CN109102494A - A kind of After Cataract image analysis method and device - Google Patents
A kind of After Cataract image analysis method and device Download PDFInfo
- Publication number
- CN109102494A CN109102494A CN201810725079.2A CN201810725079A CN109102494A CN 109102494 A CN109102494 A CN 109102494A CN 201810725079 A CN201810725079 A CN 201810725079A CN 109102494 A CN109102494 A CN 109102494A
- Authority
- CN
- China
- Prior art keywords
- cataract
- artificial intelligence
- result
- intelligence model
- feature
- 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.)
- Pending
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/34—Network arrangements or protocols for supporting network services or applications involving the movement of software or configuration parameters
-
- 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Radiology & Medical Imaging (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Medical Informatics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Probability & Statistics with Applications (AREA)
- General Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The present invention provides a kind of late coming After Cataract image analysis method and device, comprising: receives After Cataract eye picture；Position and handle the rear capsule hypertrophic region of late coming After Cataract slit-lamp retro-illumination image；The detection zone for including using artificial intelligence model extracts After Cataract region as foundation in rear capsule hypertrophic region；Wherein, artificial intelligence model is trained in advance；The feature database for including using artificial intelligence model extracts at least one After Cataract feature that After Cataract region includes as foundation；At least one After Cataract feature is handled to obtain class probability result；For the analysis result comparison library for including using artificial intelligence model as foundation, matching obtains the corresponding analysis result of class probability result.As it can be seen that the above method can simply, quickly and accurately determine the After Cataract situation of patient, and gives user and analyzed accordingly as a result, to improve the efficiency of user's medical treatment.
Description
Technical field
The present invention relates to medical fields, in particular to a kind of After Cataract image analysis method and device.
Background technique
Currently, the attention degree of medical treatment is being continuously improved in people, more and more people need to obtain the strong of itself
Health state, to guarantee work, study and the quality of life.However, people want to obtain health states in real life
There is still a need for a large amount of times, or even when a large amount of populations are checked simultaneously, the time that people need to spend is in geometric multiple
Rise；On the other hand, when people do not need to check comprehensively, as the matters such as queuing of registering influence the efficiency checked,
Cause the waste of time cost.It can be seen that how to people provide inspection result quickly and accurately or analysis conclusion just at
For a conspicuousness problem instantly.
After Cataract is as a kind of common complication, it may have problem describe above.It generally takes now
Method be that people go to a doctor to hospital in person, and are checked in sequence according to number row's code so that people exist
It is taken a substantial amount of time and energy in traffic and ranking；Meanwhile it is usually used at present manually see sheet mode, can also occupy doctor
Raw a large amount of time and energy.As it can be seen that current this test mode still cannot quickly and accurately provide inspection result or analysis
Conclusion.
Summary of the invention
In view of the above problems, the present invention provides a kind of After Cataract image analysis method and device, can it is simple,
The After Cataract situation of patient is quickly and accurately analyzed, to improve the efficiency of user's medical treatment, avoids user's wave
Time taking problem.
To achieve the goals above, the present invention adopts the following technical scheme that:
In a first aspect, the present invention provides a kind of After Cataract image analysis methods, which comprises
Obtain the eye picture for suffering from After Cataract；
Crystalline body region is positioned in the eye picture, and denoising is carried out to the crystalline body region；
Using the detection zone for including in preparatory trained artificial intelligence model as foundation, mentioned in the crystalline body region
Take After Cataract region；The feature database for including using the artificial intelligence model extracts the After Cataract as foundation
At least one After Cataract feature that region includes；
At least one described After Cataract feature is handled to obtain class probability result；
For the analysis result comparison library for including using the artificial intelligence model as foundation, matching obtains the class probability result
Corresponding analysis result.
As an alternative embodiment, the artificial intelligence model is the eye for suffering from After Cataract with several
Picture and several described corresponding labels of eye picture with After Cataract are foundation, carry out depth residual error network training
It obtains；Wherein, the label includes at least After Cataract state of an illness label and treatment recommendations label.
As an alternative embodiment, described handled to obtain at least one described After Cataract feature
Class probability result, comprising:
Analysis is iterated at least one described After Cataract feature by default iterative algorithm, obtains at least one
A characteristic parameter；
Processing integration is carried out at least one described characteristic parameter by default integration algorithm, obtains class probability result；
The class probability result includes and a variety of matched probability results of the After Cataract state of an illness.
As an alternative embodiment, the default integration algorithm is softmax multi-classification algorithm.
As an alternative embodiment, the After Cataract image analysis method further include:
In the analysis result comparison library for including using the artificial intelligence model as foundation, matching obtains the class probability knot
After the corresponding analysis result of fruit, if detecting includes operative treatment suggestion in the analysis result, the analysis knot is sent
Fruit is to medical care equipment, so that medical care equipment is put on record or worked according to the analysis result.
Second aspect, the present invention provides a kind of After Cataract image analysis apparatus, comprising:
Module is obtained, for obtaining the eye picture for suffering from After Cataract；
Preprocessing module is carried out for positioning crystalline body region in the eye picture, and to the crystalline body region
Denoising；
Artificial intelligence module is used for using preparatory trained artificial intelligence model as foundation, in the crystalline body region
After Cataract region is extracted, and according at least one After Cataract feature of the After Cataract extracted region；
The artificial intelligence module is also used to using the artificial intelligence model as foundation, at least one described late coming
Cataract feature is handled to obtain class probability as a result, obtaining the corresponding analysis result of the class probability result with matching.
As an alternative embodiment, the artificial intelligence model is the depth residual error network training by 101 layers
Obtain, wherein the depth residual error network include at least convolutional layer, over-sampling layer, full articulamentum, softmax classification layer,
Dropout algorithm, average pond algorithm and local regularization algorithm.
As an alternative embodiment, the After Cataract image analysis apparatus further includes interactive module and oneself
Update module, wherein
The interactive module for sending the analysis result to user equipment, and receives appealed user equipment feedback
Information；
The self refresh module, for obtaining the analysis as a result, and storing the analysis result to the artificial intelligence
In module, so that artificial intelligence module updates the artificial intelligence model according to the analysis result.
The third aspect, the present invention provides a kind of computer equipment, including memory and processor, the memory is used
In storage computer program, the processor runs the computer program so that the computer equipment executes first aspect public affairs
The After Cataract image analysis method opened.
Fourth aspect, the present invention provides a kind of computer readable storage mediums, are stored in above-mentioned computer equipment
The used computer program.
