Swept source OCT based comprehensive ophthalmic imaging system and acquisition method thereof

Abstract

The application discloses a comprehensive ophthalmic imaging system based on a swept source OCT and a collection method thereof, which comprises a light source module with a swept laser, a main interferometer module, a reference arm, a sample arm suitable for an eye fundus mode, a control circuit module and a collection control analysis system, the light source module emits laser, the light beam enters the reference arm and the sample arm after passing through the main interferometer module, the return light signal enters the main interferometer module again to output an interference signal, the axis point of scanning is located at the pupil position of the anterior surface of the lens, and the collection control analysis system obtains the OCT image from the front end of the eye to the eye fundus according to the interference signal. The OCT image formed by the collection control analysis system is subjected to curvature correction, that is, a fan-shaped image from the vitreous body or the front end of the eye to the eye fundus is formed, the corrected retina is identified and circular or elliptical fitting is performed, and the real curvature of the eye fundus in the current scanning meridian plane is obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to an image system, in particular to a comprehensive ophthalmic image system based on a swept source OCT and a collection method thereof. BACKGROUND

[0002] Myopia is caused by the eyeball overgrowth, the eye axis lengthening, resulting in light focusing on the retina before the retina, thus causing blurred retinal imaging. The previous myopia screening for adolescents mainly relies on traditional visual examination, including subjective visual chart examination and objective detection of automatic optometry. The biggest problem of the traditional examination method is that it is easily affected by the lens accommodation state of the measured person, the examination result is unstable and cannot distinguish false myopia. If you want to exclude the influence of lens accommodation, you must perform pupil dilation optometry. However, since pupil dilation optometry will make the measured person have blurred vision for several days, and there is a risk of blocking the angle of the eye pressure to cause glaucoma, therefore pupil dilation optometry cannot be used as a routine screening method for myopia in adolescents.

[0003] Based on the optical biometer, the distance from the corneal vertex of the measured person to the macular fovea of the retina can be measured to obtain the eye axis value with micron-level precision, which is a non-contact and non-invasive objective examination method. Since this method is not affected by the lens accommodation state, it is an ideal method for myopia screening, and in recent years, biometry has been increasingly used in myopia screening work for adolescents.

[0004] The physiological mechanism of myopia has not been fully understood, but a series of studies on the pathogenesis of myopia in recent years have shown that the occurrence and progression of myopia are related to choroidal thickness. Choroidal thickness thinning causes choroidal blood reduction, which in turn causes scleral thinning, which is likely to be an important physiological mechanism leading to myopia. The treatment method of red light irradiation of the fundus to increase choroidal thickness has also been gradually promoted. If the choroidal thickness can be obtained at the same time as the biometry, it will play an important guiding role in the prediction and treatment of myopia development of the measured person.

[0005] Traditional treatment for myopia is mainly wearing frame glasses, but after wearing myopia frame glasses, teenagers are prone to rapid deepening of myopia degree. Studies have found that this phenomenon is mainly because after wearing myopia frame glasses, the image of central vision is projected on the retina, but its periphery is projected behind the retina, which is called retinal peripheral hyperopic defocus. This defocus state leads to the elongation of the retina to the back for self-regulation, resulting in further growth of the eye axis length, thereby causing the deepening of myopia degree. Studies have shown that by pulling the focus of the retinal periphery back to the front of the retina to form retinal peripheral myopic defocus, the eye axis growth can be controlled and the deepening of myopia can be delayed. The main methods include corneal molding, specially designed peripheral defocus soft lenses and frame glasses. However, due to the inconsistency of the fundus curvature of each person, the same treatment plan will produce different amounts of peripheral defocus in different people, and the amount of defocus will directly affect the effect of myopia control. The commonly used ophthalmic B-ultrasound is greatly affected by the operator's technique and cannot accurately measure the fundus curvature, and the examination must be in contact with the eyeball; although the nuclear magnetic resonance (MRI) can accurately measure the fundus curvature, the cost is too high to be popularized.

[0006] Early ophthalmic biometry devices use ultrasonic technology. More than a decade ago, the time-domain OCT technology (also known as "weak coherent interference technology" in early literature) improved the axial measurement accuracy by more than 10 times, and gradually became the mainstream technology of ophthalmic biometry devices. The sensitivity of time-domain OCT technology is limited, and for deep cataracts often encountered in clinical practice, the signal is often too weak to cause measurement failure. The second generation of frequency-domain OCT technology has improved the signal-to-noise ratio by 2 orders of magnitude compared to time-domain OCT, but its depth is mostly limited to less than 3mm, making it difficult to apply to ophthalmic biometry. In recent years, the swept source OCT technology has been developed, with a measurement depth increased to more than 40mm in tissue, and a sensitivity exceeding that of frequency-domain OCT, making it the most ideal technology for ophthalmic biometry and gradually becoming the mainstream of clinical application.

[0007] Current swept source-based ophthalmic biometry uses telecentric or near-telecentric optical lenses, with scanning lines parallel to the optical axis and focal points in the anterior segment. Such arrangements are beneficial for traditional axial length measurement and anterior segment imaging, but the information of the fundus is very limited and cannot meet the needs of myopia prevention and control and correction described above. In addition, due to the distance of the fundus from the focus of the detection light, there is still a certain proportion of measurement failure for severe cataract eyes in the application of traditional preoperative cataract measurement.

[0008] Therefore, the above problems need to be solved urgently. SUMMARY

[0009] The first object of the present application is to provide a comprehensive ophthalmic imaging system based on a swept source OCT, which forms an image after curvature correction of an OCT image through a collection control analysis system, i.e., forms a fan-shaped image showing from the vitreous body or the anterior segment of the eye to the fundus, identifies the corrected retina and performs a circle or ellipse fitting to obtain the real curvature of the fundus in the current scanning meridian plane.

[0010] The second object of the present application is to provide a collection method of the comprehensive ophthalmic imaging system based on the swept source OCT.

[0011] Technical scheme: To achieve the above object, the present application discloses a comprehensive ophthalmic imaging system based on a swept source OCT, which comprises a light source module with a swept laser, a main interferometer module, a reference arm, a sample arm suitable for a fundus mode, a control circuit module and a collection control analysis system arranged in sequence along an optical path, the control circuit module is connected with the light source module, the reference arm and the sample arm respectively, the collection control analysis system controls the light source module to emit laser, the light beam is divided into two paths through the main interferometer module and enters the reference arm and the sample arm respectively, the light signals returned from the reference arm and the sample arm are output as sample interference signals after entering the main interferometer module again, the axis point of the scanning is located at the pupil position of the anterior surface of the lens, the collection control analysis system obtains the OCT image from the front end of the eye to the fundus through one scanning according to the sample interference signals, wherein the collection control analysis system comprises a computer, the concentric circles drawn by the computer with the scanning axis point at the pupil as the center are regarded as equal optical path planes, the physical distance between any point in the OCT image and the scanning axis point is obtained, any point in the OCT image is mapped into the polar coordinates with the scanning axis point as the center, after the polar coordinates are converted into Cartesian coordinates, the image after curvature correction of the OCT image is formed, i.e., the fan-shaped image showing from the vitreous body or the anterior segment of the eye to the fundus is formed, the corrected retina is identified and fitted as a circle or an ellipse to obtain the real curvature of the fundus in the current scanning meridian plane.

[0012] The OCT image comprises a fundus image, an anterior segment image and an image with the eye axial length from front to back, and the collection control analysis system synthesizes a full eye image according to the fundus image, the anterior segment image and the image with the eye axial length from front to back.

[0013] Preferably, the collection control analysis system adjusts the parameter configuration of the swept frequency range and the swept frequency speed of the swept laser to realize different scanning modes of image depth and axial resolution, wherein the image depth range of the OCT is inversely proportional to the spectral sampling resolution.

[0014] Further, the sample arm comprises, in sequence along the light path, a first optical fiber port, a first collimating lens, a galvanometer, a scanning mirror group, a dichroic beam splitter, an ocular group, an internal fixation lamp and a lens group, wherein the light beam is reflected by the galvanometer after being collimated by the first collimating lens, and the light rays pass through the scanning mirror group at different angles according to the angle of the galvanometer, and the light rays are focused on the intermediate phase surface after being reflected by the dichroic beam splitter, and then enter the human eye to be measured through the ocular group, and the light rays at different angles form scanning axial points at the pupil position after passing through the cornea, and the pupil and the galvanometer form an object image conjugate; the light rays pass through the lens and converge at the fundus after passing through the pupil; the returned light enters the first optical fiber port and returns to the main interferometer module; the visible light of the internal fixation lamp is imaged on the intermediate phase surface through the lens group, and then imaged on the central macular region of the fundus through the ocular group.

[0015] The acquisition method of the comprehensive ophthalmic imaging system based on the swept source OCT comprises the following steps:

[0016] (1) Fix the head of the measured person, guide the measured person to gaze at the internal fixation lamp; adjust the position of the sample arm relative to the human eye until the image of the eye to be measured is seen in the OCT image;

[0017] (2) Move the sample arm forward and backward to place the interference reference surface in front of the cornea and form an image from the cornea to the fundus that is complete and has no reflection;

[0018] (3) Adjust the distance between the scanning mirror group and the ocular group to compensate for ametropia and make the fundus image clear;

[0019] (4) Fine-tune the starting position of the scanning line and / or the up-down-left-right position of the sample arm until the central reflection of the cornea is seen, and then fine-tune the forward-backward position of the sample arm until the anterior surface of the cornea is located below the zero phase point in the OCT image and the image of the anterior surface of the lens is flat, at this time the axial point of the scanning is located at the pupil position of the anterior surface of the lens;

[0020] (5) Confirm that the measured person keeps gazing at the internal fixation lamp, the macular region of the fundus in the image is clear and visible, and a clear OCT image is obtained.

