Dome-shaped macula automatic detection system and method based on OCT image, terminal and medium

Abstract

This application provides an automatic dome-shaped macular detection system, method, terminal, and medium based on OCT images. After acquiring multiple radial OCT B-scan images of the subject centered on the fovea of ​​the macula, the system uses a pre-trained automatic layer segmentation model to segment each radial OCT B-scan image to obtain the corresponding Bruch membrane curve. Then, based on the baseline connecting the dome apex and the lowest point of each Bruch membrane curve, the corresponding dome height is calculated. Each calculated dome height is compared with a preset dome height threshold to obtain the corresponding radial image dome-shaped macular detection result. All radial image dome-shaped macular detection results are statistically analyzed to obtain the final dome-shaped macular detection result. This application uses a unified standard for automated dome-shaped macular detection and can output interpretable and verifiable quantitative values ​​of the dome apex, the baseline connecting the lowest point, and the dome height, meeting the needs of large-scale detection.

Description

Technical Field

[0001] This application relates to the field of medical image processing technology, and in particular to an automatic detection system, method, terminal and medium for dome-shaped macular based on OCT images. Background Technology

[0002] Dome-shaped macular degeneration (DSM) is a fundus structural abnormality characterized by localized bulging of the Bruch's membrane and retinal pigment epithelium in the macular region towards the vitreous humor. It is commonly seen in patients with high myopia and posterior staphyloma. DSM is closely associated with vision-threatening complications such as serous retinal detachment, choroidal neovascularization, and myopic tractional macular degeneration. Accurate detection of DSM is of great value for screening, staging, and clinical follow-up of eye diseases in highly myopic individuals. Currently, clinical identification of DSM mainly relies on ophthalmologists manually interpreting radial scan images from optical coherence tomography (OCT): the doctor observes the morphology of the Bruch's membrane layer in multiple B-scans, visually judges whether there is a localized bulge, and manually selects a baseline reference point and the dome apex to calculate the height of the bulge.

[0003] Several studies have explored the automation of DSM detection. For example, deep learning-based image classification methods exist; however, these are black-box methods, as the model does not explicitly segment and analyze the retinal structure, thus failing to output interpretable quantitative parameters such as the anatomical location and height of the dome, hindering physician review and follow-up comparisons. Another example is the dome-like macular curvature index (DSMC), which quantifies local bulging by selecting a baseline reference point for the Bruch membrane and the dome apex on an OCT B-scan and calculating the ratio of dome height to chord length. However, this method still relies on manual selection of the baseline reference point and dome boundary, limiting its application in large-scale screening. Furthermore, some studies have proposed a classification framework for dome-like macular (DSM) and ridge-like macular (RSM) based on 12 isoangular radial OCT scans, requiring physicians to interpret each of the 12 scans and provide an ocular-level classification. This method relies entirely on manual interpretation of each scan, taking several minutes for a single eye, resulting in low efficiency and poor consistency between different physicians. Summary of the Invention

[0004] In view of the shortcomings of the prior art described above, the purpose of this application is to provide an automatic dome-shaped macular detection system, method, terminal and medium based on OCT images to solve the problems of existing dome-shaped macular detection being non-fully automatic and unable to quantify.

[0005] To achieve the above and other related objectives, a first aspect of this application provides an automatic dome-like macular detection system based on OCT images, comprising: an acquisition module for acquiring multiple radial OCT B-scan images of a subject centered on the fovea of ​​the macula; a Bruch membrane curve acquisition module for segmenting each radial OCT B-scan image using a pre-trained automatic layer segmentation model to obtain a corresponding Bruch membrane curve; a dome height calculation module for calculating the corresponding dome height based on the baseline connecting the dome apex and the lowest point of each determined Bruch membrane curve; a detection module for comparing each calculated dome height with a preset dome height threshold to obtain the corresponding radial image dome-like macular detection result; and statistically analyzing all radial image dome-like macular detection results to obtain the final dome-like macular detection result.

[0006] In some embodiments of the first aspect of this application, the Bruch film curve acquisition module includes: a Bruch film segmentation unit, used to segment each radial OCT B-scan image using a pre-trained automatic layer segmentation model to obtain the corresponding Bruch film pixel curve and make an effective judgment; and a Bruch film curve generation unit, used to process each Bruch film pixel curve according to the effective judgment result of each Bruch film pixel curve to obtain the corresponding Bruch film curve.

[0007] In some embodiments of the first aspect of this application, each Bruch film pixel curve is processed according to the valid judgment result of each Bruch film pixel curve to obtain the corresponding Bruch film curve, including: if the valid judgment result is valid, then the Bruch film pixel curve is physically transformed to obtain the corresponding Bruch film curve; if the valid judgment result is invalid, then the Bruch film pixel curve is physically transformed and smoothed to obtain the corresponding Bruch film curve.

