the present invention may include the step of detecting the Y suture structure using a deep learning-based segmentation model.

[0026] A device for determining the degree of cataract progression in an animal according to one embodiment of the present invention may include a communication module, one or more processors, and a memory including one or more programs executable by the one or more processors, wherein the one or more programs may include, through the communication module, instructions for receiving an eye image of the animal from a user device, instructions for extracting a lens region from the eye image, instructions for detecting a Y-suture structure of the lens within the lens region, instructions for normalizing the image characteristics of the detected Y-suture structure with a reference value of the lens region to calculate at least one quantitative indicator representing the relative visibility of the Y-suture, instructions for determining the degree of cataract progression based on the quantitative indicator, and instructions for transmitting the determined degree of cataract progression to the user device through the communication module. Effects of the invention

[0027] According to embodiments of the present invention, a method and apparatus can be provided to determine the progression of cataracts in an animal without visiting an animal hospital, even using a video captured by a user with a smartphone or the like.

[0028] According to embodiments of the present invention, by considering together a quantitative indicator representing the visibility of the Y suture and an indicator of opacity in the lens region, a method and apparatus for determining the progression of cataracts in animals can be provided, which can distinguish a state that is not distinguishable by visibility alone.

[0029] According to embodiments of the present invention, a method and apparatus for determining the degree of cataract progression in an animal can be provided, which can predict future progression trends by calculating the rate of cataract progression based on changes in quantitative indicators derived from images taken at different times for the same individual.

[0030] According to embodiments of the present invention, a method and apparatus for determining the progression of cataracts in animals can be provided, which can differentiate between nuclear sclerosis and true cataracts based on the spatial distribution of opacity, thereby preventing misdiagnosis of age-related changes as cataracts. Brief explanation of the drawing

[0031] FIG. 1 is a drawing illustrating a device for determining the progression of cataracts in an animal and a user device according to one embodiment. FIG. 2 is a drawing illustrating a device for determining the progression of cataracts in an animal according to one embodiment. FIG. 3 is a flowchart illustrating a method for determining the progression of cataracts in an animal according to one embodiment. Figures 4 and 5 are drawings showing examples of animal eye images. Specific details for implementing the invention

[0032] Specific structural or functional descriptions regarding various embodiments are illustrative for the purpose of explaining various embodiments, and various embodiments may be implemented in various forms and should not be interpreted as being limited to the embodiments described in this specification or application.

[0033] Since various embodiments may be subject to various modifications and may take various forms, various embodiments are illustrated in the drawings and described in detail in this specification or application. However, the details disclosed in the drawings are not intended to specify or limit the various embodiments, and should be understood to include all modifications, equivalents, and substitutions that fall within the spirit and technical scope of the various embodiments.

[0034] The terms used herein are used merely to describe specific embodiments and are not intended to limit various embodiments. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this specification, terms such as “comprising” or “having” are intended to specify the existence of the described features, numbers, steps, actions, components, parts, or combinations thereof, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0035] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which this disclosure pertains. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this specification.

[0036] The operating principles of the present disclosure will be described in detail below with reference to the attached drawings. In describing the present disclosure below, specific descriptions of related known functions or configurations will be omitted if it is determined that such detailed descriptions would unnecessarily obscure the essence of the present disclosure. Furthermore, the terms described below are defined in consideration of their functions in the present disclosure, and these may vary depending on the intentions or practices of data providers or operators. Therefore, their definitions should be based on the content throughout this specification.

[0037] Meanwhile, the embodiments of the present disclosure disclosed below are merely specific examples provided to facilitate the explanation of the technical content of the present disclosure and to aid in understanding the present disclosure, and are not intended to limit the scope of the present disclosure. That is, it is obvious to those skilled in the art that other variations based on the technical concept of the present disclosure are possible. Furthermore, each embodiment may be combined and operated with one another as needed. For example, parts of one embodiment of the present disclosure and another embodiment may be combined and operated with one another. Additionally, the embodiments are exemplary, and other variations based on the technical concept of the embodiments may be implemented.