[0021] The acquisition method of the comprehensive ophthalmic imaging system based on the swept source OCT comprises the following steps:

[0022] (6) Keep the same settings of the reference arm as when the sample OCT image is obtained, and the background image that may contain ocular reflection is not included in the imaging light path;

[0023] (7) The computer subtracts the background image from the human eye image to eliminate the artifacts of possible ocular reflection.

[0024] The acquisition method of the comprehensive ophthalmic imaging system based on the swept source OCT comprises the following steps:

[0025] (1) Fix the head of the subject, and guide the subject to gaze at the internal fixation lamp; adjust the position of the sample arm relative to the human eye until the image of the eye to be measured is seen in the OCT image;

[0026] (2) Place the interference reference surface behind the choroid, and form an OCT image that is complete and free of reflections from the cornea to the fundus;

[0027] (3) Move the sample arm forward and backward so that the fundus part in the OCT image is uniformly bright;

[0028] (4) Adjust the distance between the scanning lens group and the ocular lens group to compensate for refractive errors so that the fundus image is clear;

[0029] (5) Fine-tune the starting position of the scanning line and / or the up-down-left-right position of the sample arm until the corneal center reflection is seen, and then fine-tune the forward-backward position of the sample arm until the corneal anterior surface is located below the zero phase point in the OCT image and the image of the anterior surface of the lens is flat, at which time the axis point of the scanning is located at the pupil position of the anterior surface of the lens;

[0030] (6) Confirm that the subject maintains gaze at the internal fixation lamp, and that the macular region of the fundus in the image is clearly visible, and obtain a clear OCT image.

[0031] Preferably, the computer obtains the refractive power of the measured eye according to the correspondence between the pre-calibrated compensation lens movement distance and the refractive power.

[0032] Further, the computer calculates parameters along the eye axis according to the OCT image, the parameters along the eye axis including the axial length of the measured eye, the central corneal thickness, the anterior chamber depth, and the central lens thickness.

[0033] Further, the computer corrects the fundus image, and fits the corrected retina to obtain the true curvature of the fundus retina; the fundus curvature correction method comprises the following steps:

[0034] The concentric circles drawn with the scanning axis point at the pupil as the center can be regarded as equal-optical-path surfaces; any plane in the fundus image that is perpendicular to the probe light is an equal-phase surface, which corresponds to a spherical surface in the physical space with the scanning axis point at the pupil as the center; the physical distance between any point (z0, y0) in the OCT original image and the scanning axis point can be determined according to the following formula:

[0035]

[0036] Wherein, z3 and z4 are the optical path thickness of the lens and vitreous at y=y0 in the fundus image, and n3 and n4 are the refractive index of the lens center and vitreous respectively; each z=z0 plane in the fundus image is an equal phase plane, corresponding to a concentric circle with the scanning axis point near the pupil as the center and the radius equal to the distance from the scanning axis point to the z=z0 plane; assuming that the OCT scanning is located in the meridian plane through the optical axis and the scanning axis point is located at z=0, any point (z0, y0) in the image can be mapped into the polar coordinates with the scanning axis point as the center by the following way: y0 is converted into the post-pupil incident angle through the optical model, and the post-pupil incident angle is taken as the polar coordinate After the polar coordinate conversion, the geometric correction of the fundus image is realized; then the geometrically corrected fundus image is converted from the polar coordinates back to the Cartesian coordinates; the retinal pigment epithelial layer RPE or Brush membrane of the corrected fundus image is identified, and then spherical or ellipsoidal surface fitting is performed to obtain the retinal curvature.

[0037] Advantages: Compared with the prior art, the present application has the following remarkable advantages: the present application can obtain the tomographic structure image of the fundus retina and the key information such as the axial length, the central corneal thickness, the anterior chamber depth and the lens thickness through one scanning. The biological measurement information can be used for the geometric correction of the tomographic structure of the fundus retina, so that the fundus curvature consistent with the physical structure of the eyeball is obtained. Further, through one multi-line scanning, the three-dimensional structure of the fundus can be given, and the two-dimensional thickness distribution map of the retina and choroid is obtained through the automatic segmentation algorithm, which is combined with the information such as the axial length. This device will bring great convenience for myopia prevention and control. For the diagnosis of clinical ophthalmic diseases, the present application can also bring new value for the clinical diagnosis of ophthalmology. Since the detection light focus point is moved to the fundus, the measurement capability for deep cataract can be greatly improved. The present application combines the axial information obtained in the scanning to perform geometric correction on the three-dimensional body data, so that the three-dimensional geometric structure of the fundus is obtained, and quantitative analysis can be performed to obtain the fundus curvature. This has important clinical value for the accurate evaluation of the peripheral defocus amount in myopia treatment, high myopia, posterior scleral staphyloma and fundus tumor and the like. The present application can also realize the imaging of the telecentric anterior segment, so that complete anterior segment imaging, retinal imaging and biological measurement and other important imaging and measurement functions can be realized, which brings great convenience for the clinical diagnosis of ophthalmology. Through the three-dimensional reconstruction of the high-resolution anterior segment image, the corneal curvature and the corneal epithelial thickness can be obtained. These parameters combined with the axial length and the fundus curvature obtained by the present device can provide accurate evaluation of the retinal peripheral defocus amount in the selection of the type of orthokeratology lens, so as to quickly confirm the most suitable and best myopia control effect type of orthokeratology lens. The device of the present application is compact and efficient, which maximizes the potential of the hardware, realizes powerful clinical imaging and measurement functions, and effectively reduces the equipment procurement expenditure of medical service institutions. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the system of the present invention;

[0039] Figure 2 This is a schematic diagram of the sample arm structure in this invention;

[0040] Figure 3 This is a schematic diagram illustrating the principle of fundus curvature correction.

[0041] Figure 4A The original OCT images from the cornea to the fundus obtained using Example 1, and the correspondence between the main layers of the eye and the layers in the OCT images;

[0042] Figure 4B for Figure 4A The image obtained after curvature correction of the original image;

[0043] Figure 5A This is a scanned source OCT image from the cornea to the fundus obtained using Example 1, where an ultra-wide-angle eyepiece reference plane is placed in front of the cornea;

[0044] Figure 5B To complete the collection Figure 5A After processing the original image, immediately remove the background image captured by the subject's eye;

[0045] Figure 5C for Figure 5A The image obtained by curvature correction of the original image to conform to physical proportions;

[0046] Figure 6A This refers to the original scanned source OCT image from the cornea to the fundus obtained using Example 1, where an ultra-wide-angle eyepiece reference plane is placed behind the choroid.

[0047] Figure 6B for Figure 6A The original image is processed by background subtraction, eyepiece reflection removal, and curvature correction to obtain an image that conforms to physical proportions.

[0048] Figure 6C For based on Figure 6B The RPE layer is identified and a circular fit is performed to obtain the calculated result of the fundus curvature radius.

[0049] Figure 7 This is a typical ultrasound image;

[0050] Figure 8A This is a schematic diagram of the structure of the external switching front lens group added in Embodiment 2 of the present invention;

[0051] Figure 8BThe structure diagram of adding inner switch front group lens of embodiment 2 of the application;

[0052] Figure 9A It is the high-definition fundus image of the sweep source OCT raw image obtained by using embodiment 2, which is displayed according to the direction and horizontal and vertical proportion of the traditional fundus OCT image (vertically elongated);

[0053] Figure 9B It is the high-definition fundus image of the sweep source OCT raw image obtained by using embodiment 2, which is displayed according to the direction and horizontal and vertical proportion of the traditional fundus OCT image (vertically elongated); Figure 9A

[0054] Figure 10A It is the high-definition fundus image of the sweep source OCT raw image obtained by using embodiment 2, which is displayed according to the direction and horizontal and vertical proportion of the traditional fundus OCT image (vertically elongated);

[0055] Figure 10B It is the high-definition fundus image of the sweep source OCT raw image obtained by using embodiment 2, which is displayed according to the direction and horizontal and vertical proportion of the traditional fundus OCT image (vertically elongated); Figure 10A

[0056] Figure 11 It is the high-definition fundus image of the sweep source OCT raw image obtained by using embodiment 2, which is displayed according to the direction and horizontal and vertical proportion of the traditional fundus OCT image (vertically elongated);

[0057] Figure 12 It is the high-definition fundus image of the sweep source OCT raw image obtained by using embodiment 2, which is displayed according to the direction and horizontal and vertical proportion of the traditional fundus OCT image (vertically elongated); DETAILED DESCRIPTION

[0058] The technical solutions of the application will be further described below with reference to the drawings.

[0059] ​​The application discloses a comprehensive ophthalmic imaging system based on a swept source OCT, which comprises a light source module with a swept laser, a main interferometer module, a reference arm, a sample arm suitable for an eye fundus mode, a control circuit module and a collection control analysis system arranged in sequence along an optical path; the control circuit module is connected with the light source module, the reference arm and the sample arm respectively; the collection control analysis system controls the light source module to emit laser; the light beam is divided into two paths by the main interferometer module and enters the reference arm and the sample arm respectively; the light signals returned from the reference arm and the sample arm are output as sample interference signals after entering the main interferometer module again; the collection control analysis system obtains the OCT image from the front end of the eye to the eye fundus according to the reference interference signals and the sample interference signals, and the image depth range is greater than 24mm in the tissue or 32mm in the air, the eye fundus image range covered is not less than 3mm, and the data information of the measured eye is obtained through analysis. Wherein, the OCT image is corrected in curvature to form a fan-shaped image showing from the vitreous body or the front segment of the eye to the eye fundus. The image of the OCT image corrected in curvature is used for identifying the corrected retina and fitting a circle or an ellipse to obtain the real curvature of the eye fundus in the current scanning meridian plane. The OCT image comprises an eye fundus image, a front segment image and an image with the eye axial length from the front to the back, and the collection control analysis system synthesizes a full eye image according to the eye fundus image, the front segment image and the image with the eye axial length from the front to the back. The collection control analysis system obtains the three-dimensional structure of the eye fundus through a group of scans, and calculates the curvature distribution map of the eye fundus, the retinal thickness map and the choroidal thickness map. The collection control analysis system adjusts the parameter configuration of the swept frequency range and the swept frequency speed of the swept laser to realize different image depth and axial resolution scanning modes; wherein the image depth range of the OCT is inversely proportional to the spectral sampling resolution.

[0060] The sample arm comprises a first optical fiber port, a first collimating lens, a galvanometer, a scanning mirror group, a dichroic beam splitter, an eyepiece group, an internal fixation lamp and a lens group arranged in sequence along an optical path, wherein the light beam is reflected by the galvanometer after being collimated by the first collimating lens, the light rays pass through the scanning mirror group at different angles according to the galvanometer angle, the light rays are reflected by the dichroic beam splitter and focused on an intermediate phase plane, and then enter the measured human eye through the eyepiece group; the light rays at different angles form scanning axial points at the pupil position after passing through the cornea, and the pupil and the galvanometer form an object image conjugate; the light rays pass through the lens and converge at the eye fundus after passing through the pupil; the returned light enters the first optical fiber port and returns to the main interferometer module; the visible light of the internal fixation lamp is imaged on the intermediate phase plane through the lens group, and then imaged on the central macular area of the eye fundus through the eyepiece group.

[0061] The application discloses a collection method of the comprehensive ophthalmic imaging system based on the swept source OCT.

[0062] (1) Fix the head of the measured person, guide the measured person to gaze at the internal fixation lamp, adjust the position of the sample arm relative to the human eye until the image of the measured eye is seen in the OCT image;

[0063] (2) Move the sample arm forward and backward to place the reference mirror in front of the cornea and form a complete and non-reflected image from the cornea to the fundus;

[0064] (3) Adjust the distance between the scanning lens group and the ocular lens group to compensate for refractive errors to make the fundus image clear;

[0065] (4) Fine-tune the starting position of the scanning line and / or the up-down-left-right position of the sample arm until the central reflection of the cornea is seen, and then fine-tune the forward and backward position of the sample arm until the anterior surface of the cornea is located below the zero phase point of the OCT image and the image of the anterior surface of the lens is flat, at which point the axis point of the scan is located at the pupil position of the anterior surface of the lens;

[0066] (5) Confirm that the subject maintains fixation on the fixation light, and that the macular region of the fundus in the image is clearly visible, and obtain a clear OCT image.

[0067] It also includes the following steps:

[0068] (6) Keep the same settings for the reference arm as when obtaining the sample OCT image, and do not include any sample collection in the imaging light path that may include background images of ocular reflections;

[0069] (7) The computer subtracts the background image from the human eye image to eliminate the artifacts of possible ocular reflections.

[0070] The present application is a collection method for a comprehensive ophthalmic imaging system based on a swept source OCT, comprising the following steps:

[0071] (1) Fix the head of the subject, guide the subject to fixate on the fixation light, and adjust the position of the sample arm relative to the human eye until the image of the eye to be measured is seen in the OCT image;

[0072] (2) Place the reference mirror behind the choroid and form a complete and non-reflected OCT image from the cornea to the fundus;

[0073] (3) Move the sample arm forward and backward to make the fundus part of the OCT image uniform in brightness;

[0074] (4) Adjust the distance between the scanning lens group and the ocular lens group to compensate for refractive errors to make the fundus image clear;

[0075] (5) Fine-tune the starting position of the scanning line and / or the up-down-left-right position of the sample arm until the central reflection of the cornea is seen, and then fine-tune the forward and backward position of the sample arm until the anterior surface of the cornea is located below the zero phase point of the OCT image and the image of the anterior surface of the lens is flat, at which point the axis point of the scan is located at the pupil position of the anterior surface of the lens;

[0076] (6) Confirm that the subject maintains fixation on the fixation light, and that the macular region of the fundus in the image is clearly visible, and obtain a clear OCT image.

[0077] The computer obtains the diopter of the measured eye according to the corresponding relationship between the pre-calibrated compensation lens movement distance and the diopter. The computer calculates the parameters along the eye axis according to the OCT image, and the parameters along the eye axis include the axial length of the measured eye, the central corneal thickness, the anterior chamber depth and the central lens thickness. The computer corrects the fundus image, and fits the corrected retina to obtain the real curvature of the fundus retina; the fundus curvature correction method comprises the following steps:

[0078] The concentric circle drawn with the pupil scanning axis point as the center can be regarded as an equal optical path surface. Any plane perpendicular to the detection light in the original fundus OCT image is an equal phase surface, and the equal phase surface corresponds to a spherical surface in the physical space with the pupil scanning axis point as the center. The physical distance between any point (z0, y0) in the OCT original image and the scanning axis point can be determined according to the following formula:

[0079]

[0080] Wherein z3 and z4 are the optical path thicknesses of the lens and vitreous at y=y0 in the OCT original image, and n3 and n4 are the refractive indices of the central lens and vitreous, respectively; each z=z0 plane in the original OCT image is an equal phase surface, corresponding to a concentric circle with the scanning axis point near the pupil as the center and the radius equal to the distance from the scanning axis point to the z=z0 plane; assuming that the OCT scan is located in the meridian plane through the optical axis, the scanning axis point is located at z=0, and any point (z0, y0) in the image can be mapped into the polar coordinates with the scanning axis point as the center in the following way: y0 is converted into the post-pupil incident angle through the optical model, and the post-pupil incident angle is the polar coordinate The polar coordinate conversion is completed, and the geometric correction of the fundus image is realized. Then the image is converted from the polar coordinates back to the Cartesian coordinates; the retinal pigment epithelial layer (RPE) or Brush membrane of the corrected fundus image is identified, and then spherical or ellipsoidal surface fitting is performed to obtain the retinal curvature; the computer finds and locks the corneal reflection point through the calculation of the OCT signal intensity, ensures that the measured light passes through the corneal vertex, and ensures that the measured light reaches the fovea centralis through real-time OCT image feedback.

[0081] As Figure 1 And Figure 2As shown, the application is a comprehensive ophthalmic imaging system based on a swept source OCT, which comprises a light source module 100, a main interferometer module 110, a reference arm 120, a sample arm 130, a control circuit module 140 and a collection control analysis system 150, and the light source module 100, the main interferometer module 110, the reference arm 120 and the sample arm 130 are sequentially arranged along the optical path. The light source module 100 comprises a swept laser 101, a first optical coupler 102, a reference interferometer 103, a photodetector 105 and a swept laser control circuit 104. The main interferometer module 110 comprises a second optical coupler 111, a polarization controller 112, an optical circulator 113, a coupler 114 and a photoelectric balance detector 115. The reference arm 120 comprises a second optical fiber port 123, a second collimating lens 124, a dispersion compensator 125, a reflecting prism 126, a third collimating lens 122 and a third optical fiber port 121. The sample arm 130 comprises a first optical fiber port 131, a first collimating lens 132, a galvanometer 133, a scanning mirror group 134, a dichroic beam splitter 135, an eyepiece group 137, a lens group 138 and an internal fixation lamp 139. The control circuit module 140 comprises a communication module 141, a fixation lamp control module 142, a motion control module 143 and a galvanometer control module 144. The collection control analysis system 150 comprises a dual-channel data acquisition card 151 and a computer 152.

[0082] The sample arm 130 can couple the probe light from the optical fiber to the free space, adopt one or a group of scanning galvanometers for scanning, image the scanning galvanometer to the pupil of the eye to be measured through a group of optical lenses, and realize the scanning of the fundus. The scanning light enters the eye, and the scanning center is located at the pupil position; the scanning light converges on the retina or in the vitreous body.

[0083] The inner fixation light 139 has a wavelength of visible light, which is shorter than the near-infrared wavelength used by the OCT; the dichroic mirror 135 is a short-wave transmission dichroic mirror that transmits visible light and reflects near-infrared light. The optical circulator 113 is located between the second optical coupler 111 and the first fiber port 131 of the sample arm 130, and the laser passes through the optical circulator 113 and enters the first fiber port 131 of the sample arm 130. The light returned from the sample arm 130 passes through the optical circulator 113 and enters the coupler 114. The coupler 114 is a 50:50 double-in double-out coupler. The two output ends of the coupler 114 are respectively connected to the positive and negative input ends of the photoelectric balance detector 115. The output end of the photoelectric balance detector 115 is connected to the double-channel data acquisition card 151 of the acquisition control analysis system. The sweep laser control circuit 104 is used to control the sweep laser. The motion control module 143 is used to control the deflection of the polarization controller 112 and the movement of the reflecting prism 126 in the reference arm 120. The galvanometer control module 144 is used to control the rotation of the galvanometer 133 in the sample arm 130 and to control the timing synchronization of the sweep laser, the galvanometer, and data acquisition. In order to optimize the reference arm linear signal, a polarization controller 112 is usually added before the reference arm, the sample arm, or before the photoelectric balance detector, as shown in Figure 1

[0084] Figure 2 ​​As shown, the sweep laser control circuit 104 controls the sweep laser to continuously sweep, and the wavelength continuously changes; the emitted laser is divided into two parts by the first optical coupler 102, a small part of light enters the reference interferometer 103 to generate a reference interference signal which is detected by the photodetector 105, and after filtering and amplification, it enters one channel of the dual-channel data acquisition card 151; most of the light passes through the second optical coupler 111 and is divided into two paths to enter the reference arm 120 and the sample arm 130 respectively; one path of light divided by the second optical coupler 111 passes through the optical circulator 113, is collimated by the first collimating lens 132 at the first optical fiber port 131, and is reflected by the galvanometer 133; the light rays pass through the scan lens group 134 at different angles according to the angle of the galvanometer 133, are reflected by the dichroic beam splitter 135, and are focused on the intermediate phase surface 136, and then are emitted into the human eye 200 to be measured through the eyepiece group 137; the light rays at different angles pass through the cornea 201 and form a scanning axis point at the pupil 202, and the pupil 202 and the galvanometer 133 form an object image conjugate; the light rays pass through the pupil 202 and converge at the fundus 204 through the lens 203; the light rays at different angles are focused on different positions of the fundus one by one; one-dimensional scanning of the galvanometer 133 forms a scanning line on the fundus; the returned light passes through the optical circulator 113 and enters the coupler 114 of the main interferometer module 110; in order to ensure that the scanning center is located in the macular region and to reduce the interference of eye movement during scanning, the internal fixation lamp 139 is set; the visible light of the internal fixation lamp 139 is imaged on the intermediate phase surface 116 through the lens 138, and then is imaged on the central macular region of the fundus through the eyepiece group 137; the other path of light divided by the second optical coupler 111 is collimated by the second collimating lens 124 at the second optical fiber port 123, passes through the dispersion compensator 125 and the reflecting prism 126, is coupled into the third optical fiber port 121 by the third collimating lens 122, and then enters the coupler 114 of the main interferometer module 110 again; the coupler 114 outputs the sample interference signal, which is input into the other input channel of the dual-channel data acquisition card 151 through the photoelectric balance detector 115; the dual-channel data acquisition card 151 inputs the reference interference signal and the sample interference signal into the computer; the computer obtains the OCT image from the front end of the eye to the fundus according to the reference interference signal and the sample interference signal, and the image depth range is greater than 24mm in tissue / 32mm in air, the fundus image range covered is not less than 3mm, and the data information of the eye to be measured is analyzed to obtain a series of parameters along the eye axis of the eye to be measured, including the refractive power of the measured eye, the axial length of the measured eye, the central corneal thickness, the anterior chamber depth, the central lens thickness, the fundus curvature, the retinal thickness and the choroidal thickness.

[0085] The instantaneous coherence length of the sweep laser of the application is greater than 64mm in air, and the system has an image depth greater than 32mm in air; the fundus image can be obtained by one-time scanning, the fundus image range covered is not less than 3mm, and the fundus image simultaneously contains the anterior segment structures such as cornea, anterior chamber and lens; through the OCT image, a series of parameters along the eye axis such as eye axial length, corneal central thickness, anterior chamber depth and lens central thickness can be obtained; through the measurement result of the eye axial length, the fundus curvature can be corrected to obtain a fan-shaped scanning image from the anterior segment of the eye to the fundus similar to the ultrasonic image; through the corrected image, the real curvature of the fundus in the meridian plane of the current scanning is obtained by identifying the RPE layer or Bruch membrane and fitting a circle or an ellipse.

[0086] The image containing the fundus obtained by the application can display a fan-shaped image from the vitreous or the anterior segment of the eye to the fundus after curvature correction; the application realizes clear imaging of the fundus by focusing, and the corresponding compensation lens movement range can give the refractive power of the measured eye. That is, through one-time scanning, one or more or all of the following information can be obtained: the refractive power of the measured eye; a series of parameters along the eye axis such as the eye axial length, the corneal central thickness, the anterior chamber depth and the lens central thickness; the fundus curvature, the retinal thickness and the choroidal thickness; through a group of scans such as star-shaped scanning or multi-line scanning, the three-dimensional structure of the fundus can be further obtained, and the fundus curvature distribution map, the retinal thickness map and the choroidal thickness map can be automatically calculated.

[0087] Preferably, the speed of the sweep laser of the application is adjustable, the sweep range and sweep speed of the sweep laser 101 and the corresponding parameter configuration are changed through the sweep laser control circuit 104, different image depth and axial resolution scanning modes can be realized on the basis of a set of hardware, for example, the high image depth scanning mode suitable for biological measurement and the high axial resolution scanning mode suitable for high-definition fundus image. Specifically, for the high image depth scanning mode suitable for biological measurement, the image depth is large, so a lower sweep speed is adopted, in order to obtain the largest image depth, the scanning range can be reduced, and the axial resolution of the image can be appropriately reduced to exchange for the image depth; for the high axial resolution scanning mode suitable for high-definition fundus image, the depth can be greatly reduced, a higher sweep speed can be adopted to obtain denser scanning, and a wider spectral range can be used to obtain the highest axial resolution.

[0088] Preferably, the application can combine a switchable eyepiece to converge the scanning light on the anterior segment of the eye to realize accurate imaging of the anterior segment of the eye.

[0089] By changing the sweep range and sweep speed of the sweep laser and the corresponding parameter configuration, and the optical lens with switchable front and rear section imaging, the following modes can be realized simultaneously on the basis of a set of hardware:

[0090] (1) a high image depth scanning mode suitable for biological measurement

[0091] (2) a high axial resolution scanning mode suitable for high-definition fundus image

[0092] (3) a high axial resolution scanning mode suitable for high-definition anterior segment image

[0093] The fundus image, the anterior segment image and the image with the eye axial length from front to back obtained by the above three image modes are combined to obtain a full eye image.

[0094] The acquisition method of the comprehensive ophthalmic imaging system based on the swept source OCT includes the following steps:

[0095] (1) fix the head of the measured person, guide the measured person to gaze at the internal fixation lamp; adjust the position of the sample arm relative to the human eye until the image of the measured eye is seen in the OCT image;

[0096] (2) move the sample arm forward and backward, place the interference reference surface in front of the cornea, and form a complete image from the cornea to the fundus without reflection;

[0097] (3) adjust the distance between the scanning lens group and the objective lens group to compensate for refractive errors and make the fundus image clear;

[0098] (4) fine-tune the starting position of the scanning line and / or the up-down-left-right position of the sample arm until the center of the cornea is seen, and then fine-tune the front-back position of the sample arm until the corneal anterior surface is located below the zero phase point of the OCT image and the image of the anterior surface of the lens is flat, at this time the axial point of the scanning is located at the pupil position of the anterior surface of the lens;

[0099] (5) confirm that the measured person keeps gazing at the internal fixation lamp, the fundus macular region in the image is clear and visible, and a clear OCT image is obtained;

[0100] (6) remove the measured human eye, fix the reference arm, and collect the background image that may contain the reflection of the objective lens;

[0101] (7) the computer subtracts the background image from the human eye image to eliminate the possible artifacts of the reflection of the objective lens.

[0102] The acquisition method of the comprehensive ophthalmic imaging system based on the swept source OCT includes the following steps:

[0103] (1) Fix the head of the subject, and guide the subject to gaze at the internal fixation lamp; adjust the position of the sample arm relative to the human eye until the image of the eye to be measured is seen in the OCT image;

[0104] (2) Place the interference reference surface behind the choroid, and form an OCT image that is complete and has no reflections from the cornea to the fundus;

[0105] (3) Move the sample arm forward and backward so that the fundus part in the OCT image is uniformly bright;

[0106] (4) Adjust the distance between the scanning lens group and the eyepiece group to compensate for refractive errors so that the fundus image is clear;

[0107] (5) Fine-tune the starting position of the scanning line and / or the up-down-left-right position of the sample arm until the corneal center reflection is seen, and then fine-tune the forward-backward position of the sample arm until the corneal anterior surface is located below the zero phase point of the OCT image and the image of the anterior surface of the lens is flat, at which time the axis point of the scanning is located at the pupil position of the anterior surface of the lens;

[0108] (6) Confirm that the subject maintains gaze at the internal fixation lamp, and that the macular region of the fundus in the image is clearly visible, and obtain a clear OCT image.

[0109] Steps (2) and (3) may need to be repeated several times until the cornea to fundus image is complete and has no reflections, and the fundus part in the OCT image is uniformly bright.

[0110] The computer calculates the refractive power of the measured eye according to the pre-calibrated corresponding relationship between the compensation lens movement distance and the refractive power; the computer calculates the parameters along the eye axis according to the OCT image, the parameters along the eye axis including the eye axis length of the measured eye, the corneal center thickness, the anterior chamber depth, and the lens center thickness; and the fundus image is corrected, and the corrected retina is fitted to obtain the true curvature of the fundus retina;

[0111] The fundus curvature correction method comprises the following steps:

[0112] As shown in Figure 3 Because the refractive index difference between the lens and the aqueous humor, vitreous body is much smaller than the refractive index difference from air to cornea, as a geometric approximation, the deflection of light by the lens inside the eyeball is ignored, and the concentric circles drawn with the pupil scanning axis point as the center can be regarded as equal optical path surfaces.

[0113] In the original fundus OCT image, any plane perpendicular to the probe light is an equal phase surface, corresponding to a spherical surface in the physical space with the pupil scanning axis point as the center. The physical distance between any point (z0, y0) in the OCT original image and the scanning axis point can be determined according to the following formula:

[0114]

[0115] where z3 and z4 are the optical path thickness of the lens and vitreous at y=y0 in the original OCT image, and n3 and n4 are the refractive index of the lens center and vitreous respectively; each z=z0 plane in the original OCT image is an equal phase plane, corresponding to a concentric circle with the scanning axial point near the pupil as the center and the radius equal to the distance from the scanning axial point to the z=z0 plane; assuming that the OCT scanning is located in the meridian plane through the optical axis and the scanning axial point is located at z=0, any point (z0, y0) in the image can be mapped into the polar coordinates with the scanning axial point as the center in the following way: y0 is converted into the incident angle after the pupil through the optical model as the polar coordinate Polar coordinate ρ=z0; after the polar coordinate conversion, the geometric correction of the fundus image is realized; then the image is converted from the polar coordinates back to the Cartesian coordinates. The retinal pigment epithelial layer (RPE) or Brush membrane is identified for the corrected fundus image; for the healthy human eye image, the RPE layer can be identified, and for the pathological human eye image, the Bruch membrane is identified; then the identified RPE layer or Bruch membrane is fitted with a spherical or ellipsoidal surface to obtain the retinal curvature.

[0116] The present application obtains the three-dimensional structure of the fundus through a group of scans, and calculates the fundus curvature distribution map, the retinal thickness map and the choroid thickness map.

[0117] The typical eye axial length of the human eye is 24 mm, so the imaging depth needs to be increased from several mm to at least 24 mm or more in the tissue. Myopia can cause the eye axial length to be elongated, and in extreme cases, it can reach more than 30 mm or even 40 mm. Therefore, in order to cover the most extreme case, the imaging depth needs to be about 40 mm in the tissue, which is more than 10 times of the typical fundus imaging OCT. In order to obtain such a large image depth, a very high spectral sampling resolution is required.

[0118] For the swept source OCT, the key to improving the spectral sampling resolution is:

[0119] (1) The swept source needs to have a narrow enough instantaneous line width and a long enough coherence length, and the coherence length needs to be at least twice the required image depth range;

[0120] (2) The acquisition card needs to have a fast enough acquisition speed;

[0121] (3) The detector needs to have enough bandwidth, which should be half of the highest acquisition speed of the acquisition card according to the sampling theorem;

[0122] (4) When the speed of the acquisition card reaches the technical or cost bottleneck, the scanning speed of the laser needs to be reduced accordingly, which means reducing the scanning repetition frequency or / and reducing the spectral range of the scanning.

[0123] Traditional broadband high-speed swept source for fundus is no longer suitable, low-speed swept source is needed, and the spectral width of the swept source can be reduced because the axial resolution required for bio-measurement is not as high as that required for imaging.

[0124] The following table gives an example of the parameters of ophthalmic bio-measurement compared with a regular fundus imaging scan.

[0125] Ophthalmic biometry Conventional fundus imaging 3mm OCT depth range (within tissue) 45mm 3mm Optical axial resolution 18μm 5μm Laser spectral width 30nm 100nm Laser repetition rate 20K / s 200K / s Data acquisition duty cycle 60％ 65％ Swept nonlinearity 1.15 1.15 Data acquisition speed 250MSPS 500MSPS

[0126] Table 1 Parameters of ophthalmic bio-measurement compared with a regular fundus scan

[0127] Where the axial resolution is defined as the full width at half maximum of the axial point spread function, and the swept non-linearity is defined as the ratio of the maximum swept speed to the average swept speed within the data acquisition range.

[0128] With the above parameter configuration, the ophthalmic bio-measurement can achieve an imaging depth range of more than 40 mm, an axial resolution of 18 μm, and a repeat accuracy of axial measurement of less than 10 μm.

[0129] Traditional fundus OCT usually places the interference reference plane at a position of several millimeters in front of the fundus. In the EDI (Enhanced Depth Imaging) mode, the interference reference plane is placed several millimeters behind the choroid; the typical imaging depth is 2-3 mm within the tissue. In the present application, the interference reference plane can be set in the EDI mode, or the position of the reflecting prism 126 in the reference arm 120 can be adjusted to place the interference reference plane at a position of several millimeters to several millimeters in front of the cornea. From the perspective of product usability, the latter setting can be used to accurately mark the reference plane position during the production process and then fix it. The advantage of this is that the reference plane can be used as a reference to accurately position the working distance; for example, the corneal anterior surface of the eye to be measured is accurately placed at a fixed position behind the reference zero position; or, preferably, the scanning axis point is accurately placed on the pupil.

[0130] In the present invention, the working distance (defined as the distance between the anterior surface of the cornea and the eyepiece) is no more than 45 mm. If wide-angle or ultra-wide-angle imaging of the fundus is performed, the working distance will be further reduced to below 30 mm. However, the image depth range of the swept-source OCT full-eye imaging and biometry mode in the present invention can reach 45 mm in tissue (60 mm in air), so both the front and back surfaces of the eyepiece group are within the image depth range of the swept-source OCT, which will cause obvious mirror reflection in the image. This may interfere with the identification of the ocular biological tissue interface. If the reference surface is fixed, the reflection caused by the reflection surface can be easily removed: only need to collect a set of background images without samples before the actual signal collection, and then subtract the background image from the actual collected image. Since the reference surface is fixed, the reflection caused by the lens will not move, so a simple background subtraction can be performed. Preferably, the background image can be acquired multiple times and averaged to increase the signal-to-noise ratio of the background signal. After subtracting the background, it is possible to leave a negative artifact at the position of the lens reflection. Since the purpose of such an image is to identify the biological tissue interface and perform measurement, as long as the negative artifact can be clearly distinguished from the biological tissue interface, the residual negative artifact will not have too much impact on the measurement result. Although the lens reflection can be eliminated by the background subtraction method, the number of eyepiece lenses and reflection surfaces should be minimized as much as possible. Therefore, the eyepiece group is preferably a single lens structure, such as the one shown in Figure 4 of the invention patent application 202111440377.5. The eyepiece group and the scanning mirror group after the intermediate image plane should be located outside the image depth range of the system. Figure 2

[0131] Another method to eliminate the eyepiece reflection is to place the reference plane behind the choroid of the fundus, i.e. the EDI mode. In this case, the eyepiece reflection is located outside the OCT image depth and thus does not interfere with the image. The EDI mode requires the image depth of the system to be preferably configured to cover most of the eye axial lengths, for example, more than 30 mm in tissue. In this mode, there are several options for the setting of the interference reference surface:

[0132] (1) The reference surface is fixed, and the distance from the front focal surface of the eyepiece is equal to the maximum image depth of the OCT. The advantage of this is that the system structure is simple and easy to operate, but the image quality of most eyes is in a non-optimized state.

[0133] (2) The reference surface is adjustable, and is adjusted according to the axial length of each patient; the advantage is that each image is in the best state, suitable for hospital outpatient application scenarios, but the efficiency is poor for large-scale screening.

[0134] ​(3) The reference plane is adjustable. The default position is 40mm optical path away from the front focal plane of the eyepiece (approximately 30mm tissue depth). The advantage of this is that the image quality is close to optimal in most cases, while ensuring measurement efficiency. In extreme cases where the axial length is greater than 30mm, the reference arm still needs to be adjusted.

[0135] Axial length is defined as the distance from the apex of the anterior corneal surface to the fovea centralis of the retina. During axial length measurement, to ensure accuracy, the probe light must pass through both the anterior corneal apex and the fovea centralis. This is achieved by having the subject fixate on a central fixation lamp within the instrument, ensuring the probe light falls on the fovea centralis. Galvanometer scanning allows the operator to visualize the actual optical path, ensuring the probe light passes through the anterior corneal apex and reaches the fovea centralis. In traditional ophthalmic biometers, the probe light focuses at the anterior segment of the eye, and the reflected light forms a bright column of light at the corneal apex. During alignment, the operator moves the lens assembly or scan line to observe the central column of light, thus determining the anterior corneal apex.

[0136] In this invention, if the focus of the probe light is moved to the fundus, the scanning coverage and clarity of the fundus far exceed those of conventional ophthalmic biometry systems. The corneal central reflection is not obvious but can still be clearly observed. Figure 6A As shown. To ensure measurement accuracy, a fast and convenient method for locating the corneal apex is still needed. In the application scenario of invention patent application 202210036250.5, the detection light focal point is located at the anterior segment, where the corneal reflection is saturated and broadened, simultaneously increasing the low signal of the entire AScan, manifesting as a bright light column penetrating the entire AScan. Therefore, the sum or average value of the intensity of the entire AScan is used as the intensity of the entire AScan. However, in this invention, if a method is used as... Figure 2 In optical lenses, the light column intensity is only strong near the corneal reflective surface. In this case, calculating the intensity of a particular AScan requires limiting the intensity value to the integral or average value of the intensity values ​​in the area around the cornea (e.g., within approximately ±0.5 mm), or the maximum value within that range, as the intensity of the AScan. After determining the method for calculating the AScan intensity, the method for locating and locking the light column used in this invention is consistent with that described in patent application 202210036250.5.

[0137] An important advantage of the present application is that the curvature of the fundus can be accurately measured. Traditional fundus OCT is difficult to give an accurate fundus curvature. This is because the fundus curvature presented by the traditional OCT image is actually the phase difference at each position in the horizontal direction. This includes both the real curvature of the fundus and the phase difference caused by the non-coincidence of the scanning axis point and the geometric center of the eyeball. Therefore, the traditional fundus OCT is difficult to correct the curvature of the fundus image. If the fundus curvature needs to be measured, another independent ophthalmic biometry system is needed to measure the axial length, and then the OCT fundus image is corrected. This correction method is inconvenient for clinical use and is not accurate enough. From the principle of fundus curvature correction, the distance from the scanning beam axis point to the retina of the fundus directly affects the accuracy of the fundus curvature measurement. The fundus image and the axial length are obtained on different devices, so the approximate position of the axis point is obtained. Due to individual differences in human eyes, the position of the axis point obtained from the axial length has a certain uncertainty. In addition, the scanning of the fundus has a certain change in working distance each time. Therefore, the measurement accuracy and repeatability can be improved when the axial length is measured on different devices and the fundus curvature is corrected. In the present application, the fundus image and the axial information are obtained in the same scanning of the same device, so the position of the scanning axis point can be accurately determined, which is very beneficial to the measurement accuracy and repeatability.

[0138] The traditional ophthalmic biometry cannot give the fundus curvature because the optical design of the ophthalmic biometry is a telecentric or approximately telecentric lens. The scanning beams at different positions in the anterior segment converge at one point or a small range on the fundus, and the image of the fundus in a large range cannot be obtained. In the present application, the fundus OCT image and the biometry information are obtained by the same scanning, which makes it possible to accurately correct the fundus curvature, and the clinical operation is fast and simple.

[0139] The existing biometry telecentric or near-telecentric optical system has a focal point of the light beam at the anterior chamber or the front of the vitreous. In the present application, the image system focuses the probe light beam on the retina, which greatly increases the intensity of the retinal signal, and the success rate of axial length measurement is significantly improved for deep cataracts.

[0140] The swept frequency laser 101 of the present application can realize both ophthalmic biometry and high definition fundus imaging without adding extra hardware. For the full eye mode, the image depth is large, so a lower swept frequency is used. In order to get as large image depth as possible, the spectral width of the sweep can be reduced, and the axial resolution of the image can be appropriately reduced to exchange for the image depth. For the fundus mode alone, the depth can be greatly reduced, and a higher swept frequency can be used to obtain denser scanning. The laser sweep covers the full spectrum that the laser hardware can achieve to obtain as high a system axial resolution as possible. The swept frequency laser 101, the photodetector 105, and the dual-channel data acquisition card 151 are configured in a mode suitable for high definition fundus imaging.

[0141] Embodiment 1

[0142] In embodiment 1, the swept frequency laser 101 uses a narrow-band swept light source with a center wavelength of 1060 nm and a spectral width of 30 nm, and a power greater than 15 mw. The splitting ratio of the first optical coupler 102 is 20:80. Considering the optical path loss attenuation, about 10% of the light can reach the eye, with a power of 1.5 mw, which is lower than the 1.9 mw static long-time fixation safety threshold calculated according to the international light radiation safety standard IEC 60825-1:2007 and the Chinese national standard GB7247.1-2012. The interference reference plane is calibrated during the production of the instrument, and is fixed inside the front focal point of the ocular lens, with a distance of 1.5 mm from the front surface of the ocular lens. The speed of the dual-channel data acquisition card 151 is set to 1 GSPS, and the two channels respectively collect the sample interference signal of the sample and the reference interference signal of the reference interferometer. After collection, the sample interference signal is resampled using the reference interference signal of the reference interferometer to obtain a signal in the linear optical frequency space. The swept frequency laser 101 has a swept frequency of 70,000 times per second, and the effective duty cycle available for data acquisition in each swept frequency period is about 55%, and the sampling rate of the acquisition card is fixed at 1 GPS. Such system parameter settings can obtain an OCT image range of 60 mm in air, corresponding to a tissue depth of 45 mm, which is greater than the maximum eye axial length of the human eye, and meets the clinical needs of biometry. The detailed parameters are shown in Table 2, column 1 "full eye imaging and biometry":

[0143]

[0144] Table 2 Parameter configuration examples for different image modes

[0145] Embodiment 1 can also increase the pupil camera (or "anterior segment camera module"). The pupil camera can greatly facilitate the initial alignment.

[0146] Before acquiring the image, the patient is asked to sit down, place his head in the forehead support frame, and fixate on the fixation light in the instrument. The operator adjusts the position of the optical lens group in the sample arm 130 relative to the patient's eye, making a coarse alignment until the OCT image of the cornea to the fundus is seen. The optical lens group in the sample arm 130 is fine-tuned up and down, left and right until the reflection of the center of the cornea is seen. The optical lens group in the sample arm 130 is fine-tuned forward and backward until the anterior surface of the cornea is within the target box below the zero phase point of the OCT image.

[0147] During the adjustment process, when the scan axis point is precisely located at the pupillary position of the anterior surface of the lens, the anterior surface of the lens appears as a flat surface, and in order to more accurately position the scan axis point, the flatness of the anterior surface of the lens can be used as an indication. Therefore, preferably, the optical lens group in the sample arm 130 is fine-tuned forward and backward until the anterior surface of the lens is flat. The computer can identify the anterior surface of the lens in real time and calculate its curvature, and according to the convention of optical learning, the center of curvature is on the side of the incident direction, the curvature is negative, and on the side of the outgoing direction, the curvature is positive. As real-time feedback for adjustment, it can assist in fine-tuning the position of the optical lens system relative to the human eye. The galvanometer can use a one-dimensional or two-dimensional scanning galvanometer group to ensure that the measurement light passes through the vertex of the cornea and the center of the macula by repeatedly fine-tuning the center position of the scan and / or the up and down and left and right of the optical lens group. Preferably, the system can increase the real-time imaging of the fundus, such as a line scanning fundus scope (LSO) or a confocal scanning laser fundus scope (cSLO), to facilitate macular positioning; after fine-tuning the center position of the scan and the up and down and left and right of the optical lens group in the sample arm, the macular region of the fundus in the image is again confirmed to be clearly visible, and image acquisition can be performed.

[0148] Since the depth of the OCT image in the present application is much greater than the depth of the conventional fundus OCT image, the surface reflection of the objective lens group is very likely to be located in the image. Therefore, in addition to acquiring the image of the eye, a background image should be acquired to subtract from the image of the eye. If a fixed reference arm design is used, the background image only needs to be acquired once and can be used for image acquisition of multiple patients. If a movable reference arm design is used, a background image should be acquired after each acquisition of the eye image, keeping the position of the reference arm unchanged.

[0149] Figure 4AFor the device described in Example 1, a raw OCT image from a normal eye was taken, with an image depth of 45 mm in tissue, which can cover the most extreme axial length that can be encountered in clinical practice. The correspondence of each interface to the fundus structure is marked in the figure. Based on this image, all parameters along the eye axis can be obtained, such as the axial length, the central corneal thickness, the anterior chamber depth, the lens thickness, the distance from the anterior lens surface to the fundus, and so on. In addition, the image contains detailed information in the central macular region, and the retinal thickness map and choroidal thickness map of the central macular region can be obtained. According to the distance from the anterior lens surface to the fundus obtained in the foregoing, curvature correction can be performed on the image. Figure 4B For the image after curvature correction, the curvature of the central macular region can be further obtained through the image after curvature correction.

[0150] In order to more comprehensively reflect the curvature of the fundus, a super-wide-angle objective lens can be used. Figure 5A-5C The displayed image uses the objective lens in Figure 9 of the prior patent (202111440377.5), and the reference arm is located about 2 mm in front of the cornea. Figure 5A The reflection of the middle objective lens is relatively obvious, Figure 5B For Figure 5A After the collection is completed, the measured eye is removed and the reference arm is kept stationary, and a background image is immediately collected, in which only the artifacts caused by the reflection of the objective lens exist. After simple background correction, these artifacts are basically eliminated; after curvature correction, as shown in Figure 5C .

[0151] Figure 6 is an image in which the interference reference surface is placed behind the choroid of the measured human eye, and the reflection surface of the objective lens is outside the OCT image depth range and thus does not cause obvious interference to the image. Note that in the raw image of Figure 6A , the corneal vertex reflection is obvious, and at the same time, the fundus macular depression is clearly visible, and the reflection of the corneal vertex extends from the corneal anterior surface to the fundus macular depression, which ensures the accuracy of the axial measurement. In addition, the anterior surface of the lens is flat in the raw image, which indicates that the scanning axial point is located at the pupil at the anterior surface of the lens, which ensures the accuracy of the subsequent curvature correction. Figure 6B For the image after curvature correction. According to image 6B, the RPE layer is identified and circular fitting is performed, and the result of the fundus curvature can be obtained, as shown in Figure 6C .

[0152] Figure 5C and Figure 6B the corrected image in the foregoing, due to the image depth being much larger than that of the traditional fundus OCT, a complete sector with the scanning axial point as the vertex is presented. This form is similar to the traditional ophthalmic ultrasound image, but the clarity and resolution are 1 to 2 orders of magnitude higher. As a comparison, Figure 7is a typical ophthalmic ultrasound image, the picture is from https: / / www.ftleye.com / eye-care / state-of-the-art-testing / a-scan-b-scan-ultrasonography / . Another difference from the ultrasound image is that Figure 5B and Figure 6B there is a virtual optical probe (scan axis) located at the pupil-lens anterior surface position, while the ultrasound probe is located in front of the cornea.

[0153] Example 2

[0154] Example 2 is based on Example 1, but the swept source is replaced by a broadband swept laser, and two kinds of scanning spectral width and scanning speed can be achieved. The low-speed narrow-band mode is the same as Example 1. The high-speed wide-band mode is 100 nm swept width, 100,000 or 200,000 times per second. As an example, the detailed parameter configuration is shown in Table 2, column 2 or column 3, which realizes the depth range of 12 mm and 6 mm in tissue, respectively.

[0155] The settings in Table 2, column 3 are used, and the reference arm is set in front of the fundus. All hardware is the same as Example 1, and the probe bandwidth and acquisition card speed settings are also the same. Such settings can obtain an OCT image range of 6 mm in tissue and an axial resolution of 5 μm, which is suitable for high-definition fundus scanning. Figure 9 is a high-definition fundus image of 6 mm depth collected by Example 2, in which Figure 9A is the original image, Figure 9B is the image after curvature correction

[0156] In the high-definition mode, whether the interference reference surface is in front of or behind the retina (EDI mode), the ocular reflex is located outside the depth of the OCT image, so there is no need to specially subtract the background for the ocular reflex.

[0157] The embodiment 2 realizes two image modes on the same hardware system, which can perform ophthalmic biometry and fundus curvature measurement, and also retains all the image functions of traditional fundus OCT, bringing great convenience to the clinic. The key point of the embodiment 2 is that the scanning speed of the swept laser can be adjusted according to the image depth. The speed of the data acquisition card is set to the highest acquisition speed (or can be reduced according to the need to save data volume). The acquisition speed of the fixed data acquisition card has the advantages of maximizing the data acquisition and processing capacity of the data acquisition card and the computer. In addition, the bandwidth of the balance detector and the anti-aliasing filter before the data acquisition card is fixed at half or slightly lower than the acquisition depth, and the system signal-to-noise ratio and depth in each depth mode are optimized. Otherwise, if the data acquisition card is set to different acquisition speeds at different laser speeds, the fixed balance detector and anti-aliasing filter (Anti-aliasing Filter) cannot accommodate different acquisition speeds, or the high image depth of the high acquisition speed target cannot be obtained, or aliasing occurs at low acquisition speed.

[0158] For a Fourier domain OCT system, the maximum image depth range R can be calculated by the following formula

[0159]

[0160] where δk is the optical frequency sampling resolution of the swept laser in the optical frequency k space. For a non-uniform k space sampling system, it is usually expressed in wavelength sampling resolution,

[0161]

[0162] where λ c is the center wavelength of the swept source, n is the refractive index of the medium, δλ is the wavelength sampling resolution of the swept source OCT, and δk and δλ are collectively referred to as the spectral sampling resolution.

[0163] For a swept source OCT system, the spectral sampling resolution depends on the following factors:

[0164] (1) The coherence length of the swept laser source itself, which is inversely proportional to the instantaneous linewidth of the laser;

[0165] (2) The bandwidth of the photodetector and the signal amplifier;

[0166] (3) The sampling frequency of the data acquisition card;

[0167] (4) The sweep speed of the swept laser;

[0168] Assuming the laser source has sufficient coherence, greater than 2 times the maximum target image depth range of OCT, and the photodetector and signal amplifier bandwidth is half of the highest sampling rate of the data acquisition card, then the average spectral wavelength sampling resolution δλ of the swept laser OCT:

[0169]

[0170] where Fs is the sampling frequency of the acquisition card, v is the average swept speed of the swept laser during sampling, which can be calculated by:

[0171]

[0172] where τ is the swept frequency of the swept light source, W is the spectral width of the swept laser, and S is the swept repetition frequency; then the average spectral wavelength sampling resolution δλ during sampling is:

[0173]

[0174] The actual swept speed of the swept light source will fluctuate, and the maximum available image depth range of the system is often limited by the worst spectral sampling resolution (the highest δλ value) rather than the average spectral sampling resolution. Define the scan nonlinearity ξ as the ratio of the highest swept speed to the average swept speed during data sampling, then (4) can be rewritten as

[0175]

[0176] Therefore

[0177]

[0178] From the above formula, the OCT image depth range is inversely proportional to the spectral width W of the swept light source, the swept repetition frequency S, and the scan nonlinearity ξ, and is proportional to the sampling rate Fs of the data acquisition card, and the sampling duty cycle η within each swept period.

[0179] Formula (8) gives the basic formula for parameter configuration when the mode of the multi-scan mode OCT image system is switched.

[0180] The spectral width of the swept light source is usually determined according to the target axial resolution Δz:

[0181]

[0182] where Δλis the full width at half maximum (FWHM) of the swept spectrum, λ c is the center wavelength of the swept light source, n is the refractive index of the medium, and Δz is the axial resolution (full width at half maximum of the power point spread function) in the medium. After determining the full width at half maximum Δλ of the swept spectral width, the swept spectral width W is usually selected to be between 1.5-2.5 times Δλ.

[0183] Once the sweep spectrum width W is determined, other parameters can be selected according to formula (8). It should be noted that some scanning parameters of the sweep laser, such as the scanning nonlinearity ξ and the sampling duty cycle η, are often limited by the characteristics of the hardware itself and are difficult to change flexibly. The sweep speed, however, is relatively easy to adjust. Preferably, the sampling rate of the data acquisition card can be fixed at the highest settable sampling rate. This can maximize the utilization of the system's hardware capabilities and avoid aliasing and additional noise caused by a decrease in the sampling rate while the bandwidth of the detector amplification circuit remains unchanged.

[0184] Example 3

[0185] In Example 3, the optical lens group of the sample arm 130 adopts a switchable design for front and rear segment imaging.

[0186] Based on Example 1, a switchable lens group is added to achieve integrated anterior and posterior segment imaging as well as whole-eye imaging. A switchable lens group 161 can be added between the eyepiece and the eye to achieve anterior segment imaging, such as... Figure 8A As shown. An internal switching method can also be used, such as... Figure 8B As shown, a lens group 162 is added between the scanning mirror group 134 and the intermediate image plane 136 to image the scanning galvanometer 133 onto the rear focal point of the eyepiece ("eyepiece rear focal point" refers to the focal point of the eyepiece inside the lens group), thereby forming a telecentric scan. The internal switching method is described in detail in the preceding invention patent CN107582020A.

[0187] To achieve different imaging modes and target image depth ranges, as an example, Table 2 provides a possible configuration of laser scanning parameters and acquisition card speed.

[0188] Using the device in Embodiment 3, in addition to obtaining the images in Embodiments 1 and 2, multiple image modes are added. These include one or more of the following image modes: axial length + fundus image mode, axial length + anterior segment image mode, high-definition anterior segment image mode, and high-definition fundus image mode.

[0189] For example, Figure 10 shows a scanned source OCT image from the cornea to the fundus obtained using the lens assembly in telecentric anterior segment mode in Example 3. Figure 10A This is the original image, with an image depth of 45 mm within the tissue. Figure 10B This is the image after ray tracing correction.

[0190] Figure 11 It is a high-definition panoramic anterior segment OCT image obtained by setting the lens group in telecentric anterior segment mode, from the cornea to the posterior capsule of the lens and part of the vitreous body. The image depth in the image is 12mm within the tissue and has been corrected by ray tracing.

[0191] The full eye image can be realized by one of the following image combinations:

[0192] The first combination: fundus image + axial length, high-definition anterior segment image;

[0193] The second combination: anterior segment + axial length, high-definition fundus image;

[0194] The third combination: axial length + high-definition fundus image + high-definition anterior segment image;

[0195] For example, Figure 12 The first group of combinations is adopted, the lens group is in the fundus mode and the telecentric anterior segment mode respectively, the fundus image + axial length OCT image and the high-definition anterior segment image are obtained, and the full eye image is further combined.

[0196] The application discloses a comprehensive ophthalmic imaging system based on a swept source OCT and a collection method thereof. Through one-time scanning, key information such as a tomographic structure image of a fundus retina, an axial length, a corneal central thickness, an anterior chamber depth and a lens thickness can be obtained. The biological measurement information can be used for geometric correction of the tomographic structure of the fundus retina, so that a fundus curvature conforming to the physical structure of an eyeball is obtained. Further, through one multi-line scanning, a three-dimensional structure of the fundus can be given. A retinal and choroidal two-dimensional thickness distribution map is obtained through an automatic segmentation algorithm, and is combined with information such as the axial length. The device will bring great convenience for myopia prevention and control.

[0197] For the diagnosis of clinical ophthalmic diseases, the application can also bring new value for ophthalmic clinical diagnosis. For example, the axial length measurement before cataract surgery. Compared with existing ophthalmic biological measurement instruments based on "partial coherence" (time-domain OCT) or swept source OCT plus telecentric or near-telecentric lenses, the application can greatly improve the measurement capability for deep cataracts because the focus point of the probe light is moved to the fundus.

[0198] Another application value is the reconstruction of the three-dimensional geometric structure of the fundus. If a wide-angle or super-wide-angle eyepiece is used, the posterior half of the eyeball can be accurately reconstructed in three dimensions. The specific method is as follows: the posterior half of the eyeball is scanned to obtain three-dimensional body data of the fundus structure. The three-dimensional body data is geometrically corrected in combination with the axial information obtained during the scanning, so that the three-dimensional geometric structure of the fundus is obtained, and quantitative analysis can be performed to obtain the fundus curvature. This has important clinical value for the accurate evaluation of the peripheral defocus amount in myopia treatment, high myopia, posterior scleral staphyloma, fundus tumors and other ophthalmic diseases. Compared with the existing technical means such as the three-dimensional structure reconstruction of the eyeball by using nuclear magnetic resonance MRI, the three-dimensional reconstruction based on OCT is high-speed, non-invasive, low-cost and has a resolution of more than one order of magnitude higher, and can almost completely replace the existing MRI reconstruction.

[0199] By different configurations and optimizations of the sweep speed of the swept laser and the data acquisition card, the application can realize the fundus imaging function comparable to the dedicated fundus OCT. Moreover, by matching the axial length result given by the biometry function of the same device, the fundus OCT image can be accurately geometrically corrected, which is difficult for the traditional fundus OCT to achieve.

[0200] By adding a switchable lens in the lens group, whether it is a front group lens attached in front of the eyepiece as usual, or an inner switchable lens group added in the scanning lens group as in the invention patent CN107582020A, the application can also realize the telecentric front group imaging. In this way, on the same device, complete front group imaging, retinal imaging and biometry and other important imaging and measurement functions can be realized, bringing great convenience to ophthalmic clinical diagnosis. By three-dimensional reconstruction of the high-resolution front group image, the corneal curvature and corneal epithelial thickness can be obtained. These parameters, combined with the axial length and fundus curvature obtained by the device, can provide accurate evaluation of the retinal peripheral defocus amount when selecting the type of orthokeratology lens, so as to quickly confirm the most suitable and best myopia control effect orthokeratology lens type. In terms of hardware, the core components such as the light source, main interferometer, detector and data acquisition card of the system remain unchanged, and only the corresponding parameters are configured. The entire device is compact and efficient, maximizes the potential of the hardware, realizes powerful clinical imaging and measurement functions, and effectively reduces the equipment procurement expenses of medical service institutions.

[0201] The technical points and specific embodiments described in the application can be applied to any OCT imaging device, whether the OCT imaging device is a standalone imaging system or a subsystem of a larger system. In particular, in some surgical microscopes and laser refractive surgery devices, OCT imaging can become a subsystem to provide real-time sample tomographic images for surgical microscopes and laser refractive surgery devices. The technical points and specific embodiments described in the application are still applicable; the change of the above application scenarios does not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the application, and should be covered within the protection scope of the application.

[0202] Finally, it should be noted that the above-described embodiments are merely specific embodiments of the present application, which are used to illustrate the technical solutions of the present application, but not to limit the same. The protection scope of the present application is not limited thereto. Although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those skilled in the art that any person skilled in the art can still modify or easily think of changes to the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some of the technical features, within the technical scope disclosed by the present application. The modifications, changes or replacements do not cause the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A comprehensive ophthalmic imaging system based on swept-source OCT, characterized in that: The system includes a light source module (100) with a swept-frequency laser (101) arranged sequentially along the optical path, a main interferometer module (110), a reference arm (120), a sample arm (130) suitable for fundus mode, a control circuit module (140), and an acquisition, control, and analysis system (150). The control circuit module (140) is connected to the light source module (100), the reference arm (120), and the sample arm (130), respectively. The acquisition, control, and analysis system (150) controls the light source module (100) to emit laser light. The beam passes through the main interferometer module and splits into two paths, entering the reference arm (120) and the sample arm (130), respectively. The light signals returning from the reference arm (120) and the sample arm (130) re-enter the main interferometer module and output the sample interference signal. The scanning axis is located at the pupil position on the anterior surface of the lens. The acquisition, control, and analysis system (150) transmits the sample interference signal according to the sample interference signal. An OCT image from the anterior end of the eye to the fundus is obtained through a single scan. The acquisition control and analysis system (150) includes a computer. The computer draws concentric circles with the scanning axis point at the pupil as the center, which are regarded as equioptic surfaces. The physical distance between any point in the OCT image and the scanning axis point is obtained. Any point in the OCT image is mapped to polar coordinates centered on the scanning axis point. After converting the polar coordinates to Cartesian coordinates, an image after curvature correction of the OCT image is formed, that is, a fan-shaped image showing from the vitreous body or the anterior end of the eye to the fundus is formed. The corrected retina is identified and fitted with a circle or ellipse to obtain the true curvature of the fundus in the current scanning meridional plane. The OCT image includes a fundus image, an anterior segment image, and an image from anterior to posterior with the axial length. The acquisition control and analysis system synthesizes a whole-eye image based on the fundus image, anterior segment image, and an image from anterior to posterior with the axial length. The computer corrects the fundus image, and then fits the corrected retina to obtain the true curvature of the fundus retina. The fundus curvature correction method includes the following steps: Concentric circles drawn with the scanning axis point at the pupil as the center can be regarded as equipathic surfaces; in the fundus image, any plane perpendicular to the probe light is an equiphase surface, which corresponds to a sphere in physical space with the scanning axis point at the pupil as the center; the physical distance between any point (z0, y0) in the fundus image and the scanning axis point can be determined according to the following formula: Where z3 and z4 are the optical path thicknesses of the lens and vitreous body at y=y0 in the fundus image, respectively, and n3 and n4 are the refractive indices of the lens center and vitreous body, respectively; each z=z0 plane in the fundus image is an isophase surface, corresponding to a concentric circle with the scanning axis point near the pupil as the center and the radius equal to the distance from the scanning axis point to the z=z0 plane; assuming that the OCT scan is located in the meridional plane passing through the optical axis and the scanning axis point is located at z=0, any point (z0, y0) in the image can be mapped to polar coordinates centered on the scanning axis point in the following way: y0 is converted into the incident angle after the pupil through the optical model, and the incident angle after the pupil is used as j of the polar coordinates; after completing the polar coordinate transformation, the geometric correction of the fundus image is realized; then the geometrically corrected fundus image is converted from polar coordinates back to Cartesian coordinates; the retinal pigment epithelium (RPE) or Brush membrane is identified in the corrected fundus image, and then spherical or ellipsoidal fitting is performed to obtain the retinal curvature.

2. The comprehensive ophthalmic imaging system based on swept-source OCT according to claim 1, characterized in that: The acquisition control and analysis system (150) adjusts the parameters of the sweep frequency range and sweep frequency speed of the sweep frequency laser (101) to achieve different scanning modes with different image depths and axial resolutions; wherein the image depth range of OCT is inversely proportional to the spectral sampling resolution.

3. The comprehensive ophthalmic imaging system based on swept-source OCT according to claim 1, characterized in that: The sample arm (130) includes a first fiber optic port (131), a first collimating lens (132), a galvanometer (133), a scanning mirror group (134), a dichroic beam splitter (135), an eyepiece group (137), an internal fixed-focus lamp (139), and a lens group (138) arranged sequentially along the optical path. The light beam is collimated by the first collimating lens (132) at the first fiber optic port (131) and then reflected by the galvanometer. Depending on the angle of the galvanometer, the light beam passes through the scanning mirror group (134) at different angles, and is reflected by the dichroic beam splitter (135). Then, the light is focused on the intermediate phase plane (136) and then enters the eye of the test subject (200) through the eyepiece group (137). After passing through the cornea, the light at different angles forms a scanning axis point at the pupil position. The pupil and the galvanometer form an image conjugate. After passing through the pupil, the light passes through the lens and converges at the fundus. The returning light enters the first fiber optic port (131) and returns to the main interferometer module (110). The visible light from the internal fixation lamp (139) is imaged on the intermediate phase plane (136) through the lens group (138) and then imaged on the central macular region of the fundus through the eyepiece group (137).

4. A method for acquiring images in a comprehensive ophthalmic imaging system based on swept-source OCT according to any one of claims 1 to 3, characterized in that, Includes the following steps: (1) Fix the subject’s head and guide the subject to look at the internal fixation lamp; adjust the position of the sample arm (130) relative to the human eye until the image of the eye to be tested is seen in the OCT image; (2) Move the sample arm (130) back and forth to place the interference reference plane in front of the cornea and form a complete image from the cornea to the fundus without reflection; (3) Adjust the distance between the scanning lens group (134) and the eyepiece group (137) to compensate for refractive errors and make the fundus image clear; (4) Fine-tune the starting position of the scan line and / or the up, down and left and right positions of the sample arm (130) until the corneal center reflection is seen. Then fine-tune the front and back positions of the sample arm (130) until the anterior surface of the cornea is below the zero phase point of the OCT image and the image of the anterior surface of the lens is flat. At this time, the scanning axis is located at the pupil position of the anterior surface of the lens. (5) Confirm that the subject is fixating on the fixation lamp and that the macular region of the fundus is clearly visible in the image, and obtain a clear OCT image.

5. The acquisition method of the comprehensive ophthalmic imaging system based on swept-source OCT according to claim 4, characterized in that: It also includes the following steps: (6) Keep the reference arm in the same settings as when acquiring the sample OCT image, and ensure that the imaging optical path does not contain any background image that may contain eyepiece reflections from the sample acquisition; (7) The computer subtracts the background image from the human eye image to eliminate possible artifacts of eyepiece reflection.

6. A method for acquiring images in a comprehensive ophthalmic imaging system based on swept-source OCT according to any one of claims 1 to 3, characterized in that, Includes the following steps: (1) Fix the subject’s head and guide the subject to look at the internal fixation lamp; adjust the position of the sample arm (130) relative to the human eye until the image of the eye to be tested is seen in the OCT image; (2) The interference reference plane is placed behind the choroid to form a complete OCT image from the cornea to the fundus without reflection; (3) Move the sample arm (130) back and forth to make the brightness of the fundus in the OCT image uniform; (4) Adjust the distance between the scanning lens group (134) and the eyepiece group (137) to compensate for refractive errors and make the fundus image clear; (5) Fine-tune the starting position of the scan line and / or the up, down and left and right positions of the sample arm (130) until the corneal center reflection is seen. Then fine-tune the front and back positions of the sample arm (130) until the anterior surface of the cornea is below the zero phase point of the OCT image and the image of the anterior surface of the lens is flat. At this time, the scanning axis is located at the pupil position of the anterior surface of the lens. (6) Confirm that the subject is fixating on the fixation lamp and that the macular region of the fundus is clearly visible in the image, and obtain a clear OCT image.

7. The acquisition method of the comprehensive ophthalmic imaging system based on swept-source OCT according to claim 4 or 6, characterized in that: The computer obtains the refractive power of the tested eye based on a pre-calibrated correspondence between the compensation lens movement distance and refractive power.

8. The acquisition method of the comprehensive ophthalmic imaging system based on swept-source OCT according to claim 4 or 6, characterized in that: The computer calculates parameters along the axial length of the eye based on OCT images. These parameters include the axial length of the eye being measured, the central corneal thickness, the anterior chamber depth, and the central lens thickness.

Images