CN109859214B - Automatic retina layer segmentation method and device with CSC lesion - Google Patents
Automatic retina layer segmentation method and device with CSC lesion Download PDFInfo
- Publication number
- CN109859214B CN109859214B CN201910085912.6A CN201910085912A CN109859214B CN 109859214 B CN109859214 B CN 109859214B CN 201910085912 A CN201910085912 A CN 201910085912A CN 109859214 B CN109859214 B CN 109859214B
- Authority
- CN
- China
- Prior art keywords
- image
- information
- boundary
- oct image
- segmenting
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Abstract
The invention discloses a method and a device for automatically segmenting retina layers with CSC pathological changes, which receive a sample group of normal retina images, and calculate the thickness mean value and standard deviation of each layer of the normal retina images in the sample group as prior information; receiving an SD-OCT image to be segmented, and segmenting a first SD-OCT image in each group of image data based on prior information and gradient information; and calculating the mean value and the standard deviation of the thickness of the retina of the segmented SD-OCT image as coupling information, and segmenting the next SD-OCT image based on the coupling information and the gradient information until the segmentation of all SD-OCT images is completed.
Description
Technical Field
The disclosure belongs to the technical field of automatic image segmentation, and relates to a method and a device for automatically segmenting a retina layer with CSC lesions.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
Optical Coherence Tomography (OCT) is a new type of Tomography technology with the greatest development prospect in recent years, especially has attractive application prospects in biopsy and imaging of biological tissues, and has been applied to clinical diagnosis in ophthalmology, dentistry and dermatology.
Among retinopathy, Central Serous Chorioretinopathy (CSC) is abbreviated as "mesoplasmic disease". After the retina is scanned by adopting the optical coherence tomography, the method has important significance for tissue layering of an SD-OCT image with CSC pathological changes, and on one hand, the method can quantitatively analyze pathological change regions (the volume, the position and the outline of the CSC regions); on the other hand, manual segmentation of doctors is very time-consuming, automatic segmentation can effectively improve the working efficiency of doctors, and subjective errors of doctors can be avoided.
Currently, there are many algorithms that can be applied to retinal layer segmentation, such as level set-based algorithms, graph theory-based algorithms, and machine learning and deep learning-based algorithms. However, due to the influence of the lesion area, the existing retinal layer segmentation algorithms do not have a good effect on retinal layer segmentation with CSC lesions.
Disclosure of Invention
In view of the deficiencies in the prior art, one or more embodiments of the present disclosure provide a method and an apparatus for automatically segmenting a retinal layer with a CSC lesion, which effectively automatically segments the retinal layer with the CSC lesion based on graph theory and interlayer coupling information.
According to an aspect of one or more embodiments of the present disclosure, there is provided a method for automatic segmentation of retinal layers with CSC lesions.
A method of automatically segmenting retinal layers with CSC lesions, the method comprising:
receiving a sample group of a normal retina image, and calculating the thickness mean value and the standard deviation of each layer of the normal retina image in the sample group as prior information;
receiving an SD-OCT image to be segmented, constructing a weight function for a first SD-OCT image in each group of image data based on prior information and gradient information, and segmenting the image;
and calculating the mean value and the standard deviation of the thickness of the retina of the segmented SD-OCT image as coupling information, and reconstructing a weight function based on the coupling information and the gradient information to segment the next SD-OCT image until the segmentation of all SD-OCT images is completed.
Further, in the method, the received SD-OCT image to be segmented is subjected to image preprocessing, and an improved bilateral filtering algorithm is adopted to suppress speckle noise of the SD-OCT image.
Further, in the method, the segmenting the first SD-OCT image in each set of image data based on the prior information and the gradient information specifically includes:
and constructing a weight function by adopting the prior information and the gradient information, calculating the weight between nodes in the first SD-OCT image in each group of image data, and segmenting the image.
Further, in the method, the segmenting the next SD-OCT image based on the coupling information and the gradient information specifically includes:
and constructing a weight function by adopting the coupling information and the gradient information, calculating the weight of the coupling information corresponding to the image between nodes in the next SD-OCT image in each group of image data, and segmenting the next SD-OCT image.
