CN108257126B - Blood vessel detection and registration method, equipment and application of three-dimensional retina OCT image - Google Patents
Blood vessel detection and registration method, equipment and application of three-dimensional retina OCT image Download PDFInfo
- Publication number
- CN108257126B CN108257126B CN201810072546.6A CN201810072546A CN108257126B CN 108257126 B CN108257126 B CN 108257126B CN 201810072546 A CN201810072546 A CN 201810072546A CN 108257126 B CN108257126 B CN 108257126B
- Authority
- CN
- China
- Prior art keywords
- image
- blood vessel
- retina
- layer
- registered
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
- A61B3/10—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions
- A61B3/12—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions for looking at the eye fundus, e.g. ophthalmoscopes
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4887—Locating particular structures in or on the body
- A61B5/489—Blood vessels
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/003—Reconstruction from projections, e.g. tomography
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10101—Optical tomography; Optical coherence tomography [OCT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20068—Projection on vertical or horizontal image axis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30041—Eye; Retina; Ophthalmic
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Veterinary Medicine (AREA)
- Biophysics (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Heart & Thoracic Surgery (AREA)
- Public Health (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Vascular Medicine (AREA)
- Pathology (AREA)
- Ophthalmology & Optometry (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Multimedia (AREA)
- Eye Examination Apparatus (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method, equipment and application for detecting and registering blood vessels of a three-dimensional retina OCT image, wherein the method comprises an image preprocessing step, a vertical projection image of the blood vessels on the corresponding position of the image is generated according to the position of the blood vessels in the retina, and the image is denoised and enhanced; a blood vessel skeleton extraction step, which is to extract the skeleton structure of the blood vessel based on the vertical projection image after denoising and enhancement; a characteristic extraction step, which is used for finding out matched characteristic points of the target image and the image to be registered; and a registration step, namely generating a rigid registration model between the target image and the image to be registered based on the extracted characteristic points. The method can well eliminate the influence of speckle noise, realize automatic detection and accurate registration of blood vessels in the retina SD-OCT image, and play an important auxiliary role in the evaluation of the development of common ophthalmologic diseases and the determination of a preoperative and postoperative treatment scheme of a doctor.
Description
Technical Field
The invention belongs to the field of medical image processing, and particularly relates to a method, equipment and application for detecting and registering blood vessels of a three-dimensional retina OCT image.
Background
The retina is the light-sensitive tissue located at the back of the eyeball, an important component of the human visual system. Currently, the SD-OCT technique has become a powerful tool for non-destructive evaluation of retinal diseases, providing rapid, high resolution, three-dimensional images showing internal layers of the retina. The registration of the longitudinal retinal data of the same patient can effectively evaluate the disease development of the patient and provide effective help for the determination of the treatment scheme before and after the operation of the doctor. Due to the strong speckle noise of the SD-OCT retinal image, the unique imaging mode of the SD-OCT retinal image and other factors, the existing mature method for registering longitudinal image data of MRI, CT and the like is directly used for registering the SD-OCT retinal image, and the ideal effect is difficult to obtain. The main technical defects are that (1) the registration method based on gray scale is affected by speckle noise and cannot accurately register (2) the discontinuity of B scanning direction data caused by eye movement causes great difficulty for the existing registration technology; (3) more importantly, SD-OCT data has strong correlation in the A scanning direction, but the current registration method does not consider the imaging characteristic, thereby causing the result of overlarge registration freedom. To date, no relevant report exists on a complete method for extracting and registering blood vessels of the retinal SD-OCT image. Therefore, the invention provides a blood vessel automatic detection and registration method based on a three-dimensional retina OCT image, and the method is applied to the evaluation and treatment scheme of the progression of ophthalmic diseases.
Disclosure of Invention
Aiming at the problems, the invention provides a three-dimensional retina OCT image blood vessel automatic detection and registration method and application.
