OCT (optical coherence tomography) scanning method, device, equipment and storage medium
By using an AI recognition model to locate lesion areas and determine the OCT scanning protocol, the OCT device is controlled to scan and stitch images, solving the problem of long scanning time in existing technologies and achieving fast and accurate lesion area scanning and stitching.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SVISION IMAGING LTD
- Filing Date
- 2025-12-31
- Publication Date
- 2026-04-21
AI Technical Summary
Existing OCT equipment requires a full scan of the eye when scanning small lesion areas, resulting in long scanning times and high requirements for patient cooperation, making it difficult to quickly and accurately locate and stitch together lesion areas.
The lesion area is located by a preset AI recognition model, the OCT scanning protocol is determined according to the boundary range of the lesion, the OCT device is controlled to perform scanning, and the images of multiple scan areas are stitched together to generate a stitched image.
It enables rapid and accurate location and stitching of lesion areas, improving the scanning efficiency of eye examinations and solving the problem of long scanning time in traditional scanning methods.
Smart Images
Figure CN121890934A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of OCT scanning technology, and more specifically, to an OCT scanning method, apparatus, device, and storage medium. Background Technology
[0002] Currently, for eye examinations, optical coherence tomography (OCT) equipment is commonly used in the market to obtain information about the health of the eyes.
[0003] However, in existing technologies, if the lesion area is at the edge and the lesion area is relatively small, a full scan of the eye is required to scan the lesion area clearly. The scanning method used in this case takes a long time and requires a high degree of cooperation from the patient. There may be problems such as the patient not cooperating, which may lead to a long scanning time. Summary of the Invention
[0004] The purpose of this application is to provide an OCT scanning method, apparatus, device, and storage medium to solve the above-mentioned problems existing in the prior art and to greatly improve the scanning time of eye examinations.
[0005] Firstly, an OCT scanning method is provided, which may include: Acquire a fundus image of the target eye; perform recognition processing on the fundus image using a preset AI recognition model to locate at least one lesion region; wherein, the lesion region includes the boundary range of the lesion; Based on the boundary range of the lesion, an OCT scanning protocol is determined; based on the OCT scanning protocol, a preset OCT device is controlled to scan the target eye and generate at least one scan area image; If there are multiple scan area images, then the multiple scan area images are stitched together to generate a stitched image.
[0006] Secondly, an OCT scanning device is provided, which may include: The positioning module is used to acquire fundus images of the target eye; and to perform recognition processing on the fundus images using a preset AI recognition model to locate at least one lesion region; wherein, the lesion region includes the boundary range of the lesion; The generation module is used to determine the OCT scanning protocol based on the boundary range of the lesion; and to control a preset OCT device to scan the target eye according to the OCT scanning protocol, thereby generating at least one scan area image. The stitching module is used to stitch together multiple scan area images to generate a stitched image if there are multiple scan area images.
[0007] Thirdly, an electronic device is provided, which includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; When a processor executes a program stored in memory, it implements any of the steps described in the first aspect above.
[0008] Fourthly, a computer-readable storage medium is provided, wherein a computer program is stored therein, and when executed by a processor, the computer program implements the steps of any of the methods described in the first aspect above.