[0038] Meanwhile, the order of description in the drawings illustrating the method of the present disclosure does not necessarily correspond to the order of execution, and the order of execution may be changed or executed in parallel.

[0039] Alternatively, the drawings describing the method of the present disclosure may omit some components and include only some components to the extent that the essence of the present invention is not impaired.

[0040] In addition, the method of the present disclosure may be implemented by combining some or all of the contents included in each embodiment to the extent that it does not impair the essence of the invention.

[0042] FIG. 1 is a drawing illustrating a device (100) for determining the progression of cataracts in an animal and a user device (200 or 201) according to one embodiment.

[0043] A device (100) for determining the degree of progression of cataracts is a computing device that receives an image of an animal's eye from a user device (200 or 201), processes the received image of the eye to determine the degree of progression of cataracts, and transmits the determination result to the user device (200 or 201). A device (100) according to one embodiment may be implemented as, for example, a server, a cloud server, or a workstation, and may be physically implemented as a single device or have its functions distributed across multiple devices.

[0044] A user device (200) is a device that acquires or receives an image of an animal's eye and transmits it to a device (100) according to one embodiment, and receives and displays a judgment result from the device (100) according to one embodiment. As illustrated in FIG. 1, the user device (200 or 201) may include various types of devices such as a smartphone, a desktop computer, or a tablet. The user device (200 or 201) may be, for example, a smartphone equipped with a camera and a display, and an image of an animal's eye may be displayed on the display.

[0045] A device (100) and a user device (200 or 201) according to one embodiment may communicate with each other through a network. The network may include wired and / or wireless networks, for example, the Internet, a mobile communication network (LTE, 5G, etc.), a short-range wireless communication network (Wi-Fi, etc.), or a combination thereof. In one embodiment, the user device (200 or 201) may be provided with an application or web interface that interacts with the device (100) according to one embodiment, and a guardian or veterinarian may upload an eye image and check the judgment result through the application or web interface.

[0046] In this way, by separating the acquisition (user device side) and analysis (device side) of the eye image, high-performance image processing and learning model computation are performed on the device (100), and the user device (200 or 201) can use the service with low computational resources.

[0048] FIG. 2 is a drawing illustrating a device (100) and a user device (201) for determining the progression of cataracts in an animal according to one embodiment.

[0049] A device (100) according to one embodiment includes a processor (101), a memory (102), and a communication module (103). The processor (101) and the memory (102) are interconnected, and the communication module (103) is connected to communicate with a user device (201).

[0050] The processor (101) controls the overall operation of the device (100) and performs calculations for determining the progression of cataracts, which will be described later. The processor (101) may include, for example, a central processing unit (CPU), a graphics processing unit (GPU), a neural network processing unit (NPU), a digital signal processing unit (DSP), or a combination thereof, and may be composed of one or more processors. In one embodiment, for the calculation of deep learning-based segmentation, which will be described later, the processor (101) may include an accelerator such as a GPU or an NPU.

[0051] The memory (102) stores one or more programs (instructions) executable by the processor (101) and stores data such as eye images, intermediate calculation results, learned model parameters, and judgment results. The memory (102) may include, for example, RAM, ROM, flash memory, SSD, HDD, or a combination thereof. One or more programs may include instructions that, when executed by the processor (101), perform operations such as receiving an eye image from a user device (201) via a communication module (103), extracting a lens region, detecting a Y suture structure, calculating a quantitative index through normalization, determining the degree of cataract progression, and transmitting a judgment result to the user device (201) via the communication module (103).

[0052] The communication module (103) can transmit and receive data with the user device (201). The communication module (103) may include a wired communication interface and / or a wireless communication interface. The communication module (103) can receive an eye image from the user device (201) and transmit it to the processor (101), and transmit the degree of cataract progression determined by the processor (101) to the user device (201).

[0053] A user device (201) is a terminal that communicates with a device (100) according to one embodiment, and may include a camera for acquiring an eye image, a display for displaying a judgment result, an input unit for user input, and a communication unit for communicating with the device (100). The cataract progression determination operation described in this specification is mainly performed in the device (100), but according to the embodiment, some operations (e.g., preprocessing of the eye image or primary selection of frames to be analyzed) may be performed in the user device (201).