The technical purpose is achieved, the technical effect is achieved, and the invention is realized through the following technical scheme:
a blood vessel automatic detection and registration method based on three-dimensional retina OCT images comprises the following steps:
an image preprocessing step, namely generating a vertical projection image of the blood vessel at the corresponding position of the image according to the position of the blood vessel in the retina, and denoising and enhancing the image;
a blood vessel skeleton extraction step, which is to extract the skeleton structure of the blood vessel based on the vertical projection image after denoising and enhancement;
a characteristic extraction step, namely finding out matched characteristic points of the target image and the image to be registered;
and a registration step, namely generating a rigid registration model between the target image and the image to be registered based on the extracted characteristic points.
As a further improvement of the present invention, wherein the image preprocessing step comprises:
an internal retina layering step, wherein a three-dimensional image is layered by adopting a three-dimensional image cutting technology corresponding to a physiological anatomy image of the retina; and
and (3) vertical projection and denoising of the image, namely selecting a gray projection image of the layered image of the position of the blood vessel in the vertical direction, and then sequentially adopting a histogram equalization method to perform enhancement processing on the image and a wiener filtering method to denoise the projection image.
As a further improvement of the invention, the method is characterized in that:
the three-dimensional graph cutting technology adopts a boundary cost function method to segment an image into 11 surfaces corresponding to the inner part of the retina.
As a further improvement of the present invention, wherein the step of extracting the blood vessel skeleton comprises:
a blood vessel detection step, wherein a multi-scale blood vessel enhancement filter based on a Hessian matrix detects the tubular structure of a blood vessel;
a blood vessel skeleton extraction step, wherein a circular structural body is adopted to corrode a blood vessel until a blood vessel tubular structure is one pixel wide; and
and judging the extraction result of the blood vessel skeleton, and purifying the extracted blood vessel skeleton by adopting a threshold value method.
As a further improvement of the present invention, the step of determining the result of the blood vessel skeleton extraction includes:
(1) the blood vessel skeleton size is smaller than the set threshold value, and the result is an error extraction result;
(2) and if the size of the blood vessel skeleton is larger than the set threshold value, the correct extraction result is obtained.
As a further improvement of the invention, the target image and the image to be registered are processed simultaneously by adopting the image preprocessing step and the blood vessel skeleton extracting step.
As a further improvement of the invention: the characteristic extraction step comprises the steps of extracting characteristic points matched with a target image and an image to be registered by adopting an accelerated robust characteristic algorithm; and optimizing the characteristic points by adopting a random sampling consistency method.
As a further improvement of the invention, the minimum distance between the target image and the feature point of the image to be registered is calculated through the acceleration algorithm, and the feature point matched with any feature point A in the target image in the image to be registered is determined.
The application of the blood vessel automatic detection and registration method based on the three-dimensional retina OCT image in the evaluation and treatment scheme of the ophthalmologic disease course development is disclosed.
An automatic blood vessel detection and registration device based on three-dimensional retina OCT images, comprising:
the image preprocessing module is configured to generate a vertical projection image of the blood vessel at the corresponding position of the image according to the position of the blood vessel in the retina, and perform image denoising and enhancement;
a vessel skeleton extraction module configured to extract a skeleton structure of the vessel based on the denoised and enhanced vertical projection image;
the feature extraction module is configured to find out matched feature points of the target image and the image to be registered;
and the registration module is configured to generate a rigid registration model between the target image and the image to be registered based on the extracted feature points.
The invention has the beneficial effects that: the invention integrates the three-dimensional image cutting technology to carry out internal layering on the retinal SD-OCT, carry out histogram equalization and enhance and remove noise on a vertical projection image by a wiener filtering method, eliminates the influence of speckle noise, then uses a multi-scale blood vessel enhancement filter to detect the tubular structure of a blood vessel, carries out optimized selection on the characteristics based on the characteristic calculation and matching of an SURF characteristic algorithm and a random sampling consistency method, and the like, realizes the automatic detection and registration of the retinal SD-OCT image blood vessel for the first time, and plays an important auxiliary role in the evaluation of the clinical common ophthalmological disease course development and the determination of a doctor preoperative and postoperative treatment scheme.
Drawings
FIG. 1 is a process block diagram of a retinal OCT image vessel registration method according to the present invention;
FIG. 2 is a diagram of the effects of layering of the retina;
FIG. 3(a) is a vertical projection gray scale of the acquired layer 7 to layer 11 images;
FIGS. 3(b) and 3(c) are the projection views of FIG. 3(a) after image enhancement and image de-noising, respectively;
FIG. 3(d), FIG. 3(e) and FIG. 3(f) are graphs of the variation of the extracted vascular skeleton during the vascular skeleton extraction step;
FIGS. 4(a) and 4(b) are a target image and an image to be registered, respectively;
FIG. 4(c) is a diagram of the matching result of the feature points of the target image and the image to be registered;
FIGS. 4(d) and 4(e) are overlay images of the vessel skeleton of the target image and the image to be registered before and after registration, respectively;
fig. 5 is a functional schematic block diagram of an apparatus for implementing the method of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The following detailed description of the principles of the invention is provided in connection with the accompanying drawings.
For simplicity of explanation, the "three-dimensional retinal OCT image" is hereinafter simply referred to as "SD-OCT image"
The basic block diagram of the method of the invention is shown in the attached figure 1, and mainly comprises 4 steps: image preprocessing, blood vessel skeleton extraction, feature extraction and registration. The specific description is as follows.
(1) Image pre-processing
The image preprocessing mainly comprises the following two steps: retina internal layering and vertical projection and denoising of images.
(a) Internal retinal delamination
The retinal internal layering has an important role in analyzing retinopathy such as the severity of ocular trauma, the formation of macular edema, and the like. The SD-OCT image of the retina was automatically segmented into 10 layers using a multi-scale three-dimensional graph segmentation technique in combination with the physiological anatomy map of the retina, resulting in 11 surfaces, as shown in fig. 2, corresponding from top to bottom to the nerve fiber layer, ganglion cell layer, inner plexiform layer, inner nuclear layer, outer plexiform layer, outer nuclear layer, inner ganglion layer, outer ganglion layer, vervelvets membrane and retinal pigment epithelium layer of the physiological anatomy map of the retina. The basic idea of the three-dimensional graph cutting technology is to detect the inner surfaces of the retina by adopting different resolutions from coarse resolution to fine resolution based on a graph theory method. The technique employs a boundary-based cost function, and when the cost function is minimal, the respective surfaces are found.
(b) Vertical projection and denoising of images
After the segmentation of each layer is completed, since the blood vessels to be detected are mainly located from the outer nuclear layer to the retinal pigment epithelium layer of the retina, the gray value data between the 7 th surface and the 11 th surface of the SD-OCT image can be extracted, and the average value of the gray values in the vertical direction between the 7 th surface and the 11 th surface is calculated to obtain the vertical projection image containing the projection information of the blood vessels, as shown in fig. 3 (a). Because the vertical projection image also contains noise due to the influence of the speckle noise of the SD-OCT image, the invention firstly uses a histogram equalization method to carry out image enhancement processing on the projection image, as shown in a figure 3(b), and then uses wiener filtering to carry out denoising on the projection image, so as to obtain a denoised image, as shown in a figure 3 (c).
(2) Vascular scaffolding extraction
Accurate detection and extraction of blood vessels is an important step in SD-OCT image registration. However, due to the influence of speckle noise of the SD-OCT image, it is much more difficult to extract blood vessel information from the SD-OCT image than to extract blood vessels from the fundus image. After image enhancement and denoising are carried out on the SD-OCT vertical projection image, a multiscale vascular enhancement filter based on Hessian matrix is adopted to detect the tubular structure of the blood vessel, and the detection result is shown in figure 3 (d). And then corroding the detected tubular blood vessel by adopting a morphological processing method, thereby extracting the skeleton structure of the blood vessel. The present invention uses a circular structure to erode the blood vessel until the vascular tubular structure is eroded to one pixel wide, as shown in fig. 3 (e). And finally, judging the extraction result of the blood vessel framework by adopting a threshold value method. Wherein, the blood vessel skeleton size is a wrong extraction result when smaller than the threshold, and a correct extraction result when larger than the set threshold, so as to obtain a more accurate processing diagram after the blood vessel skeleton extraction result, as shown in fig. 3 (f).
The two steps comprise processing the target image and the image to be registered simultaneously, and then placing the processed target image and the image to be registered together for registration.
(3) Feature extraction
Feature extraction is a technique commonly used in pattern recognition, and aims to reduce the dimension of a feature space and select main features from a large feature set. The method comprises the steps of extracting characteristic points of the extracted blood vessel skeleton by using an SURF (speeded up robust feature) algorithm, and then performing optimized selection on the characteristics by using a random sampling consistency method.
The SURF characteristic algorithm adopted by the invention is a local invariant characteristic, has better stability and moderate operation complexity. The invention uses SURF characteristic algorithm to find out the characteristic points which meet the conditions, and matches the characteristic points according to the minimum distance between the calculated target image and the characteristic points of the image to be registered. Specifically, the characteristic point in the image to be matched with the minimum distance to a certain characteristic point A in the target image is calculated and determined as the characteristic point matched with the characteristic point A in the image to be registered. The matching feature points found in the target image and the image to be registered are shown in fig. 4 (c). In order to further accelerate the operation speed and reduce wrong matching points, the invention adopts a random sampling consistency method to optimally select the characteristics on the basis. The random sampling consistency method is an iterative algorithm, and calculates mathematical model parameters of data according to a group of sample data sets containing abnormal data, so as to obtain effective sample data.
(4) Registration
And according to the optimized matching feature points, finding out transformation parameters, such as rotation parameters, translation parameters and the like, between the target image and the image to be registered by adopting a rigid registration model. The rigid registration model used is shown below:
wherein a is rcos θ, b is rsin θ; [ x, y ]]And [ x ', y']Respectively are coordinate positions before and after registration; r, theta, tx,tyScaling parameters, rotation parameters and translation parameters, respectively.
The experiments were carried out according to the methods described above, and the experimental results are as follows:
the method is tested by taking 10 groups of sample data, wherein each group of sample data is SD-OCT images from the same person at three different time points, the image of the first time point in each group of sample is taken as a target image by farmers, and the images of the subsequent time points are images to be registered respectively to the image of the first time point.
After the registration is finished according to the method, a Dice coefficient and a Structural Similarity Index (SSIM) are used as objective indexes of the evaluation method, and the higher the two indexes are, the higher the registration accuracy is.
Experimental results show that the average Dice coefficient for registering longitudinal retina SD-OCT image data by adopting the method is 77.69%, and the average structural similarity index is 98.97%. The invention achieves better registration effect. As shown in fig. 4(d) and (e), it can be seen from the figure that the registration corrects the position deviation between the target image and the image to be registered, and the registration effect is good.
The results show that the invention can realize automatic detection and registration of the retinal SD-OCT image blood vessels.
The method provided by the invention can play an important auxiliary role in clinically applying the method to the evaluation of the development of common ophthalmologic diseases and the determination of a treatment scheme before and after a doctor operation.
Meanwhile, the method is configured into an image processing device to realize the work of assisting in completing the evaluation and the determination of the treatment scheme. The equipment comprises an image preprocessing module, a blood vessel skeleton extraction module, a feature extraction module and a registration module. The image preprocessing module is configured to generate a vertical projection image of the blood vessel at a corresponding position of the image according to the position of the blood vessel in the retina, and perform image denoising and enhancing; the blood vessel skeleton extraction module is configured to extract the skeleton structure of the blood vessel based on the denoised and enhanced vertical projection image; the feature extraction module is configured to find out matched feature points of the target image and the image to be registered; the registration module is configured to generate a rigid registration model between the target image and the image to be registered based on the extracted feature points.
The foregoing shows and describes the general principles and broad features of the present invention and advantages thereof. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (6)
1. A blood vessel automatic detection and registration method based on three-dimensional retina OCT images comprises the following steps:
(1) an image preprocessing step, namely generating a vertical projection image of the blood vessel at the corresponding position of the image according to the position of the blood vessel in the retina, and denoising and enhancing the image; the method comprises the following steps:
(1.1) a retina internal layering step of layering a three-dimensional map by using a three-dimensional map cutting technology corresponding to a physiological anatomical map of a retina, wherein the divided membranous layers sequentially comprise a nerve fiber layer, a ganglion cell layer, an inner plexiform layer, an inner nuclear layer, an outer plexiform layer, an outer nuclear layer, an inner ganglion layer, an outer ganglion layer, a Wilhoff's membrane and a retinal pigment epithelium layer from top to bottom;
and
(1.2) vertical projection and denoising steps of images: extracting gray value data of an outer nuclear layer, an inner node layer, an outer node layer, a Wilhoff's membrane and a retinal pigment epithelium layer, and calculating a gray average value in a vertical direction to obtain a vertical projection image containing blood vessel projection information;
(1.3) sequentially adopting a histogram equalization method to enhance the image and a wiener filtering method to denoise the projection image;
(2) a blood vessel skeleton extraction step, which is to extract the skeleton structure of the blood vessel based on the vertical projection image after denoising and enhancement; the method comprises the following steps:
(2.1) a blood vessel detection step, namely detecting the tubular structure of the blood vessel by a multi-scale blood vessel enhancement filter based on a Hessian matrix;
(2.2) a blood vessel skeleton extraction step, wherein a circular structure body is adopted to corrode a blood vessel until a blood vessel tubular structure is one pixel wide; and
(2.3) judging the extraction result of the blood vessel skeleton, and purifying the extracted blood vessel skeleton by adopting a threshold value method;
(3) a characteristic extraction step, namely finding out matched characteristic points of the target image and the image to be registered;
(4) a registration step, namely generating a rigid registration model between the target image and the image to be registered based on the extracted feature points, wherein the adopted rigid registration model is shown as the following formula:
wherein a is rcos θ and b is rsin θ; [ x, y ]]And [ x ', y']Respectively are coordinate positions before and after registration; r, θ, tx，tyScaling parameters, rotation parameters and translation parameters, respectively.
2. The method for automatically detecting and registering the blood vessel based on the three-dimensional retina OCT image as claimed in claim 1, wherein: the three-dimensional graph cutting technology adopts a boundary cost function method to segment an image into 11 surfaces corresponding to the inner part of the retina.
3. The method for automatically detecting and registering blood vessels based on the three-dimensional retina OCT image as claimed in claim 1, wherein the step of determining the result of the blood vessel skeleton extraction comprises:
(1) the blood vessel skeleton size is smaller than the set threshold value, and the result is an error extraction result;
(2) the accurate extraction result is obtained when the size of the blood vessel skeleton is larger than a set threshold value.
4. The method for automatically detecting and registering blood vessels based on the three-dimensional retina OCT image according to any one of claims 1-3, wherein the image preprocessing step and the blood vessel skeleton extraction step are adopted to simultaneously process a target image and an image to be registered.
5. The method for automatically detecting and registering the blood vessel based on the three-dimensional retina OCT image as claimed in claim 1, wherein: the characteristic extraction step comprises the steps of extracting characteristic points matched with a target image and an image to be registered by adopting an accelerated robust characteristic algorithm; and optimizing the characteristic points by adopting a random sampling consistency method.
6. The method for automatically detecting and registering the blood vessel based on the three-dimensional retina OCT image as claimed in claim 5, wherein: and calculating the minimum distance between the target image and the feature point of the image to be registered through the accelerated robust feature algorithm, and determining the feature point matched with any feature point A in the target image in the image to be registered.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810072546.6A CN108257126B (en) | 2018-01-25 | 2018-01-25 | Blood vessel detection and registration method, equipment and application of three-dimensional retina OCT image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810072546.6A CN108257126B (en) | 2018-01-25 | 2018-01-25 | Blood vessel detection and registration method, equipment and application of three-dimensional retina OCT image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108257126A CN108257126A (en) | 2018-07-06 |
| CN108257126B true CN108257126B (en) | 2022-06-24 |
Family
ID=62741860
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810072546.6A Active CN108257126B (en) | 2018-01-25 | 2018-01-25 | Blood vessel detection and registration method, equipment and application of three-dimensional retina OCT image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108257126B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110033496B (en) * | 2019-03-27 | 2023-07-11 | 江苏理工学院 | Motion artifact correction method for time sequence three-dimensional retina SD-OCT image |
| CN110264472B (en) * | 2019-05-29 | 2021-05-04 | 浙江科技学院 | SD-OCT-based fruit hypodermal cell nondestructive imaging method |
| CN110211111A (en) * | 2019-05-31 | 2019-09-06 | 上海联影医疗科技有限公司 | A kind of method, apparatus of vessel extraction, image processing equipment and storage medium |
| CN110448267B (en) * | 2019-09-06 | 2021-05-25 | 重庆贝奥新视野医疗设备有限公司 | Multimode fundus dynamic imaging analysis system and method |
| CN111862114A (en) * | 2020-07-10 | 2020-10-30 | 温州医科大学 | Choroidal three-dimensional blood vessel imaging and quantitative analysis method and device based on optical coherence tomography system |
| CN114119519A (en) * | 2021-11-16 | 2022-03-01 | 高峰 | Collateral circulation assessment method |
| CN116309391B (en) * | 2023-02-20 | 2023-09-05 | 依未科技(北京)有限公司 | Image processing method and device, electronic equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090216794A1 (en) * | 2008-02-27 | 2009-08-27 | General Electric Company | Method and system for accessing a group of objects in an electronic document |
| CN106934761A (en) * | 2017-02-15 | 2017-07-07 | 苏州大学 | A kind of method for registering of three-dimensional non-rigid optical coherence tomographic image |
-
2018
- 2018-01-25 CN CN201810072546.6A patent/CN108257126B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090216794A1 (en) * | 2008-02-27 | 2009-08-27 | General Electric Company | Method and system for accessing a group of objects in an electronic document |
| CN106934761A (en) * | 2017-02-15 | 2017-07-07 | 苏州大学 | A kind of method for registering of three-dimensional non-rigid optical coherence tomographic image |
Non-Patent Citations (1)
| Title |
|---|
| 基于Hessian矩阵的冠脉造影图像分割与骨架提取;秦红星等;《数据采集与处理》;20160930;第911-918页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108257126A (en) | 2018-07-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108257126B (en) | Blood vessel detection and registration method, equipment and application of three-dimensional retina OCT image | |
| CN110276356B (en) | Fundus image microaneurysm identification method based on R-CNN | |
| CN109472781B (en) | Diabetic retinopathy detection system based on serial structure segmentation | |
| JP5582772B2 (en) | Image processing apparatus and image processing method | |
| US9418423B2 (en) | Motion correction and normalization of features in optical coherence tomography | |
| CN108961261B (en) | Optic disk region OCT image hierarchy segmentation method based on space continuity constraint | |
| CN109685809B (en) | Liver infusorian focus segmentation method and system based on neural network | |
| CN108272434B (en) | Method and device for processing fundus images | |
| CN110415216B (en) | CNV automatic detection method based on SD-OCT and OCTA retina images | |
| Kumari et al. | Feature extraction for early detection of diabetic retinopathy | |
| Kafieh et al. | An accurate multimodal 3-D vessel segmentation method based on brightness variations on OCT layers and curvelet domain fundus image analysis | |
| CN109325955B (en) | Retina layering method based on OCT image | |
| Hatanaka et al. | Improvement of automatic hemorrhage detection methods using brightness correction on fundus images | |
| CN108470348A (en) | Slit-lamp anterior ocular segment faultage image feature extracting method | |
| CN114627067A (en) | Wound area measurement and auxiliary diagnosis and treatment method based on image processing | |
| Tavakoli et al. | Radon transform technique for linear structures detection: application to vessel detection in fluorescein angiography fundus images | |
| Aruchamy et al. | Automated glaucoma screening in retinal fundus images | |
| CN116030042B (en) | Diagnostic device, method, equipment and storage medium for doctor's diagnosis | |
| Hezekiah et al. | A Review on cornea imaging and processing techniques | |
| CN110033496B (en) | Motion artifact correction method for time sequence three-dimensional retina SD-OCT image | |
| Liu et al. | Retinal vessel segmentation using densely connected convolution neural network with colorful fundus images | |
| CN115841472A (en) | Method, device, equipment and storage medium for identifying high-density characteristics of middle cerebral artery | |
| CN111292285A (en) | Automatic screening method for diabetes mellitus based on naive Bayes and support vector machine | |
| Ashame et al. | Abnormality Detection in Eye Fundus Retina | |
| Zhang et al. | Blood vessel segmentation based on digital subtraction angiography sequence |
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 |