Further, in the method, the image layer divided by the image includes ILM, IB-OPR _ BMEIS Complex or BMEIS, OB _ RPE, and IB _ OPR.
Further, in the method, the image segmentation specifically comprises the following steps:
searching a dark-bright boundary by adopting a Dijkstra algorithm, and judging whether the type of the searched boundary is ILM, IB-OPR _ BMEIS Complex or BMEIS;
for the obtained BMEIS or IB-OPR _ BMEIS Complex, the upper boundary is moved downwards by 9.77 mu m to be the lower boundary to obtain a narrow band;
extracting another 'dark-bright' boundary in the narrow band by adopting Dijkstra algorithm to obtain IB _ OPR;
and searching the whole image by adopting Dijkstra algorithm to search the 'bright-dark' boundary to obtain OB _ RPE.
Further, in the method, the method further comprises: for image layers with insignificant gradient information in the image, including OPL-HFL, RNFL-GCL, INL-OPL, IPL-INL and GCL-IPL, using the coupling information to constrain the position of the retina boundary, the specific steps include:
adding a coupling information item into the weight function;
constructing a narrow band according to the obtained boundary of the current SD-OCT image or the boundary obtained by the previous SD-OCT image;
and extracting the bright-dark or dark-bright boundary of the SD-OCT image by adopting a Dijkstra algorithm to obtain a final segmentation result.
Further, in the method, the specific step of constructing the narrow band comprises:
one boundary above the boundary to be solved is shifted down by half of the average thickness;
moving the lower boundary of the boundary to be solved by half of the average thickness;
a narrow band is constructed.
According to an aspect of one or more embodiments of the present disclosure, there is provided a computer-readable storage medium.
A computer readable storage medium having stored therein a plurality of instructions adapted to be loaded by a processor of a terminal device and to execute a method of automatic segmentation of retinal layers with CSC lesions.
According to an aspect of one or more embodiments of the present disclosure, there is provided a terminal device.
A terminal device comprising a processor and a computer-readable storage medium, the processor being configured to implement instructions; the computer readable storage medium is for storing a plurality of instructions adapted to be loaded by a processor and to perform the method for automatic segmentation of retinal layers with CSC lesions.
The beneficial effect of this disclosure:
according to the automatic retina layer segmentation method and device with the CSC lesion, the coupling information and the vertical gradient information of the image are used for constructing the weight function, then the coupling information is used for constructing the self-adaptive narrow-band region, and finally the accurate boundary of the retina layer is extracted from the self-adaptive narrow band. The retina layer with the CSC pathological changes is automatically segmented by the aid of the computer and other terminal devices, the result of the segmentation of the retina layer can be obtained by the computer and other terminal devices and corresponding operating environments, changes of the thickness of the retina layer caused by the CSC pathological changes can be effectively analyzed, and evolution of disease conditions such as the volume of a pathological change area can be quantitatively analyzed, so that the working efficiency of ophthalmologists can be greatly improved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the application and, together with the description, serve to explain the application and are not intended to limit the application.
Fig. 1 is a flow diagram of a method for automatic segmentation of retinal layers with CSC lesions, in accordance with one or more embodiments;
FIG. 2 is a detailed method flow diagram in accordance with one or more embodiments;
FIG. 3 is an illustration of layers of a retina according to one or more embodiments;
FIG. 4(A) is a graph of pre-noise suppression effects according to one or more embodiments;
FIG. 4(B) is a graph of post-noise suppression effects in accordance with one or more embodiments;
FIG. 5 is a schematic illustration of retinal layer thickness in accordance with one or more embodiments;
FIG. 6 is a schematic illustration of coupling information in accordance with one or more embodiments;
fig. 7(a) is a schematic diagram of a retinal image with CSC lesions, according to one or more embodiments;
FIG. 7(B) is a schematic illustration of a normal retinal image in accordance with one or more embodiments;
FIG. 8(A) is a schematic diagram of extracting boundaries in a segmentation based on coupling information and gradient information boundaries, according to one or more embodiments;
FIG. 8(B) is a schematic diagram of a segmentation result from a segmentation based on coupling information and gradient information boundaries in accordance with one or more embodiments.