[0008] In some embodiments of the first aspect of this application, the system further includes a curve feature determination module; wherein the curve feature determination module includes: a retrieval range determination unit, configured to divide each Bruch film curve into a dome vertex retrieval range and a minimum point retrieval range; a feature determination unit, configured to determine the point in the dome vertex retrieval range of each Bruch film curve whose tangent slope is closest to zero as the dome vertex of each Bruch film curve, and to determine the point in the minimum point retrieval range of each Bruch film curve whose tangent slope is closest to zero as the minimum point of each Bruch film curve; and to determine the line connecting the minimum points of each Bruch film curve as the baseline connecting the minimum points of each Bruch film curve.

[0009] In some embodiments of the first aspect of this application, the corresponding dome height is calculated based on the baseline connecting the dome apex and the lowest point of each Bruch membrane curve, including: calculating the orthogonal distance from the dome apex to the baseline connecting the lowest point of each Bruch membrane curve and using it as the dome height.

[0010] In some embodiments of the first aspect of this application, the calculated height of each dome is compared with a preset dome height threshold to obtain the corresponding radial image dome-like macular detection result, including: if the calculated dome height is greater than or equal to the preset dome height threshold, the obtained radial image dome-like macular detection result is positive; if the calculated dome height is less than the preset dome height threshold, the obtained radial image dome-like macular detection result is negative.

[0011] In some embodiments of the first aspect of this application, the final dome-shaped macular detection result is obtained by statistically analyzing all radial image dome-shaped macular detection results, including: if at least one positive radial image dome-shaped macular detection result exists among all radial image dome-shaped macular detection results, the final dome-shaped macular detection result is positive; if all radial image dome-shaped macular detection results are negative, the final dome-shaped macular detection result is negative.

[0012] To achieve the above and other related objectives, a second aspect of this application provides an automatic dome-like macular detection method based on OCT images, comprising: acquiring multiple radial OCT B-scan images of a subject centered on the fovea of ​​the macula; segmenting each radial OCT B-scan image using a pre-trained automatic layer segmentation model to obtain the corresponding Bruch membrane curve; calculating the corresponding dome height based on the baseline connecting the dome apex and the lowest point of each Bruch membrane curve; comparing each calculated dome height with a preset dome height threshold to obtain the corresponding radial image dome-like macular detection result; and statistically analyzing all radial image dome-like macular detection results to obtain the final dome-like macular detection result.

[0013] To achieve the above and other related objectives, a third aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the dome-shaped macular automatic detection method based on OCT images.

[0014] To achieve the above and other related objectives, a fourth aspect of this application provides an electronic terminal, including a memory, a processor, and a computer program stored in the memory; the processor executes the computer program to implement the automatic detection method for dome-like macular based on OCT images.

[0015] As described above, the dome-shaped macular automatic detection system, method, terminal and medium based on OCT images of this application have the following beneficial effects: This application adopts a unified standard for automated detection of dome-shaped maculars, and can output interpretable and verifiable quantitative values ​​of the baseline connecting the dome apex and the lowest point, as well as the dome height, which can meet the needs of large-scale detection. Attached Figure Description

[0016] Figure 1 The diagram shown is a schematic block diagram of an automatic dome-shaped macular detection system based on OCT images, according to one embodiment of this application.

[0017] Figure 2 The diagram shown is a schematic representation of the curve region division in one embodiment of this application.

[0018] Figure 3 The diagram shown is a schematic diagram of dome height calculation in one embodiment of this application.

[0019] Figure 4 (a) shows the Bruch membrane segmentation result of a normal eye in a specific embodiment of this application.

[0020] Figure 4 (b) Shows the Bruch membrane segmentation results of a dome-like macular positive eye in a specific embodiment of this application.

[0021] Figure 5 (a) shows the Bruch membrane curve of a normal eye in a specific embodiment of this application.

[0022] Figure 5 (b) Shows the Bruch membrane curve of a dome-like macular positive eye in a specific embodiment of this application.

[0023] Figure 6 The diagram shown is a schematic representation of dome height measurement in a specific embodiment of this application.

[0024] Figure 7 The diagram shown is a flowchart illustrating an automatic detection method for dome-shaped macular based on OCT images in one embodiment of this application.

[0025] Figure 8 The diagram shown is a structural schematic of an electronic terminal according to an embodiment of this application. Detailed Implementation

[0026] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.

[0027] In the embodiments of this application, the terms "first" and "second" are used to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and that the terms "first" and "second" do not necessarily imply that they are different.

