CN105411525A - Eye ground picture and image intelligent obtaining and recognizing system - Google Patents
Eye ground picture and image intelligent obtaining and recognizing system Download PDFInfo
- Publication number
- CN105411525A CN105411525A CN201510758675.7A CN201510758675A CN105411525A CN 105411525 A CN105411525 A CN 105411525A CN 201510758675 A CN201510758675 A CN 201510758675A CN 105411525 A CN105411525 A CN 105411525A
- Authority
- CN
- China
- Prior art keywords
- photograph image
- fundus photograph
- region
- optic disc
- blood vessel
- 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.)
- Granted
Links
Landscapes
- Eye Examination Apparatus (AREA)
Abstract
The invention discloses an eye ground picture and image intelligent obtaining and recognizing system which comprises an eye ground camera, a computer and artificial intelligent processing program installed in the computer. The artificial intelligent processing program comprises an eye ground picture and image feather recognition unit, an eye ground picture and image feature marking unit, an eye ground image feature measuring unit and an eye ground picture and image evaluation unit. According to the obtaining and recognizing system, after eye ground pictures are taken, a normal grading report is directly given for eye ground pictures more than constant quantity automatically through an eye ground picture artificial intelligent processing program built in the obtaining and recognizing system, cases of suspected lesion are sent to an artificial intelligent processing cloud platform of the obtaining and the recognizing system, and cloud automatic evaluation is provided through a larger sample library and higher computing power. Meanwhile, ophthalmologists are allowed to carry out manual classification through a cloud computing platform, and the system is an effective solution of remarkably increasing eye ground disease diagnosis speed on the premise that accuracy is guaranteed.
Description
Technical field
The present invention relates to a kind of diagnostic device of oculopathy, specifically a kind of fundus photograph Image Acquisition and recognition system.
Background technology
Eyes are one of most important organs of human body, and the information that people obtains from the external world has 80% to be realized by eyes.Once generation ophthalmic, gently then affect visual quality and outward appearance, heavy then blinding is disabled.The defect of vision certainly will allow the quality of life of patient have a greatly reduced quality.Curing eye diseases key is not only seeking medical attention early, more crucially obtains the Accurate Diagnosis of authoritative expert and the treatment suggestion of science.
invisible oculopathy is as the symptom that changes of anopsia in early days such as glaucoma, diabetic renal papillary necrosis, and fundus photography is the effective ways of the invisible oculopathy of early discovery, all significant to instructing the diagnosis and treatment of fundus oculi disease, assessment whole body health situation.Fundus photography is in recent years in the clinical examination of ocular fundus adopting often, it is special together as digital camera to utilize exactly, is connected, is shown on computers by the image on optical fundus with ophthalmofundoscope, can print and get off to be kept at case history, before and after can also treating again, do a contrast.It objectively can record optical fundus Posterior pole retinal morphology change, has good objectivity, repeatability and comparability.Carry out optical fundus examination with eye fundus image, by the ophthalmologist oculist at read tablet center, read tablet classification is carried out to optical fundus picture, patient can be made to obtain early stage treatment, delay disease progression, realize the transformation from disease treatment to disease prevention.
But; adopt the optical fundus examination project of fundus photography technology usually can produce the eye fundus image needing the classification of read tablet person's read tablet in a large number; wherein; the fundus photograph of more than 80% is normal retina, and this situation will cause most classification work time loss without any in the normal fundus photograph of ocular disease sign.
Summary of the invention
The object of the invention is to overcome deficiency of the prior art, the fundus photograph image intelligent significantly improving fundus oculi disease diagnosis speed under providing a kind of prerequisite ensureing accuracy rate obtains recognition system.