The specific implementation mode is as follows:
technical solutions in one or more embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings in one or more embodiments of the present disclosure, and it is apparent that the described embodiments are only a part of the embodiments of the present disclosure, and not all embodiments. All other embodiments, which can be derived by one of ordinary skill in the art based on one or more embodiments of the disclosure without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
It is noted that the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of methods and systems according to various embodiments of the present disclosure. It should be noted that each block in the flowchart or block diagrams may represent a module, a segment, or a portion of code, which may comprise one or more executable instructions for implementing the logical function specified in the respective embodiment. It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted 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 special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
Without conflict, the embodiments and features of the embodiments in the present disclosure may be combined with each other, and the present disclosure will be further described with reference to the drawings and the embodiments.
As shown in fig. 1, to address the deficiencies of the prior art, one or more embodiments of the present disclosure provide an automatic retina layer segmentation method and apparatus with CSC lesion, which effectively automatically segments the retina layer with CSC lesion based on graph theory and interlayer coupling information.
According to an aspect of one or more embodiments of the present disclosure, there is provided a method for automatic segmentation of retinal layers with CSC lesions.
A method of automatically segmenting retinal layers with CSC lesions, the method comprising:
receiving a sample group of a normal retina image, and calculating the thickness mean value and the standard deviation of each layer of the normal retina image in the sample group as prior information;
receiving an SD-OCT image to be segmented, constructing a weight function for a first SD-OCT image in each group of image data based on prior information and gradient information, and segmenting the image;
and calculating the mean value and the standard deviation of the thickness of the retina of the segmented SD-OCT image as coupling information, and reconstructing a weight function based on the coupling information and the gradient information to segment the next SD-OCT image until the segmentation of all SD-OCT images is completed.
Fig. 2 shows a detailed flowchart of an automatic retinal layer segmentation method with CSC lesions. The method firstly learns 10 groups of normal retina average thickness and standard deviation data (1280 images in total) and guides the segmentation of a first SD-OCT image (B-scan) of each group of data by taking the data as prior information. The mean retinal thickness and variance of the B-scan that each set of data has segmented is then analyzed to guide the segmentation of the next B-scan.
The first step is as follows: and building a hardware platform and configuring an operating environment. A computer was configured and Matlab was installed.
And secondly, counting and calculating the thickness mean value and standard deviation of each layer of 10 groups of normal retina images (1280 OCT images in total, hereinafter referred to as B-scan), and using the information as prior information to guide the segmentation of the first B-scan of each group of data.
As shown in fig. 3, which is the boundary of each of the nine retinal layers to be segmented in the present embodiment.
The third step: and (5) image preprocessing. And the improved bilateral filtering algorithm is used for inhibiting the speckle noise of the OCT image. The method comprises the steps of preprocessing an original image, wherein in the preprocessing, a bilateral filtering algorithm is adopted to suppress speckle noise in the image. As shown in fig. 4, the effect of image noise suppression is schematically shown.
The original image is preprocessed, the first B-scan of each group of data is segmented by adopting statistical information and gradient information, and each subsequent B-scan is segmented by adopting coupling information and gradient information, wherein the segmentation sequence is ILM, IB-OPR _ BMEIS Complex/BMEIS, OB _ RPE, IB _ OPR, OPL-HFL, RNFL-GCL, INL-OPL, IPL-INL and GCL-IPL, and finally the segmentation result is obtained.
The fourth step: for the first B-scan, the weights between nodes in the graph are calculated using statistical prior information and gradient information to construct a weight function to guide the segmentation of the first B-scan for each set of data.
The fifth step: the thicknesses and standard deviations (coupling information) of the retinas of the layers of the segmented B-scan in the group are calculated and used as prior information to guide the segmentation of the next B-scan. Fig. 5 is a schematic diagram of the retinal layer thickness. Fig. 6 is a schematic diagram of coupling information.
The weight function is constructed using the coupling information and the gradient information of the image. The weighting values between nodes in the graph are calculated using this weighting function.
And a sixth step: ILM, IB-OPR _ BMEIS Complex/BMEIS, OB _ RPE and IB _ OPR are divided. As shown in fig. 6, the boundaries are segmented based on gradient information.
The boundary has a large image gradient, so that only gradient information is used to calculate the weight in the segmentation of the boundary,
wherein the Dijkstra algorithm is used to search for "dark-light" boundaries on the image. The boundary searched for in normal retinal images is ILM or BMEIS, and the boundary searched for in images with CSC lesions is ILM or IB-OPR _ BMEIS Complex, which can distinguish the two boundaries according to the location characteristics.
For the BMEIS or IB-OPR _ BMEIS Complex that has been obtained, a narrow band can be obtained with this as the upper boundary and a downward shift of 9.77 μm as the lower boundary, and then another "dark-light" boundary, i.e., IB _ OPR, is extracted using dijkstra's algorithm.
And finally searching the boundary of 'light-dark' on the whole image by using Dijkstra algorithm to obtain OB _ RPE.
Fig. 7(a) is a retina image with CSC lesions, and fig. 7(B) is a normal retina image. The boundary has a large image gradient, so that only gradient information is used to calculate the weight in the boundary segmentation. The "dark-light" boundaries are searched on the image using dijkstra's algorithm. The boundary searched for in normal retinal images is ILM or BMEIS, and the boundary searched for in images with CSC lesions is ILM or IB-OPR _ BMEIS Complex, which can distinguish the two boundaries according to the location characteristics. For the BMEIS or IB-OPR _ BMEIS Complex that has been obtained, a narrow band can be obtained with this as the upper boundary and a downward shift of 9.77 μm as the lower boundary, and then another "dark-light" boundary, i.e., IB _ OPR, is extracted using dijkstra's algorithm. And finally searching the boundary of 'light-dark' on the whole image by using Dijkstra algorithm to obtain OB _ RPE.
The seventh step: segmentation based on the coupling information and the gradient information. As shown in FIG. 8, the gradient information is not significant for the OPL-HFL, RNFL-GCL, INL-OPL, IPL-INL and GCL-IPL, so in the above-described segmentation of the boundary, we use the coupled thickness information to constrain the position of the retinal boundary.
First a coupling information item is added to the weight function value. Then we construct a narrow band according to the boundary obtained by the current B-scan or the boundary obtained by the previous frame B-scan. The method specifically comprises the following steps: pending boundary (B)i) Upper one boundary (B)i-1) Moving down by one half the average thickness, the boundary to be determined (B)i) Next one boundary (B)i+1) The average thickness was shifted up by one-half and the upper and lower boundaries for the different boundaries are shown in table 1 to obtain a narrow band.
Then, Dijkstra algorithm is used to extract the 'bright-dark' or 'dark-bright' boundary, and finally, the segmentation result can be obtained. The Dijkstra algorithm is then used to extract the "light-dark" or "dark-light" boundaries, as in FIG. 8(B), and the segmentation results are finally obtained.
TABLE 1BiCorresponding to Bi+1And Bi-1
According to an aspect of one or more embodiments of the present disclosure, there is provided a computer-readable storage medium.
A computer readable storage medium having stored therein a plurality of instructions adapted to be loaded by a processor of a terminal device and to execute a method of automatic segmentation of retinal layers with CSC lesions.
According to an aspect of one or more embodiments of the present disclosure, there is provided a terminal device.
A terminal device comprising a processor and a computer-readable storage medium, the processor being configured to implement instructions; the computer readable storage medium is for storing a plurality of instructions adapted to be loaded by a processor and to perform the method for automatic segmentation of retinal layers with CSC lesions.
These computer-executable instructions, when executed in a device, cause the device to perform methods or processes described in accordance with various embodiments of the present disclosure.