[0055] FIG. 3 is a flowchart illustrating a method for determining the progression of cataracts in an animal according to one embodiment.

[0056] According to one embodiment, an image of an animal's eye can be received from a user device (3001). Here, the eye image is an image obtained by photographing the animal's eye, and may include a still image or a video image that includes at least an area where the lens is visible through the pupil. The eye image may be received through a communication module after being captured by a camera of the user device, and may have a still image format such as JPEG, PNG, HEIC, etc., or a video format such as MP4, MOV, etc.

[0057] Meanwhile, according to one embodiment, if the received eye image is a video, a frame to be analyzed can be selected based on the visibility or sharpness of the Y suture line among a plurality of frames of the video (3011). Here, the frame to be analyzed may refer to one or more frames selected for use in subsequent analysis among a plurality of frames. Sharpness refers to the degree to which the image is in focus and can be quantified by a focus scale such as Laplacian variance or edge intensity, for example. Visibility refers to the degree to which the Y suture line is distinguished from the surrounding background. Sharpness and visibility are calculated for each frame, and the frame with the highest score based on these is selected as the frame to be analyzed, and frames with motion, blur, blinking, or excessive reflection may be excluded.

[0058] Meanwhile, according to one embodiment, when the eye image received from the user device is a video, the frame to be analyzed can be selected based on the visibility or clarity of the Y suture line among a plurality of frames of the video (3011).

[0059] In addition, according to one embodiment, a lens region can be extracted from an eye image (3002). Here, the lens region refers to an area corresponding to the lens within the eye image, that is, an area occupied by the lens observed through the pupil. The lens region can be extracted by dividing a generally circular or elliptical part inside the pupil using at least one of circle / ellipse fitting, threshold-based segmentation, edge detection, or deep learning-based segmentation, and in this case, specular reflection (light source reflection) of the eyelid, eyelash, iris, sclera, and corneal surface can be excluded from the lens region or masked separately.

[0060] In addition, according to one embodiment, the Y suture structure of the lens can be detected within the extracted lens region (3003). Here, the Y suture structure is a suture formed by the ends of the lens fibers coming into contact with each other, and refers to a structure in which three branches extend radially at intervals of about 120 degrees from a single branch point. The Y suture structure can be detected by at least one of ridge detection (e.g., Hessian-based detection, Frangi filter), template matching for a shape of three branches at about 120 degrees, or deep learning-based segmentation.

[0061] Here, according to one embodiment, a Y suture structure can be detected using a deep learning-based segmentation model. Here, the deep learning-based segmentation model is a neural network model trained to predict, on a pixel-by-pixel basis, whether each pixel of an input image belongs to the Y suture, and may have, for example, a U-Net, FCN, or DeepLab structure and be trained on animal eye images labeled with the Y suture. This model outputs a pixel-by-pixel probability map corresponding to the Y suture for the lens region, and the Y suture structure and branching point can be determined from the probability map.

[0062] Here, according to one embodiment, the detection reliability of the Y suture structure can be calculated, and uncertainty information can be generated in response to the detection reliability being less than a reference value (3031). Here, the detection reliability is a measure indicating how well the detected Y suture structure corresponds to the actual Y suture, and can be expressed as, for example, a value between 0 and 1. The detection reliability can be calculated based on at least one of the mean or maximum probability of a segmentation probability map, a template matching score, or the degree to which the detected branch shape corresponds to approximately 120-degree symmetry. The reference value is a threshold value compared with the detection reliability, and can be pre-set, for example, as a value where the false detection rate and the non-detection rate are balanced for a set of verification images with known reliability, or set as a predetermined fixed value (e.g., 0.5). The uncertainty information is information indicating that the judgment result is difficult to trust, and may include, for example, a message requesting re-shooting, a confidence interval of the judgment, or an indication indicating low reliability. If the detection reliability is below a threshold value, instead of determining the degree of cataract progression as is, uncertainty information is generated and provided to the user device, allowing the user device to induce re-imaging.