For achieving the above object, the technical solution adopted in the present invention is:
A kind of fundus photograph image intelligent obtains recognition system, comprises fundus camera, computer and is arranged on the artificial intelligence process program in computer; Described artificial intelligence process program comprises:
1) feature identification unit of fundus photograph image, this unit comprises:
(1) optic disc identification module: search optic disc region in order in the fundus photograph image that sends at fundus camera;
(2) optic cup identification module: search optic cup region in order in the optic disc region that finds at optic disc identification module;
(3) macula lutea identification module: in order to search macular region near optic disc region;
(4) ooze out identification module: in order to search whether have exudate in fundus photograph image, if find to there is sharpness of border but irregular yellow-white point or wax yellow region in fundus photograph image, then there is exudate;
(5) blood vessel recognition module: in order to search the distribution of blood vessel in fundus photograph image, in elongated strip and the darker pattern of color is then blood vessel in fundus photograph image;
(6), as presenting block kermesinus region or the kermesinus region of strip in fundus photograph image, then there is bleeding in hemorrhage identification module: in order to search whether have bleeding in fundus photograph image;
2) the signature unit of fundus photograph image, this unit comprises:
(1) optic cup optic disc scanning line mark module: in order to labelling VCDR scanning line and Rim scanning line; Scan from left to right
Optic disc and optic cup region, for each vertical scan line, search following intersection point: the intersection point of scanning line and optic disc top edge, is designated as A, the intersection point of scanning line and optic cup top edge, is designated as B, the intersection point of scanning line and optic cup lower limb, be designated as C, the intersection point of scanning line and optic disc lower limb, is designated as D; When | when BC| obtains maximum, current scan line is designated as VCDR scanning line, and labelling BC line segment; When MIN (| AB|, | CD|) obtains minima, current scan line is designated as Rim scanning line, and line segment shorter in labelling AB or CD;
(2) arteriovenous mark module: in order to carry out labelling to arteries and vein blood vessel, in the blood vessel that blood vessel recognition module finds, thicker blood vessel is arteries, and thinner blood vessel is vein blood vessel, and arteries and the different color of vein blood vessel are carried out labelling;
3) characteristic measuring unit of fundus photograph image, this unit comprises:
(1) vertical cup disc ratio measurement module: utilize VCDR scanning line, calculate the ratio of BC/AD, be designated as vertical cup disc ratio;
(2) coil edge thickness measurement module: utilize Rim scanning line, calculate the value of MIN (AB, CD), be designated as dish edge thickness;
(3) arteriovenous is than measurement module: being calculated as of arteriovenous ratio:
AVR(arteriovenous than) maximum of the width of the average/vein blood vessel of=many arterial vascular width,
Wherein: the length of the area/blood vessel of the width=blood-vessel image of blood vessel, described blood-vessel image refers to optic disc center for the center of circle, the angiosomes in the annular region that 2.67 to 3 times of radiuses are formed; Described length of vessel refers to carries out the pixel number after skeleton calculating to blood-vessel image;
4) evaluation unit of fundus photograph image: whether normal according to the retina that the recognition result of vertical cup disc ratio, arteriovenous ratio, hemorrhage identification module and the recognition result that oozes out identification module are evaluated in fundus photograph image, and fundus photograph image is classified.
Further, described artificial intelligence process program also comprises plain splice unit, when fundus photograph image is imperfect, plain splice unit is piece image in order to several to be pointed to the fundus photograph image mosaic of different azimuth shooting, to obtain the complete fundus photograph comprising each region.
Further, also comprise cloud platform, the improper fundus photograph image that the evaluation unit of fundus photograph image is chosen by described computer sends cloud platform to by the Internet.
Further, in described cloud platform, described artificial intelligence process program is installed.
Further, the recognition methods that adopts of described optic disc identification module is as follows: A, tentatively determine the candidate region at optic disc place; B, Fuzzy processing is carried out to candidate region, meanwhile homogenization process is carried out to candidate region; C, the candidate region of Fuzzy processing adopted in expand the optic disc that algorithm searches out Fuzzy processing, adopt overhanging algorithm to search out the optic disc of homogenization process to the candidate region of homogenization process; The overlapping region of D, the searching optic disc of Fuzzy processing and the optic disc of homogenization process; E, in overlapping region, found the center of optic disc by blood vessel, thus finally determine the region of optic disc.
Beneficial effect of the present invention:
A kind of fundus photograph image intelligent is provided to obtain recognition system, after shooting completes fundus photograph, automatically by its built-in optical fundus picture artificial intelligence process program, normal graded reporting is directly provided for the fundus photograph of super quantity, and only the case of suspected abnormality is sent to its artificial intelligence process cloud platform, by larger Sample Storehouse, stronger computing capability provides the automatic evaluation in high in the clouds, allow ophthalmologist oculist to carry out manual classification by cloud computing platform simultaneously, it is the effective solution significantly improving fundus oculi disease diagnosis speed under a kind of prerequisite ensureing accuracy rate.