[0028] It should be noted that, in the embodiments of this application, the words "exemplary" or "for example" indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0029] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0030] Before providing a further detailed description of the present invention, the nouns and terms used in the embodiments of the present invention are explained, and the nouns and terms used in the embodiments of the present invention are subject to the following interpretations:

[0031] <1> OCT (Optical Coherence Tomography): OCT is a non-invasive imaging technique based on the principle of low coherence optical interference, which can perform tomographic scans of biological tissues or materials with micron-level precision.

[0032] <2> B-scan (Brightness scan, B-mode scan): B-scan is the most commonly used two-dimensional tomographic imaging method in OCT, which intuitively presents the cross-sectional structure inside the target.

[0033] <3> Bruch's membrane (BM): An elastic membrane located beneath the retinal pigment epithelium, it is the primary anatomical reference layer for measuring macular bulge in DSM testing.

[0034] <4> Dome-shaped macular (DSM): A fundus structural abnormality characterized by a localized bulge of the Bruch membrane in the macular region toward the vitreous body, which is more common in patients with high myopia.

[0035] <5> The fovea centralis, or simply fovea, is a tiny depression in the center of the macula of the retina. It is the core area where human vision is most acute and has the highest resolution.

[0036] To facilitate understanding of the embodiments of this application, firstly, in conjunction with Figure 1 Detailed explanation. Figure 1 A schematic block diagram of an automatic dome-shaped macular detection system based on OCT images, according to an embodiment of the present invention, is shown. The automatic dome-shaped macular detection system 100 based on OCT images in this embodiment includes:

[0037] The acquisition module 101 is used to acquire multiple radial OCT B-scan images of the subject centered on the fovea of ​​the macula;

[0038] The Bruch film curve acquisition module 102 is used to segment each radial OCT B-scan image using a pre-trained automatic layer segmentation model to obtain the corresponding Bruch film curve.

[0039] The dome height calculation module 103 is used to calculate the corresponding dome height based on the baseline connecting the dome apex and the lowest point of each Bruch membrane curve.

[0040] The detection module 104 is used to compare the calculated height of each dome with a preset dome height threshold to obtain the corresponding radial image dome-like macular detection result; and to count all radial image dome-like macular detection results to obtain the final dome-like macular detection result.

[0041] It should be understood that the module division in the embodiments of this application is illustrative and only represents a logical functional division. In actual implementation, there may be other division methods. Furthermore, the functional modules in the various embodiments of this application can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0042] In one embodiment, the OCT device scans the macula of the subject in radial scanning mode to obtain multiple isoangular radial OCT B-scan images centered on the fovea of ​​the macula. It should be understood that an isoangular radial B-scan is a two-dimensional cross-sectional imaging mode that uses a radially arranged, evenly spaced angular pattern. Its core lies in the fact that the probe or detection beam rotates around a fixed central axis, sequentially acquiring B-scan data at equal angular intervals, ultimately forming a fan-shaped or ring-shaped cross-sectional view. In OCT imaging, each scan line corresponds to one axial sampling path of the probe or detection beam, and the cross-sectional B-scan is formed by a linear arrangement of axial A-scans. Each scan line covers a predetermined range on both sides of the fovea of ​​the macula.

[0043] In one embodiment, the training process of the automatic layer segmentation model includes: acquiring radial OCTB-scan images of multiple macular regions and annotating them with Bruch membrane layers to construct a training set; and training a convolutional encoder-decoder network using the contrast-normalized training set to obtain a trained automatic layer segmentation model. The convolutional encoder-decoder network includes an encoder, a decoder, and skip connections. The encoder consists of multiple convolutional blocks, extracting image features layer by layer through downsampling. The decoder recovers segmentation accuracy layer by layer through upsampling. Skip connections directly copy and stitch feature maps of the same size from the encoder to the corresponding layers of the decoder. It should be noted that the convolutional encoder-decoder network can refer to existing technologies, and will not be elaborated further here.

[0044] In one embodiment, the Bruch film curve acquisition module includes: a Bruch film segmentation unit, used to segment each radial OCT B-scan image using a pre-trained automatic layer segmentation model to obtain the corresponding Bruch film pixel curve and make an effective judgment; and a Bruch film curve generation unit, used to process each Bruch film pixel curve according to the effective judgment result of each Bruch film pixel curve to obtain the corresponding Bruch film curve.