[0063] In addition, according to one embodiment, one or more quantitative indicators representing the relative visibility of the Y suture line can be calculated by normalizing the image characteristics of the detected Y suture line structure to a reference value of the lens area (3004). Here, image characteristics refer to values ​​measured from the image, such as brightness, contrast, or edge intensity of pixels corresponding to the Y suture line. The reference value is a representative value derived from the lens area, and may be, for example, the average brightness of the entire lens area, a median, or a predetermined quantile value. Normalization means converting the image characteristics of the Y suture line into a relative value (e.g., ratio) with respect to the reference value, and relative visibility refers to the degree to which the Y suture line stands out relative to the brightness of the lens background. The quantitative indicator is a value that quantifies this relative visibility and can be calculated by dividing or subtracting the image characteristics of the Y suture line by the reference value of the lens area. By normalizing to the value of the lens itself in this way, an indicator robust to deviations in lighting, exposure, and white balance during shooting can be obtained.

[0064] Here, quantitative indicators may include one or more of Y suture length, Y suture contrast, Y suture edge sharpness, Y suture branch count, or Y suture visibility index. Y suture length refers to the total length of the detected suture or the sum of the lengths of each branch, and may be normalized to the lens size; Y suture contrast refers to the brightness difference between the pixel corresponding to the suture and the surrounding background pixels. Y suture edge sharpness refers to the degree of steepness of the brightness change at the suture boundary (edge ​​slope); and Y suture branch count refers to the number of detected branches and may be represented as 3. The Y suture visibility index may be an indicator representing the relative visibility of the suture as a ratio.

[0065] The Y suture visibility index can be calculated based on the value obtained by dividing the average brightness or average contrast value of the Y suture area by the overall average brightness of the lens area (3041). For example, the Y suture visibility index can be defined as (average contrast of the Y suture area) / (overall average brightness of the lens area). By setting the denominator to the overall average brightness of the lens area, the value can be corrected for varying values ​​depending on the shooting brightness even for the same suture, thereby allowing for the calculation of a visibility index independent of exposure deviation.

[0066] Meanwhile, according to one embodiment, a branch point of the Y suture is set as the origin, and a suture-based coordinate system is established with multiple branches of the Y suture as reference axes. By mapping the turbidity of the lens region in the coordinate system as a function of angle and radius, a normalized index for the difference in eye size or imaging magnification can be calculated (3042). Here, the branch point refers to the point where three branches of the Y suture meet, and the suture-based coordinate system refers to a coordinate system (e.g., a polar coordinate system) with the branch point as the origin and the three branches as reference axes. Turbidity refers to the degree of opacity, that is, the degree of cloudiness, at each location of the lens. The turbidity of the lens region is angle If mapped as a function of radius r and normalized by dividing radius r by the lens radius, an index comparable to the same standard can be obtained even if the size of the eyeball or the magnification of the image changes.

[0067] According to one embodiment, a symmetry score can be calculated by comparing the shape of the Y suture line with a reference shape, or a branch-specific asymmetry can be calculated for each of the multiple branches of the Y suture line (3043). Here, the reference shape refers to the shape of a normal Y suture line, that is, a shape in which three branches are symmetrically spaced at intervals of approximately 120 degrees. The symmetry score is a value indicating how close the detected shape of the Y suture line is to the reference shape; for example, it can be calculated to have a higher value as the deviation between the branch angle and the branch length becomes smaller. Branch-specific asymmetry refers to a value indicating how much the visibility or opacity around each branch is skewed compared to other branches. Furthermore, according to one embodiment, localized or early cataract lesions can be detected based on the symmetry score or branch-specific asymmetry. Here, a localized lesion refers to a lesion that appears confined to a part of the lens, and an early cataract lesion refers to a lesion in the stage where opacity begins to be minimal. For example, if turbidity increases only around a specific branch and the degree of asymmetry is significant, it can be detected that a localized initial lesion exists at that location.