In the present embodiments, a computer program product may include a computer-readable storage medium having computer-readable program instructions embodied thereon for performing various aspects of the present disclosure. The computer readable storage medium may be a tangible device that can hold and store the instructions for use by the instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic memory device, a magnetic memory device, an optical memory device, an electromagnetic memory device, a semiconductor memory device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a Static Random Access Memory (SRAM), a portable compact disc read-only memory (CD-ROM), a Digital Versatile Disc (DVD), a memory stick, a floppy disk, a mechanical coding device, such as punch cards or in-groove projection structures having instructions stored thereon, and any suitable combination of the foregoing. Computer-readable storage media as used herein is not to be construed as transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission medium (e.g., optical pulses through a fiber optic cable), or electrical signals transmitted through electrical wires.
The computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to a respective computing/processing device, or to an external computer or external storage device over a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. The network adapter card or network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in the respective computing/processing device.
The computer program instructions for carrying out operations of the present disclosure may be assembly instructions, Instruction Set Architecture (ISA) instructions, machine related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, the electronic circuitry can execute computer-readable program instructions to implement aspects of the present disclosure by utilizing state information of the computer-readable program instructions to personalize the electronic circuitry, such as a programmable logic circuit, a Field Programmable Gate Array (FPGA), or a Programmable Logic Array (PLA).
It should be noted that although several modules or sub-modules of the device are mentioned in the above detailed description, such division is merely exemplary and not mandatory. Indeed, the features and functionality of two or more of the modules described above may be embodied in one module in accordance with embodiments of the present disclosure. Conversely, the features and functions of one module described above may be further divided into embodiments by a plurality of modules.
The beneficial effect of this disclosure:
the invention discloses an automatic retina layer segmentation method and device with CSC lesions. The automatic segmentation of the retina with the CSC lesion plays an important role in quantitatively analyzing lesion areas including the volume, the position, the contour and the like of the CSC lesion, so that the working efficiency of a clinician can be effectively improved; on the other hand, the method has important significance for evaluating the lesion degree and lesion evolution. The CSC lesion appears in the SD-OCT image as a lesion area which arches the ISOS layer, is darker in gray scale and appears for continuous multiple frames. The algorithm therefore uses the average thickness information (coupling information) of already segmented images as a priori knowledge to guide the segmentation of the next frame image. Firstly, constructing a weight function by using coupling information and vertical gradient information of an image, then constructing a self-adaptive narrow-band region by using the coupling information, and finally extracting an accurate boundary of a retina layer from the self-adaptive narrow band.
According to the automatic retina layer segmentation method and device with the CSC lesion, the coupling information and the vertical gradient information of an image are used for constructing a weight function, then the coupling information is used for constructing a self-adaptive narrow-band region, and finally the accurate boundary of the retina layer is extracted from the self-adaptive narrow band. The retina layer with the CSC pathological changes is automatically segmented by the aid of the computer and other terminal devices, the result of the segmentation of the retina layer can be obtained by the computer and other terminal devices and corresponding operating environments, changes of the thickness of the retina layer caused by the CSC pathological changes can be effectively analyzed, and evolution of disease conditions such as the volume of a pathological change area can be quantitatively analyzed, so that the working efficiency of ophthalmologists can be greatly improved.
The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (8)
1. A method for automatically segmenting retinal layers with CSC lesions, the method comprising:
receiving a sample group of a normal retina image, and calculating the thickness mean value and the standard deviation of each layer of the normal retina image in the sample group as prior information;
receiving an SD-OCT image to be segmented, and segmenting a first SD-OCT image in each group of image data based on prior information and gradient information;
when image segmentation is carried out, constructing a weight function for a first SD-OCT image in each group of image data based on prior information and gradient information, and carrying out image segmentation; reconstructing a weight function based on the coupling information and the gradient information, and segmenting the next SD-OCT image; segmenting ILM, IB-OPR _ BMEIS Complex, OB _ RPE and IB _ OPR or ILM, BMEIS, OB _ RPE and IB _ OPR based on gradient information; based on the coupling information and gradient information segmentation, segmenting OPL-HFL, RNFL-GCL, INL-OPL, IPL-INL and GCL-IPL, constructing a self-adaptive narrow-band region by using the coupling information, and finally extracting an accurate boundary of the retina layer from the self-adaptive narrow band;
and calculating the mean value and the standard deviation of the thickness of the retina of the segmented SD-OCT image as coupling information, and segmenting the next SD-OCT image based on the coupling information and the gradient information until the segmentation of all SD-OCT images is completed.
2. The method as claimed in claim 1, wherein the received SD-OCT image to be segmented is pre-processed, and a modified bilateral filtering algorithm is used to suppress speckle noise of the SD-OCT image.
3. The method according to claim 1, wherein the segmenting based on the prior information and the gradient information for the first SD-OCT image in each set of image data comprises:
and constructing a weight function by adopting the prior information and the gradient information, calculating the weight between nodes in the first SD-OCT image in each group of image data, and segmenting the image.
4. The method according to claim 1, wherein the segmenting the next SD-OCT image based on the coupling information and the gradient information comprises:
and constructing a weight function by adopting the coupling information and the gradient information, calculating the weight of the coupling information corresponding to the image between nodes in the next SD-OCT image in each group of image data, and segmenting the next SD-OCT image.
5. A method according to claim 1, wherein the image segmentation comprises the following steps:
searching a dark-bright boundary by adopting a Dijkstra algorithm, and judging whether the type of the searched boundary is ILM, IB-OPR _ BMEIS Complex or BMEIS;
for the obtained BMEIS or IB-OPR _ BMEIS Complex, the upper boundary is moved downwards by 9.77 mu m to be the lower boundary to obtain a narrow band;
extracting another 'dark-bright' boundary by adopting a Dijkstra algorithm to obtain IB _ OPR;
and searching the whole image by adopting Dijkstra algorithm to search the 'bright-dark' boundary to obtain OB _ RPE.
6. The method according to claim 1, wherein the method further comprises: for image layers with insignificant gradient information in the image, including OPL-HFL, RNFL-GCL, INL-OPL, IPL-INL and GCL-IPL, using the coupling information to constrain the position of the retina boundary, the specific steps include:
adding a coupling information item into the weight function;
constructing a narrow band according to the obtained boundary of the current SD-OCT image or the boundary obtained by the previous SD-OCT image;
extracting the 'bright-dark' or 'dark-bright' boundary of the SD-OCT image by adopting a Dijkstra algorithm to obtain a final segmentation result;
or, in the method, the specific steps of narrow-band construction include:
one boundary above the boundary to be solved is shifted down by half of the average thickness;
moving the lower boundary of the boundary to be solved by half of the average thickness;
a narrow band is constructed.
7. A computer-readable storage medium having stored thereon a plurality of instructions adapted to be loaded by a processor of a terminal device and to perform a method for automatic segmentation of retinal layers with CSC lesions according to any one of claims 1 to 6.
8. A terminal device comprising a processor and a computer-readable storage medium, the processor being configured to implement instructions; a computer readable storage medium for storing a plurality of instructions adapted to be loaded by a processor and to perform a method for automatic segmentation of retinal layers with CSC lesions according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910085912.6A CN109859214B (en) | 2019-01-29 | 2019-01-29 | Automatic retina layer segmentation method and device with CSC lesion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910085912.6A CN109859214B (en) | 2019-01-29 | 2019-01-29 | Automatic retina layer segmentation method and device with CSC lesion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109859214A CN109859214A (en) | 2019-06-07 |
| CN109859214B true CN109859214B (en) | 2021-02-23 |
Family
ID=66896628
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910085912.6A Expired - Fee Related CN109859214B (en) | 2019-01-29 | 2019-01-29 | Automatic retina layer segmentation method and device with CSC lesion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109859214B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103514605A (en) * | 2013-10-11 | 2014-01-15 | 南京理工大学 | Choroid layer automatic partitioning method based on HD-OCT retina image |
| US8811745B2 (en) * | 2010-01-20 | 2014-08-19 | Duke University | Segmentation and identification of layered structures in images |
| CN106408558A (en) * | 2016-09-05 | 2017-02-15 | 南京理工大学 | Analysis method of hard exudates and high-reflection signals in diabetic retinopathy image |
| CN107392918A (en) * | 2017-06-22 | 2017-11-24 | 苏州大学 | OCT image layer dividing method based on random forest Yu composite reactive curve |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9877646B2 (en) * | 2011-10-11 | 2018-01-30 | Carl Zeiss Meditec, Inc. | Assessment of retinal disruption |
-
2019
- 2019-01-29 CN CN201910085912.6A patent/CN109859214B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8811745B2 (en) * | 2010-01-20 | 2014-08-19 | Duke University | Segmentation and identification of layered structures in images |
| CN103514605A (en) * | 2013-10-11 | 2014-01-15 | 南京理工大学 | Choroid layer automatic partitioning method based on HD-OCT retina image |
| CN106408558A (en) * | 2016-09-05 | 2017-02-15 | 南京理工大学 | Analysis method of hard exudates and high-reflection signals in diabetic retinopathy image |
| CN107392918A (en) * | 2017-06-22 | 2017-11-24 | 苏州大学 | OCT image layer dividing method based on random forest Yu composite reactive curve |
Non-Patent Citations (3)
| Title |
|---|
| Automatic Identification of Pathology-Distorted Retinal Layer Boundaries Using SD-OCT Imaging;Md Akter Hussain et.al;《IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING》;20170731;第64卷(第7期);第1638-1649页 * |
| Automatic segmentation of seven retinal layers in SDOCT images congruent with expert manual segmentation;Stephanie J. Chiu et.al;《OPTICS EXPRESS》;20100830;第18卷(第18期);第19413-19428页 * |
| 基于光学相干层析成像的视网膜图像自动分层方法;贺琪欲 等;《光学学报》;20161031;第36卷(第10期);第1-10页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109859214A (en) | 2019-06-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110033456B (en) | Medical image processing method, device, equipment and system | |
| Zeng et al. | Identification of glomerular lesions and intrinsic glomerular cell types in kidney diseases via deep learning | |
| Li et al. | Optimal surface segmentation in volumetric images-a graph-theoretic approach | |
| CN108171692B (en) | Lung image retrieval method and device | |
| CN108765385B (en) | Double-source CT coronary artery automatic extraction method | |
| WO2008097520A2 (en) | System and method for computer aided detection of pulmonary embolism in tobogganing in ct angiography | |
| CN111488921A (en) | Panoramic digital pathological image intelligent analysis system and method | |
| JP2017510427A (en) | Radiation image lung segmentation technology and bone attenuation technology | |
| JP2009517163A (en) | Method, system and computer program for segmenting structure associated with reference structure in image | |
| WO2018176319A1 (en) | Ultrasound image analysis method and device | |
| CN112862835A (en) | Coronary vessel segmentation method, device, equipment and computer readable storage medium | |
| Dong et al. | Simultaneous segmentation of multiple organs using random walks | |
| CN112508941A (en) | Three-dimensional ultrasonic scanning integrity detection method and device | |
| CN102855483B (en) | Handle the method and apparatus and breast cancer diagnosis apparatus of ultrasonoscopy | |
| CN117373070B (en) | Method and device for labeling blood vessel segments, electronic equipment and storage medium | |
| CN113971728B (en) | Image recognition method, training method, device, equipment and medium for model | |
| WO2020168647A1 (en) | Image recognition method and related device | |
| CN110706217A (en) | Deep learning-based lung tumor automatic delineation method | |
| CN109859214B (en) | Automatic retina layer segmentation method and device with CSC lesion | |
| CN113269764A (en) | Automatic segmentation method and system for intracranial aneurysm, sample processing method and model training method | |
| CN108198213B (en) | Method and device for marking tubular structure and processing terminal of tubular structure | |
| Selvan et al. | Extracting tree structures in CT data by tracking multiple statistically ranked hypotheses | |
| US11830622B2 (en) | Processing multimodal images of tissue for medical evaluation | |
| CN115482261A (en) | Blood vessel registration method, device, electronic equipment and storage medium | |
| CN115064270A (en) | Liver cancer recurrence prediction method based on image omics image characteristics |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210223 Termination date: 20220129 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |