Apparatus, method and system for evaluating exophthalmos used in exophthalmos examination device

Abstract

The present invention relates to an apparatus and method for evaluating the degree of exophthalmos used in an exophthalmos inspection apparatus. The apparatus includes a left displacement platform, a left eye corner locking point, a left mirror, a camera, a near-infrared light source, a right eye corner locking point, a right mirror, a right displacement platform, a visible light beam column, and a housing. The method includes adjusting and fixing the outer corner point of the eye, controlling the near-infrared light source to sequentially turn on multiple columns of visible light beam columns, capturing and obtaining a front view video of the eyeball by a camera, and obtaining a video frame when the longest reflected light of the visible light beam column appears on the eyeball in the mirror from the video; identifying the corneal apex from the video frame and identifying the pupil center using a neural network; and calculating the degree of exophthalmos based on a known position, a mirror tilt angle, the corneal apex, and the pupil center.

Description

[Technical Field]

[0001] The present invention relates to the technical field of eye detection, and more particularly to an apparatus, method and system for assessing exophthalmos used in exophthalmos examination devices. [Background technology]

[0002] Exophthalmos refers to the distance from the corneal vertex to the temporal orbital rim. The average exophthalmos in healthy individuals ranges from 12 to 14 mm, with an average of 13 mm, and the difference between the two eyes does not exceed 2 mm. Patients with orbital diseases often have increased exophthalmos. Currently, Hertel exophthalmos and CT scanning are the main methods used to measure exophthalmos clinically.

[0003] The Hertel exophthalmometer is a contact-type measurement system, requiring manual recording of power during the experimental process. Inter-individual evaluation results are closely related to the experience and measurement method of the diagnosing physician, which can easily lead to individual differences. In complex clinical settings, where patients and physicians are in close contact, this increases the risk of infection for both the examiner and the examiner. Research has also shown that when using a Hertel exophthalmometer for detection, the lateral orbital rim is not completely symmetrical, and the degree of soft tissue subsidence varies. There are no unified application standards or operating rules related to the examiner's visual error and experience, resulting in significant differences in readings between different examiners. Readings by experienced examiners are thought to be 1 mm larger than those by less experienced observers. CT scanning measurement is more accurate than the Hertel exophthalmometer method, but suffers from problems such as large radiation, high measurement costs, and slow result generation.

[0004] To address the problems of low accuracy or high cost of the above methods, the prior art also has technical means for determining exophthalmos using automated measuring devices and methods. As described in Chinese Patent CN104720738A (Application No.: 201510155640.4, Publication Date: June 24, 2015), a first anterior-posterior distance X1 from the system reference point St of an ophthalmic device to the lateral edge Ek of a person's eye socket, a second anterior-posterior distance X2 from the system probe of the ophthalmic device to the system reference point St, and a third anterior-posterior distance X3 from the corneal apex Ec to the system probe are calculated, and the exophthalmos are calculated by calculating X1 + X2 - X3. As described in Chinese Patent CN112806956A (Application No.: 202110139034.9, Publication Date: May 18, 2021), a distance measuring sensor moves on a driving member to identify the corneal apex, and the distance between a light beam emitter and the distance measuring sensor is calculated to determine the exophthalmos. All of the above methods have the drawbacks of requiring complicated equipment and inaccurate identification of the corneal apex. Summary of the Invention [Problem to be solved by the invention]

[0005] In order to solve the problems of large errors and high costs in measuring exophthalmos in the prior art, the present invention provides an exophthalmos evaluation device, method and system for use in an exophthalmos examination device to improve the accuracy of exophthalmos measurement. [Means for solving the problem]

[0006] The device for evaluating the degree of exophthalmos used in the exophthalmos examination device according to the first aspect of the present invention comprises: The device includes a left displacement platform, a left eye corner locking point, a left mirror, a camera, a near-infrared light source, a right eye corner locking point, a right mirror, a right displacement platform, a visible light light pole, and a housing, wherein the left mirror and the right mirror are respectively provided at the left eye corner locking point and the right eye corner locking point, the left eye corner locking point and the right eye corner locking point are respectively provided on the left displacement platform and the right displacement platform, the left displacement platform and the right displacement platform are respectively provided on both sides of the housing, the left mirror and the right mirror are both provided at an angle, the camera is located in front of the left mirror and the right mirror, the near-infrared light source is located below the camera, and the visible light light poles are in multiple rows and are arranged vertically along the inside of the housing.