[0045] Specifically, using a pre-trained automatic layer segmentation model, each radial OCT B-scan image is segmented to obtain the corresponding Bruch film pixel curve. This includes: inputting each radial OCT B-scan image into the pre-trained automatic layer segmentation model after contrast normalization to determine the Bruch film in each radial OCT B-scan image; obtaining all pixel coordinates of the Bruch film in each radial OCT B-scan image; and connecting these pixel coordinates in sequence to form the Bruch film pixel curve.

[0046] In one embodiment, the valid determination of the Bruch film pixel curve includes: if the Bruch film pixel curve is a continuous curve, the Bruch film pixel curve is determined to be valid, and the generated valid determination result is valid; if the Bruch film pixel curve is a discontinuous curve (e.g., there are breakpoints or gaps), the Bruch film pixel curve is determined to be invalid, and the generated valid determination result is invalid.

[0047] In one embodiment, based on the valid judgment result of each Bruch film pixel curve, each Bruch film pixel curve is processed accordingly to obtain the corresponding Bruch film curve, including: if the valid judgment result is valid, then the Bruch film pixel curve is physically transformed to obtain the corresponding Bruch film curve; if the valid judgment result is invalid, then the Bruch film pixel curve is physically transformed and smoothed to obtain the corresponding Bruch film curve.

[0048] Specifically, for each Bruch film pixel curve, if the validity determination result of the Bruch film pixel curve is valid, then the Bruch film pixel curve is physically transformed according to the scanning resolution parameter to obtain the corresponding Bruch film curve; if the validity determination result of the Bruch film pixel curve is invalid, then the Bruch film pixel curve is physically transformed and smoothed according to the scanning resolution parameter to obtain the corresponding Bruch film curve. It should be noted that this embodiment does not limit the order of physical coordinate transformation and smoothing. The Bruch film pixel curve can be physically transformed according to the scanning resolution parameter first and then smoothed, or the Bruch film pixel curve can be smoothed first and then the physical coordinate transformation can be performed according to the scanning resolution parameter.

[0049] In one embodiment, the scan resolution parameters are automatically read from the OCT device metadata after scanning is completed. The scan resolution parameters include vertical resolution and horizontal resolution. It should be understood that horizontal resolution refers to the pixel density of the scanned image in the horizontal direction, and vertical resolution refers to the pixel density of the scanned image in the vertical direction. Physical coordinate transformation of the Bruch film pixel curve based on the scan resolution parameters means converting each pixel coordinate of the Bruch film pixel curve into its corresponding actual physical coordinates according to the scan resolution parameters, and connecting these actual physical coordinates to obtain the corresponding curve. Specifically, for any pixel coordinate (width pixels, height pixels) of the Bruch film pixel curve, the actual physical coordinates are (horizontal coordinate, vertical coordinate), where width pixels = horizontal coordinate × horizontal resolution, and height pixels = vertical coordinate × vertical resolution. It should be understood that the horizontal coordinate, with the midpoint of the scan line as the origin, represents the physical distance from the fovea centralis, and the total coordinate represents the depth.

[0050] In one embodiment, a smoothing filter is used for smoothing, which suppresses high-frequency noise while preserving the morphological characteristics of the curve. It should be noted that the smoothing filter can refer to existing technologies, and will not be described in detail here.

[0051] In one embodiment, the dome-like macular automatic detection system based on OCT images further includes a curve feature determination module; wherein, the curve feature determination module includes: a retrieval range determination unit, used to divide each Bruch film curve into a dome vertex retrieval interval and a minimum point retrieval interval; a feature determination unit, used to determine the point in the dome vertex retrieval interval of each Bruch film curve whose tangent slope is closest to zero as the dome vertex of each Bruch film curve, and to determine the point in the minimum point retrieval interval of each Bruch film curve whose tangent slope is closest to zero as the minimum point of each Bruch film curve; and to determine the baseline of the line connecting the minimum points of each Bruch film curve as the minimum point connecting line of each Bruch film curve. It should be understood that the tangent slope refers to the inclination of the tangent line of a function at a certain point, and is usually calculated using the derivative at that point.

[0052] Specifically, such as Figure 2 As shown, the curve intervals corresponding to the preset search range on both sides of the origin (representing the fovea of ​​the macula) of each Bruch film curve are divided into the dome apex search intervals of the Bruch film curve, and the remaining two curve intervals are divided into the lowest point search intervals of the Bruch film curve. For ease of explanation, these two lowest point search intervals are referred to as lowest point search interval 1 and lowest point search interval 2, respectively.

[0053] The point whose tangent slope is closest to zero within the retrieval interval of the Bruch film curve's dome apex is determined as the dome apex of the Bruch film curve. The point whose tangent slope is closest to zero within retrieval interval 1 of the Bruch film curve's lowest point is determined as one of the Bruch film curve's lowest points. The point whose tangent slope is closest to zero within retrieval interval 2 of the Bruch film curve's lowest point is determined as the other lowest point of the Bruch film curve. The line connecting the two lowest points is determined as the baseline of the lowest point line of the Bruch film curve.

[0054] In one embodiment, such as Figure 3 As shown, based on the baseline connecting the dome apex and the lowest point of each Bruch membrane curve, the corresponding dome height is calculated, including: calculating the orthogonal distance from the dome apex of each Bruch membrane curve to the baseline connecting the lowest point of that Bruch membrane curve and using this distance as the dome height. It should be understood that the orthogonal distance is the shortest distance along the vertical direction from the dome apex to the baseline connecting the lowest point.

[0055] In one embodiment, the calculated height of each dome is compared with a preset dome height threshold to obtain the corresponding radial image dome-like macular detection result, including: if the calculated dome height is greater than or equal to the preset dome height threshold, the obtained radial image dome-like macular detection result is positive; if the calculated dome height is less than the preset dome height threshold, the obtained radial image dome-like macular detection result is negative.

[0056] Specifically, after processing each radial OCT B-scan image in the manner described in the above embodiments, a dome height is calculated. By comparing the dome height calculated based on each radial OCT B-scan image with a preset dome height threshold, the radial image dome-like macular detection result corresponding to each radial OCT B-scan image can be obtained.

[0057] It should be understood that the dome height threshold can be set according to requirements and is not limited here.

[0058] In one embodiment, the final dome-shaped macular detection result is obtained by statistically analyzing all radial image dome-shaped macular detection results, including: if there is at least one positive radial image dome-shaped macular detection result among all radial image dome-shaped macular detection results, the final dome-shaped macular detection result is positive; if all radial image dome-shaped macular detection results are negative, the final dome-shaped macular detection result is negative.

[0059] To better illustrate the automatic dome-shaped macular detection system based on OCT images of the present invention, specific embodiments are provided. Embodiment 1: A detection process of an automatic dome-shaped macular detection system based on OCT images.

[0060] First, a sweep-frequency OCT device was used to acquire 12 isoangular radial B-scan images centered on the fovea of ​​the macula, with scanning angle intervals of 15° (0°, 15°, 30°, ..., 165° sequentially). Each scan line covered approximately 4.5 mm on each side of the fovea. The longitudinal resolution (approximately 2.609 μm / pixel) and lateral resolution (approximately 8.789 μm / pixel) were automatically read from the device's metadata. Next, each B-scan image was contrast-normalized and input into a pre-trained automatic segmentation model to obtain the Bruch membrane curve for each B-scan image. Then, based on the resolution parameters, the pixel coordinates were converted to physical coordinates (in μm) with the fovea as the origin, and the curves were smoothed to suppress high-frequency noise. Subsequently, the dome apex was detected within the central constraint area, a baseline was constructed connecting the two lowest points, and the dome height was calculated. Finally, the dome height was compared with a preset dome height threshold to determine whether each B-scan image was DSM-positive.

[0061] Example 2: A result of automatic layer segmentation.

[0062] Figure 4 (a) shows the Bruch membrane segmentation results of a normal eye. Figure 4 (b) illustrates the Bruch membrane segmentation results of a dome-shaped macular-positive eye (DSM-positive eye). The segmentation network used in this embodiment is a convolutional encoder-decoder network. The encoder of the convolutional encoder-decoder network consists of four convolutional blocks (each layer contains two 3×3 convolutions, batch normalization, ReLU activation, and max-pooling downsampling), and the decoder consists of four corresponding upsampling blocks (bilinear upsampling and skip connection fusion). The output layer corresponds to a pixel-wise probability map of the Bruch membrane. As can be seen from the figure, the Bruch membrane of a normal eye has a gentle arc shape, while the Bruch membrane of a DSM-positive eye exhibits a significant upward bulge near the fovea.

[0063] Figure 5 (a) shows the Bruch membrane curve of a normal eye. Figure 5 (b) The Bruch membrane curve of a dome-shaped macular-positive eye (DSM-positive eye) is shown. After converting the segmented Bruch membrane pixel coordinates to physical coordinates, the differences between the normal eye and the DSM-positive eye can be clearly compared. The Bruch membrane curve of the normal eye has a uniform U-shaped curvature, while the curve of the DSM-positive eye shows a local bulge (dome) near the fovea. The highest point of this bulge is the dome apex.

[0064] Example 3: A specific process for measuring the height of a dome.

[0065] like Figure 6 As shown, firstly, the points on the Bruch membrane curves with the slopes closest to zero on both sides of the dome apex are searched as the lowest points on both sides. Then, the line connecting the two lowest points is used as the baseline. Finally, the orthogonal distance from the dome apex to this baseline is calculated, which is the dome height. In this embodiment, the measured dome height of the DSM-positive eye is 171 μm, which exceeds the positive threshold (set to 50 μm in this embodiment), therefore it is determined to be DSM-positive.

[0066] The beneficial effects of this invention are:

[0067] (1) Automation and efficiency (solving the problem of low efficiency): It realizes the full-process automation from OCT image input to DSM detection and quantitative height output. The processing time of a single scan is completed in seconds, which significantly improves efficiency compared to the several minutes required for manual interpretation. It can directly support large-scale DSM screening of highly myopic populations.

[0068] (2) Accuracy: The system was validated on DSM cases manually annotated by doctors. The system had a high detection rate for cases with larger dome heights and showed good consistency with the results of manual measurement.

[0069] (3) Objectivity and interpretability (solving the problem of strong subjectivity): The system outputs all verifiable intermediate quantities such as the layer curve, the physical coordinate position of the dome apex, and the quantitative value of the dome height, rather than just providing uninterpretable classification labels.

[0070] (4) Quantification and standardization (solving the problem of inconsistent quantitative standards): All dome heights are output in physical units (e.g., μm). Baseline construction adopts automated deterministic rules, eliminating measurement inconsistencies caused by different operators selecting different baselines, and supporting longitudinal follow-up comparisons.

[0071] Figure 7 This is a schematic flowchart of the automatic detection method for dome-shaped macular based on OCT images provided in an embodiment of this application. Figure 7 As shown, the automatic detection method for dome-shaped macular based on OCT images includes:

[0072] Step S701: Acquire multiple radial OCT B-scan images of the subject centered on the fovea of ​​the macula.

[0073] Step S702: Using a pre-trained automatic layer segmentation model, segment each radial OCT B-scan image to obtain the corresponding Bruch film curve.

[0074] Step S703: Calculate the corresponding dome height based on the baseline connecting the dome apex and the lowest point of each Bruch membrane curve.

[0075] Step S704: Compare the calculated height of each dome with the preset dome height threshold to obtain the corresponding radial image dome-like macular detection result; count all radial image dome-like macular detection results to obtain the final dome-like macular detection result.

[0076] It should be understood that the specific implementation process of each step of the method has been described in detail in the above embodiments, and will not be repeated here for the sake of brevity.

[0077] In one embodiment, a pre-trained automatic layer segmentation model is used to segment each radial OCT B-scan image to obtain the corresponding Bruch film curve. This includes: segmenting each radial OCT B-scan image using the pre-trained automatic layer segmentation model to obtain the corresponding Bruch film pixel curve and making an effective judgment; and processing each Bruch film pixel curve according to the effective judgment result of each Bruch film pixel curve to obtain the corresponding Bruch film curve.

[0078] In one embodiment, based on the valid judgment result of each Bruch film pixel curve, each Bruch film pixel curve is processed accordingly to obtain the corresponding Bruch film curve, including: if the valid judgment result is valid, then the Bruch film pixel curve is physically transformed to obtain the corresponding Bruch film curve; if the valid judgment result is invalid, then the Bruch film pixel curve is physically transformed and smoothed to obtain the corresponding Bruch film curve.

[0079] In one embodiment, the specific method for determining the baseline connecting the dome apex and the lowest point of each Bruch membrane curve includes: dividing each Bruch membrane curve into a dome apex retrieval interval and a lowest point retrieval interval; determining the point in the dome apex retrieval interval of each Bruch membrane curve whose tangent slope is closest to zero as the dome apex of each Bruch membrane curve, and determining the point in the lowest point retrieval interval of each Bruch membrane curve whose tangent slope is closest to zero as the lowest point of each Bruch membrane curve; and determining the line connecting the lowest points of each Bruch membrane curve as the baseline connecting the lowest points of each Bruch membrane curve.

[0080] In one embodiment, the dome height is calculated based on the baseline connecting the dome apex and the lowest point of each Bruch membrane curve, including: calculating the orthogonal distance from the dome apex to the baseline connecting the lowest point of each Bruch membrane curve and using it as the dome height.

[0081] In one embodiment, the calculated height of each dome is compared with a preset dome height threshold to obtain the corresponding radial image dome-like macular detection result, including: if the calculated dome height is greater than or equal to the preset dome height threshold, the obtained radial image dome-like macular detection result is positive; if the calculated dome height is less than the preset dome height threshold, the obtained radial image dome-like macular detection result is negative.

[0082] In one embodiment, the final dome-shaped macular detection result is obtained by statistically analyzing all radial image dome-shaped macular detection results, including: if there is at least one positive radial image dome-shaped macular detection result among all radial image dome-shaped macular detection results, the final dome-shaped macular detection result is positive; if all radial image dome-shaped macular detection results are negative, the final dome-shaped macular detection result is negative.

[0083] Figure 8 This is a schematic block diagram of the electronic terminal provided in an embodiment of this application. Figure 8 As shown, the electronic terminal 800 includes at least one processor 801, a memory 802, at least one network interface 803, and a user interface 805. The various components in the electronic terminal 800 are coupled together via a bus system 804. It is understood that the bus system 804 is used to implement communication between these components. In addition to a data bus, the bus system 804 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in… Figure 8 The general will label all buses as bus systems.

[0084] The user interface 805 may include a monitor, keyboard, mouse, trackball, clicker, button, touchpad, or touch screen.

[0085] It is understood that memory 802 can be volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM) or programmable read-only memory (PROM), which serves as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM) and synchronous static random access memory (SSRAM). The memories described in the embodiments of this invention are intended to include, but are not limited to, these and any other suitable categories of memory.

[0086] In this embodiment of the invention, the memory 802 is used to store various types of data to support the operation of the electronic terminal 800. Examples of this data include: any executable program for operation on the electronic terminal 800, such as the operating system 8021 and application program 8022; the operating system 8021 contains various system programs, such as the framework layer, core library layer, driver layer, etc., for implementing various basic services and handling hardware-based tasks. The application program 8022 may contain various applications, such as a media player, browser, etc., for implementing various application services. The dome-shaped macular automatic detection method based on OCT images provided in this embodiment of the invention can be included in the application program 8022.

[0087] The methods disclosed in the above embodiments of the present invention can be applied to or implemented by processor 801. Processor 801 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware in processor 801 or by instructions in software form. The processor 801 may be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Processor 801 can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. General-purpose processor 801 may be a microprocessor or any conventional processor, etc. The steps of the accessory optimization method provided in the embodiments of the present invention can be directly reflected as being executed by a hardware decoding processor, or being executed by a combination of hardware and software modules in the decoding processor. The software module may be located in a storage medium, which is located in memory. The processor reads the information in the memory and combines it with its hardware to complete the steps of the aforementioned method.

[0088] In an exemplary embodiment, the electronic terminal 800 may be used by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), or complex programmable logic devices (CPLDs) to perform the aforementioned method.

[0089] According to the method provided in the embodiments of this application, this application also provides a computer program product, which includes: computer program code, which, when run on a computer, causes the computer to execute the dome-shaped macular automatic detection method based on OCT images in the above embodiments.

[0090] According to the method provided in the embodiments of this application, this application also provides a computer-readable storage medium storing program code, which, when run on a computer, causes the computer to execute the dome-shaped macular automatic detection method based on OCT images in the above embodiments.

[0091] As used in this specification, the terms "component," "module," "system," etc., are used to refer to computer-related entities, hardware, firmware, combinations of hardware and software, software, or software in execution. For example, a component can be, but is not limited to, a process running on a processor, a processor, an object, an executable file, an execution thread, a program, and / or a computer. As illustrated, applications running on computing devices and computing devices can both be components. One or more components may reside in a process and / or an execution thread, and components may be located on a single computer and / or distributed among two or more computers. Furthermore, these components can be executed from various computer-readable media on which various data structures are stored. Components can communicate, for example, via local and / or remote processes based on signals having one or more data packets (e.g., data from two components interacting with another component between a local system, a distributed system, and / or a network, such as the Internet interacting with other systems via signals).

[0092] Those skilled in the art will recognize that the various illustrative logical blocks and steps described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0093] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0094] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0095] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0096] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0097] In the above embodiments, the functions of each functional unit can be implemented entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. A computer program product includes one or more computer instructions (programs). When the computer program instructions (programs) are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., high-density digital video discs (DVDs)), or semiconductor media (e.g., solid-state disks (SSDs)).

[0098] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0099] In summary, this application provides an automatic dome-shaped macular detection system, method, terminal, and medium based on OCT images. After acquiring multiple radial OCT B-scan images of the subject centered on the fovea of ​​the macula, the system uses a pre-trained automatic layer segmentation model to segment each radial OCT B-scan image to obtain the corresponding Bruch membrane curve. Then, based on the baseline connecting the dome apex and the lowest point of each Bruch membrane curve, the corresponding dome height is calculated. Each calculated dome height is compared with a preset dome height threshold to obtain the corresponding radial image dome-shaped macular detection result. All radial image dome-shaped macular detection results are statistically analyzed to obtain the final dome-shaped macular detection result. This application uses a unified standard for automated dome-shaped macular detection and can output interpretable and verifiable quantitative values ​​of the dome apex, the baseline connecting the lowest point, and the dome height, meeting the needs of large-scale detection. Therefore, this application effectively overcomes the various shortcomings of existing technologies and has high industrial application value.

[0100] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. An automatic dome-shaped macular detection system based on OCT images, characterized in that, include: The acquisition module is used to acquire multiple radial OCT B-scan images of the subject centered on the fovea of ​​the macula; The Bruch film curve acquisition module is used to segment each radial OCTB-scan image using a pre-trained automatic layer segmentation model to obtain the corresponding Bruch film curve. The dome height calculation module is used to calculate the corresponding dome height based on the baseline connecting the dome apex and the lowest point of each Bruch membrane curve. The detection module is used to compare the calculated height of each dome with a preset dome height threshold to obtain the corresponding radial image dome-like macular detection result; and to count all radial image dome-like macular detection results to obtain the final dome-like macular detection result.

2. The automatic detection system for dome-shaped macular based on OCT images according to claim 1, characterized in that, The Bruch membrane curve acquisition module includes: The Bruch film segmentation unit is used to segment each radial OCT B-scan image using a pre-trained automatic layer segmentation model, obtain the corresponding Bruch film pixel curve, and make effective judgments. The Bruch film curve generation unit is used to process each Bruch film pixel curve according to the valid judgment result of each Bruch film pixel curve to obtain the corresponding Bruch film curve.

3. The automatic detection system for dome-shaped macular based on OCT images according to claim 2, characterized in that, Based on the valid judgment result of each Bruch film layer pixel curve, each Bruch film layer pixel curve is processed accordingly to obtain the corresponding Bruch film layer curve, including: If the valid judgment result is valid, then the Bruch film pixel curve is physically transformed to obtain the corresponding Bruch film curve. If the valid judgment result is invalid, then the Bruch film pixel curve is subjected to physical coordinate transformation and smoothing to obtain the corresponding Bruch film curve.

4. The automatic detection system for dome-shaped macular based on OCT images according to claim 1, characterized in that, The system further includes a curve feature determination module; wherein, the curve feature determination module includes: The search range determination unit is used to divide each Bruch film curve into a dome apex search range and a minimum point search range. The feature determination unit is used to determine the point in the retrieval interval of the dome vertex of each Bruch membrane curve that has the tangent slope closest to zero as the dome vertex of each Bruch membrane curve, and to determine the point in the retrieval interval of the utmost point of each Bruch membrane curve that has the tangent slope closest to zero as the utmost point of each Bruch membrane curve; and to determine the baseline of the line connecting the utmost points of each Bruch membrane curve.

5. The automatic detection system for dome-shaped macular based on OCT images according to claim 1, characterized in that, Based on the baseline connecting the dome apex and the lowest point of each Bruch membrane curve, the corresponding dome height is calculated, including: Calculate the orthogonal distance from the baseline of the line connecting the dome apex to the lowest point of each Bruch membrane curve and use it as the dome height.

6. The automatic detection system for dome-shaped macular based on OCT images according to claim 1, characterized in that, The calculated height of each dome is compared with a preset dome height threshold to obtain the corresponding radial image dome-like macular detection results, including: If the calculated dome height is greater than or equal to the preset dome height threshold, the obtained radial image dome-like macular detection result is positive. If the calculated dome height is less than the preset dome height threshold, the obtained radial image dome-like macular detection result is negative.

7. The automatic detection system for dome-shaped macular based on OCT images according to claim 6, characterized in that, The final dome-shaped macular detection results are obtained by statistically analyzing all radial image dome-shaped macular detection results, including: If at least one radial image dome-like macular detection result is positive among all radial image dome-like macular detection results, then the final dome-like macular detection result obtained is positive. If all radial image dome-shaped macular detection results are negative, then the final dome-shaped macular detection result is negative.

8. An automatic detection method for dome-shaped macular region based on OCT images, characterized in that, include: Acquire multiple radial OCT B-scan images of the subject centered on the fovea of ​​the macula; Using a pre-trained automatic layer segmentation model, each radial OCT B-scan image is segmented to obtain the corresponding Bruch film curve; Calculate the corresponding dome height based on the baseline connecting the dome apex and the lowest point of each Bruch membrane curve. The calculated height of each dome is compared with a preset dome height threshold to obtain the corresponding radial image dome-like macular detection result; all radial image dome-like macular detection results are statistically analyzed to obtain the final dome-like macular detection result.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in claim 8.

10. An electronic terminal, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the method as described in claim 8.

Images