[0068] According to one embodiment, radial opacity within a fan-shaped area set along the branch of the Y suture can be quantified (3044). Here, the fan-shaped area refers to a fan-shaped area set within a certain angular range along a branch centered on the branch point, and radial opacity refers to opacity that spreads outward from the branch point around the suture (opacity around the suture). Radial opacity can be quantified by calculating the degree of opacity according to the radius within each fan-shaped area. Furthermore, the location of opacity progression can be localized based on the quantitative value of radial opacity. That is, by calculating which radius range of which branch the opacity is prominent, the direction and range of progression of cataract opacity can be determined.

[0069] In addition, according to one embodiment, the degree of cataract progression can be determined based on a quantitative indicator (3005). The determination can be performed by a threshold rule for the quantitative indicator, or by a classifier that takes the quantitative indicator as input (e.g., decision tree, support vector machine, multilayer perceptron). Here, the degree of cataract progression can be classified into one or more of normal, early cataract, immature cataract, mature cataract, or hypermature cataract. Normal means a state in which there is no significant opacity in the lens; early cataract means a state in which mild localized opacity appears; immature cataract means a state in which part of the lens is opaque but part of the fundus is observed; mature cataract means a state in which the entire lens is opaque and the fundus is not observed; and hypermature cataract means a more advanced state, such as when lens proteins liquefy or shrink.

[0070] Furthermore, according to one embodiment, a turbidity index indicating the turbidity of the lens region can be calculated, and normal or mature cataracts can be distinguished based on the quantitative index and the turbidity index. Here, the turbidity index is an indicator of how opaque the entire lens region is, independently of the visibility of the suture line, and is calculated, for example, based on the average brightness or average opacity of the entire lens region. The quantitative index indicating the visibility of the Y suture line changes non-monotonically with the progression of the cataract and may appear as a low value in both normal lenses and mature cataracts. Therefore, by classifying it as normal when both the quantitative index and the turbidity index are low, and as mature cataract when the quantitative index is low but the turbidity index is high, the two states that cannot be distinguished by visibility alone can be reliably distinguished.

[0071] Meanwhile, according to one embodiment, nuclear sclerosis and cataracts can be differentiated based on the spatial distribution of opacity indicated by a quantitative indicator. Here, nuclear sclerosis refers to a phenomenon in which the lens nucleus hardens and becomes slightly opaque due to aging; it is distinguished from true cataracts and generally has a small impact on vision. The spatial distribution of opacity refers to the pattern of opacity distribution by location within the lens. If the opacity is concentrated in the center of the lens (nucleus) and the opacity in the cortical region around the sutures is relatively small, it can be differentiated as nuclear sclerosis; if the opacity around the sutures and in the cortical region is prominent, it can be differentiated as true cataracts.

[0072] Meanwhile, according to one embodiment, one or more of the following may be output as auxiliary indicators based on the degree of cataract progression and quantitative indicators: whether there is an indication for surgery, the risk of lens-induced uveitis, and the fragility of the lens capsule (3061). Here, the indication for surgery refers to information indicating whether a condition requiring cataract surgery is present, the risk of lens-induced uveitis refers to the degree of risk that protein from the cloudy lens may leak and cause intraocular inflammation, and the fragility of the lens capsule refers to the degree to which the lens capsule handled during surgery is susceptible to damage. These auxiliary indicators are calculated through a rule or regression model from the determined degree of progression and quantitative indicators (e.g., whether there is an overmaturity stage, the degree of cloudiness around the suture line and its change) and may be output as grades such as low, medium, or high. For example, if the overmaturity stage is determined and the change in cloudiness around the suture line is significant, the risk of uveitis and the fragility of the lens capsule may be output as high grades.

[0073] Meanwhile, according to one embodiment, a correction parameter can be applied depending on the breed of animal (3051). Here, the correction parameter is a value for correcting the brightness of the normal lens, the basic visibility of the suture line, the pattern of cataract prevalence, etc., which vary depending on the breed, and refers to a coefficient or offset applied to a normalized reference value or a judgment threshold. By identifying the breed of animal and selecting a correction parameter corresponding to that breed to apply to the calculation of quantitative indicators or the judgment of cataract progression, misjudgment due to differences between breeds can be reduced.

[0074] Furthermore, according to one embodiment, the cataract progression rate can be calculated based on the time-to-time change of a quantitative indicator derived from each of a plurality of ocular images captured at different times for the same individual. Here, the cataract progression rate refers to the speed at which the cataract progresses over time. By calculating a quantitative indicator from the ocular image at each time point and calculating the difference in the quantitative indicator between times or the rate of change (slope) over time, the cataract progression rate can be calculated, thereby allowing for the prediction of future progression trends or the time at which a specific stage is reached.

[0075] In addition, according to one embodiment, the determined cataract progression can be transmitted to a user device (3006). The determined cataract progression is transmitted to the user device through a communication module, and the previously calculated auxiliary indicator or cataract progression speed may be transmitted together. The user device can display the received information on a display.

[0077] Figures 4 and 5 are drawings showing examples of images of an animal's eye.

[0078] In FIG. 4, reference numeral 4001 represents the lens region, and reference numeral 4002 represents the branch point. FIG. 4 is an example of an eye image having a relatively less advanced bluish-gray cataract, in which the Y suture line and its branch point (4002) are observed relatively clearly within the lens region (4001). The lens region (4001) corresponds to the region extracted in step (3002), and its overall average brightness, etc., is used as the normalization reference value in step (3004). The branch point (4002) is the point where the three branches of the Y suture line meet, and is used as the origin of the suture-based coordinate system in step (3042).

[0079] In FIG. 5, reference numeral 5001 represents the lens region, and reference numeral 5002 represents the branching point. FIG. 5 is an example of an eye image having a mature or brunescent cataract. Even in the stage as shown in FIG. 5, a Y suture line structure based on the branching point (5002) can be identified within the lens region (5001), and the branching point (5002) is likewise used as the origin of the coordinate system of step (3042).

[0080] Comparing FIG. 4 (an example with relatively high visibility) and FIG. 5 (an example of maturation and overall opacity), it can be seen that the visibility of the Y suture line changes non-monotonically with respect to cataract progression, as described above. Therefore, the device (100) can reliably distinguish between a normal state and a mature cataract state by considering together the quantitative index of the Y suture line normalized to the reference value of the lens region (4001, 5001) and the opacity index of the lens region (e.g., step 3005). Meanwhile, specular reflection (light source reflection) on the corneal surface that may appear in FIG. 4 and FIG. 5 is excluded from or masked from the lens region in step (3002) so as not to act as an error factor when calculating the quantitative index.

[0082] Meanwhile, the embodiments of the present disclosure disclosed in this specification and drawings are merely specific examples provided to facilitate the explanation of the technical content of the present disclosure and to aid in understanding the present disclosure, and are not intended to limit the scope of the present disclosure. That is, it is obvious to those skilled in the art that other variations based on the technical concept of the present disclosure are possible. Furthermore, each embodiment may be combined with one another as needed. For example, parts of one embodiment of the present disclosure and another embodiment may be combined to be implemented in the form of an embodiment not specified in this specification.

[0083] Meanwhile, in the drawings illustrating the method of the present disclosure, the order of description does not necessarily correspond to the order of execution, and the order of precedence may be changed or executed in parallel.

[0084] Alternatively, drawings describing the method of the present disclosure may omit some components and include only some components to the extent that the essence of the present disclosure is not impaired.

[0085] Additionally, the method of the present disclosure may be practiced by combining some or all of the contents included in each embodiment to the extent that it does not impair the essence of the disclosure.

Claims

Claim 1 A method for determining the degree of cataract progression in an animal—the animal having a Y-suture structure in the lens of its eye—performed by a computing device comprising a communication module, one or more processors, and memory, comprising: receiving an image of the animal’s eye from a user device; extracting a lens region from the image of the eye; detecting the Y-suture structure within the lens region; normalizing the image characteristics of the detected Y-suture structure with a reference value of the lens region to calculate one or more quantitative indicators representing the relative visibility of the Y-suture; determining the degree of cataract progression based on the quantitative indicators; and transmitting the determined degree of cataract progression to the user device. Claim 2 A method according to claim 1, wherein the quantitative indicator comprises one or more of: Y suture length, Y suture contrast, Y suture edge sharpness, Y suture branch count, or Y suture visibility index. Claim 3 In claim 2, the Y suture line visibility index is calculated based on the value obtained by dividing the average brightness or average contrast value of the Y suture line region by the overall average brightness of the lens region. Claim 4 A method according to claim 1, wherein the step of calculating the quantitative index comprises: setting a branch point of the Y suture as the origin and setting a suture-based coordinate system with a plurality of branches of the Y suture as reference axes; and mapping the opacity of the lens region on the coordinate system as a function of angle and radius to calculate a normalized index for the difference in eye size or imaging magnification. Claim 5 The method according to claim 1, wherein the step of calculating the quantitative index comprises: calculating a symmetry score by comparing the shape of the Y suture with a reference shape or calculating the degree of asymmetry for each of the plurality of branches of the Y suture, and the step of determining the degree of cataract progression comprises: detecting a localized or early cataract lesion based on the symmetry score or the degree of asymmetry for each branch. Claim 6 A method according to claim 1, wherein the step of calculating the quantitative index comprises: quantifying radial opacity within a fan-shaped area set along the branch of the Y suture line, and the step of determining the degree of cataract progression comprises: localizing the location of progression of opacity based on the quantitative value of the radial opacity. Claim 7 A method according to claim 1, wherein the degree of cataract progression is classified into one or more of normal, early cataract, immature cataract, mature cataract, or hypermature cataract. Claim 8 A method according to claim 7, further comprising the step of calculating a turbidity index indicating the turbidity of the lens region, and the step of determining the degree of cataract progression comprising: distinguishing between normal or mature cataracts based on the quantitative index and the turbidity index. Claim 9 The method of claim 1, wherein the step of determining the degree of cataract progression comprises: a step of differentiating nuclear sclerosis and cataract based on the spatial distribution of opacity indicated by the quantitative indicator. Claim 10 A method according to claim 1, further comprising the step of outputting one or more of the indication for surgery, risk of lens-induced uveitis, and lens capsule fragility as auxiliary indicators based on the degree of cataract progression and the quantitative indicator. Claim 11 A method according to claim 1, further comprising the step of calculating the rate of progression of cataracts based on the time-interval change of the quantitative indicator calculated from each of a plurality of ocular images taken at different times for the same individual. Claim 12 A method according to claim 1, wherein, when the eye image is a video, the method further comprises the step of selecting a frame to be analyzed based on the visibility or clarity of the Y suture line among a plurality of frames of the video. Claim 13 A method according to claim 1, wherein the step of detecting the Y suture structure comprises: a step of calculating the detection reliability of the Y suture structure, and further comprising a step of generating uncertainty information in response to the detection reliability being less than a reference value. Claim 14 The method of claim 1, wherein the step of determining the degree of cataract progression comprises: applying a correction parameter according to the breed of the animal. Claim 15 The method of claim 1, wherein the step of detecting the Y suture structure comprises: detecting the Y suture structure using a deep learning-based segmentation model. Claim 16 A device for determining the degree of cataract progression in an animal—the animal having a Y-sutral structure in the lens of the eyeball—comprising a communication module; one or more processors; and a memory comprising one or more programs executable by the one or more processors, wherein the one or more programs include: instructions for receiving an image of the animal’s eyeball from a user device via the communication module; instructions for extracting a lens region from the image of the eyeball; instructions for detecting the Y-sutral structure within the lens region; instructions for normalizing the image characteristics of the detected Y-sutral structure to a reference value of the lens region to calculate at least one quantitative index representing the relative visibility of the Y-sutral; instructions for determining the degree of cataract progression based on the quantitative index; and instructions for transmitting the determined degree of cataract progression to the user device via the communication module.

Images