The After Cataract image analysis method provided according to the present invention, this method can preferentially be obtained with late coming
The eye picture of cataract, the crystalline body region in eye picture of the positioning with After Cataract, obtains the crystalline lens
The picture in region, and denoising is carried out to the picture；After completing to the denoising of crystalline lens region picture, this method
It is foundation according to artificial intelligence model, After Cataract region is extracted in the picture of crystalline lens region, and further extract
After Cataract feature in After Cataract region, so that above-mentioned After Cataract feature can be by artificial intelligence
Handled by the neural network that model includes, obtain class probability as a result, and match obtain analyzing result accordingly.As it can be seen that implementing
This embodiment can simply, quickly and accurately determine the After Cataract situation of patient, and it is corresponding to give user
Analysis as a result, to improve the acquisition efficiency of user's After Cataract information.
To enable the above objects, features and advantages of the present invention to be clearer and more comprehensible, preferred embodiment is cited below particularly, and cooperate
Appended attached drawing, is described in detail below.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not construed as pair
The restriction of the scope of the invention.
Fig. 1 is the flow diagram for the After Cataract image analysis method that the embodiment of the present invention one provides；
Fig. 2 is the flow diagram of After Cataract image analysis method provided by Embodiment 2 of the present invention；
Fig. 3 is the structural schematic diagram for the After Cataract image analysis apparatus that the embodiment of the present invention three provides.
Specific embodiment
Below in conjunction with attached drawing in the embodiment of the present invention, technical solution in the embodiment of the present invention carries out clear, complete
Ground description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Usually exist
The component of the embodiment of the present invention described and illustrated in attached drawing can be arranged and be designed with a variety of different configurations herein.Cause
This, is not intended to limit claimed invention to the detailed description of the embodiment of the present invention provided in the accompanying drawings below
Range, but it is merely representative of selected embodiment of the invention.Based on the embodiment of the present invention, those skilled in the art are not doing
Every other embodiment obtained under the premise of creative work out, shall fall within the protection scope of the present invention.
For the problems of the prior art, the present invention provides a kind of After Cataract image analysis methods, in this kind
In method, user can input or upload the eye picture with After Cataract of oneself, so that analytical equipment can be right
Eye picture with After Cataract carries out crystalline lens zone location, and extracts After Cataract in crystalline body region
Region, with more classification processings, obtains final so that artificial intelligence model can carry out feature extraction to After Cataract region
Analysis result；The analysis result can also be output to user equipment simultaneously, so that user can quickly and accurately get
Analyze result.As it can be seen that implementing this embodiment, the After Cataract feelings of patient can simply, be quickly and accurately determined
Condition, and give user and analyzed accordingly as a result, to improve the efficiency of user's medical treatment.
Wherein, above-mentioned technical method can also be realized using relevant software or hardware, in this present embodiment
No longer add to repeat.For the After Cataract image analysis method and device, it is described below by embodiment.
Embodiment 1
Referring to Fig. 1, Fig. 1 is a kind of process signal of After Cataract image analysis method provided in this embodiment
Figure.
As shown in Figure 1, this kind of After Cataract image analysis method the following steps are included:
S101, the eye picture for suffering from After Cataract is obtained.
As an alternative embodiment, obtaining the eye picture for suffering from After Cataract, may include:
Receive picture, judge whether picture is eye picture, if eye picture then judge eye picture whether be with
The eye picture of After Cataract, if the eye picture with After Cataract thens follow the steps S102.
Implement this embodiment, the eye picture with After Cataract can be accurately obtained, avoid non-suffer from
Being mixed into for the eye picture of After Cataract, to ensure that the matching degree of picture and artificial intelligence model, and then ensure that
The stability and efficiency of the analysis method.
In the present embodiment, the equipment for receiving the eye picture with After Cataract can be After Cataract image
Analytical equipment is also possible to have corresponding artificial intelligence model and other devices with transmission-receiving function, in this present embodiment
It is not construed as limiting.
In the present embodiment, obtaining the eye picture with After Cataract is executed by analytical equipment；From another angle
For degree, the device for uploading the eye picture with After Cataract is not construed as limiting, which can be user and set
It is standby.Meanwhile in the present embodiment, the mode for uploading the eye picture with After Cataract for user is not construed as limiting, for
The method that analytical equipment obtains the eye picture with After Cataract is also not construed as limiting.
In the present embodiment, analytical equipment is mainly to be analyzed and processed to the eye picture with After Cataract, weight
Point is not intended to the transmission of the eye picture with After Cataract, as long as therefore getting the eye with After Cataract
The obtained result of portion's picture, as this step.
In the present embodiment, the eye picture with After Cataract can be user by the mobile phone of oneself or other people or
Person's plate is shot, and is also possible to user and is acquired to obtain using fixed eyes image acquisition device within the hospital
's.
In the present embodiment, the eye picture with After Cataract can be by USB flash disk or cloud transmission, can also
To be that analytical equipment is downloaded automatically, it is not construed as limiting in the present embodiment.
In the present embodiment, After Cataract refer to because a variety of causes (such as aging, heredity, local dystrophia, it is immune with
Metabolic disorder, wound, poisoning, radiation etc.), cause crystalline lens metabolic disorder, causes crystallins to be denaturalized and occur muddy.
Light is obstructed by muddy crystalline lens and can not be projected on the retina at this time, leads to blurred vision.
In the present embodiment, artificial intelligence model mainly for object be late coming After Cataract, then issue after
After cataract refers to After Cataract extracapsular cataract extraction or features of lens in ocular trauma, remaining cortex and fall off in posterior lens capsule
On epithelial hyperplasia, lesser ring of Merkel formed semi-transparent film.
In the present embodiment, the judgment mode of the eye picture with After Cataract is pre-stored in analytical equipment, specifically
Can to be compared roughly by database to the eye picture memory of After Cataract, with judge whether it is suffer from after send out
The eye picture of cataract.Meanwhile other judging means can also be used in the present embodiment, to not making in this present embodiment
Any restriction.
S102, crystalline body region is positioned in eye picture, and denoising is carried out to crystalline body region.
In the present embodiment, crystalline body region is rear capsule hypertrophic region.
In the present embodiment, step S102 is the pretreatment to the eye picture with After Cataract, in the step, point
Analysis apparatus first positions the crystalline body region of slit-lamp retro-illumination, makes an uproar around above-mentioned crystalline body region being removed according to preset algorithm
Sound point, thus the crystalline body region after being processed.
In the present embodiment, the crystalline body region of slit-lamp retro-illumination can be determined according to human body eye structure, because
This, analytical equipment positioning is to be automatically positioned out by crystalline body region, and the method for positioning can be in artificial intelligence model and instruct
The empirical method got；Or the method for positioning can be the method positioned according to factors such as color, contrasts.It can
See, the positioning method in the present embodiment can be it is above-mentioned any, be also possible to that relevant to slit-lamp retro-illumination other are fixed
Position method, above-mentioned localization method come under the protection scope of this programme.