Accompanying drawing explanation
Below in conjunction with the drawings and specific embodiments, the present invention is described in further detail:
Fig. 1 is structural schematic block diagram of the present invention;
Fig. 2 is the structured flowchart of end photograph image artificial intelligence process program modularity;
The flow chart of the recognition methods that Fig. 3 adopts for optic disc identification module;
Fig. 4 is the flow chart of the specific algorithm of optic cup identification module;
Fig. 5 is the schematic diagram of optic disc region and optic cup span scan line;
Fig. 6 is the flow chart of the specific algorithm of the evaluation unit of fundus photograph image;
Fig. 7 is the spliced complete fundus photograph image comprising each region;
Fig. 8 is the comparison diagram of one group of fundus photograph image before and after optic disc identification;
Fig. 9 is the comparison diagram of one group of fundus photograph image before and after optic cup identification;
Figure 10 is the comparison diagram of one group of fundus photograph image before and after macula lutea identification;
Figure 11 is the comparison diagram oozing out one group of fundus photograph image before and after identifying;
Figure 12 is the comparison diagram of one group of fundus photograph image before and after hemorrhage identification;
Figure 13 is the comparison diagram of one group of fundus photograph image before and after arteriovenous labelling.
Detailed description of the invention
As shown in Figure 1, a kind of fundus photograph image intelligent obtains recognition system, the artificial intelligence process program comprising fundus camera, computer, cloud platform and be arranged in computer.The improper fundus photograph image that the evaluation unit of fundus photograph image is chosen by described computer sends cloud platform to by the Internet, is provided with described artificial intelligence process program in described cloud platform.
As shown in Figure 2, described artificial intelligence process program comprises:
1) plain splice unit: in order to present more local detail, the personnel that take pictures utilize fundus camera to take pictures to region, optical fundus usually centered by several important orientation, optical fundus, every pictures region, optical fundus, cover part, plain splice unit is piece image in order to several to be pointed to the fundus photograph image mosaic of different azimuth shooting, to obtain the complete fundus photograph comprising each region, spliced image as shown in Figure 7.If taken pictures, personnel have taken the fundus photograph comprising complete area, skip this step.
2) feature identification unit of fundus photograph image, this unit comprises:
(1) optic disc identification module: search optic disc region in order in the fundus photograph image that sends at fundus camera.
Optic disc full name optic disc or papilla of optic nerve, retina is about 3mm by macula lutea to nasal side has a diameter to be about 1.5mm, and boundary is pale red disc-shaped structure clearly.
The feature of optic disc: the yellow or the white portion that 1, are usually expressed as sub-circular; 2, blood vessel extends out from optic disc center.
As shown in Figure 3, the recognition methods that adopts of described optic disc identification module is as follows: A, tentatively determine the candidate region at optic disc place; B, Fuzzy processing is carried out to candidate region, meanwhile homogenization process is carried out to candidate region; C, the candidate region of Fuzzy processing adopted in expand the optic disc that algorithm searches out Fuzzy processing, adopt overhanging algorithm to search out the optic disc of homogenization process to the candidate region of homogenization process; The overlapping region of D, the searching optic disc of Fuzzy processing and the optic disc of homogenization process; E, in overlapping region, found the center of optic disc by blood vessel, thus finally determine the region of optic disc.As shown in Figure 8, the left figure in figure is the fundus photograph image needing to identify to fundus photograph image, and the right figure in figure is that in fundus photograph image, red area is the optic disc region recognized.
(2) optic cup identification module: search optic cup region in order in the optic disc region that finds at optic disc identification module.
Optic cup is the white cup-shaped region at optic disc center.The ratio of optic cup size and optic disc size diagnoses glaucomatous important evidence.
The feature of optic cup: 1, be positioned at optic disc inside; 2, brightness is usually high than optic disc mean flow rate; 3, transfer at edge's blood vessel.Optic disc identification module is mainly searched optic cup according to the feature of optic cup.As shown in Figure 4, as shown in Figure 9, the left figure in figure is the fundus photograph image needing to identify to fundus photograph image to the flow chart of the specific algorithm of optic cup identification module, and the right figure in figure is that in fundus photograph image, red area is the optic cup region recognized.
(3) macula lutea identification module: in order to search macular region near optic disc region, according to the feature of macula lutea to macula lutea
Search.
Macular area is below at the 0.35cm place, temporo side of optical fundus optic disc and slightly, and being in the optical center district of human eye, is the subpoint of vision axis.Macular area is rich in phylloxanthin, and color is darker compared with peripheral retinal tear.Macular area is an amphiblestroid important area, is positioned at a Posterior pole, main relevant with the visual function such as epicritic vision and colour vision.Once pathological changes appears in macular area, usually there is visual deterioration, muscae genetic vision or metamorphopsia.As shown in Figure 10, the left figure in figure is the fundus photograph image needing to identify to fundus photograph image, and the right figure in figure is fundus photograph image Green region is macular area.Macular area is divided into 9 regions, and division rule is: 1, by with macular area central point for the center of circle, with 1/3 of disc diameter times, 1 times and 2 times for radius makes 3 concentric circulars; 2, make 4 rays from macular area central point to 45 °, 135 °, 225 ° and 315 ° of directions, it drops on the part in the middle of roundlet and great circle.
Feature: 1, macula lutea is in the optical center of human eye, be about in optic disc temporo side 0.35cm place and slightly below; 2, macular region is comparatively dark, in bronzing, and sub-circular; 3, there are a very little depression in the central authorities of macula lutea, are the sharpest place of vision, are the darkest center of macular region in eye fundus image.
(4) ooze out identification module: in order to search whether have exudate in fundus photograph image, if find to there is sharpness of border but irregular yellow-white point or wax yellow region in fundus photograph image, then there is exudate; As shown in figure 11, the left figure in figure is the fundus photograph image needing to identify to fundus photograph image, and the right figure in figure is the yellow area in fundus photograph image is the seepage areas recognized;
(5) blood vessel recognition module: in order to search the distribution of blood vessel in fundus photograph image, in elongated strip and the darker pattern of color is then blood vessel in fundus photograph image;
(6), as presenting block kermesinus region or the kermesinus region of strip in fundus photograph image, then there is bleeding in hemorrhage identification module: in order to search whether have bleeding in fundus photograph image; As shown in figure 12, the left figure in figure is the fundus photograph image needing to identify to fundus photograph image, and the right figure in figure is the pink region in fundus photograph image is the hemorrhagic areas recognized;
3) the signature unit of fundus photograph image, this unit comprises:
(1) optic cup optic disc scanning line mark module: in order to labelling VCDR scanning line and Rim scanning line; Scan optic disc and optic cup region from left to right, for each vertical scan line, search following intersection point: as shown in Figure 5, in figure, great circle is optic disc region, and roundlet is optic cup region, and the region between 1 in figure, 5,7,11 is sweep limits, the intersection point of scanning line and optic disc top edge, be designated as A, the intersection point of scanning line and optic cup top edge, is designated as B, the intersection point of scanning line and optic cup lower limb, be designated as C, the intersection point of scanning line and optic disc lower limb, is designated as D; When | when BC| obtains maximum, current scan line is designated as VCDR scanning line, and labelling BC line segment; When MIN (| AB|, | CD|) obtains minima, current scan line is designated as Rim scanning line, and line segment shorter in labelling AB or CD;
(2) arteriovenous mark module: in order to carry out labelling to arteries and vein blood vessel, in the blood vessel that blood vessel recognition module finds, thicker blood vessel is arteries, and thinner blood vessel is vein blood vessel, and arteries and the different color of vein blood vessel are carried out labelling; Fundus photograph image as shown in figure 13, left figure in figure is the fundus photograph image needing to identify, the scattergram of to be white line in fundus photograph image the be blood vessel that blood vessel recognition module identifies out of the right figure in figure, red line is described on white line, represent arteries, blue line is described on white line, is expressed as vein blood vessel.
4) characteristic measuring unit of fundus photograph image, this unit comprises:
(1) vertical cup disc ratio measurement module: utilize VCDR scanning line, calculate the ratio of BC/AD, be designated as vertical cup disc ratio;
(2) coil edge thickness measurement module: utilize Rim scanning line, calculate the value of MIN (AB, CD), be designated as dish edge thickness;
(3) arteriovenous is than measurement module: being calculated as of arteriovenous ratio:
AVR(arteriovenous than) maximum of the width of the average/vein blood vessel of=many arterial vascular width,
Wherein: the length of the area/blood vessel of the width=blood-vessel image of blood vessel, described blood-vessel image refers to optic disc center for the center of circle, the angiosomes in the annular region that 2.67 to 3 times of radiuses are formed; Described length of vessel refers to carries out the pixel number after skeleton calculating to blood-vessel image;
5) evaluation unit of fundus photograph image:
According to vertical cup disc ratio, dish edge thickness, arteriovenous ratio, hemorrhage, ooze out, whether retina that the recognition result of macula lutea identification module is evaluated in fundus photograph image normal, and classification and marking is carried out to fundus photograph image, idiographic flow as shown in Figure 6, normal fundus photograph image is super quantity image, improper fundus photograph image is non-super quantity image, by the evaluation of the evaluation unit of fundus photograph image, image is divided into two classes: super quantity image and non-super quantity image, further, macula lutea focus whether is had for non-super quantity image tagged.
The above is the preferred embodiment of the present invention; certainly the interest field of the present invention can not be limited with this; should be understood that; for those skilled in the art; technical scheme of the present invention is modified or equivalent replacement, do not depart from the protection domain of technical solution of the present invention.
Claims (5)
1. fundus photograph image intelligent obtains a recognition system, it is characterized in that: comprise fundus camera, computer and be arranged on the artificial intelligence process program in computer; Described artificial intelligence process program comprises:
The feature identification unit of fundus photograph image, this unit comprises:
(1) optic disc identification module: search optic disc region in order in the fundus photograph image that sends at fundus camera;
(2) optic cup identification module: search optic cup region in order in the optic disc region that finds at optic disc identification module;
(3) macula lutea identification module: in order to search macular region near optic disc region;
(4) ooze out identification module: in order to search whether have exudate in fundus photograph image, if find to there is sharpness of border but irregular yellow-white point or wax yellow region in fundus photograph image, then there is exudate;
(5) blood vessel recognition module: in order to search the distribution of blood vessel in fundus photograph image, in elongated strip and the darker pattern of color is then blood vessel in fundus photograph image;
(6), as presenting block kermesinus region or the kermesinus region of strip in fundus photograph image, then there is bleeding in hemorrhage identification module: in order to search whether have bleeding in fundus photograph image;
The signature unit of fundus photograph image, this unit comprises:
(1) optic cup optic disc scanning line mark module: in order to labelling VCDR scanning line and Rim scanning line; Scan from left to right
Optic disc and optic cup region, for each vertical scan line, search following intersection point: the intersection point of scanning line and optic disc top edge, is designated as A, the intersection point of scanning line and optic cup top edge, is designated as B, the intersection point of scanning line and optic cup lower limb, be designated as C, the intersection point of scanning line and optic disc lower limb, is designated as D; When | when BC| obtains maximum, current scan line is designated as VCDR scanning line, and labelling BC line segment; When MIN (| AB|, | CD|) obtains minima, current scan line is designated as Rim scanning line, and line segment shorter in labelling AB or CD;
(2) arteriovenous mark module: in order to carry out labelling to arteries and vein blood vessel, in the blood vessel that blood vessel recognition module finds, thicker blood vessel is arteries, and thinner blood vessel is vein blood vessel, and arteries and the different color of vein blood vessel are carried out labelling;
The characteristic measuring unit of fundus photograph image, this unit comprises:
(1) vertical cup disc ratio measurement module: utilize VCDR scanning line, calculate the ratio of BC/AD, be designated as vertical cup disc ratio;
(2) coil edge thickness measurement module: utilize Rim scanning line, calculate the value of MIN (AB, CD), be designated as dish edge thickness;
(3) arteriovenous is than measurement module: being calculated as of arteriovenous ratio:
AVR(arteriovenous than) maximum of the width of the average/vein blood vessel of=many arterial vascular width,
Wherein: the length of the area/blood vessel of the width=blood-vessel image of blood vessel, described blood-vessel image refers to optic disc center for the center of circle, the angiosomes in the annular region that 2.67 to 3 times of radiuses are formed; Described length of vessel refers to carries out the pixel number after skeleton calculating to blood-vessel image;
The evaluation unit of fundus photograph image: whether normal according to the retina that the recognition result of vertical cup disc ratio, arteriovenous ratio, hemorrhage identification module and the recognition result that oozes out identification module are evaluated in fundus photograph image, and fundus photograph image is classified.
2. fundus photograph image intelligent according to claim 1 obtains recognition system, it is characterized in that: described artificial intelligence process program also comprises plain splice unit, when fundus photograph image is imperfect, plain splice unit is piece image in order to several to be pointed to the fundus photograph image mosaic of different azimuth shooting, to obtain the complete fundus photograph comprising each region.
3. fundus photograph image intelligent according to claim 1 obtains recognition system, and it is characterized in that: also comprise cloud platform, the improper fundus photograph image that the evaluation unit of fundus photograph image is chosen by described computer sends cloud platform to by the Internet.
4. fundus photograph image intelligent according to claim 3 obtains recognition system, it is characterized in that: be provided with described artificial intelligence process program in described cloud platform.