[0009] This application provides an OCT scanning method, apparatus, device, and storage medium to acquire fundus images of a target eye; the fundus images are processed using a preset AI recognition model to locate at least one lesion region; wherein the lesion region includes the boundary range of the lesion. An OCT scanning protocol is determined based on the boundary range of the lesion; according to the OCT scanning protocol, a preset OCT device is controlled to scan the target eye, generating at least one scan area image. If there are multiple scan area images, the multiple scan area images are stitched together to generate a stitched image. In this solution, fundus images, due to their fast imaging speed, only display surface morphology and cannot fully display the three-dimensional structure of the lesion; OCT images, with their slower imaging speed, can display three-dimensional images (can present cross-sections) and can fully display the lesion. Therefore, fundus images are captured quickly first, and lesion regions are identified within the fundus images. Then, based on the size and type of the lesion, different OCT scanning protocols are selected for OCT imaging. If multiple lesions are involved, the OCT images of multiple lesions can be stitched together, allowing focus on the three-dimensional location of the lesion and enabling rapid image capture. Therefore, it can accurately locate the lesion area, enabling rapid scanning and stitching of the lesion area, and providing a complete OCT image of the lesion region, thus solving the problem of long scanning time in traditional stitching methods for eye examinations. Attached Figure Description
[0010] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0011] Figure 1 A flowchart illustrating an OCT scanning method provided in an embodiment of this application; Figure 2A schematic diagram illustrating the training process of an AI recognition model provided in an embodiment of this application; Figure 3 A flowchart illustrating an OCT scanning method provided in this application; Figure 4 A schematic diagram of the total scanning range provided in this application; Figure 5 A schematic diagram of yet another total scanning range provided in this application; Figure 6 A schematic diagram of yet another total scanning range provided in this application; Figure 7 A schematic diagram of yet another total scanning range provided in this application; Figure 8 A schematic diagram of yet another total scanning range provided in this application; Figure 9 A schematic diagram of yet another total scanning range provided in this application; Figure 10 A flowchart illustrating an OCT scanning method provided in this application; Figure 11 A flowchart illustrating an OCT scanning method provided in this application; Figure 12 This is a schematic diagram of the structure of an OCT scanning device provided in an embodiment of this application; Figure 13 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation
[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application. Unless otherwise defined, the technical or scientific terms used in this application should have the ordinary meaning understood by those skilled in the art. The words "first," "second," and similar terms used in this application do not indicate any order, quantity, or importance, but are only used to distinguish different components. The words "comprising" or "including," etc., mean that the element or object preceding the word covers the element or object listed after the word and its equivalents, but do not exclude other elements or objects. The words "connected," "coupled," or "connected," etc., are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up," "down," "left," "right," etc., are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.
[0013] For ease of understanding, the terms used in the embodiments of this application are explained below: OCT protocol: refers to a set of preset scanning parameters. Just like a camera has different shooting modes, OCT devices also have multiple preset scanning schemes for efficiently acquiring images of specific shapes and qualities.
[0014] OCT scanning protocol: Defines the shape of the scan path, such as line scan, cross scan, circular scan, raster scan or 3D volume scan, and can also define the scan range, number of scan repetitions, etc.
[0015] The OCT scanning method provided in this application can be applied to electronic devices, terminal devices, OCT scanning processing devices or equipment, or other devices or equipment that can execute this embodiment, and there are no limitations on this.
[0016] The preferred embodiments of this application are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit this application. Furthermore, the embodiments and features in the embodiments of this application can be combined with each other without conflict.
[0017] Figure 1 This is a flowchart illustrating an OCT scanning method provided in an embodiment of this application. Figure 1 As shown, the method may include: Step S101: Obtain a fundus image of the target eye; use a preset AI recognition model to process the fundus image and locate at least one lesion area; wherein, the lesion area includes the boundary range of the lesion.
[0018] For example, fundus images of the target eye are acquired and used as data for AI recognition model identification. These fundus images are obtained by the user through the acquisition of the patient's fundus images using a preset acquisition tool, such as an ultra-wide-angle fundus camera. In this embodiment, the OCT device may include both a fundus camera module and an OCT imaging module; that is, the OCT device can capture both fundus images and OCT images. Then, a pre-trained AI recognition model is used to identify and locate one or more lesion regions, including the boundary range of the lesion.
[0019] Optional, Figure 2 This is a schematic diagram illustrating the training process of an AI recognition model provided in an embodiment of this application. Figure 2 As shown, the process includes: collecting case datasets; training models using pyTorch; developing an AI recognition model; applying the model when collecting data; and automatically identifying and locating lesion areas.
[0020] Step S102: Determine the OCT scanning protocol based on the boundary range of the lesion; according to the OCT scanning protocol, control the preset OCT device to scan the target eye and generate at least one scan area image.
[0021] For example, an OCT scanning protocol is determined based on the size of the lesion's boundary. Then, according to the OCT scanning protocol, a preset OCT device is controlled to scan the target eye, generating at least one scanned area image. The OCT scanning protocol includes a single OCT scanning protocol and a combined OCT scanning protocol. A single OCT scanning protocol includes a single OCT protocol corresponding to the total scan range. A combined OCT scanning protocol includes a combination of multiple OCT protocols corresponding to the total scan range, or it includes a preset stitching method, which can be a mosaic stitching method used to cover the total scan range, which is the area covering the lesion and a preset surrounding area of the lesion.
[0022] Step S103: If there are multiple scan area images, then the multiple scan area images are stitched together to generate a stitched image.
[0023] For example, if there are multiple scan area images, they are mosaicked together to generate a stitched image, such as an OCT / OCTA image. If there is only one scan area image, no stitching is needed, and the scan area image can be output directly. Furthermore, the stitched image or the directly output scan area image can be analyzed for lesion regions using an AI recognition model to determine the analysis results.
[0024] The method provided in this application embodiment acquires a fundus image of the target eye; the fundus image is processed by a preset AI recognition model to locate at least one lesion region; wherein, the lesion region includes the boundary range of the lesion. An OCT scanning protocol is determined based on the boundary range of the lesion; according to the OCT scanning protocol, a preset OCT device is controlled to scan the target eye, generating at least one scan area image. If there are multiple scan area images, the multiple scan area images are stitched together to generate a stitched image. In this solution, the lesion region is accurately located, with the lesion as the main center, and a faster OCT scanning protocol is selected to scan the edges, achieving rapid scanning and stitching of the lesion region. A complete OCT image of the lesion region can be stitched together, solving the problem of long scanning time in traditional stitching methods for eye examinations, thereby improving the efficiency of eye scanning.
[0025] Figure 3 This application provides a flowchart illustrating an OCT scanning method, as shown below. Figure 3 As shown, in this embodiment... Figure 1 Based on the embodiments, the method is described in detail below, and the method includes: Step S201: Obtain a fundus image of the target eye; use a preset AI recognition model to recognize and process the fundus image to locate at least one lesion area; wherein, the lesion area includes the boundary range of the lesion.
[0026] For example, this step is described in step S101, and will not be repeated here.
[0027] Step S202: Determine the OCT scanning protocol based on the boundary range of the lesion.
[0028] In one example, S202 includes: if one or more lesion areas are identified, calculating the total scanning range covering the lesion and a preset surrounding area of the lesion, and determining the OCT scanning protocol based on the total scanning range.
[0029] In one example, "determining the OCT scanning protocol based on the total scanning range" includes: if the total scanning range can be covered by one OCT protocol, then the OCT scanning protocol is determined to be a single OCT scanning protocol; if the total scanning range can be covered by at least two OCT protocols, then the OCT scanning protocol is determined to be a combined OCT scanning protocol; wherein, a single OCT scanning protocol includes a single OCT protocol corresponding to the total scanning range; a combined OCT scanning protocol includes a camera panoramic shooting mode and a checkerboard partitioning mode, and the checkerboard partitioning mode includes a combination scheme of multiple OCT protocols corresponding to the total scanning range.
[0030] For example, if a lesion area is identified, the total scanning range covering the lesion and its preset surrounding area is calculated, and the OCT scanning protocol is determined based on the total scanning range. Alternatively, if multiple lesion areas are identified, the center and extent of each individual lesion are first determined. Based on the center and extent of each individual lesion, the overall scanning range consisting of multiple lesion areas is determined. The method for determining the overall scanning range is to obtain a polygon (or circle) by using the maximum tangent. Then, the overall scanning range is added to the user-preset preset surrounding area of the lesion to obtain the final total scanning range, and the OCT scanning protocol is determined based on the total scanning range. If no lesion area is identified, the user is prompted that no lesion was detected, and the preset default scanning protocol is used. The default mode can be a suggested normal scanning range or a regular mosaic scan, which is not limited.
[0031] Optionally, for a lesion area, the size of the total scanning range is calculated and matched against the device's built-in protocol library. This library includes multiple preset OCT protocols, each with a different scanning size. If a preset OCT protocol is found whose scanning range is completely greater than or equal to the size of the total scanning range, it is considered "coverable," meaning that complete three-dimensional data of the lesion can be obtained in a single scan without stitching. Therefore, the OCT scanning protocol is determined to be a single OCT scanning protocol, which includes the individual OCT protocol corresponding to the total scanning range. For example, if the total scanning range is a circular area with a diameter of 2mm, and the matched preset OCT protocol is a 3mm × 3mm 3D scanning protocol, since 3mm × 3mm is greater than or equal to a circular area with a diameter of 2mm, the OCT scanning protocol is a single OCT scanning protocol.
[0032] Alternatively, if the total scan area is too large to be covered by a single OCT protocol, it needs to be covered by at least two OCT protocols using different methods, thus defining a combined OCT scanning protocol. This combined OCT scanning protocol includes a panoramic camera mode (i.e., Mosaic stitching mode) and a checkerboard partitioning mode. The checkerboard partitioning mode includes a combination of multiple OCT protocols corresponding to the total scan area. In Mosaic stitching mode, the center point of the lesion area is used as the scan center, and the Mosaic stitching mode covers the preset surrounding area of the lesion area and the lesion area itself, ensuring complete coverage. In checkerboard partitioning mode, the total scan area is divided into multiple sub-regions, ensuring that each sub-region overlaps with a preset area, and that the range of each sub-region is covered by a single OCT protocol. Scanning is performed based on the divided sub-regions, thus ensuring scan optimization. Optionally, for multiple lesion areas, it is first determined whether the total scan area corresponding to the multiple lesion areas can be covered by a single OCT protocol. If it can, it is recommended to use a single OCT scanning protocol. If coverage is not possible, it is recommended to use a combined OCT scan protocol. For an explanation of single and combined OCT scan protocols, please refer to the above explanation regarding a single lesion area; it will not be repeated here.
[0033] Step S203: According to the OCT scanning protocol, control the preset OCT device to scan the target eye and generate at least one scan area image.
[0034] In one example, step S203 includes multiple implementations: The first implementation of step S203: If the OCT scanning protocol is a single OCT scanning protocol, then the center point of the lesion area is determined based on the OCT protocol corresponding to the single OCT scanning protocol, which is the scanning center; according to the single OCT scanning protocol, the fixation lamp of the preset OCT device is controlled to move the lesion area to the main optical axis of the fixation lamp, and the data of the target eye within the total scanning range is collected around the scanning center, and the scanning area image centered on the lesion is output; wherein, the total scanning range is the range covering the lesion and the preset surrounding area of the lesion.
[0035] The second implementation of step S203: If the OCT scanning mode scanning protocol is a camera panoramic imaging mode in the combined OCT scanning mode scanning protocol, then based on the camera panoramic imaging mode, the center of the lesion area is determined as the scanning center; according to the camera panoramic imaging mode, the fixation lamp of the preset OCT device is controlled to move the lesion area to the main optical axis of the fixation lamp, and the target eye area within the total scanning range is scanned around the scanning center to generate at least two scanning area images; wherein, the total scanning range is the range covering the lesion and the preset peripheral area of the lesion.
[0036] The third implementation of step S203: If the OCT scanning protocol is a checkerboard partitioning pattern in a combined OCT scanning protocol, then based on the multiple OCT protocols corresponding to the checkerboard partitioning pattern, the target eye area within the total scanning range is divided into multiple sub-regions; the fixation lamp of the preset OCT device is controlled to move the sub-regions to the main optical axis of the fixation lamp, and scan around the center of each sub-region, generating scan area images corresponding to each sub-region; wherein, the total scanning range is the range covering the lesion and the preset surrounding area of the lesion; there are overlapping areas between adjacent sub-regions. For example, with the center point of the lesion area as the scanning center, the preset OCT device is controlled to scan the target eye according to the OCT scanning protocol, generating at least one scan area image. The scanning beam of the OCT device will perform a complete scan with the scanning center as the center point, according to the pattern specified by the OCT protocol (such as grid pattern), generating at least one scan area image. Optionally, the generated scan area image includes a set of acquired center point data, which is continuous three-dimensional data (not just a single point) centered on the lesion and including all of its surrounding tissues.
[0037] Furthermore, the fixation lamp is a guide light inside the OCT device that the patient needs to focus on. When controlling the OCT device to scan the target eye, the fixation lamp can be adjusted according to the location of the lesion area. Moving the fixation lamp to the center point of the lesion area ensures that the lesion area is on the main optical axis. Essentially, moving the fixation lamp guides the patient's eye movement, thus moving the lesion area to the main optical axis of the OCT scan (i.e., the optimal imaging position). Therefore, by moving the fixation lamp, precise positioning can be achieved. If the eye position is incorrect, even if the OCT protocol and center point settings are correct, the scanning beam will be off-target. This step ensures precise alignment between the theoretical coordinates (the center coordinates of the lesion area) and the actual physiological position (the eyeball).
[0038] Optionally, in the first implementation of step S202, if the OCT scanning protocol is a single OCT scanning protocol, the center point of the lesion area is determined as the scanning center based on the OCT protocol corresponding to the single OCT scanning protocol. Then, according to the single OCT scanning protocol, the fixation lamp of the preset OCT device is controlled to move the lesion area to the main optical axis of the fixation lamp, scan around the scanning center, collect data of the target eye within the total scanning range, and output the scan area image centered on the lesion. The total scanning range includes the area covering the lesion and the preset surrounding area of the lesion. Therefore, if the lesion area is relatively large and concentrated, and the corresponding OCT scanning protocol is a single OCT scanning protocol, the OCT protocol with a larger range can be directly selected to scan the results.
[0039] In the second implementation of step S203, if the OCT scanning protocol is a camera panoramic imaging mode within a combined OCT scanning protocol, then the center of the lesion area is determined as the scanning center based on the camera panoramic imaging mode. Then, according to the camera panoramic imaging mode, the fixation lamp of the preset OCT device is controlled to move the lesion area onto the main optical axis of the fixation lamp, and the target eye area within the total scanning range is scanned around the scanning center, generating at least two scan area images. The total scanning range includes the area covering the lesion and its preset surrounding area.
[0040] Therefore, if the lesion area is small and scattered, and the corresponding OCT scanning protocol is the panoramic camera mode in the combined OCT scanning protocol, a scan (one image) can be performed first with the center of the lesion area as the origin. Then, the fixation lamp or scanning mirror will move according to a preset path (such as up, down, left, or right) to perform the next scan at adjacent positions, and there will be partial overlap between the sub-regions of each scan, thus generating multiple scan area images with overlapping parts. This scanning method has a straightforward scanning logic, is easy to plan, and all sub-images are unfolded around a central reference point, which facilitates image stitching later.
[0041] In the third implementation of step S203, Figure 4 A schematic diagram of the total scanning range provided in this application, such as... Figure 4 As shown, the displayed image represents the total scanning area, including sub-regions 1, 2, 3, 4, and 5. Sub-region 1 includes lesions with irregular shapes; the shape of the lesions here is merely an example. Figure 5 Another schematic diagram of the total scanning range provided in this application, such as Figure 5 As shown, the bold quadrilateral displayed is sub-region 1. Figure 6 Another schematic diagram of the total scanning range provided in this application, such as Figure 6 As shown, the bold quadrilateral displayed is sub-region 2. Figure 7 Another schematic diagram of the total scanning range provided in this application, such as Figure 7 As shown, the bold quadrilateral displayed is sub-region 3. Figure 8 Another schematic diagram of the total scanning range provided in this application, such as Figure 8 As shown, the bold quadrilateral displayed is sub-region 4. Figure 9 Another schematic diagram of the total scanning range provided in this application, such as Figure 9 As shown, the bold quadrilateral displayed is sub-region 5.
[0042] If the OCT scanning protocol is a checkerboard partitioning pattern within a combined OCT scanning protocol, the target eye area within the total scanning range is divided into multiple sub-regions 1, 2, 3, 4, and 5 based on the multiple OCT protocols corresponding to the checkerboard partitioning pattern. These sub-regions overlap. Optionally, if multiple lesion areas are irregular and sub-region 1 is enclosed by sub-regions 2, 3, 4, and 5, a regular shape such as a rectangle can be set. Sub-regions are then divided within this regular shape, which includes sub-regions 1, 2, 3, 4, and 5. The edges of the regular shape can coincide with the edges of sub-regions 2, 3, 4, and 5, or a preset distance can be maintained between them to prevent lesions from being located at the edges, thus increasing scanning difficulty. Alternatively, if multiple lesion areas are irregular, they can be directly divided according to a preset shape to obtain sub-regions of that preset shape; this is not limited. Optionally, sub-regions 2, 3, 4, and 5 can be divided equally or according to the actual situation; the division method is not limited.
[0043] Figure 10 This application provides a flowchart illustrating an OCT scanning method, as shown below. Figure 10 As shown, the process includes: identifying the lesion area using a recognition model and determining the total scanning range; determining N sub-regions based on the total scanning range (N is a natural number greater than or equal to 2); controlling the preset OCT fixation lamp to sequentially move the N sub-regions onto the main optical axis and perform scanning. Finally, scan images corresponding to each sub-region are generated. The total scanning range covers the lesion and its preset surrounding area; adjacent sub-regions overlap; the recognition model can be an AI recognition model or other models capable of recognizing lesion areas, without limitation.
[0044] Therefore, if the lesion area is small and scattered, scanning with a single "large-area" OCT protocol will result in a long scan time. In this case, the checkerboard partitioning mode from a combined OCT scanning protocol can be used. This checkerboard partitioning mode can select a protocol with a smaller data volume for data acquisition, generating multiple scan area images with overlapping parts. Therefore, this method of scanning is more flexible and can efficiently cover large, irregular areas. Each sub-region uses a standard OCT protocol, ensuring uniform and controllable scan quality. Step S204: If there are multiple scan area images, then the multiple scan area images are stitched together to generate a stitched image.
[0045] In one example, step S204 includes: preprocessing the images of each scan area; wherein the preprocessing includes filtering and contrast enhancement; extracting the same feature points in each scan area image using a preset feature extraction algorithm; determining matching feature point pairs between the scan area images based on the same feature points in each scan area image; determining a transformation matrix based on the matching feature point pairs; performing position transformation on the pixels in the scan area image based on the transformation matrix, and stitching and fusing the position-transformed scan area images to generate a stitched image.
[0046] For example, if there are multiple scanned area images, then the multiple scanned area images are mosaicked to generate a stitched image. Specifically, Figure 11 This application provides a flowchart illustrating an OCT scanning method, as shown below. Figure 11 The process includes: acquiring images from multiple locations, i.e., scanned area images obtained through scanning; image preprocessing; feature extraction and matching; and image stitching. In this step, after acquiring scanned area images of multiple sub-regions, the images are preprocessed using a filtering algorithm, and histogram equalization and other methods are used to enhance image contrast, thereby making image features more prominent. Feature points are extracted from the scanned area images using a pre-defined feature extraction algorithm. By calculating the distance between feature points, matching feature point pairs are found between different scanned area images. Based on the matching feature point pairs, a transformation matrix is determined, which is used to calculate the precise positional relationship between the scanned area images. Based on the transformation matrix, the translation, rotation, and scaling relationships between the scanned area images are determined, and the pixel positions in the scanned area images are transformed according to the calculated transformation matrix. Then, the transformed scanned area images are mosaicked and fused to generate a stitched image.
[0047] The method provided in this application acquires a fundus image of the target eye; it then uses a preset AI recognition model to process the fundus image and locate at least one lesion region, wherein the lesion region includes the boundary range of the lesion. An OCT scanning protocol is determined based on the boundary range of the lesion. According to the OCT scanning protocol, a preset OCT device is controlled to scan the target eye, generating at least one scan area image. If there are multiple scan area images, they are stitched together to generate a stitched image. In this solution, the lesion region is accurately located, with the lesion as the main center, and a faster OCT scanning protocol is selected to scan the edges, achieving rapid scanning and stitching of the lesion region. This allows for the stitching of a complete OCT image of the lesion region, solving the problem of long scanning time in traditional stitching methods for eye examinations, thereby improving the efficiency of eye scanning.
[0048] Corresponding to the above method, embodiments of this application also provide an OCT scanning device, such as... Figure 12 As shown, the device includes: The positioning module 41 is used to acquire a fundus image of the target eye; and to perform recognition processing on the fundus image using a preset AI recognition model to locate at least one lesion area; wherein, the lesion area includes the boundary range of the lesion; The generation module 42 is used to determine the OCT scanning protocol based on the boundary range of the lesion; and to control a preset OCT device to scan the target eye according to the OCT scanning protocol, thereby generating at least one scan area image. The stitching module 43 is used to stitch together multiple scan area images to generate a stitched image if there are multiple scan area images.
[0049] The functions of each functional unit of the OCT scanning device provided in the above embodiments of this application can be implemented through the above method steps. Therefore, the specific working process and beneficial effects of each unit in the OCT scanning device provided in the embodiments of this application will not be repeated here.
[0050] This application also provides an electronic device, such as... Figure 13 As shown, it includes a processor 510, a communication interface 520, a memory 530, and a communication bus 540, wherein the processor 510, the communication interface 520, and the memory 530 communicate with each other through the communication bus 540.
[0051] Memory 530 is used to store computer programs; The processor 510 performs the above steps when executing the program stored in the memory 530.
[0052] The communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.
[0053] The communication interface is used for communication between the aforementioned electronic devices and other devices.
[0054] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.
[0055] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.
[0056] The implementation methods and beneficial effects of the various components of the electronic device in the above embodiments for solving the problem can be found in [reference needed]. Figure 1 The steps in the illustrated embodiments are used to implement the electronic device. Therefore, the specific working process and beneficial effects of the electronic device provided in this application will not be repeated here.
[0057] In another embodiment provided in this application, a computer-readable storage medium is also provided, which stores instructions that, when executed on a computer, cause the computer to perform any of the OCT scanning methods described in the above embodiments.
[0058] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to perform any of the OCT scanning methods described in the above embodiments.
[0059] Those skilled in the art will understand that the embodiments in this application can be provided as methods, systems, or computer program products. Therefore, the embodiments in this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the embodiments in this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0060] This application describes embodiments of methods, apparatus (systems), and computer program products according to embodiments of this application with reference to flowchart illustrations and / or block diagrams. It will 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 program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0061] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0062] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0063] Although preferred embodiments have been described in this application, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of this application.
[0064] Obviously, those skilled in the art can make various modifications and variations to the embodiments of this application without departing from the spirit and scope of the embodiments of this application. Therefore, if these modifications and variations to the embodiments of this application fall within the scope of the claims in this application and their equivalents, then this application also intends to include these modifications and variations.
Claims
1. An OCT scanning method, characterized in that, The method includes: Acquire a fundus image of the target eye; perform recognition processing on the fundus image using a preset AI recognition model to locate at least one lesion region; wherein, the lesion region includes the boundary range of the lesion; Based on the boundary range of the lesion, an OCT scanning protocol is determined; based on the OCT scanning protocol, a preset OCT device is controlled to scan the target eye and generate at least one scan area image; If there are multiple scan area images, then the multiple scan area images are stitched together to generate a stitched image.
2. The method as described in claim 1, characterized in that, The step of determining the OCT scanning protocol based on the boundary range of the lesion includes: If one or more lesion areas are identified, the total scanning range covering the lesion and the preset surrounding area of the lesion is calculated, and the OCT scanning protocol is determined based on the total scanning range.
3. The method as described in claim 2, characterized in that, The determination of the OCT scanning protocol based on the total scanning range includes: If the total scanning range can be covered by a single OCT protocol, then the OCT scanning protocol is determined to be a single OCT scanning protocol; If the total scanning range can be covered by at least two OCT protocols, then the OCT scanning protocol is determined to be a combined OCT scanning protocol; The single OCT scanning protocol includes a single OCT protocol corresponding to the total scanning range; the combined OCT scanning protocol includes a camera panoramic shooting mode and a checkerboard partitioning mode, and the checkerboard partitioning mode includes a combination scheme of multiple OCT protocols corresponding to the total scanning range.
4. The method as described in claim 1, characterized in that, According to the OCT scanning protocol, a preset OCT device is controlled to scan the target eye, generating at least one scanned area image. include, If the OCT scanning protocol is a single OCT scanning protocol, then the center point of the lesion area is determined based on the OCT protocol corresponding to the single OCT scanning protocol, which is the scanning center; According to the single OCT scanning protocol, the fixation lamp of the preset OCT device is controlled to move the lesion area to the main optical axis of the fixation lamp, collect data of the target eye within the total scanning range around the scanning center, and output the scanning area image centered on the lesion; wherein, the total scanning range is the range covering the lesion and the preset surrounding area of the lesion.
5. The method as described in claim 1, characterized in that, According to the OCT scanning protocol, a preset OCT device is controlled to scan the target eye, generating at least one scanned area image, including: If the OCT scanning protocol is a camera panoramic imaging mode in a combined OCT scanning protocol, then the center of the lesion area is determined based on the camera panoramic imaging mode, which is the scanning center; According to the panoramic shooting mode of the camera, the fixation lamp of the preset OCT device is controlled to move the lesion area to the main optical axis of the fixation lamp, and scan the target eye area within the total scanning range around the scanning center to generate at least two scanning area images; wherein, the total scanning range is the range covering the lesion and the preset surrounding area of the lesion.
6. The method as described in claim 1, characterized in that, According to the OCT scanning protocol, a preset OCT device is controlled to scan the target eye, generating at least one scanned area image, including: If the OCT scanning protocol is a checkerboard partitioning pattern in a combined OCT scanning protocol, then based on the multiple OCT protocols corresponding to the checkerboard partitioning pattern, the target eye area within the total scanning range is divided into multiple sub-regions; the fixation lamp of the preset OCT device is controlled to move the sub-regions to the main optical axis of the fixation lamp, and scan around the center of each sub-region to generate scan area images corresponding to each sub-region; wherein, the total scanning range is the range covering the lesion and the preset surrounding area of the lesion; there are overlapping areas between adjacent sub-regions.
7. The method according to any one of claims 1-6, characterized in that, If there are multiple scan area images, then the multiple scan area images are stitched together to generate a stitched image, including: The images of each scan area are preprocessed; wherein, the preprocessing includes filtering and contrast enhancement; Using a pre-defined feature extraction algorithm, the same feature points are extracted from images in each scanned area; Based on the common feature points in each scanned area image, determine the matching feature point pairs between each scanned area image; Based on the matched feature point pairs, a transformation matrix is determined; based on the transformation matrix, the positions of the pixels in the scanned area image are transformed, and the transformed scanned area images are stitched together to generate a stitched image.
8. An OCT scanning device, characterized in that, The device includes: The positioning module is used to acquire fundus images of the target eye; and to perform recognition processing on the fundus images using a preset AI recognition model to locate at least one lesion region; wherein, the lesion region includes the boundary range of the lesion; The generation module is used to determine the OCT scanning protocol based on the boundary range of the lesion; and to control a preset OCT device to scan the target eye according to the OCT scanning protocol, thereby generating at least one scan area image. The stitching module is used to stitch together multiple scan area images to generate a stitched image if there are multiple scan area images.
9. An electronic device, characterized in that, The electronic device includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the method of any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method described in any one of claims 1-7.