[0007] Furthermore, the two cameras are provided so as to be positioned in front of the left mirror and the right mirror, respectively.

[0008] Furthermore, the visible light columns in each row are composed of discrete point light sources.

[0009] A method for evaluating exophthalmos according to a second aspect of the present invention, using an exophthalmos evaluation device used in the exophthalmos examination device according to the first aspect of the present invention, includes: Step S1: adjusting and fixing the corner of the eye, controlling the near-infrared light source to sequentially turn on multiple rows of visible light pillars, taking and acquiring a video of the front view of the eyeball with a camera, and acquiring a video frame from the video when the longest reflected light of the visible light pillars appears on the eyeball in the mirror; Step S2 of identifying the corneal apex from the video frame and identifying the pupil center using a neural network; and step S3 of calculating the degree of exophthalmos using the known position, the mirror tilt angle, the corneal apex, and the pupil center, The known positions are the position of the optical center of the camera, the distance from the optical center of the camera to the video frame, and the position of the corner of the eye, and the mirror tilt angle is the included angle between the mirror and the imaging plane.

[0010] Furthermore, in step S1, lighting up the visible light pillars in a plurality of rows in sequence means lighting up the visible light pillars in sequence from the center to the outside, or lighting up the visible light pillars in sequence from the outside to the center.

[0011] Furthermore, in step S1, when the visible light pillar on one side is turned on, a video of the opposite eye is taken.

[0012] Furthermore, in step S3, the optical center is O, the outer corner of the eye at the corner of the eye is E, the distance from the optical center O to the video frame is OB, the image point in the video frame of the corneal apex Z is C, the image point in the video frame of the reflection point U of the corneal apex Z on the mirror is A, the perpendicular distances between the image points C and A and the optical axis are AB and CB, respectively, the inclination angle between the mirror and the eye is ∠UED, the distance from the optical center O to the facial plane where the outer corner of the eye at the corner of the eye E is located is OF, the distance from the outer corner of the eye at the corner of the eye E to the optical axis is EF, the degree of exophthalmos is the perpendicular distance ZG from the corneal apex Z to the straight line DE, and the lengths of OB, OF, EF, AB, and CB and the angle of ∠UED are known, The step of calculating exophthalmos comprises: Step S31: A line passing through the reflection point U on the mirror at the corneal vertex Z and parallel to the line DE is drawn so that it intersects with the optical axis at point W, intersects with ZG at point Y, and intersects with an extension of the line OC at point X; Step S32: calculating the length of the line segment WF using the following formula; WF=(OFtan∠UOW-EF) / (tan∠UOW+1 / (tan∠UED)) Step S33: determining the length of the line segment ZY by iterative calculation; and step S34 of obtaining the exophthalmos according to ZG=WF+ZY.

[0013] An evaluation system for implementing the method for evaluating exophthalmos according to the second aspect of the present invention, which is related to another aspect of the present invention, comprises: a video acquisition module that acquires a video of the frontal view of the eye when the rows of visible light columns of the near-infrared light source are sequentially illuminated; an image capture module for capturing a video frame from the video in which the visible light column has a longest reflected light from a mirror; a feature extraction module for extracting the corneal apex and pupil center from the video frames; and a calculation module for calculating the degree of exophthalmos based on a known position, a mirror tilt angle, the corneal apex, and the pupil center, wherein the known positions are the position of the optical center of the camera, the distance from the optical center of the camera to the video frame, and the position of the corner of the eye, and the mirror tilt angle is the included angle between the mirror and the imaging plane.

[0014] A computer-readable storage medium according to another aspect of the present invention stores at least one program code that, when loaded by a processor, executes the exophthalmos assessment method according to the second aspect of the present invention. [Effects of the Invention]

[0015] The technical means of the present invention has at least the following beneficial effects compared with the prior art.

[0016] The evaluation device captures a video of the eye in a near-infrared light field, and uses a neural network to effectively distinguish between the iris, pupil, and sclera, thereby accurately identifying the pupil center. A series of vertical visible light pillars are placed around the user's face, and the visible light pillars are sequentially turned on to find the longest reflected light of the visible light pillars that appear on the eye in the mirror, thereby accurately and easily identifying the corneal apex. After identifying the pupil center and corneal apex, the algorithm in this invention is used to accurately calculate the exophthalmos without contacting the cornea, combining the known positions and the mirror tilt angle based on the optical imaging principle, thereby ensuring the accuracy and safety of exophthalmos measurement. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a perspective view of an evaluation device for evaluating the degree of exophthalmos used in the exophthalmos examination device of the present invention. FIG. [Figure 2] 1 is a flowchart of a method for evaluating exophthalmos according to the present invention. [Figure 3] This is a video frame taken when the light reflection in the mirror is at its longest point. [Figure 4] FIG. 1 is a diagram illustrating the calculation principle of exophthalmos. [Figure 5] FIG. 1 is a schematic diagram illustrating the configuration of an exophthalmos evaluation system. DETAILED DESCRIPTION OF THE INVENTION

[0018] Specific embodiments of the present invention will be described in more detail below with reference to the drawings and examples. The following examples are intended to illustrate the present invention but are not intended to limit the scope of the present invention.

[0019] In the present invention, OB, AB, CB, EF, etc. used in the calculation all refer to the lengths of line segments.

[0020] Example 1 As shown in FIG. 1, the exophthalmos evaluation device used in the exophthalmos examination device according to this embodiment includes a left displacement platform 3, a left eye corner locking point 4, a left mirror 5, a camera 8, a near-infrared light source 9, a right eye corner locking point 10, a right mirror 11, a right displacement platform 12, a visible light beam 13, and a housing 14. The left mirror 5 and the right mirror 11 are respectively provided at the left eye corner locking point 4 and the right eye corner locking point 10, and the left eye corner locking point 4 and the right eye corner locking point 10 are respectively provided at the left displacement platform 12. The left and right displacement platforms 3 and 12 are respectively provided on both sides of the housing 14, the left mirror 5 and the right mirror 11 are both provided at an angle, the camera 8 is located in front of the left mirror 5 and the right mirror 11, the near-infrared light source 9 is located below the camera 8, and the visible light pillars 13 are arranged in multiple rows and are aligned vertically along the inside of the housing 14.

[0021] To further support and adjust the evaluation device and ensure its fixation to the user, the evaluation device further includes a lower support frame 1, a tray 2, a forehead rest 6, and an upper support frame 7. The lower support frame 1 and the upper support frame 7 are located at the bottom and top of the housing 14, respectively, and support and rotate the entire evaluation device, while the tray 2 and the forehead rest 6 fix the user.

[0022] Furthermore, the two cameras 8 are positioned in front of the left mirror 5 and the right mirror 11, respectively, to capture videos of the left and right eyes. To provide a uniform near-infrared light field, the near-infrared light sources 9 may be provided as multiple light sources, uniformly distributed below the cameras 8, and each vertical visible light column 13 may be provided as an array of point light sources, wrapped in silica gel.

[0023] To measure exophthalmos using the above device, first adjust the left and right displacement platforms 3 and 12, and use the left and right canthus anchoring points 4 and 10 to fix the outer corners of the user's eyes to accommodate different facial features. After fixing the user's eyes and face, turn on the near-infrared light source 9, and then turn on a series of visible light columns 13 arranged vertically from the center to the outside or from the outside to the center. The visible light columns 13 sequentially form reflections of different lengths on the eyeball in the mirror. From the captured video, the video frame in which the visible light columns 13 form the longest projection on the cornea is found, and this projection passes through the corneal apex. Using the video frame captured in the light field of the near-infrared light source 9, the pupil center can be accurately located using neural network analysis. Using the corneal apex and pupil center, combined with known distances and parameters such as the mirror tilt angle, the exophthalmos can be calculated.

[0024] Example 2 2 and 3, this embodiment provides a method for evaluating exophthalmos, and will be described below in conjunction with an apparatus for evaluating exophthalmos used in an exophthalmos examination device. The method for evaluating exophthalmos includes the following steps S1 to S3.

[0025] In step S1, the outer corner point of the eye is adjusted and fixed, the near-infrared light source is controlled to sequentially turn on multiple rows of visible light pillars, and a video of the front view of the eyeball is captured by a camera, and a video frame is captured from the video when the longest reflected light of the visible light pillars appears on the eyeball in the mirror; In step S1, when one vertical visible light pillar is turned on, a video of the opposite eye is taken. Specifically, when the left visible light pillar 13 is turned on, a video of the right eye is taken, and when the right visible light pillar 13 is turned on, a video of the left eye is taken. Because a series of vertical visible light pillars 13 are provided and the left mirror 5 and right mirror 11 are tilted, by turning on the visible light pillars 13 sequentially from the outside to the inside, the visible light pillars 13 will present a series of reflected light of different lengths on the side of the eyeball in the mirror, and the longest projection of the visible light pillar 13 on the cornea can be found by the reflected light on the eye in the mirror. For example, when the rightmost visible light pillar 13 is turned on, the reflected light of this pillar on the side surface of the eyeball in the left mirror 5 of the left eye approaches the canthus of the eye; when the turned-on visible light pillar 13 gradually moves inward, the reflected light of this pillar on the side surface of the eyeball in the left mirror 5 gradually becomes longer until it passes through the apex of the cornea, and the reflected light at this time is the longest, and a video frame at this time is obtained from the video.

[0026] In step S2, the corneal apex is identified from the video frame, and the pupil center is identified using a neural network; As shown in Figure 3, the corneal apex can be identified by the video frame in which the light reflected by the eyeball in the mirror is the longest. The corneal apex is the point where the light reflected by the eyeball in the left mirror intersects with a vertical line in Figure 3. In the normal visible light wavelength range of 400-700 nm, the colors of different parts of the eye, namely the pupil, iris, and sclera, have little effect on image formation. The gradient structure of the corneal limbus at the interface between the iris and sclera makes it difficult to accurately identify the center of the eyeball. Human melanin has an absorption peak at approximately 335 nm and exhibits almost no absorption in wavelengths above 700 nm. The reflectance of the iris is fairly stable in the near-infrared wavelength range above 700 nm. Therefore, the present invention can effectively distinguish the boundaries of the sclera, iris, and pupil by using near-infrared light fields. This can be combined with a neural network training model to accurately identify the center of the pupil. In one possible implementation, this embodiment uses the RITNet neural network model.

[0027] The exophthalmos evaluation device used in the exophthalmos examination device of this Example 1 further provides a function of manually identifying the pupil center after identifying the corneal apex from the video frame, and the user can select between a neural network or manual operation to identify the pupil center as needed.

[0028] In step S3, the degree of exophthalmos is calculated from the known position, the mirror tilt angle, the corneal apex, and the pupil center.

[0029] In step S3, the known positions are the position of the optical center of the camera, the distance from the optical center of the camera to the video frame, and the position of the corner of the eye at the angle of the eye, and the mirror tilt angle is the angle between the mirror and the imaging plane.The distance can be obtained from the known positions, and this distance refers to the distance from the optical center of the camera to the imaging plane (i.e., the photoreceptor), which is called the image distance in photography.The above distance and mirror tilt angle can be combined with the pupil center in the video frame and the imaging point at the corneal apex to determine the degree of exophthalmos.

[0030] Example 3 Based on the above method example 2, this example 3 further describes how to perform mathematical calculation of exophthalmos.

[0031] As shown in FIG. 4 , the optical center is O, the outer corner of the eye at the canthus is E, the vertical distance from the optical center O to the video frame is OB, the image point of the corneal apex Z in the video frame is C, the image point of the reflection point U of the corneal apex Z on the mirror in the video frame is A, the vertical distances between the image points C and A and the optical axis are AB and CB, respectively, the inclination angle between the mirror and the eye is ∠UED, the distance from the optical center O to the facial plane where the outer corner of the eye at the canthus is E is OF, the distance from the outer corner of the eye at the canthus E to the optical axis is EF, the degree of exophthalmos is the vertical distance ZG from the corneal apex Z to the line DE, and the lengths of OB, OF, EF, AB, and CB and the angle of ∠UED are known. The step of calculating the exophthalmos includes the following steps S31 to S33.

[0032] In step S31, a line passing through the reflection point U on the mirror at the corneal vertex Z and parallel to the line DE intersects with the optical axis at point W, with ZG at point Y, and with the extension of the line OC at point X, respectively.

[0033] According to the imaging principle, the captured video frame has an imaging point C of the corneal vertex Z, and also an imaging point A of the reflection point U on the mirror at the corneal vertex Z. In the video frame, the distances CB and AB from the optical axis of imaging point C and imaging point A can be read. According to the light reflection principle, the incident angle of the corneal vertex Z on the mirror is equal to the reflection angle, that is,

number

number

[0034] In step S32, the lengths of OB, OF, AB, CB, and EF and the magnitude of ∠UED are known, and the following can be obtained from analytical geometric relationships:

number

number

number

number

number

number

number

[0035] From analytical geometric relationships we can obtain:

number

number

number

[0036] In step S34, the degree of exophthalmos is determined as the distance ZG from the corneal vertex Z to the line DE.

[0037] ZG=WF+ZY(9) The degree of exophthalmos is calculated according to equations (4) and (8).

[0038] Thus, to calculate the degree of exophthalmos, it is first necessary to calculate the lengths of line segments YG and ZY in Figure 4. Since the length of line segment YG is equal to the length of line segment WF, the degree of exophthalmos can be calculated by adding the distance between line segments WF and ZY.

[0039] In step S32, the length of the line segment WF and the distance of the line segment UX can be obtained by solving the linear equation with three unknowns in equation (3). In step S33, first, it is assumed that UY is equal to UX, that is, YX is equal to zero, and then, using trigonometric functions, an estimated value of ZY, that is,

number

number

number

number

number

number

number

number

[0040] The above calculation process can be performed automatically using a computer program. Simply obtain the video frame with the longest reflected light from the visible light column on the mirror and input it into the computer, and the result of the exophthalmos can be automatically output.

[0041] Example 4 5 is a schematic diagram of an exophthalmos evaluation system according to an embodiment of the present invention. As shown in FIG. 5, the system includes: a video acquisition module 501 for acquiring a video of the frontal view of the eye when multiple rows of visible light columns of near-infrared light sources are illuminated in sequence; an image capture module 502 for capturing the video frame from the collected video in which the visible light column has the longest reflected light from the mirror; a feature extraction module 503 for identifying the corneal apex from the video frames acquired from the image acquisition module and extracting the pupil center using a neural network; and a calculation module 504 that calculates the degree of exophthalmos using the parameters of the known position, the mirror tilt angle, the corneal apex, and the pupil center, where the known positions include the position of the optical center of the camera, the distance from the optical center of the camera to the video frame, and the position of the corner of the eye, and the mirror tilt angle is the included angle between the mirror and the imaging plane.

[0042] In an exemplary embodiment, a computer-readable storage medium including a memory storing at least one program code that is loaded and executed by a processor to implement the exophthalmos evaluation method in the above embodiment is further provided. For example, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CDROM), a magnetic tape, a floppy disk, an optical data storage device, etc.

[0043] Those skilled in the art will understand that the realization of all or part of the steps in the above embodiments may be completed by hardware, or may be completed by hardware associated with at least one program code, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk, an optical disk, etc.

[0044] The above is merely a preferred embodiment of the present invention, and is not intended to limit the present invention. All modifications, equivalent replacements, improvements, etc. made within the concept and principle of the present invention should be included within the protection scope of the present invention. [Explanation of symbols]

[0045] 1 Lower support frame 2 trays 3. Left displacement platform 4 Left canthus anchoring point 5 Left mirror 6 Forehead protector 7 Upper support frame 8. Camera 9 Near-infrared light source 10 Right canthus locking point 11 Right mirror 12 Right displacement platform 13 Visible light column 14 Housing 501 Video Acquisition Module 502 Image Acquisition Module 503 Feature Extraction Module 504 Computational Module

Claims

1. The device includes a left displacement platform (3), a left canthus anchoring point (4), a left mirror (5), a camera (8), a near-infrared light source (9), a right canthus anchoring point (10), a right mirror (11), a right displacement platform (12), a visible light column (13), and a housing (14), The left mirror (5) and the right mirror (11) are provided at the left eye corner anchoring point (4) and the right eye corner anchoring point (10), respectively; The left eye corner anchoring point (4) and the right eye corner anchoring point (10) are provided on the left displacement platform (3) and the right displacement platform (12), respectively; The left displacement platform (3) and the right displacement platform (12) are provided on both sides of the housing (14), respectively; The left mirror (5) and the right mirror (11) are both provided at an angle, The camera (8) is located in front of the left mirror (5) and the right mirror (11), The near-infrared light source (9) is located below the camera (8), The visible light columns (13) are arranged in multiple rows, aligned vertically along the inside of the housing (14).

1. An apparatus for evaluating the degree of exophthalmos, which is used in an exophthalmos examination apparatus.

2. The two cameras (8) are provided so as to be positioned in front of the left mirror (5) and the right mirror (11), respectively.

2. An apparatus for evaluating the degree of exophthalmos, using the apparatus for examining exophthalmos according to claim 1.

3. The visible light columns (13) of each row are composed of discrete point light sources.

2. An apparatus for evaluating the degree of exophthalmos, which is used in the apparatus for examining exophthalmos according to claim 1.

4. A computer program used in the device for evaluating exophthalmos according to claim 1, By causing a computer to execute the computer program, Step S1: controlling the near-infrared light source to sequentially light up multiple rows of visible light pillars, capturing a video of the front view of the eyeball with a camera, and capturing a video frame from the video when the longest reflected light of the visible light pillars appears on the eyeball in a mirror; Step S2: determining the corneal apex from the video frames and identifying the pupil center using a neural network; Step S3: calculating the degree of exophthalmos based on the known position, the mirror tilt angle, the corneal apex, and the pupil center; The known positions are the position of the optical center of the camera, the distance from the optical center of the camera to the video frame, and the position of the corner of the eye; The mirror tilt angle is the angle between the mirror and the image plane. A computer program characterized by:

5. In step S1, the sequential lighting of the visible light columns in the plurality of rows is performed by sequentially lighting the visible light columns from the center to the outside or from the outside to the center.

5. A computer program according to claim 4.

6. In step S1, when the visible light pillar on one side is turned on, a video of the opposite eye is taken.

6. A computer program according to claim 5.

7. In step S3, the optical center is O, The outer corner of the eye is E, the distance from the optical center O to the video frame is OB; The image point in the video frame of the corneal vertex Z is C, The image point in the video frame of the reflection point U of the corneal vertex Z on the mirror is A, The perpendicular distances between the image forming point C and the optical axis and the image forming point A are AB and CB, respectively. The angle of inclination between the mirror and the eye is ∠UED, The distance from the optical center O to the face plane on which the outer corner point E of the eye is located is OF, The distance from the outer corner of the eye E to the optical axis is EF, The degree of exophthalmos is the vertical distance ZG from the corneal vertex Z to the line DE, The lengths of OB, OF, EF, AB, and CB and the angle ∠UED are known, The step of calculating exophthalmos comprises: Step S31: A line passing through the reflection point U on the mirror at the corneal vertex Z and parallel to the line DE is intersected with the optical axis at point W, with point ZG at point Y, and with an extension of the line OC at point X; Step S32: calculating the length of the line segment WF using the following formula; [Equation 1] Step S33: determining the length of the line segment ZY by iterative calculation; and step S34 of obtaining the exophthalmos according to ZG=WF+ZY.

5. A computer program according to claim 4.

8. An evaluation system that realizes each step processed by causing a computer to execute the computer program according to any one of claims 4 to 7, a video acquisition module that acquires video from a frontal view of the eye when the rows of visible light columns of the near-infrared light source are sequentially illuminated; an image capture module for capturing a video frame from the video that has the longest reflected light from a mirror of the visible light column; a feature extraction module for extracting the corneal apex and pupil center from the video frames; a calculation module for calculating the degree of exophthalmos using a known position, a mirror tilt angle, the corneal apex, and the pupil center; An evaluation system characterized by:

Images