5. fundus photograph image intelligent according to claim 1 obtains recognition system, it is characterized in that: the recognition methods that described optic disc identification module adopts is as follows: A, tentatively determine the candidate region at optic disc place; B, Fuzzy processing is carried out to candidate region, meanwhile homogenization process is carried out to candidate region; C, the candidate region of Fuzzy processing adopted in expand the optic disc that algorithm searches out Fuzzy processing, adopt overhanging algorithm to search out the optic disc of homogenization process to the candidate region of homogenization process; The overlapping region of D, the searching optic disc of Fuzzy processing and the optic disc of homogenization process; E, in overlapping region, found the center of optic disc by blood vessel, thus finally determine the region of optic disc.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510758675.7A CN105411525B (en) | 2015-11-10 | 2015-11-10 | A kind of fundus photograph image intelligent obtains identifying system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510758675.7A CN105411525B (en) | 2015-11-10 | 2015-11-10 | A kind of fundus photograph image intelligent obtains identifying system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105411525A true CN105411525A (en) | 2016-03-23 |
| CN105411525B CN105411525B (en) | 2017-05-31 |
Family
ID=55490417
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510758675.7A Active CN105411525B (en) | 2015-11-10 | 2015-11-10 | A kind of fundus photograph image intelligent obtains identifying system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105411525B (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106725295A (en) * | 2016-11-29 | 2017-05-31 | 瑞达昇科技(大连)有限公司 | A kind of miniature check-up equipment, device and its application method |
| CN106725292A (en) * | 2016-12-30 | 2017-05-31 | 深圳市新产业眼科新技术有限公司 | Multispectral fundus imaging remote diagnosis system and its operation method |
| CN107680683A (en) * | 2017-10-09 | 2018-02-09 | 上海睦清视觉科技有限公司 | A kind of AI eye healths appraisal procedure |
| CN107713981A (en) * | 2017-10-09 | 2018-02-23 | 上海睦清视觉科技有限公司 | A kind of AI ophthalmology health detection equipment and its detection method |
| CN108392174A (en) * | 2018-04-19 | 2018-08-14 | 梁建宏 | A kind of automatic check method and system of retinopathy of prematurity |
| CN108492872A (en) * | 2017-07-21 | 2018-09-04 | 湖州师范学院 | Intelligent auxiliary diagnosis system |
| CN108765379A (en) * | 2018-05-14 | 2018-11-06 | 深圳明眸科技有限公司 | Computational methods, device, Medical Devices and the storage medium of eyeground pathological changes region area |
| CN109199322A (en) * | 2018-08-31 | 2019-01-15 | 福州依影健康科技有限公司 | A kind of macula lutea detection method and a kind of storage equipment |
| CN110037651A (en) * | 2018-01-15 | 2019-07-23 | 江威 | The method of quality control and device of eye fundus image |
| CN110327013A (en) * | 2019-05-21 | 2019-10-15 | 北京至真互联网技术有限公司 | Eye fundus image detection method, device and equipment and storage medium |
| WO2019206209A1 (en) * | 2018-04-26 | 2019-10-31 | 上海鹰瞳医疗科技有限公司 | Machine learning-based fundus image detection method, apparatus, and system |
| WO2020042500A1 (en) * | 2018-08-31 | 2020-03-05 | 福州依影健康科技有限公司 | Health big data service method and system based on remote fundus screening |
| WO2020042405A1 (en) * | 2018-08-31 | 2020-03-05 | 福州依影健康科技有限公司 | Preprocessing method for performing quantitative analysis on fundus image, and storage device |
| CN112053780A (en) * | 2020-08-24 | 2020-12-08 | 宁波市眼科医院 | Artificial intelligence cataract screening referral system based on eye ground picture |
| CN112168135A (en) * | 2019-07-05 | 2021-01-05 | 顺天乡大学校产学协力团 | Eyeball disease diagnosis device and method based on artificial intelligence |
| CN113724857A (en) * | 2021-08-27 | 2021-11-30 | 清华大学深圳国际研究生院 | Automatic diagnosis device for eye ground disease based on eye ground image retina blood vessel |
| CN115497229A (en) * | 2022-09-15 | 2022-12-20 | 江苏至真健康科技有限公司 | Using method and system of unmanned self-service portable mydriasis-free fundus camera |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1189535A1 (en) * | 1999-04-29 | 2002-03-27 | Torsana Diabetes Diagnostics A/S | Analysis of fundus images |
| AU2008356237A1 (en) * | 2008-05-14 | 2009-11-19 | Agency For Science, Technology And Research | Automatic cup-to-disc ratio measurement system |
| WO2011059409A1 (en) * | 2009-11-16 | 2011-05-19 | Jiang Liu | Obtaining data for automatic glaucoma screening, and screening and diagnostic techniques and systems using the data |
| CN102283634A (en) * | 2011-07-12 | 2011-12-21 | 曹米莎 | Method and system for testing eye ground image based on image identification |
| CN102843957A (en) * | 2009-08-24 | 2012-12-26 | 新加坡保健服务集团有限公司 | Method and system for detecting disc haemorrhages |
| CN103870838A (en) * | 2014-03-05 | 2014-06-18 | 南京航空航天大学 | Eye fundus image characteristics extraction method for diabetic retinopathy |
-
2015
- 2015-11-10 CN CN201510758675.7A patent/CN105411525B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1189535A1 (en) * | 1999-04-29 | 2002-03-27 | Torsana Diabetes Diagnostics A/S | Analysis of fundus images |
| AU2008356237A1 (en) * | 2008-05-14 | 2009-11-19 | Agency For Science, Technology And Research | Automatic cup-to-disc ratio measurement system |
| CN102843957A (en) * | 2009-08-24 | 2012-12-26 | 新加坡保健服务集团有限公司 | Method and system for detecting disc haemorrhages |
| WO2011059409A1 (en) * | 2009-11-16 | 2011-05-19 | Jiang Liu | Obtaining data for automatic glaucoma screening, and screening and diagnostic techniques and systems using the data |
| CN102283634A (en) * | 2011-07-12 | 2011-12-21 | 曹米莎 | Method and system for testing eye ground image based on image identification |
| CN103870838A (en) * | 2014-03-05 | 2014-06-18 | 南京航空航天大学 | Eye fundus image characteristics extraction method for diabetic retinopathy |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106725295A (en) * | 2016-11-29 | 2017-05-31 | 瑞达昇科技(大连)有限公司 | A kind of miniature check-up equipment, device and its application method |
| CN106725292A (en) * | 2016-12-30 | 2017-05-31 | 深圳市新产业眼科新技术有限公司 | Multispectral fundus imaging remote diagnosis system and its operation method |
| CN106725292B (en) * | 2016-12-30 | 2018-07-24 | 深圳市新产业眼科新技术有限公司 | Multispectral fundus imaging remote diagnosis system and its operation method |
| CN108492872A (en) * | 2017-07-21 | 2018-09-04 | 湖州师范学院 | Intelligent auxiliary diagnosis system |
| CN107680683A (en) * | 2017-10-09 | 2018-02-09 | 上海睦清视觉科技有限公司 | A kind of AI eye healths appraisal procedure |
| CN107713981A (en) * | 2017-10-09 | 2018-02-23 | 上海睦清视觉科技有限公司 | A kind of AI ophthalmology health detection equipment and its detection method |
| CN110037651B (en) * | 2018-01-15 | 2022-03-25 | 江威 | Method and device for controlling quality of fundus image |
| CN110037651A (en) * | 2018-01-15 | 2019-07-23 | 江威 | The method of quality control and device of eye fundus image |
| CN108392174A (en) * | 2018-04-19 | 2018-08-14 | 梁建宏 | A kind of automatic check method and system of retinopathy of prematurity |
| WO2019206209A1 (en) * | 2018-04-26 | 2019-10-31 | 上海鹰瞳医疗科技有限公司 | Machine learning-based fundus image detection method, apparatus, and system |
| CN108765379B (en) * | 2018-05-14 | 2019-11-19 | 深圳明眸科技有限公司 | Calculation method, device, Medical Devices and the storage medium of eyeground pathological changes region area |
| CN108765379A (en) * | 2018-05-14 | 2018-11-06 | 深圳明眸科技有限公司 | Computational methods, device, Medical Devices and the storage medium of eyeground pathological changes region area |
| CN109199322A (en) * | 2018-08-31 | 2019-01-15 | 福州依影健康科技有限公司 | A kind of macula lutea detection method and a kind of storage equipment |
| WO2020042500A1 (en) * | 2018-08-31 | 2020-03-05 | 福州依影健康科技有限公司 | Health big data service method and system based on remote fundus screening |
| WO2020042405A1 (en) * | 2018-08-31 | 2020-03-05 | 福州依影健康科技有限公司 | Preprocessing method for performing quantitative analysis on fundus image, and storage device |
| CN110875092A (en) * | 2018-08-31 | 2020-03-10 | 福州依影健康科技有限公司 | Health big data service method and system based on remote fundus screening |
| CN110875092B (en) * | 2018-08-31 | 2023-10-20 | 福州依影健康科技有限公司 | Health big data service method and system based on remote fundus screening |
| TWI719587B (en) * | 2018-08-31 | 2021-02-21 | 大陸商福州依影健康科技有限公司 | Pre-processing method and storage device for quantitative analysis of fundus image |
| GB2591910A (en) * | 2018-08-31 | 2021-08-11 | Fuzhou Yiying Health Tech Co Ltd | Preprocessing method for performing quantitative analysis on fundus image, and storage device |
| CN110327013B (en) * | 2019-05-21 | 2022-02-15 | 北京至真互联网技术有限公司 | Fundus image detection method, device and equipment and storage medium |
| CN110327013A (en) * | 2019-05-21 | 2019-10-15 | 北京至真互联网技术有限公司 | Eye fundus image detection method, device and equipment and storage medium |
| CN112168135A (en) * | 2019-07-05 | 2021-01-05 | 顺天乡大学校产学协力团 | Eyeball disease diagnosis device and method based on artificial intelligence |
| CN112053780A (en) * | 2020-08-24 | 2020-12-08 | 宁波市眼科医院 | Artificial intelligence cataract screening referral system based on eye ground picture |
| CN113724857A (en) * | 2021-08-27 | 2021-11-30 | 清华大学深圳国际研究生院 | Automatic diagnosis device for eye ground disease based on eye ground image retina blood vessel |
| CN115497229A (en) * | 2022-09-15 | 2022-12-20 | 江苏至真健康科技有限公司 | Using method and system of unmanned self-service portable mydriasis-free fundus camera |
| CN115497229B (en) * | 2022-09-15 | 2023-09-12 | 江苏至真健康科技有限公司 | Use method and system of unmanned self-help portable mydriasis-free fundus camera |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105411525B (en) | 2017-05-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105411525A (en) | Eye ground picture and image intelligent obtaining and recognizing system | |
| US12114929B2 (en) | Retinopathy recognition system | |
| CN109528155B (en) | Intelligent screening system suitable for high myopia complicated with open angle glaucoma and establishment method thereof | |
| WO2018095994A1 (en) | Method and system for classifying optic nerve head | |
| AU2018438719A1 (en) | Fundus image automatic analysis and comparison method and storage device | |
| US20230022921A1 (en) | System and method for analyzing corneal lesion using anterior ocular segment image, and computer-readable recording medium | |
| CN108392174B (en) | Automatic examination method and system for retinopathy of prematurity | |
| US20210391056A1 (en) | Health big data service method and system based on remote fundus screening | |
| CN109691979A (en) | A kind of diabetic retina image lesion classification method based on deep learning | |
| Suero et al. | Locating the Optic Disc in Retinal Images Using Morphological Techniques. | |
| CN104933715A (en) | Registration method applied to retina fundus image | |
| CN110210357A (en) | A kind of ptosis image measuring method based on still photo face recognition | |
| Dong et al. | Deep learning-based estimation of axial length and subfoveal choroidal thickness from color fundus photographs | |
| Sivaraman et al. | A novel, smartphone-based, teleophthalmology-enabled, widefield fundus imaging device with an autocapture algorithm | |
| Ramaswamy et al. | A study and comparison of automated techniques for exudate detection using digital fundus images of human eye: a review for early identification of diabetic retinopathy | |
| GB2593824A (en) | Analysis method and system for feature data change of diabetic retinopathy fundus, and storage device | |
| Gairola et al. | Keratoconus classifier for smartphone-based corneal topographer | |
| Besenczi et al. | Automatic optic disc and optic cup detection in retinal images acquired by mobile phone | |
| Sinha et al. | Prospects of artificial intelligence in ophthalmic practice | |
| Pinos-Velez et al. | Implementation of support tools for the presumptive diagnosis of Glaucoma through identification and processing of medical images of the human eye | |
| Perez-Rovira et al. | Robust optic disc location via combination of weak detectors | |
| CN108230287B (en) | Method and device for detecting crystal region of anterior segment image | |
| Sumathy et al. | Feature extraction in retinal fundus images | |
| Zhou | [Retracted] Design of Intelligent Diagnosis and Treatment System for Ophthalmic Diseases Based on Deep Neural Network Model | |
| Sharma et al. | A system for grading diabetic maculopathy severity level |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |