CN108231165A - A kind of auxiliary system of eye disease diagnosis - Google Patents
A kind of auxiliary system of eye disease diagnosis Download PDFInfo
- Publication number
- CN108231165A CN108231165A CN201810092101.4A CN201810092101A CN108231165A CN 108231165 A CN108231165 A CN 108231165A CN 201810092101 A CN201810092101 A CN 201810092101A CN 108231165 A CN108231165 A CN 108231165A
- Authority
- CN
- China
- Prior art keywords
- retina
- image
- module
- dimensional
- field
- 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.)
- Withdrawn
Links
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/193—Preprocessing; Feature extraction
-
- 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)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Ophthalmology & Optometry (AREA)
- Theoretical Computer Science (AREA)
- Medical Informatics (AREA)
- General Physics & Mathematics (AREA)
- Veterinary Medicine (AREA)
- Quality & Reliability (AREA)
- Surgery (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Animal Behavior & Ethology (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Heart & Thoracic Surgery (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Eye Examination Apparatus (AREA)
Abstract
The present invention provides a kind of auxiliary systems of eye disease diagnosis, examine including retina scanning module, two dimensional image processing module, three-dimensional structure module, correcting module and in advance module, the retina scanning module and the two dimensional image processing module wired connection, for acquiring retina two dimension wide-field image；The two dimensional image processing module and the three-dimensional structure module wired connection, for optimizing processing to retina two dimension wide-field image, obtain retina two dimension wide field optimization image；The three-dimensional structure module and the correcting module wired connection, for being built to retina three-dimensional image；The correcting module with it is described it is pre- examine module wired connection, for retina three-dimensional image to be modified retina two dimension wide-field image, obtain retina two dimensional modified image；The pre- module of examining tentatively assert conditions of patients according to retina two dimensional modified image.
Description
Technical field
The present invention relates to medical auxiliary system fields, and in particular to a kind of auxiliary system of eye disease diagnosis.
Background technology
Containing abundant structure in retinal vessel, when these structural informations change, often have with ophthalmology disease
It closes, and the eye illness of identical type often has similar feature, therefore pathological characters that are true, accurately observing retina
Particularly important, it can help the state of an illness of oculist accurate judgement patient, and most accurate therapeutic scheme is provided for patient.
In the prior art, the retinal images of acquisition still have than better quality in central area quality, but by
In human eye be spherical or ellipsoid, the edge for the two-dimentional retinal images that common fundus camera or scanning eye-checking instrument device obtain
Area characteristic information is often unintelligible, and the difficulty of diagnosis is virtually brought to medical staff.
Invention content
In view of the above-mentioned problems, a kind of the present invention is intended to provide auxiliary system of eye disease diagnosis.
The purpose of the present invention is realized using following technical scheme：
A kind of auxiliary system of eye disease diagnosis, including retina scanning module, two dimensional image processing module, three-dimensional structure
It models block, correcting module and examines module in advance, the retina scanning module and the two dimensional image processing module wired connection are used
In acquisition retina two dimension wide-field image；The two dimensional image processing module and the three-dimensional structure module wired connection, are used for
Processing is optimized to retina two dimension wide-field image, obtains retina two dimension wide field optimization image；The three-dimensional structure module
With the correcting module wired connection, for being built to retina three-dimensional image；The correcting module pre- examines mould with described
Block wired connection for retina three-dimensional image to be modified retina two dimension wide-field image, obtains retina two dimension
Correct image；The pre- module of examining tentatively assert conditions of patients according to retina two dimensional modified image.
Beneficial effects of the present invention are：The present invention carries out optimization processing after obtaining patient's retina two dimension wide-field image,
And 3-D view is built to be modified to retina two dimensional image, then revised retina two dimensional image is subjected to state of an illness knowledge
Not, be conducive to oculist and accurate judgement is made to the state of an illness of eye disease patient.
Description of the drawings
Using attached drawing, the invention will be further described, but the embodiment in attached drawing does not form any limit to the present invention
System, for those of ordinary skill in the art, without creative efforts, can also obtain according to the following drawings
Other attached drawings.
Fig. 1 is the frame construction drawing of the present invention；
Fig. 2 is the frame construction drawing of the two dimensional image processing module of the present invention.
Reference numeral：
Retina scanning module 01, three-dimensional structure module 03, correcting module 04, examines mould at two dimensional image processing module 02 in advance
Block 05, enhancing processing submodule 0201 and smoothing processing submodule 0202.
Specific embodiment
With reference to following application scenarios, the invention will be further described.
Referring to Fig. 1, including retina scanning module 01, two dimensional image processing module 02, three-dimensional structure module 03, mould is corrected
Block 04 examines module 05, the retina scanning module 01 and 02 wired connection of two dimensional image processing module with pre-, for adopting
Collect retina two dimension wide-field image；The two dimensional image processing module 02 and three-dimensional structure 03 wired connection of module, are used for
Processing is optimized to retina two dimension wide-field image, obtains retina two dimension wide field optimization image；The three-dimensional structure module
03 with 04 wired connection of correcting module, for being built to retina three-dimensional image；The correcting module 04 with it is described
05 wired connection of module is examined in advance, for retina three-dimensional image to be modified retina two dimension wide-field image, depending on
Nethike embrane two dimensional modified image；The pre- module 05 of examining tentatively assert conditions of patients according to retina two dimensional modified image.
Preferably, the retina scanning module irradiates the back side of patient's eyeball using laser light source.
The above embodiment of the present invention carries out optimization processing, and build three after obtaining patient's retina two dimension wide-field image
Dimension image carries out state of an illness identification to be modified to retina two dimensional image, then by revised retina two dimensional image, favorably
Accurate judgement is made to the state of an illness of eye disease patient in oculist.
Preferably, referring to Fig. 2, the two dimensional image processing module includes enhancing processing submodule and smoothing processing submodule
Block；
The enhancing processing submodule extracts the side of retina two dimension wide-field image using improved annular Gabor filter
To Invariance feature, the inclined field of class of retina two dimension wide-field image is calculated, specially：
K=∫ ∫ s0(j,k)[Gabor1(j,k)-Gabor2(j,k)]djdk
Wherein, Gabor (j, k) is improved annular Gabor filter, Gabor1(j, k) represents improved annular Gabor
The real part of wave filter, Gabor2(j, k) represents the imaginary part of improved annular Gabor filter, and γ is improved ring
The centre frequency of shape Gabor filter, j, k represent the horizontal stroke of retina two dimension wide-field image, ordinate, S respectively0(j, k) is represented
Retina two dimension wide-field image；
Then such inclined field from retina two dimension wide-field image is removed, the gray value of obtained retinal images is returned
One changes to 1 to 255, specially：
S1=S0-K
Wherein, S2For the gray value of the retina two dimension wide-field image after normalization, S1Represent the view behind the inclined field of removal class
The gray value of film two dimension wide-field image, S1∈ [- 255,255], K represent the inclined field of class of retina two dimension wide-field image.
The above embodiment of the present invention carries out enhancing processing to retina two dimension wide-field image, is conducive to the convex of image detail
It shows, it is more quick and accurate during feature point extraction in subsequent module, while enhance processing not to retina two dimension wide field
Image information is destroyed, and maintains the integrality of retina two dimension wide-field image information well, and wide to retina two dimension
The gray value of field picture, which is normalized, to be conducive to widen contrast, improves the accurate of subsequent step processing indirectly
Property.
Preferably, the smoothing processing submodule is removed enhanced noise using the method for smoothing processing, main
It to be selected using the spread function of monotone decreasing come the brightness to retinal images and gradient, the side for keeping Grad larger
Edge region, smoothing process are expressed as with mathematical formulae：
Wherein, div [] represents divergence operator, vectorRepresentation space coordinate, i represent the wheel number of diffusion process,It represents
Brightness,It is the diffusion strength control function of a monotone decreasing；
Retina two dimension wide field optimization figure is obtained after the enhancing processing submodule and the processing of smoothing processing submodule
Picture.
The above embodiment of the present invention is removed the noise enhanced in the enhancing processing submodule, is conducive to follow-up
Module determines that calculating speed gets a promotion during the characteristic point of retinal images and accuracy rate is improved, and what is obtained after denoising regards
Nethike embrane two dimension wide field optimization image is more clear, significantly, and marginal information is also obtained compared to the image detail before non-denoising
Effectively preserve, when being conducive to be modified retina two dimension wide field optimization image in subsequent correction module, obtained retina
Two dimensional modified image is more true, accurate.
Preferably, the three-dimensional structure module builds the 3-D view of the eyeball of patient, and determine retina phase
For the position of eyeball, retina three-dimensional image is obtained.
Preferably, the correcting module comes to carry out position to retina two dimension wide field optimization image with retina three-dimensional image
It corrects, obtains retina two dimensional modified image.
Preferably, the pre- module of examining further includes memory, has the retinal feature of various ophthalmology diseases in memory
Information；
The pre- module of examining first is detected the characteristic point of retina two dimensional modified image, then by these characteristic points with
The retinal feature information for having various ophthalmology diseases in memory carries out matching comparison, specially：
(1) feature point extraction：16 × 16 fields of each pixel of the retinal images after smoothing processing are divided into
A series of 4 × 4 unit counts on the gradient direction in each unit in 8 principal directions, that is, obtains the spy of one 128 dimension
Sign vector, in Gauss subspace, using the extreme point of difference of Gaussian image detection feature vector, selects stable extreme point to make
It is characterized a little, and calculates the principal direction of these characteristic points；
(2) characteristic matching is evaluated：Using self-defined matching score function P to the characteristic point of retina two dimensional modified image with
The retinal feature information for having various ophthalmology diseases in memory is matched：
Wherein, μ is pixel neighborhoods, (j1,k1)、(jx,kx) respectively represent retina two dimensional modified image pixel coordinate and
The pixel coordinate of the retinal images of x kind ophthalmology diseases, g1、gxThe feature vector square of retina two dimensional modified image is represented respectively
The eigenvectors matrix of the retinal images of battle array and certain ophthalmology disease, y (j1,k1) represent retina two dimensional modified image
With pixel displacement, m represents horizontal translation amount, the vertical translational movements of n, and δ is translational movement parameter, δ=345；
IfThen illustrate, which suffers from x kind ophthalmology diseases；
IfThis module (2) step is then re-executed, until there is the retina for having suitable ophthalmology disease met
Characteristic information with the characteristic point of retina two dimensional modified image match so thatFor preset matching score threshold.
The above embodiment of the present invention utilizes characteristic point of the self-defined scoring function to retina two dimensional modified image and storage
There is the many-sided of the retinal feature information of various ophthalmology diseases in device and carry out matching primitives, to ensure that medical staff can be accurate
Really judge ophthalmology disease that patient is suffered from by the present invention, substantially reduce misdiagnosis rate, enable the invention to be truly realized mitigation
The effect of medical staff's workload.
Finally it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention rather than the present invention is protected
The limitation of range is protected, although being explained in detail with reference to preferred embodiment to the present invention, those of ordinary skill in the art should
Work as understanding, technical scheme of the present invention can be modified or replaced equivalently, without departing from the reality of technical solution of the present invention
Matter and range.
Claims (7)
1. a kind of auxiliary system of eye disease diagnosis, it is characterized in that, including retina scanning module, two dimensional image processing mould
Block, three-dimensional build module, correcting module and examine module in advance, and the retina scanning module has with the two dimensional image processing module
Line connects, for acquiring retina two dimension wide-field image；The two dimensional image processing module and the three-dimensional structure module are wired
Connection for optimizing processing to retina two dimension wide-field image, obtains retina two dimension wide field optimization image；The three-dimensional
Module and the correcting module wired connection are built, for being built to retina three-dimensional image；The correcting module and institute
State it is pre- examine module wired connection, for retina three-dimensional image to be modified retina two dimension wide-field image, depending on
Nethike embrane two dimensional modified image；The pre- module of examining tentatively assert conditions of patients according to retina two dimensional modified image.
2. a kind of auxiliary system of eye disease diagnosis according to claim 1, it is characterized in that, the retina scanning mould
Block irradiates the back side of patient's eyeball using laser light source.
3. a kind of auxiliary system of eye disease diagnosis according to claim 1, it is characterized in that, the two dimensional image processing
Module includes enhancing processing submodule and smoothing processing submodule；
The enhancing processing submodule extracts the direction of retina two dimension wide-field image not using improved annular Gabor filter
Vertic features calculate the inclined field of class of retina two dimension wide-field image, specially：
K=∫ ∫ S0(j,k)[Gabor1(j,k)-Gabor2(j,k)]djdk
Wherein, Gabor (j, k) is improved annular Gabor filter, Gabor1(j, k) represents improved annular Gabor filtering
The real part of device, Gabor2(j, k) represents the imaginary part of improved annular Gabor filter, and γ is improved annular
The centre frequency of Gabor filter, j, k represent the horizontal stroke of retina two dimension wide-field image, ordinate, S respectively0(j, k) expression regards
Nethike embrane two dimension wide-field image；
Then such inclined field from retina two dimension wide-field image is removed, the gray value of obtained retinal images is normalized
To 1 to 255, specially：
S1=S0-K
Wherein, S2For the gray value of the retina two dimension wide-field image after normalization, K represents the class of retina two dimension wide-field image
Inclined field, S1Represent the gray value of the retina two dimension wide-field image behind the inclined field of removal class, S1∈ [- 255,255].
4. a kind of auxiliary system of eye disease diagnosis according to claim 3, it is characterized in that, the smoothing processing submodule
Block is removed enhanced noise using the method for smoothing processing, mainly using the spread function of monotone decreasing come to view
The brightness of film image and gradient are selected, the fringe region for keeping Grad larger, smoothing process mathematical formulae table
It is shown as：
Wherein, div [] represents divergence operator, vectorRepresentation space coordinate, i represent the wheel number of diffusion process,Represent brightness,It is the diffusion strength control function of a monotone decreasing；
Optimization image in retina two dimension wide field is obtained after the enhancing processing submodule and the processing of smoothing processing submodule.
5. a kind of auxiliary system of eye disease diagnosis according to claim 4, it is characterized in that, the three-dimensional structure module
The 3-D view of the eyeball of patient is built, and determines that retina relative to the position of eyeball, obtains retina three-dimensional figure
Picture.
6. a kind of auxiliary system of eye disease diagnosis according to claim 5, it is characterized in that, the correcting module is to regard
Nethike embrane 3-D view to carry out position correction to retina two dimension wide field optimization image, obtains retina two dimensional modified image.
7. a kind of auxiliary system of eye disease diagnosis according to claim 6, it is characterized in that, the pre- module of examining also is wrapped
Memory is included, the retinal feature information for having various ophthalmology diseases in memory；
The pre- module of examining first is detected the characteristic point of retina two dimensional modified image, then by these characteristic points with storing
The retinal feature information for having various ophthalmology diseases in device carries out matching comparison, specially：
(1) feature point extraction：One system is divided into 16 × 16 fields of each pixel of the retinal images after smoothing processing
The unit of row 4 × 4 counts on the gradient direction in each unit in 8 principal directions, that is, obtain one 128 dimension feature to
Amount in Gauss subspace, using the extreme point of difference of Gaussian image detection feature vector, selects stable extreme point as spy
Point is levied, and calculates the principal direction of these characteristic points；
(2) characteristic matching is evaluated：Utilize characteristic points of the self-defined matching score function P to retina two dimensional modified image and storage
The retinal feature information for having various ophthalmology diseases in device is matched：
Wherein, μ is pixel neighborhoods, (j1,k1)、(jx,kx) pixel coordinate of retina two dimensional modified image and x kind eyes are represented respectively
The pixel coordinate of the retinal images of section's disease, g1、gxRespectively represent retina two dimensional modified image eigenvectors matrix and certain
The eigenvectors matrix of the retinal images of kind ophthalmology disease, y (j1,k1) represent retina two dimensional modified image matched pixel
Displacement, m represent horizontal translation amount, the vertical translational movements of n, δ be translational movement parameter, δ=345；
IfThen illustrate, which suffers from x kind ophthalmology diseases；
IfThis module (2) step is then re-executed, until there is the retinal feature for having suitable ophthalmology disease met to believe
Breath with the characteristic point of retina two dimensional modified image match so thatFor preset matching score threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810092101.4A CN108231165A (en) | 2018-01-30 | 2018-01-30 | A kind of auxiliary system of eye disease diagnosis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810092101.4A CN108231165A (en) | 2018-01-30 | 2018-01-30 | A kind of auxiliary system of eye disease diagnosis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108231165A true CN108231165A (en) | 2018-06-29 |
Family
ID=62669904
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810092101.4A Withdrawn CN108231165A (en) | 2018-01-30 | 2018-01-30 | A kind of auxiliary system of eye disease diagnosis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108231165A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113990455A (en) * | 2021-11-05 | 2022-01-28 | 杭州安幂医学科技有限公司 | Operation navigation system and method based on focus image projection |
| CN116646079A (en) * | 2023-07-26 | 2023-08-25 | 武汉大学人民医院（湖北省人民医院） | Auxiliary diagnosis method and device for ophthalmologic symptoms |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101534701A (en) * | 2006-11-09 | 2009-09-16 | 奥普托斯股份有限公司 | Improvements in or relating to retinal scanning |
-
2018
- 2018-01-30 CN CN201810092101.4A patent/CN108231165A/en not_active Withdrawn
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101534701A (en) * | 2006-11-09 | 2009-09-16 | 奥普托斯股份有限公司 | Improvements in or relating to retinal scanning |
Non-Patent Citations (1)
| Title |
|---|
| 孟宪静: "血管类图像分割与识别方法研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113990455A (en) * | 2021-11-05 | 2022-01-28 | 杭州安幂医学科技有限公司 | Operation navigation system and method based on focus image projection |
| CN116646079A (en) * | 2023-07-26 | 2023-08-25 | 武汉大学人民医院（湖北省人民医院） | Auxiliary diagnosis method and device for ophthalmologic symptoms |
| CN116646079B (en) * | 2023-07-26 | 2023-10-10 | 武汉大学人民医院（湖北省人民医院） | Auxiliary diagnosis method and device for ophthalmologic symptoms |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11253171B2 (en) | System and method for patient positioning | |
| US10111583B1 (en) | System, method, and non-transitory computer-readable storage media related to correction of vision defects using a visual display | |
| CN109656373B (en) | Fixation point positioning method and positioning device, display equipment and storage medium | |
| Fuhl et al. | Non-intrusive practitioner pupil detection for unmodified microscope oculars | |
| CN108618749B (en) | Retina blood vessel three-dimensional reconstruction method based on portable digital fundus camera | |
| CN105094337A (en) | Three-dimensional gaze estimation method based on irises and pupils | |
| EP3905943B1 (en) | System and method for eye tracking | |
| Richa et al. | Fundus image mosaicking for information augmentation in computer-assisted slit-lamp imaging | |
| CN111524175A (en) | Depth reconstruction and eye movement tracking method and system for asymmetric multiple cameras | |
| TWI515609B (en) | Eyeball locating method and system | |
| JP5935146B2 (en) | Ophthalmic disease image analysis apparatus, ophthalmic image analysis method, and ophthalmic image analysis program | |
| CN108231165A (en) | A kind of auxiliary system of eye disease diagnosis | |
| CN113951813A (en) | Retinal blood vessel branch angle calculation method and device and electronic equipment | |
| De Zanet et al. | Retinal slit lamp video mosaicking | |
| US20240221423A1 (en) | Method and apparatus for measuring gaze difference based on gaze estimation | |
| CN116309661B (en) | Method for extracting OCT (optical coherence tomography) image contour of anterior segment of eye | |
| ES2977594T3 (en) | Method for automatic quantification of the shape of an optic nerve head | |
| CN116407080A (en) | Evolution identification and 3D visualization system and method for fundus structure of myopic patient | |
| CN115424319A (en) | Strabismus recognition system based on deep learning | |
| Ahdi et al. | A hybrid method for 3D mosaicing of OCT images of macula and optic nerve head | |
| TWI524876B (en) | Image stitching method and camera system | |
| CN108182425A (en) | A kind of retina identifying system | |
| Song et al. | 3D Reconstruction of the retina based on un-calibrated fundus images | |
| CN113011286B (en) | Squint discrimination method and system based on deep neural network regression model of video | |
| US11676255B2 (en) | Image correction for ophthalmic images |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20180629 |