Methods, devices, multi-path OCT systems, and media for determining axial eye length

Abstract

According to an example embodiment of the present disclosure, a method, device, system and medium for determining axial length of an eye are provided. The method comprises: obtaining a first OCT image and three second OCT images of a target eye taken via a multi-optical-path OCT system, the first OCT image comprising an anterior segment of the target eye, one of the three second OCT images comprising a retina of the target eye; determining a corneal vertex position of the target eye based on the first OCT image; for each of the second OCT images: performing lateral pixel accumulation on the second OCT image to generate a one-dimensional result; performing local peak searching on the one-dimensional result to obtain a maximum local peak position and a corresponding forward gradient value; determining a maximum local peak position corresponding to a maximum value of the three forward gradient values as a RPE layer position; and determining an axial length of the target eye based on the corneal vertex position, the RPE layer position and a predetermined calibration parameter. Thus, the axial length of the eye can be accurately determined.

Description

Technical Field

[0001] The embodiments disclosed herein generally relate to the field of image processing, and more specifically to a method for determining axial length, an electronic device, a multi-path OCT system, and a computer-readable medium. Background Technology

[0002] Optical coherence tomography (OCT) was proposed in 1991 by Huang et al. at MIT and has seen tremendous development and application over the decades. This technique utilizes the principle of a Michelson interferometer, replacing one of the mirrors with the sample. Depth information of the sample is obtained by coherently imaging the reflected light from the sample arm and the reference arm. OCT technology features rapid imaging, deep imaging depth, non-invasiveness, non-contact operation, and low cost, leading to its widespread development and application in medical testing. Compared to other imaging methods in terms of resolution and imaging depth, OCT imaging technology fills the gap between confocal microscopy and ultrasound.

[0003] Based on specific imaging principles and system design, OCT systems can be divided into time-domain OCT (TD-OCT) and frequency-domain OCT (FD-OCT). Frequency-domain OCT can be further divided into spectral-domain OCT (SD-OCT) and swept-source OCT (SS-OCT) depending on the light source and receiver used. Currently, frequency-domain OCT, because it can directly obtain the depth information of the sample and eliminates the need for reference arm movement, has greatly improved imaging speed and has essentially replaced time-domain OCT as the main equipment for ophthalmic examination. Currently, the maximum imaging depth of traditional OCT is around 7mm, while the axial length of the eye can reach nearly 30mm. Therefore, a single optical path cannot measure the axial length (the distance from the corneal apex to the RPE layer (retinal pigment epithelium)). In current OCT systems, multiple mirrors are added to the reference arm to extend the depth range of OCT. In practical systems, OCT images at different depths can be reconstructed. While this design solves the current measurement range problem of OCT, the inherent system design and noise of spectral domain OCT, especially the characteristic of the reconstructed signal weakening with depth, pose certain difficulties and challenges for the actual localization of the RPE layer. Using traditional signal-to-noise ratio methods may lead to misinterpretations of the actual image, resulting in incorrect axial lengths. Summary of the Invention

[0004] Embodiments of this disclosure provide a method, electronic device, multi-path OCT system, and computer-readable medium for determining axial length, thereby enabling accurate calculation of the retinal (RPE layer) position and correct selection of the imaging optical path under the condition that the background signal value decreases with axial depth, eliminating interference from low signal, noise, and negative frequency during OCT image reconstruction, and thus calculating axial length more accurately.

[0005] In a first aspect of this disclosure, a method for determining axial length is provided. The method includes: acquiring a first OCT image of a target eye and three second OCT images captured by a multi-path OCT system, the first OCT image including the anterior segment of the target eye, and one of the three second OCT images including the retina of the target eye; determining the corneal vertex position of the target eye based on the first OCT image; for each of the three second OCT images, performing the following steps: horizontally accumulating pixels in the second OCT image to generate a one-dimensional result; performing a local peak search on the one-dimensional result to obtain the maximum local peak position and the corresponding forward gradient value; determining the maximum local peak position corresponding to the maximum of the three forward gradient values ​​corresponding to the three second OCT images as the RPE layer position; and determining the axial length of the target eye based on the corneal vertex position, the RPE layer position, and predetermined calibration parameters.

[0006] In a second aspect of this disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method according to the first aspect of this disclosure.

[0007] In a third aspect of this disclosure, a multi-path OCT system is provided, including the electronic equipment described in the second aspect.

[0008] In a fourth aspect of this disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, characterized in that the computer instructions are used to cause a computer to perform the method described according to a first aspect of this disclosure.

[0009] The summary section is provided to present the chosen concepts in a simplified form, which will be further described in the detailed description below. The summary section is not intended to identify key or essential features of this disclosure, nor is it intended to limit the scope of this disclosure. Attached Figure Description

[0010] The above and other objects, features and advantages of this disclosure will become more apparent from the accompanying drawings, in which like reference numerals generally denote like parts.

[0011] Figure 1 A schematic diagram of an example environment 100 according to an embodiment of the present disclosure is shown;

[0012] Figure 2 A schematic diagram of a method 200 for determining axial length according to an embodiment of the present disclosure is shown;

[0013] Figure 3 A schematic diagram of a target eye imaging result 300 according to an embodiment of the present disclosure is shown;

[0014] Figure 4 A schematic diagram of a standard imaging result 400 according to an embodiment of the present disclosure is shown;

[0015] Figure 5 A schematic diagram of a method 500 for performing local peak search on a one-dimensional result according to an embodiment of the present disclosure is shown;

[0016] Figure 6 A schematic diagram is shown as an example of a method 600 for determining the location of the maximum local peak corresponding to the maximum of three forward gradient values ​​corresponding to three second OCT images, as the location of the RPE layer, according to an embodiment of the present disclosure.

[0017] Figure 7 A schematic diagram illustrating an example of a method 700 for determining the axial length of a target eye according to an embodiment of the present disclosure is shown.

[0018] Figure 8 A block diagram schematically illustrates an electronic device 800 suitable for implementing embodiments of the present disclosure;

[0019] Figure 9 A schematic diagram showing the position a of the corneal apex, the position b of the RPE layer, the vertical length f of the image, and the spacing S12 between images img_0 and img_1 is shown.

[0020] Figure 10 A schematic diagram of the front and rear surface imaging of the standard is shown in different images.

[0021] In the various figures, the same or corresponding reference numerals indicate the same or corresponding parts. Detailed Implementation

[0022] Preferred embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While preferred embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure will be thorough and complete, and will fully convey the scope of the present disclosure to those skilled in the art.

[0023] As mentioned above, due to the inherent system design and noise effects of spectral domain OCT, especially the characteristic of the reconstructed signal weakening with depth, locating the RPE layer in practice presents certain difficulties and challenges. If traditional signal-to-noise ratio methods are used, misjudgments of the actual image may occur, resulting in incorrect axial lengths.

[0024] To address the aforementioned problems or other issues not described, this disclosure provides a scheme for determining axial length. In this scheme, a computing device acquires a first OCT image of a target eye and three second OCT images captured by a multi-path OCT system. The first OCT image includes the anterior segment of the target eye, and one of the three second OCT images includes the retina of the target eye. Subsequently, the computing device determines the corneal apex position of the target eye based on the first OCT image. For each of the three second OCT images, the computing device performs the following steps: horizontal pixel accumulation of the second OCT image to generate a one-dimensional result; and a local peak search is performed on the one-dimensional result to obtain the maximum local peak position and the corresponding forward gradient value. Next, the computing device determines the maximum local peak position corresponding to the maximum of the three forward gradient values ​​corresponding to the three second OCT images as the RPE layer position. Finally, the computing device determines the axial length of the target eye based on the corneal apex position, the RPE layer position, and pre-determined calibration parameters.

[0025] Therefore, under the condition that the background signal value decreases with axial depth, the position of the retina (RPE layer) can be accurately calculated and the correct imaging optical path can be selected, eliminating the interference of low signal, noise and negative frequency during OCT image reconstruction, thus calculating the axial length of the eye more accurately.

[0026] The following is a detailed explanation with reference to the accompanying drawings.

[0027] Figure 1 A schematic diagram of an example environment 100 according to an embodiment of the present disclosure is shown. (As...) Figure 1 As shown, the example environment 100 includes a computing device 110, a first OCT image 120, three second OCT images 130-1 to 130-3 (hereinafter collectively referred to as 130), and an axial length 140.

[0028] The computing device 110 may include, but is not limited to, a multi-path OCT system, a personal computer, a personal digital assistant, a wearable device, a tablet computer, a smartphone, etc. In some embodiments, the computing device 110 may have or be coupled to an image acquisition device for acquiring images of the human eye.

[0029] The computing device 110 can acquire a first OCT image 120 and three second OCT images 130 of a target eye, captured by a multi-path OCT system. The first OCT image includes the anterior segment of the target eye, and one of the three second OCT images includes the retina of the target eye. The computing device 110 can determine the corneal apex position of the target eye based on the first OCT image 120. For each of the three second OCT images 130, the computing device 110 can also perform the following steps: lateral pixel accumulation of the second OCT image 130 to generate a one-dimensional result; and local peak search of the one-dimensional result to obtain the maximum local peak position and the corresponding forward gradient value. Subsequently, the computing device 110 can determine the maximum local peak position corresponding to the maximum of the three forward gradient values ​​corresponding to the three second OCT images 130 as the RPE layer position. Finally, the computing device 110 can determine the axial length 140 of the target eye based on the corneal apex position, the RPE layer position, and predetermined calibration parameters.

[0030] Therefore, under the condition that the background signal value decreases with axial depth, the present invention can achieve accurate calculation of the retinal (RPE layer) position and correct selection of the imaging optical path, eliminate interference from low signal, noise and negative frequency during OCT image reconstruction, and thus calculate the axial length of the eye more accurately.

[0031] Figure 2 A schematic diagram illustrating an example of a method 200 for determining axial length according to embodiments of the present disclosure is shown. Figure 2 In the middle, each action, for example, can be generated by... Figure 1 The computing device shown performs the operation. It should be understood that method 200 may also include additional actions not shown and / or the actions shown may be omitted, and the scope of this disclosure is not limited in this respect.

[0032] At frame 202, computing device 110 acquires a first OCT image of the target eye and three second OCT images captured by a multi-path OCT system. The first OCT image includes the anterior segment of the target eye, and one of the three second OCT images includes the retina (also called the RPE layer) of the target eye.

[0033] See Figure 3, the multi - optical - path OCT system can capture a first OCT image img01 of the target eye and three second OCT images img02, img03, and img04, which correspond to three channel modes (also known as optical paths). The anterior segment part (front surface) appears in the first OCT image img01. Depending on the axial length of different subjects or the specifications of the standardizer, the retina (rear surface) may appear in one of the three second OCT images img02, img03, or img04. Among the three second OCT images, one is a real retinal reconstruction image, at least one is an empty image, and there may be one reverse artifact (image of negative frequency, such as Figure 3 in img03).

[0034] Returning to Figure 2 , at block 204, the computing device 110 determines the corneal vertex position of the target eye based on the first OCT image.

[0035] In some embodiments, the computing device 110 can obtain a predetermined number of central points from the first OCT image to generate a first matrix. For example, the resolution of the first OCT image is 2048 ×1024 (m1 ×n), and 600 central points can be intercepted. After interception, it is 2048 ×600 (m2 ×n, m2 < m1), denoted as the first matrix M1. In some examples, the first OCT image can be filtered to remove noise before generating the first matrix.

[0036] After generating the first matrix, for any m - th row in the first matrix, the computing device 110 can determine the difference between the value of the n - th column in the m - th row and the value of the n - th column in the (m - 1) - th row in the first matrix as the value of the n - th column in the m - th row in the second matrix. For example, take the second matrix M2, with the same size as the first M1, and all initial pixel values are 0. The values of M2 start from the row of 1 + 10(1 + num) to 2024 - 10(2024 - num) and are calculated as M2[m,n]=M1[m,n]-M1[m - 1,n]. The value of num can be preset, such as 0, 5, 10, 15, etc.

[0037] After obtaining the second matrix, the computing device 110 can perform horizontal accumulation on the second matrix to obtain the vertical accumulation result. Finally, the computing device 110 can determine the position corresponding to the maximum value in the vertical accumulation result as the corneal vertex position.

[0038] Thus, under the conditions of low signal - to - noise ratio and strong interference, the corneal vertex position can be accurately calculated.

[0039] Returning to Figure 2For each of the three second OCT images, the computing device 110 can perform horizontal pixel accumulation on the second OCT image at box 206 to generate a one-dimensional result. The one-dimensional result can be represented as a one-dimensional curve, for example, with the vertical position of the second OCT image as the independent variable and the accumulated pixel value as the dependent variable. The computing device 110 can also perform local peak search on the one-dimensional result to obtain the location of the maximum local peak and the corresponding forward gradient value. In some examples, median filtering can be performed on the second OCT image before performing horizontal pixel accumulation. Before box 208, considering boundary noise, the pixel values ​​of the four boundary portions can be removed when converting to a one-dimensional result; in addition, smoothing filtering can be performed on the one-dimensional result.

[0040] At frame 208, computing device 110 determines the location of the maximum local peak corresponding to the maximum of the three forward gradient values ​​corresponding to the three second OCT images, as the location of the RPE layer.

[0041] At frame 210, computing device 110 determines the axial length of the target eye based on the corneal apex position, RPE layer position, and pre-determined calibration parameters.

[0042] Specifically, the computing device 110 determines the spacing between the second OCT image and the first OCT image corresponding to the RPE layer position. Subsequently, based on the determined spacing, corneal apex position, RPE layer position, average refractive index of the target eye, calibration parameters between pixels and true distance, and image longitudinal distance, the axial length is calculated using a predetermined formula.

[0043] Reference Figure 3 When the position of the corneal vertex in img01 is obtained, denoted as 'a' (which can be understood as the distance between the corneal vertex and the bottom of the img01 image), and the positions of the RPE layers in img02-04 are denoted as 'b', 'c', and 'd' respectively (which can be understood as the distances between the RPE layers and the tops of img02, img03, and img04), if the anterior and posterior surfaces (cornea and retina) are imaged on img01 and img02, then the axial length of the eye is length = (a + b). p / nx+S12(1). If the images on the anterior and posterior surfaces are at img01 and img03, then the axial length lengh = [(a+c)]. p+f] / nx+S12+S23(2). If the images of the front and back surfaces are at img01 and img04, then the axial length lengh = [(a+d] [p+2f] / nx+S12+S23+S34(3). Where nx is the average refractive index of the target eye (a known parameter). S is the spacing between images, S12 represents the spacing between img01 and img02, S23 represents the spacing between img02 and img03, and S34 represents the spacing between img03 and img04. The spacing between different images can be obtained in advance through calibration. p is the calibration parameter between the pixel and the true distance, which can be obtained in advance through calibration, and f is the vertical length of the image (a known parameter).

[0044] Therefore, under the condition that the background signal value decreases with axial depth, the present invention can achieve accurate calculation of the retinal (RPE layer) position and correct selection of the imaging optical path, eliminate interference from low signal, noise and negative frequency during OCT image reconstruction, and thus calculate the axial length of the eye more accurately.

[0045] In some embodiments, prior to block 212, the computing device 110 may further determine whether the maximum value among the three forward gradient values ​​corresponding to the three second OCT images is greater than a predetermined gradient threshold. For example, a testing instrument can be used to determine the lowest and most acceptable forward gradient value as the predetermined gradient threshold. If the value is below this predetermined gradient threshold, it indicates that the image is unclear or the signal is unreasonable, and the computing device 110 may indicate an error or issue a warning.

[0046] If the computing device 110 determines that the maximum value of the three forward gradient values ​​corresponding to the three second OCT images is greater than a predetermined gradient threshold, then at box 212, it determines the axial length of the target eye based on the corneal vertex position, the RPE layer position, and predetermined calibration parameters.

[0047] Therefore, axial length can be calculated only when the forward gradient value is within the normal range, further improving the accuracy of axial length calculation.

[0048] The following describes the process of determining the calibration parameters, which can be obtained through calibration using a standard instrument.

[0049] First, the computing device 110 can acquire the standard axial length of the standard, as well as a third OCT image of the standard and three fourth OCT images taken by a multi-path OCT system. The third OCT image includes the anterior segment of the standard, and one of the three fourth OCT images includes the retina of the standard.

[0050] Next, the computing device 110 can determine the standard corneal vertex position of the standard based on the third OCT image.

[0051] For each of the three fourth OCT images, the computing device 110 can perform horizontal pixel accumulation on the fourth OCT image to generate a standard one-dimensional result; and perform local peak search on the one-dimensional result to obtain the standard maximum local peak position and the corresponding standard forward gradient value.

[0052] Subsequently, the computing device 110 can determine the standard maximum local peak position corresponding to the maximum of the three standard forward gradient values ​​corresponding to the three fourth OCT images, as the standard RPE layer position. These steps are similar to those for the target eye, and can be found in other parts of this document, so they will not be repeated here.

[0053] Finally, the computing device 110 can determine the calibration parameters based on the standard corneal apex position, the standard RPE layer position, and the standard axial length.

[0054] Reference Figure 4 Taking the imaging of the anterior and posterior surfaces (cornea and retina) in img_0 and img_1 as examples, the axial length is (a+b). p / nx+S12(1), where the axial length, a, b, nx, and f are all known, and p and S12 can be obtained by changing the axial length of the standard instrument. Taking the imaging of the anterior and posterior surfaces (cornea and retina) in img_0 and img_1 as examples, Figure 9 The diagram illustrates the position of the corneal apex (a), the position of the RPE layer (b), the longitudinal length of the image (f), and the spacing S12 between images img_0 and img_1. Similarly, the imaging of the anterior and posterior surfaces (cornea and retina) at img_0 and img_2, and the imaging of the anterior and posterior surfaces (cornea and retina) at img_0 and img_3, can be deduced and can also be expressed using the formula length = [(a + c)]. p+f] / nx+S12+S23(2) and length= [(a+d)] p+2f] / nx+S12+S23+S34(3) S23 and S34 are obtained through calibration. Figure 10 As shown, by changing the axial length of the standard eye, the anterior and posterior surfaces (cornea and retina) are imaged on img_0 and img_1 (arrows in the first row of the figure), on img_0 and img_2 (arrows in the second row of the figure), and on img_0 and img_3 (arrows in the third row of the figure), respectively. p, S12, S23 and S34 can be calibrated using the above formula.

[0055] Therefore, relevant calibration parameters can be obtained through pre-calibration using a standard instrument in order to accurately calculate the axial length.

[0056] Figure 5A schematic diagram illustrating an example of a method 500 for performing local peak search on one-dimensional results according to embodiments of the present disclosure is shown. Figure 5 In the middle, each action, for example, can be generated by... Figure 1 The computing device shown performs the operation. It should be understood that method 500 may also include additional actions not shown and / or the actions shown may be omitted, and the scope of this disclosure is not limited in this respect.

[0057] For each point in the one-dimensional result, the computing device 110 can determine, at frame 502, a first mean of a first predetermined number of points surrounding the point, a second mean of a third predetermined number of points surrounding the point at a forward distance of a second predetermined number of points, and a third mean of a third predetermined number of points surrounding the point at a backward distance of a second predetermined number of points.

[0058] For example, the average of n1 points near each point is denoted as Peak.

[0059] For example, calculate the mean of n² points around a forward interval of m points, denoted as side1.

[0060] For example, calculate the mean of n² points around a backward interval of m points, denoted as side2.

[0061] In some embodiments, reference coefficients coff1 and coff2 can be preset on both sides to correct the values ​​of side1 and side2, for example, side1 = coff1. side1, side2 = coff2 side2.

[0062] At box 504, computing device 110 determines whether the first mean is greater than the second mean and whether the first mean is greater than the third mean.

[0063] For example, we can determine whether Peak>side1 and Peak>side2, and if both conditions are met, it is equivalent to the true local peak value.

[0064] If, at box 504, the computing device 110 determines that the first mean is greater than the second mean and also greater than the third mean, then at box 506, the difference between the first and second means is determined as the forward gradient value corresponding to that point. For example, for a point with a true local peak, the forward gradient value at that point can be calculated as Peak-side1. Otherwise, at box 508, the forward gradient value corresponding to that point is determined to be zero.

[0065] After calculating the forward gradient value for each point in the one-dimensional result, at frame 510, the computing device 110 determines the point corresponding to the maximum value among the multiple forward gradient values ​​corresponding to the one-dimensional result as the maximum local peak position of the one-dimensional result. The maximum local peak position can be denoted as posMax, and its corresponding forward gradient value can be denoted as peakMax.

[0066] Therefore, the location of the maximum local peak can be determined from the one-dimensional result, thereby determining the location of the maximum local peak in the OCT image.

[0067] Figure 6 A schematic diagram illustrating an example of a method 600 according to embodiments of the present disclosure for determining the location of the maximum local peak corresponding to the maximum of three forward gradient values ​​corresponding to three second OCT images, as the location of the RPE layer. Figure 6 In the middle, each action, for example, can be generated by... Figure 1 The computing device shown performs the operation. It should be understood that method 600 may also include additional actions not shown and / or the actions shown may be omitted, and the scope of this disclosure is not limited in this respect.

[0068] At frame 602, computing device 110 determines the second OCT image with the lowest signal-to-noise ratio from the three second OCT images.

[0069] For example, determine Figure 3 The second OCT image img02 has the lowest signal-to-noise ratio.

[0070] At box 604, the computing device 110 uses the vertical position and pixel accumulation result in the one-dimensional result corresponding to the second OCT image with the lowest signal-to-noise ratio as the independent and dependent variables, respectively, to perform linear fitting and obtain the linear fitting result.

[0071] In some cases, the beginning and end data of the one-dimensional result can be removed before performing linear fitting. The elements in the one-dimensional result can be represented as (x, y), where x is the vertical position of the second OCT image and y is the accumulated pixel value. A linear fit of y = ax + b can be performed on the one-dimensional result to remove data oscillations and unevenness.

[0072] At box 606, the computing device 110 normalizes the strain in the straight-line fitting result using the maximum strain in the straight-line fitting result to obtain the normalized result.

[0073] For example, if the maximum dependent variable is represented as y0, then the elements in the normalization result can be represented as (x, y / y0), and y / y0 can be called the depth correction factor coff.

[0074] At box 608, computing device 110 corrects the three forward gradient values ​​based on the three normalized strains corresponding to the three maximum local peak positions in the normalized result, and obtains three corrected forward gradient values.

[0075] In some cases, the forward gradient value can be divided by the normalized dependent variable (also known as the depth correction factor) to obtain the corrected forward gradient value. For example, if the maximum local peak position is posMax, the corresponding forward gradient value is peakMax, and the depth correction factor corresponding to posMax in the normalized result is coff, then the corrected forward gradient value is peakMax / coff.

[0076] At box 610, computing device 110 determines the location of the maximum local peak corresponding to the maximum of the three modified forward gradient values ​​as the location of the RPE layer.

[0077] Therefore, under the condition that the background signal value decreases with axial depth, the forward gradient value is corrected by the depth correction factor to eliminate the influence of signal axial depth attenuation and more accurately determine the position of the RPE layer.

[0078] Figure 7 A schematic diagram illustrating an example of a method 700 for determining the axial length of a target eye according to an embodiment of the present disclosure is shown. Figure 7 In the middle, each action, for example, can be generated by... Figure 1 The illustrated computing device performs the operation. It should be understood that method 700 may also include additional actions not shown and / or the actions shown may be omitted, and the scope of this disclosure is not limited in this respect.

[0079] For each of the three second OCT images, at frame 702, the computing device 110 acquires a first portion of the image from the second OCT image, centered on a predetermined number of points before and after the location of the maximum local peak.

[0080] At frame 704, computing device 110 performs vertical pixel accumulation on the first part of the image to obtain a horizontal one-dimensional result.

[0081] At box 706, computing device 110 determines a weighted result of the mean and maximum of the horizontal one-dimensional result as a boundary threshold.

[0082] For example, a scaling factor can be set, denoted as shapeRatio1 and shapeRatio2. The boundary threshold thred = shapeRatio1 is then calculated. shapeMean+shapeRatio2 shapeMax, where shapeMean represents the mean and shapeMax represents the maximum value.

[0083] At box 708, computing device 110 determines the first position in the horizontal one-dimensional result that is greater than the boundary threshold from the left, as the left boundary; and determines the first position in the horizontal one-dimensional result that is greater than the boundary threshold from the right, as the right boundary.

[0084] At frame 710, computing device 110 acquires a second portion of the image from the second OCT image, with a predetermined number of points above and below the maximum local peak position and extending from the left boundary to the right boundary.

[0085] At box 712, computing device 110 fits a straight line based on the position of the maximum value of each column of the second part of the image.

[0086] At box 714, computing device 110 determines the slope of the straight line as the retinal tilt angle corresponding to the location of the maximum local peak.

[0087] At frame 716, computing device 110 determines whether the retinal tilt angle corresponding to the RPE layer position is less than a predetermined tilt angle threshold.

[0088] If the computing device 110 determines at frame 716 that the retinal tilt angle corresponding to the RPE layer position is less than a predetermined tilt angle threshold, then at frame 718, the computing device 110 determines the axial length of the target eye based on the corneal apex position, the RPE layer position, and predetermined calibration parameters. Otherwise, an error or warning may be displayed. The predetermined tilt angle threshold may be pre-calibrated, for example, through instrument testing.

[0089] Therefore, axial length can be calculated only when the retinal tilt angle is within the normal range, further improving the accuracy of axial length calculation.

[0090] Figure 8 A block diagram schematically illustrates an electronic device 800 suitable for implementing embodiments of the present disclosure. Device 800 can be used to implement... Figure 1 The computing device 100, as shown, includes a central processing unit (CPU) 801, which can perform various appropriate actions and processes based on computer program instructions stored in read-only memory (ROM) 802 or loaded from storage unit 808 into random access memory (RAM) 803. The RAM 803 can also store various programs and data required for the operation of the device 800. The CPU 801, ROM 802, and RAM 803 are interconnected via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0091] Multiple components in device 800 are connected to I / O interface 805, including: input unit 806, such as keyboard, mouse, etc.; output unit 807, such as various types of monitors, speakers, etc.; storage unit 808, such as disk, optical disk, etc.; and communication unit 809, such as network card, modem, wireless transceiver, etc. Communication unit 809 allows device 800 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0092] Processing unit 801 executes the various methods and processes described above, such as methods 200, 500-700. For example, in some embodiments, methods 200, 500-700 may be implemented as computer software programs stored in a machine-readable medium, such as storage unit 808. In some embodiments, part or all of the computer program may be loaded and / or installed on device 800 via ROM 802 and / or communication unit 809. When the computer program is loaded into RAM 803 and executed by CPU 801, one or more operations of methods 200, 500-700 described above may be performed. Alternatively, in other embodiments, CPU 801 may be configured to perform one or more actions of methods 200, 500-700 by any other suitable means (e.g., by means of firmware).

[0093] This disclosure can be a method, apparatus, system, and / or computer program product. A multi-path OCT system may include the aforementioned electronic device 800. The computer program product may include a computer-readable storage medium loaded with computer-readable program instructions for performing various aspects of this disclosure.

[0094] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0095] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0096] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0097] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0098] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processing unit of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0099] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0100] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0101] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, and are not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical applications, or technical improvements to the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for determining axial length, comprising: Acquire a first OCT image of the target eye and three second OCT images captured by a multi-path OCT system. The first OCT image includes the anterior segment of the target eye, and one of the three second OCT images includes the retina of the target eye. Based on the first OCT image, the position of the corneal apex of the target eye is determined; For each of the three second OCT images, perform the following steps: The second OCT image is then subjected to horizontal pixel accumulation to generate a one-dimensional result; A local peak search is performed on the one-dimensional result to obtain the location of the maximum local peak and the corresponding forward gradient value; The position of the maximum local peak value among the three forward gradient values ​​corresponding to the three second OCT images is determined as the position of the RPE layer. as well as Based on the corneal apex position, the RPE layer position, and predetermined calibration parameters, the axial length of the target eye is determined; The local peak search for the one-dimensional result includes: For each point in the one-dimensional result, perform the following steps: Determine the first mean of a first predetermined number of points surrounding the stated point; Determine the second mean of a third predetermined number of points surrounding a point that is a second predetermined number of points forward from the given point; Determine the third mean of the third predetermined number of points surrounding the point at a second predetermined number of points after the point is determined; If it is determined that the first mean is greater than the second mean and the first mean is greater than the third mean, then the difference between the first mean and the second mean is determined as the forward gradient value corresponding to the point; and Otherwise, the forward gradient value corresponding to the point is set to zero; and The point corresponding to the maximum value among the multiple forward gradient values ​​corresponding to the one-dimensional result is determined as the location of the maximum local peak of the one-dimensional result.

2. The method according to claim 1, wherein determining the location of the maximum local peak corresponding to the maximum value among the three forward gradient values ​​corresponding to the three second OCT images, as the RPE layer location, includes: The second OCT image with the lowest signal-to-noise ratio is determined from the three second OCT images; Using the vertical position and pixel accumulation result in the one-dimensional result corresponding to the second OCT image with the lowest signal-to-noise ratio as the independent and dependent variables, a straight line fitting is performed to obtain the straight line fitting result; Using the largest dependent variable in the linear fitting result, normalize the dependent variable in the linear fitting result to obtain the normalized result; Based on the three normalized dependent variables corresponding to the three maximum local peak positions in the normalized result, the three forward gradient values ​​are corrected respectively to obtain three corrected forward gradient values. as well as The location of the maximum local peak corresponding to the maximum value among the three corrected forward gradient values ​​is determined as the location of the RPE layer.

3. The method according to any one of claims 1-2, wherein determining the axial length of the target eye comprises: For each of the three second OCT images, perform the following steps: Obtain a first portion of the image from the second OCT image, centered on the maximum local peak position and taking a predetermined number of points before and after it; Vertical pixel accumulation is performed on the first part of the image to obtain a horizontal one-dimensional result; The weighted sum of the mean and maximum values ​​of the horizontal one-dimensional results is determined as the boundary threshold; The first position in the horizontal one-dimensional result that is greater than the boundary threshold from the left is determined as the left boundary, and the first position in the horizontal one-dimensional result that is greater than the boundary threshold from the right is determined as the right boundary. A second portion of the image is obtained from the second OCT image, with a predetermined number of points above and below the maximum local peak position as the center, and extending from the left boundary to the right boundary; Based on the position of the maximum value in each column of the second part of the image, a straight line is fitted; as well as The slope of the straight line is determined as the retinal tilt angle corresponding to the location of the maximum local peak. as well as If it is determined that the retinal tilt angle corresponding to the RPE layer position is less than a predetermined tilt angle threshold, then the axial length of the target eye is determined based on the corneal apex position, the RPE layer position, and predetermined calibration parameters.

4. The method according to any one of claims 1-2, wherein determining the axial length of the target eye comprises: If the maximum value among the three forward gradient values ​​corresponding to the three second OCT images is determined to be greater than a predetermined gradient threshold, the axial length of the target eye is determined based on the corneal vertex position, the RPE layer position, and the predetermined calibration parameters.

5. The method according to any one of claims 1-2, wherein determining the corneal apex position of the target eye comprises: A predetermined number of points are obtained from the center of the first OCT image to generate a first matrix; For any m-th row in the first matrix, determine the difference between the value in the m-th row and n-th column of the first matrix and the value in the (m-1)-th row and n-th column, and use it as the value in the m-th row and n-th column of the second matrix; The second matrix is ​​then summed horizontally to obtain the vertical summation result; as well as The position corresponding to the maximum value in the longitudinal accumulation result is determined as the corneal vertex position.

6. The method according to any one of claims 1-2, further comprising: The standard axial length of the standard device is obtained, along with a third OCT image and three fourth OCT images of the standard device captured by the multi-path OCT system. The third OCT image includes the anterior segment of the standard device, and one of the three fourth OCT images includes the retina of the standard device. Based on the third OCT image, the standard corneal vertex position of the standard instrument is determined; For each of the three fourth OCT images, perform the following steps: The fourth OCT image is then subjected to horizontal pixel accumulation to generate a standard one-dimensional result; A local peak search is performed on the one-dimensional result to obtain the standard maximum local peak position and the corresponding standard forward gradient value; The standard maximum local peak position corresponding to the maximum value among the three standard forward gradient values ​​corresponding to the three fourth OCT images is determined as the standard RPE layer position. as well as The calibration parameters are determined based on the standard corneal vertex position, the standard RPE layer position, and the standard axial length.

7. An electronic device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.

8. A multi-path OCT system, comprising the electronic device according to claim 7.

9. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-6.

Images