In the present embodiment, crystalline body region is the sub-pictures of the eye picture with After Cataract, therefore, crystalline lens
Whether region is also one, and there are actual pictures, still, need to store in practice, will be not construed as limiting in the present embodiment.
In the present embodiment, denoising is carried out to crystalline lens region picture.Wherein, denoising refers to reduction number
The process of noise in image.Digital picture in reality is subjected to imaging device and external rings in digitlization and transmission process
Border noise jamming etc. influences, and therefore, this picture is referred to as noisy image or noise image.Meanwhile noise is the weight of image interference
Reason is wanted, in practical applications there may be various noises, these noises may generate piece image in the transmission,
It may be generated in the processing such as quantization.
S103, the detection zone for including using preparatory trained artificial intelligence model mention in crystalline body region as foundation
Take After Cataract region.
In the present embodiment, the extraction target of artificial intelligence model is the crystalline body region that obtains that treated, and non-primary is obtained
The crystalline body region obtained, therefore this is explained herein.
In the present embodiment, artificial intelligence model be in advance it is trained, the artificial intelligence model be used for After Cataract
Analysis, specific for the analysis of late coming After Cataract.
In the present embodiment, After Cataract region is to carry out distinguishing judgement, and artificial intelligence by artificial intelligence model
Can model be to carry out distinguishing judgement by the obtained experience of training, therefore, the detection zone that artificial intelligence includes be it is dynamic,
It is according to detection target, i.e. crystalline lens regional dynamics adjust.
In the present embodiment, After Cataract region is medically for determining the judgement of user's After Cataract state of an illness
Region, for example, the After Cataract region can be complete ocular.Wherein, when After Cataract region
There are when subjective determination deviation, which takes the maximum region of range or takes multiple subjective determination regions
Maximum set region.For example, when multiple doctors illustrate that After Cataract region is different zones, maximum region is taken
(when maximum After Cataract region includes After Cataract region described in other doctors), or take multiple
The set (obtaining After Cataract region described in all doctors, generate maximum region) in After Cataract region.
In the present embodiment, After Cataract region includes at least crystalline lens (i.e. lens tissue region), retina, view
Nipple and macular area, to the region for After Cataract inspection for including in After Cataract region, the present embodiment
In be not limited in any way.
In the present embodiment, artificial intelligence model is a kind of a kind of model with autonomous judgement, and the model from
Main judgement is from the study to big data.Using this artificial intelligence model can simply, accurately by machine into
The thing done required for the previous people of row, and can also guarantee accuracy and efficiency.
As foundation, extracting After Cataract region includes at least for S104, the feature database for including using artificial intelligence model
One After Cataract feature.
In the present embodiment, the feature database that artificial intelligence model includes refers to empirical features library in artificial intelligence model, citing
For, the empirical features that empirical features library includes are that artificial intelligence model is rule of thumb made after being learnt by big data
Empirical judgement, the specific physical contents of the empirical judgement exist with artificial intelligence model, but the physical contents can not
It reads and identifies.But, in practice, artificial intelligence model can identify After Cataract feature by the experience, together
When, it can also be ensured that good accuracy.
In the present embodiment, After Cataract region may include multiple After Cataract features, and it is multiple after hair
Cataract feature can be different；After artificial intelligence model extracts at least one After Cataract feature, people
Work model of mind can store at least one above-mentioned After Cataract characteristic model.
In the present embodiment, After Cataract feature can make characteristics of image corresponding with After Cataract symptom,
Can be auxiliary the After Cataract state of an illness determine other characteristics of image, or even can also be with the After Cataract state of an illness without
It closes still artificial intelligence and thinks useful characteristics of image.As it can be seen that After Cataract feature can be a variety of, wherein late coming
Cataract feature is there are in After Cataract region, and After Cataract is characterized in the training knot with artificial intelligence model
Fruit carries out judging extraction.Therefore, for the specific form of After Cataract feature, the study with artificial intelligence model is needed
As a result subject to, wherein After Cataract feature includes at least the corresponding characteristics of image of After Cataract symptom from principle.
In the present embodiment, artificial intelligence model is trained in advance, therefore user can suffer from late coming by input
The eye picture of cataract directly obtains analysis result；The benefit of this artificial intelligence model is strongly professional, accuracy height,
It is high-efficient.
In the present embodiment, artificial intelligence model is After Cataract artificial intelligence model, stronger specific aim with sentence
Cutting capacity allows the model that work is effectively performed.Meanwhile the artificial intelligence model is by algorithm and big data training
It obtains, specifically self study can be carried out by artificial intelligence deep neural network and obtained.Wherein, big data includes having label
After Cataract profession picture, this kind of picture accurate label added by artificial or machine so that information definitely,
So that it is determined that the learning direction of artificial intelligence model with study after major field；On the other hand, artificial in the present embodiment
Algorithm used in model of mind is the algorithm for After Cataract, and specific algorithm can be carried out by specific implementation step
Understand.
In the present embodiment, the training of the artificial intelligence model is preceding, and training method is Neural Network Self-learning, in this reality
It applies in example, the training method is it can be appreciated that have the intelligence learning method artificially interfered, wherein artificial interference refers to phase
The label answered.And in the present solution, the training for artificial intelligence model explains, but not as intermediate portions, this
The emphasis of scheme is to the use of artificial intelligence model (After Cataract artificial intelligence model) and machinery equipment to model
Application method, will no longer repeat in this present embodiment.
In the present embodiment, feature extraction is carried out to the eye picture with After Cataract and refers to acquisition with late coming
Related feature is judged with the After Cataract state of an illness in the eye picture of cataract, wherein is sentenced with the After Cataract state of an illness
Related feature of breaking refers to that the feature stored in artificial intelligence model or medical staff sentence in the progress After Cataract state of an illness
The disconnected fixed character observed often；For example, what feature extraction obtained is wrapped in the eye picture with After Cataract
It includes After Cataract or may include the part of After Cataract.
In the present embodiment, feature extraction can also be it is a kind of will with After Cataract eye picture according to rear hair
Cataract pathological characters carry out piecemeal subregion, obtain multiple characteristic areas, and the method for carrying out signature analysis according to region.
Implement this embodiment, the eye picture with After Cataract can be divided into multiple representative
Characteristic area perhaps multiple characteristic blocks so that artificial intelligence model can be by targetedly to features described above region or feature
Block is analyzed, to improve efficiency；Wherein, above-mentioned representative characteristic area can refer to the eye with After Cataract
With the region of pathological characters in portion's picture.
In the present embodiment, feature extracting method is provided by artificial intelligence model, the feature extraction side in artificial intelligence model
Method is obtained by artificial intelligence self study training, to repeating no more in this present embodiment.Also, artificial intelligence self study is
The self study cooperatively formed by multi-classification algorithm and cost function and gradient optimization algorithm.
S105, at least one After Cataract feature is handled to obtain class probability result.
In the present embodiment, what at least one After Cataract was characterized in being extracted according to artificial intelligence model.
In the present embodiment, class probability the result is that one group of probability value embodiment, for determining this with After Cataract
Eye picture the After Cataract state of an illness, wherein the corresponding After Cataract state of an illness of most probable value is used as the trouble
There is the After Cataract state of an illness of the eye picture of After Cataract.
For example, the acquisition process of class probability result is as follows: artificial intelligence model is to an After Cataract spy
Levy the multiple level-one probability carried out, wherein multiple level-one probability respectively correspond an After Cataract state of an illness, i.e. the level-one is general
What rate was used to indicate the corresponding After Cataract state of an illness obtains disease probability, meanwhile, multiple level-one probability and value can be any
Numerical value is not fixed as " 1 "；After getting multiple level-one probability, the multiple second levels similar with multiple level-one probability are obtained
Probability, multiple second level probability is similar with level-one probability, and the difference is that second level probability it is corresponding be second late coming
Cataract feature；And so on, artificial intelligence will acquire the equiprobability such as level-one probability, second level probability, three-level probability, these are general
Rate is all in ad eundem, and the calculating for participating in class probability result as radix；The class probability is the result is that based on artificial
The more classification layers of intelligence are handled, and display can learn class probability the result is that one group of probability value, the group in processing result
Probability value is " 1 " and interrelated influence, therefore, a maximum corresponding After Cataract in this group of probability value with value
The state of an illness will be judged as the After Cataract state of an illness of the eye picture with After Cataract.Wherein, the part is to classification
The explanation of probability results is for illustrating a kind of feasible calculation method of class probability and obtaining process, but the part is retouched
It states and any restriction is not made to the acquisition of class probability result.
In the present embodiment, compressive classification processing refers to that one feature of input, output this feature meet the probability of various situations
As a result.Wherein, above-mentioned various situations are fixed a variety of situations, will not increase reduction with classification processing, meanwhile, classification is general
Rate result means that features described above corresponds to the probability results of various situations, i.e. probability is bigger, and degree of conformity is higher.
In the present embodiment, compressive classification processing is carried out at least one After Cataract feature, i.e., to be got all
Probability results of the feature for all situations.And class probability result as described in this embodiment is above-mentioned all situations
Probability results.
In the present embodiment, in multiple probability results of a feature, when being much larger than other probability there are a probability, sentence
The probability is corresponding calmly is classified as multi-class classification, which includes After Cataract severity, After Cataract
The information such as range.Wherein, information included by above-mentioned multi-class classification is stored in corresponding analysis result.
As foundation, matching obtains class probability result pair for S106, the analysis result comparison library for including using artificial intelligence model
The analysis result answered.
In the present embodiment, the analysis result comparison library that artificial intelligence model includes refers to preset various results, wherein pre-
If the sums of various results can be fixed and immutable.
It include various results in the analysis result comparison library that artificial intelligence model includes, wherein Mei Gejie in the present embodiment
Fruit can be made of multiple entries, and in artificial intelligence model, each feature can correspond to one or more corresponding entry,
Obtain class probability after completing signature analysis as a result, and the class probability result includes multiple probability values, and it is multiple general
Rate value corresponds to multiple features, and multiple features correspond to corresponding entry, and in multilayer neural network, artificial intelligence model can basis
A large amount of comparing obtains the analysis result being accurately made of multiple entries.Wherein, analysis result comparison library is people
Work intelligent training obtains, meanwhile, which can be dynamically.
In the present embodiment, the After Cataract state of an illness that analysis result can be artificial intelligence model generation refers to the state of an illness
It leads, to not being limited in any way in this present embodiment.On the other hand, for the acquisition modes of analysis result, with phase described above
Together, therefore, it will not repeat herein.
In the present embodiment, artificial intelligence model provides matching process, and class probability result provides matching basic data, citing
For, analytical equipment get all class probabilities as a result, according to artificial intelligence model train come method, to classify it is general
Rate result is summarized analysis, and a correspondence multi-class classification the most appropriate is obtained.
In the present embodiment, artificial intelligence model is obtained by artificial intelligence, machine learning and neural network learning,
It is inputted as largely with the picture of label, wherein the calibration mode of label does not do any restriction；Receiving a large amount of tool
After having the picture of label, equipment remembers (storage) all pictures, and is understood that (aspect ratio is to according to label
Practise) so that equipment completes the understanding to picture self by a large amount of study and self-correcting, eventually disengage from picture formed it is specific
Model, and model at this time is the experience and understanding learnt, so that subsequent can pass through when reuse the model
The artificial intelligence model borrows the experience that it learns to solve corresponding things, to complete the use of artificial intelligence.
But above-mentioned content be basic concept, rather than specific implementation method, and above-mentioned basic concept be based on this programme, because
This does not cause to inspire to this programme, to will not be described in great detail in this present embodiment.
In the present embodiment, which pays attention to the use in terms of late coming After Cataract, meanwhile, artificial intelligence
Energy model is also the artificial intelligence model in terms of late coming After Cataract.But the technical method can be used for it
His adaptable field, it is seen then that the technical method, technological means can play the role of guidance and enlightenment for other aspects.
As shown in Figure 1, the After Cataract image analysis method provided according to the present invention, available to suffer from late coming
The eye picture of cataract, so that analytical equipment can pre-process the eye picture with After Cataract, and
After pretreatment, feature extraction and more classification processings are carried out based on artificial intelligence, obtain class probability as a result, so that artificial intelligence
Model can carry out result matching to class probability result, obtain correspondence analysis as a result, to rapidly and accurately provide to user
Analyze result.As it can be seen that implementing method described in Fig. 1, analysis can be efficiently and accurately provided the user with as a result, so that user
It can judge oneself whether need further to see a doctor based on the analysis results.As it can be seen that implement this embodiment, can it is simple,
The After Cataract situation of patient is quickly and accurately obtained, and gives user and replies accordingly, to improve user's medical treatment
Efficiency.
Embodiment 2
Referring to Fig. 2, Fig. 2 is a kind of process signal of After Cataract image analysis method provided in this embodiment
Figure.As shown in Fig. 2, the After Cataract image analysis method the following steps are included:
S201, the eye picture for suffering from After Cataract is obtained.
Implement this embodiment, it can be to avoid the eye figure with After Cataract of non-post cataract information
The reception of piece improves the treatment effeciency of artificial intelligence model.
S202, crystalline body region is positioned in eye picture, and denoising is carried out to crystalline body region.
Implement this embodiment, the eye picture with After Cataract can be marked off, so that suffering from late coming
The eye picture of cataract can be divided into the crystalline lens region picture of After Cataract information.As it can be seen that implementing this reality
Mode is applied, the region picture including After Cataract information can be accurately obtained, realizes the eye for suffering from After Cataract
It is subsequent to improve the eye picture with After Cataract to reduce subsequent useless analysis for the pretreatment of portion's picture
Analysis efficiency.
S203, the detection zone for including using preparatory trained artificial intelligence model mention in crystalline body region as foundation
Take After Cataract region.
In the present embodiment, after artificial intelligence model is eye picture with several with After Cataract and several suffer from
The corresponding label of eye picture of cataract is foundation, carries out what depth residual error network training obtained；Wherein, label is at least
Including After Cataract state of an illness label and treatment recommendations label.
In the present embodiment, artificial intelligence model is the eye with After Cataract according to several with disease label
Picture carries out what the training of depth convolutional neural networks was formed；Wherein, above-mentioned label can also include that After Cataract is different tight
The corresponding transparency label of weight degree, the corresponding range tag of After Cataract range size and for indicating that late coming is white
The location tags of both cataract or glaucoma and the optical axis area relative position.
Implement this embodiment, effective feature can be got, by artificial intelligence convenient for the execution of subsequent step.
As foundation, extracting After Cataract region includes at least for S204, the feature database for including using artificial intelligence model
One After Cataract feature.
Implement this embodiment, can accurately obtain a large amount of After Cataract feature, avoid omitting, and improve
The Effective Probability of feature.
S205, analysis is iterated at least one After Cataract feature by default iterative algorithm, obtained at least
One characteristic parameter.
In the present embodiment, default iterative algorithm is the calculation for meeting After Cataract feature what state of an illness condition and being analyzed
Method, wherein characteristic parameter represents the probability for some state of an illness.
For example, step S205 is become it is to be understood that at least one After Cataract feature by mathematical formulae
The continuous iteration of swap-in row, and process and be converted at least one 0~1 probability value.Wherein at least one above-mentioned probability value is then
At least one characteristic parameter.
S206, processing integration is carried out at least one characteristic parameter by default integration algorithm, obtains class probability result；
Class probability result includes and a variety of matched probability results of the After Cataract state of an illness.
In the present embodiment, the After Cataract state of an illness may include that severe After Cataract is badly in need of the case where seeing a doctor, light
The case where degree After Cataract needs diagnosis and treatment, include thes case where that slight After Cataract needs drug therapy etc.,
In, above-mentioned probability results are the probability for above-mentioned various situations.For example, the probability results are for indicating that user is current
The probability that whether meets of After Cataract disease condition, after illustrating that user is current if the probability highest of certain situation
Cataract disease condition is certain above-mentioned situation.As it can be seen that the probability results are actually and the After Cataract state of an illness
Matching probability, rather than other probabilistic information such as deteriorates probability, blind probability or genetic probability.
In the present embodiment, presetting integration algorithm is softmax multi-classification algorithm.
In the present embodiment, softmax classifier refers to one layer of neural network including softmax function, and softmax is to use
Polytypic means are solved, are usually used in solving two classification problems from our common logistic regression, SVM etc. different.softmax
Corresponding more classification problems, for example, can be abstracted and be considered to identify that handwritten numeral needs 10 classification (more classification).
For example, neural network includes input layer, characteristic layer with softmax classifier and class probability result.Manually
Intelligence is by carrying out judgement statistical learning to input layer, then is subject to the multiple complicated comparison study of characteristic layer, and corresponding general
Rate calculates, and obtains the probability results for more results, which is class probability result.Wherein, for concrete example,
Input layer is one layer, and two layers of characteristic layer, softmax classifier is one layer, and class probability result is output layer, wherein altogether five
Layer, still, by way of example only, specific embodiment can be used in similar neural network this figure.Above-mentioned softmax classification
Device is also one layer of neural network, polytypic for solving the problems, such as, wherein for example, if there is weight degree, middle degree with
And if light degree, it can be understood as there are four classification for tool, and one is inputted, and artificial intelligence model can carry out the input
Probabilistic determination, so that probability results correspond to aforementioned four classification, to complete the output of class probability result.However, in model
During (artificial intelligence model, following model can be understood as artificial intelligence model) learns, this is basic study side
Method, and non-final learning outcome, to no longer adding to repeat in this present embodiment.
As foundation, matching obtains class probability result pair for S207, the analysis result comparison library for including using artificial intelligence model
The analysis result answered.
As an alternative embodiment, the analysis result comparison library for including using artificial intelligence model as foundation, matches
The corresponding analysis of class probability result is obtained as a result, may include:
It is according to acquisition disease type corresponding with class probability result with artificial intelligence model and class probability result；Disease
Whether sick type includes at least After Cataract illness and the After Cataract extent depth, and disease type is matched
Corresponding analysis result.
In the present embodiment, disease type may include whether After Cataract, After Cataract range size, journey
The depth, After Cataract position etc. are spent, to being not construed as limiting in this present embodiment.
It in the present embodiment, include that user does not suffer from After Cataract and user whether After Cataract illness
Two kinds of situations of After Cataract, and the extent depth refers to that After Cataract extent is deep or late coming is white interior
It is shallow to hinder extent, wherein the depth of extent is to predefine well.In artificial intelligence model, corresponding mark is posted
Label, label substance are that extent is deep, and extent is shallow, and label substance can determine for machine or human intervention machine
, and emphasize to use artificial intelligence model in the present embodiment, it is seen then that the degree depth is the noun determined, in the present embodiment not
It repeats again；On the other hand, to being not construed as limiting in establishment process the present embodiment of artificial intelligence, therefore the establishment process of artificial intelligence
Also this programme should not be constituted and is limited.
In the present embodiment, the extent depth may include three aspects, wherein first aspect is transparency (transparency
Be further divided into two classes, one kind be it is transparent, one kind is opaque), second aspect be range (range can also be divided into two classes,
One kind be it is a wide range of, one kind is small range), the third aspect be position (position can also be divided into two classes, and one kind is center, one
Class is surrounding).In the present embodiment, three aspects of the extent depth, six subclassifications can be used as artificial intelligence model
Trained label, so that artificial intelligence model can independently be trained according to the label, to have to the reason in terms of these
Solution and judgement.
It can have corresponding treatment recommendations in the present embodiment, in disease type, so that user is getting disease
Corresponding treatment recommendations can also be obtained while sick type.In the present embodiment, treatment recommendations may include two classification, one
To need the suggestion performed the operation, one is the suggestion for needing to carry out follow-up.In the present embodiment, the training of artificial intelligence model can also
To include the label of above two suggestion, so that artificial intelligence model can establish treatment recommendations corresponding with disease type
Database, to obtain corresponding analysis result.
In the present embodiment, according to disease type, built-in analysis result corresponding with disease type is matched.
Implement this embodiment, can be matched according to After Cataract disease type it is corresponding analysis as a result, from
And make the output of artificial intelligence equipment accurate and stablize, avoid analysis result as the iteration of artificial intelligence changes.
S208, when including that operative treatment is suggested in detecting analysis result, analysis result is sent to medical care equipment, so that
Medical care equipment is put on record or is worked based on the analysis results.
In the present embodiment, medical care equipment can be surgical apparatus, can be service equipment, can also be doctor's equipment, right
It is not construed as limiting in this present embodiment.Wherein, for example, this method can carry out network pass with the equipment for executing laser surgey
Connection will need the patient information performed the operation to be synchronized in surgical apparatus automatically, so as to avoid repetitive operation and artificial careless omission.
In the present embodiment, artificial intelligence model can according to user input the eye picture with After Cataract into
Row analysis, and based on the analysis results carry out artificial intelligence model learn again and iteration, to constantly improve the artificial intelligence mould
Type.
In After Cataract image analysis method as shown in Figure 2, the eye for suffering from After Cataract is obtained first
Picture, and the eye picture with After Cataract is pre-processed, obtain corresponding After Cataract administrative division map
Piece obtains corresponding disease type (including rear hair so that artificial intelligence model handles After Cataract region picture
The information such as the cataract degree depth), and match to obtain corresponding analysis as a result, and learning user according to disease type
Analysis is as a result, in order to see a doctor in time or trust immediately.As it can be seen that implement method described in Fig. 2, it can be simple, quick, quasi-
It really determines the After Cataract situation of patient, and gives user and reply accordingly, to improve the effect of user's medical treatment
Rate.
Embodiment 3
Referring to Fig. 3, the structure that Fig. 3 is a kind of After Cataract image analysis apparatus provided in an embodiment of the present invention is shown
It is intended to.
As shown in figure 3, the After Cataract image analysis apparatus, comprising:
Module 301 is obtained, for obtaining the eye picture for suffering from After Cataract.
Preprocessing module 302 carries out at denoising for positioning crystalline body region in eye picture, and to crystalline body region
Reason.
Artificial intelligence module 303, for being mentioned in crystalline body region using preparatory trained artificial intelligence model as foundation
After Cataract region is taken, and according at least one After Cataract feature of After Cataract extracted region.
Artificial intelligence module 303 is also used to using artificial intelligence model as foundation, at least one After Cataract feature
It is handled to obtain class probability as a result, obtaining the corresponding analysis result of class probability result with matching.
As an alternative embodiment, artificial intelligence model is obtained by 101 layers of depth residual error network training
, wherein depth residual error network includes at least convolutional layer, over-sampling layer, full articulamentum, softmax classification layer, dropout and calculates
Method, average pond algorithm and local regularization algorithm.
In the present embodiment, dropout algorithm can be regarded as a kind of model average algorithm, which can manage
Solution for from different models estimated value or predicted value got up by certain weighted average.The model is averagely also referred to as mould
Type combination, generally comprises combinational estimation and combined prediction." different models " is embodied in dropout, is in practice exactly us
What the method that hidden node is ignored in random selection generated.Because the hidden node ignored at random in each training process is not
Together, so it is different for allowing for the network trained every time all, so that training can singly make the model of one " new " every time.
In addition, above-mentioned implicit node is all to occur at random with certain probability, therefore cannot be guaranteed that every 2 implicit nodes go out simultaneously every time
Existing, the update of such weight is no longer dependent on the collective effect that fixed relationship implies node, and certain features is prevented only to exist
Situation just effective under other special characteristics.As it can be seen that dropout process is exactly that a very effective neural network model is flat
Equal method carrys out consensus forecast probability by a large amount of different network of training.Wherein, in essence, dropout algorithm is just
It is to train (each training data is all random selection) on different training sets by different models, finally in each mould
Type is with identical weight come the algorithm of " fusion ".
In the present embodiment, both dropout and global average pondization are combined into one, it can be to entire network in structure
On do regularization and prevent over-fitting.To directly eliminate the feature of black box in full articulamentum, it is real directly to impart each node
The interior other meaning on border.
As an alternative embodiment, After Cataract image analysis apparatus further includes interactive module and self refresh
Module, wherein
Interactive module for transmission analysis result to user equipment, and receives the information of appealed user equipment feedback.
Self refresh module, for obtain analysis as a result, and store analysis result into artificial intelligence module 303, to make one
Work intelligent object 303 updates artificial intelligence model based on the analysis results.
As an alternative embodiment, artificial intelligence model is to suffer from late coming with disease label according to several
The eye picture of cataract carries out what the training of depth convolutional neural networks was formed；Disease label may include After Cataract not
Corresponding transparency label and the corresponding range tag of After Cataract range size with severity.
Implement this embodiment, the source of artificial intelligence model can be embodied, so that this programme not only includes artificial
The use of model of mind further includes the training method of artificial intelligence model.
Implement this embodiment, intelligence learning can be effectively performed by 101 layer depth residual error networks, to mention
High efficiency and improve order of accuarcy.
As it can be seen that After Cataract image analysis apparatus described in the present embodiment, it can be simple, quickly and accurately true
The After Cataract situation of patient is made, and gives user and replies accordingly, to improve the efficiency of user's medical treatment.
In addition, the computer equipment may include smart phone, put down the present invention also provides another computer equipment
Plate computer, vehicle-mounted computer, intelligent wearable device etc..The computer equipment includes memory and processor, and memory can be used for depositing
Store up computer program, processor by running above-mentioned computer program, thus make computer equipment execute the above method or on
State the function of each unit in device.
Memory may include storing program area and storage data area, wherein storing program area can storage program area, at least
Application program needed for one function (such as sound-playing function, image player function etc.) etc.；Storage data area can store root
Created data (such as audio data, phone directory etc.) etc. are used according to computer equipment.In addition, memory may include height
Fast random access memory, can also include nonvolatile memory, a for example, at least disk memory, flush memory device,
Or other volatile solid-state parts.
The present embodiment additionally provides a kind of computer storage medium, for storing calculating used in above-mentioned computer equipment
Machine program.
In several embodiments provided herein, it should be understood that disclosed device and method can also pass through
Other modes are realized.The apparatus embodiments described above are merely exemplary, for example, flow chart and structure in attached drawing
Figure shows the system frame in the cards of the device of multiple embodiments according to the present invention, method and computer program product
Structure, function and operation.In this regard, each box in flowchart or block diagram can represent a module, section or code
A part, a part of above-mentioned module, section or code includes one or more for implementing the specified logical function
Executable instruction.It should also be noted that function marked in the box can also be to be different from the implementation as replacement
The sequence marked in attached drawing occurs.For example, two continuous boxes can actually be basically executed in parallel, they are sometimes
It can execute in the opposite order, this depends on the function involved.It is also noted that in structure chart and/or flow chart
The combination of each box and the box in structure chart and/or flow chart, can function or movement as defined in executing it is dedicated
Hardware based system realize, or can realize using a combination of dedicated hardware and computer instructions.
In addition, each functional module or unit in each embodiment of the present invention can integrate one independence of formation together
Part, be also possible to modules individualism, an independent part can also be integrated to form with two or more modules.
If described function is realized and when sold or used as an independent product in the form of software function module, can
To be stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention substantially or
Say that the part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, it should
Computer software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be
Smart phone, personal computer, server or network equipment etc.) execute the complete of method described by each embodiment of the present invention
Portion or part steps.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. are various can store journey
The medium of sequence code.
Content described above, only a specific embodiment of the invention, but protection scope of the present invention is not limited to
In this, anyone skilled in the art in the technical scope disclosed by the present invention, can readily occur in variation or replace
It changes, should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with scope of protection of the claims
Subject to.
Claims (10)
1. a kind of After Cataract image analysis method, which is characterized in that the described method includes:
Obtain the eye picture for suffering from After Cataract；
Crystalline body region is positioned in the eye picture, and denoising is carried out to the crystalline body region；
Using the detection zone for including in preparatory trained artificial intelligence model as foundation, after being extracted in the crystalline body region
Cataract region；The feature database for including using the artificial intelligence model extracts the After Cataract region as foundation
Including at least one After Cataract feature；
At least one described After Cataract feature is handled to obtain class probability result；
For the analysis result comparison library for including using the artificial intelligence model as foundation, it is corresponding that matching obtains the class probability result
Analysis result.
2. After Cataract image analysis method according to claim 1, which is characterized in that the artificial intelligence model
Be with several with After Cataract eye picture and it is described several with After Cataract eye pictures it is corresponding
Label is foundation, carries out what depth residual error network training obtained；Wherein, the label includes at least After Cataract state of an illness mark
Label and treatment recommendations label.
3. After Cataract image analysis method according to claim 1, which is characterized in that described to described at least one
A After Cataract feature is handled to obtain class probability result, comprising:
Analysis is iterated at least one described After Cataract feature by default iterative algorithm, obtains at least one spy
Levy parameter；
Processing integration is carried out at least one described characteristic parameter by default integration algorithm, obtains class probability result；It is described
Class probability result includes and a variety of matched probability results of the After Cataract state of an illness.
4. After Cataract image analysis method according to claim 3, which is characterized in that the default integration algorithm
For softmax multi-classification algorithm.
5. After Cataract image analysis method according to claim 1, which is characterized in that the After Cataract
Image analysis method further include:
In the analysis result comparison library for including using the artificial intelligence model as foundation, matching obtains the class probability result pair
After the analysis result answered, if detecting includes operative treatment suggestion in the analysis result, the analysis result is sent extremely
Medical care equipment, so that medical care equipment is put on record or worked according to the analysis result.
6. a kind of After Cataract image analysis apparatus, which is characterized in that the After Cataract image analysis apparatus packet
It includes:
Module is obtained, for obtaining the eye picture for suffering from After Cataract；
Preprocessing module is denoised for positioning crystalline body region in the eye picture, and to the crystalline body region
Processing；
Artificial intelligence module, for being extracted in the crystalline body region using preparatory trained artificial intelligence model as foundation
After Cataract region, and according at least one After Cataract feature of the After Cataract extracted region；
The artificial intelligence module is also used to using the artificial intelligence model as foundation, white at least one described late coming interior
Barrier feature is handled to obtain class probability as a result, obtaining the corresponding analysis result of the class probability result with matching.
7. After Cataract image analysis apparatus according to claim 6, which is characterized in that the artificial intelligence model
It is to be obtained by 101 layers of depth residual error network training, wherein the depth residual error network includes at least convolutional layer, crosses and adopt
Sample layer, full articulamentum, softmax classification layer, Dropout algorithm, average pond algorithm and local regularization algorithm.
8. After Cataract image analysis apparatus according to claim 6, which is characterized in that the After Cataract
Image analysis apparatus further includes interactive module and self refresh module, wherein
The interactive module for sending the analysis result to user equipment, and receives the letter of appealed user equipment feedback
Breath；
The self refresh module, for obtaining the analysis as a result, and storing the analysis result to the artificial intelligence module
In, so that artificial intelligence module updates the artificial intelligence model according to the analysis result.
9. a kind of computer equipment, which is characterized in that including memory and processor, the memory is for storing computer
Program, the processor runs the computer program so that the computer equipment executes according to claim 1 to any in 5
After Cataract image analysis method described in.
10. a kind of computer readable storage medium, which is characterized in that it is stored in computer equipment as claimed in claim 9
Used computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810725079.2A CN109102494A (en) | 2018-07-04 | 2018-07-04 | A kind of After Cataract image analysis method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810725079.2A CN109102494A (en) | 2018-07-04 | 2018-07-04 | A kind of After Cataract image analysis method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109102494A true CN109102494A (en) | 2018-12-28 |
Family
ID=64845682
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810725079.2A Pending CN109102494A (en) | 2018-07-04 | 2018-07-04 | A kind of After Cataract image analysis method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109102494A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109872322A (en) * | 2019-02-27 | 2019-06-11 | 电子科技大学 | A kind of nuclear cataract lesion region localization method based on cascade detection model |
| CN110013216A (en) * | 2019-03-12 | 2019-07-16 | 中山大学中山眼科中心 | A kind of artificial intelligence cataract analysis system |
| CN110176297A (en) * | 2019-05-24 | 2019-08-27 | 南京博视医疗科技有限公司 | A kind of Brilliant Eyes bottom laser surgey assistant diagnosis system and its method |
| CN111598866A (en) * | 2020-05-14 | 2020-08-28 | 四川大学 | Lens key feature positioning method based on eye B-ultrasonic image |
| CN111658308A (en) * | 2020-05-26 | 2020-09-15 | 首都医科大学附属北京同仁医院 | In-vitro focusing ultrasonic cataract treatment operation system |
| CN112006651A (en) * | 2020-09-10 | 2020-12-01 | 孙礼华 | Cataract surgery auxiliary diagnosis system and method thereof |
| CN115969311A (en) * | 2023-03-09 | 2023-04-18 | 汕头大学·香港中文大学联合汕头国际眼科中心 | Laser capsulotomy auxiliary system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102984997A (en) * | 2009-08-24 | 2013-03-20 | 新加坡保健服务集团有限公司 | A Method and system of determining a grade of nuclear cataract |
| CN103458772A (en) * | 2011-04-07 | 2013-12-18 | 香港中文大学 | Method and device for retinal image analysis |
| CN104881683A (en) * | 2015-05-26 | 2015-09-02 | 清华大学 | Cataract eye fundus image classification method based on combined classifier and classification apparatus |
| US20160063893A1 (en) * | 2014-09-03 | 2016-03-03 | Aira Tech Corporation | Media streaming methods, apparatus and systems |
| CN106897268A (en) * | 2017-02-28 | 2017-06-27 | 科大讯飞股份有限公司 | Text semantic understanding method, device and system |
| CN108198621A (en) * | 2018-01-18 | 2018-06-22 | 中山大学 | A kind of database data synthesis dicision of diagnosis and treatment method based on neural network |
-
2018
- 2018-07-04 CN CN201810725079.2A patent/CN109102494A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102984997A (en) * | 2009-08-24 | 2013-03-20 | 新加坡保健服务集团有限公司 | A Method and system of determining a grade of nuclear cataract |
| CN103458772A (en) * | 2011-04-07 | 2013-12-18 | 香港中文大学 | Method and device for retinal image analysis |
| US20160063893A1 (en) * | 2014-09-03 | 2016-03-03 | Aira Tech Corporation | Media streaming methods, apparatus and systems |
| CN104881683A (en) * | 2015-05-26 | 2015-09-02 | 清华大学 | Cataract eye fundus image classification method based on combined classifier and classification apparatus |
| CN106897268A (en) * | 2017-02-28 | 2017-06-27 | 科大讯飞股份有限公司 | Text semantic understanding method, device and system |
| CN108198621A (en) * | 2018-01-18 | 2018-06-22 | 中山大学 | A kind of database data synthesis dicision of diagnosis and treatment method based on neural network |
Non-Patent Citations (1)
| Title |
|---|
| HAOTIAN LIN等: "an artificial intelligence platform for the multihospital collaborative management of congenital cataracts", 《NATURE BIOMEDICAL ENGINEERING》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109872322A (en) * | 2019-02-27 | 2019-06-11 | 电子科技大学 | A kind of nuclear cataract lesion region localization method based on cascade detection model |
| CN110013216A (en) * | 2019-03-12 | 2019-07-16 | 中山大学中山眼科中心 | A kind of artificial intelligence cataract analysis system |
| CN110013216B (en) * | 2019-03-12 | 2022-04-22 | 中山大学中山眼科中心 | Artificial intelligence cataract analysis system |
| CN110176297A (en) * | 2019-05-24 | 2019-08-27 | 南京博视医疗科技有限公司 | A kind of Brilliant Eyes bottom laser surgey assistant diagnosis system and its method |
| CN110176297B (en) * | 2019-05-24 | 2020-09-15 | 南京博视医疗科技有限公司 | Intelligent auxiliary diagnosis system for fundus laser surgery |
| CN111598866A (en) * | 2020-05-14 | 2020-08-28 | 四川大学 | Lens key feature positioning method based on eye B-ultrasonic image |
| CN111598866B (en) * | 2020-05-14 | 2023-04-11 | 四川大学 | Lens key feature positioning method based on eye B-ultrasonic image |
| CN111658308A (en) * | 2020-05-26 | 2020-09-15 | 首都医科大学附属北京同仁医院 | In-vitro focusing ultrasonic cataract treatment operation system |
| CN111658308B (en) * | 2020-05-26 | 2022-06-17 | 首都医科大学附属北京同仁医院 | In-vitro focusing ultrasonic cataract treatment operation system |
| CN112006651A (en) * | 2020-09-10 | 2020-12-01 | 孙礼华 | Cataract surgery auxiliary diagnosis system and method thereof |
| CN115969311A (en) * | 2023-03-09 | 2023-04-18 | 汕头大学·香港中文大学联合汕头国际眼科中心 | Laser capsulotomy auxiliary system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109102494A (en) | A kind of After Cataract image analysis method and device | |
| Hacisoftaoglu et al. | Deep learning frameworks for diabetic retinopathy detection with smartphone-based retinal imaging systems | |
| Vega et al. | Retinal vessel extraction using lattice neural networks with dendritic processing | |
| Shankar et al. | Deep neural network with moth search optimization algorithm based detection and classification of diabetic retinopathy images | |
| Zulkifley et al. | Pterygium-Net: a deep learning approach to pterygium detection and localization | |
| Khan et al. | Deep learning for ocular disease recognition: an inner-class balance | |
| CN111667490B (en) | Fundus picture cup optic disc segmentation method | |
| WO2020176039A1 (en) | System and method for classifying eye images | |
| Tahvildari et al. | Application of artificial intelligence in the diagnosis and management of corneal diseases | |
| Mai et al. | Few-shot transfer learning for hereditary retinal diseases recognition | |
| Ghamsarian et al. | LensID: a CNN-RNN-based framework towards lens irregularity detection in cataract surgery videos | |
| CN108806778A (en) | A kind of cataract state of an illness method for monitoring and system | |
| BAKIR et al. | Using transfer learning technique as a feature extraction phase for diagnosis of cataract disease in the eye | |
| Shulha et al. | Deep learning with metadata augmentation for classification of diabetic retinopathy level | |
| Kind et al. | An explainable AI-based computer aided detection system for diabetic retinopathy using retinal fundus images | |
| Muijzer et al. | Automatic evaluation of graft orientation during Descemet membrane endothelial keratoplasty using intraoperative OCT | |
| CN113330522A (en) | System and method for selecting intraocular lens using frontal view zone prediction | |
| CN109036550A (en) | A kind of cataract analysis method and device based on artificial intelligence | |
| Ullaha et al. | Optic disc segmentation and classification in color fundus images: a resource-aware healthcare service in smart cities | |
| Sherbakov | Computational principles for an autonomous active vision system | |
| Tanvir et al. | Clinical Insights Through Xception: A Multiclass Classification of Ocular Pathologies | |
| CN113378794A (en) | Information correlation method for elephant and symptom information | |
| Sujithra et al. | Adaptive cluster-based superpixel segmentation and BMWMMBO-based DCNN classification for glaucoma detection | |
| Angeline et al. | Automated Detection of Cataract Using a Deep Learning Technique | |
| Bhattacharjee et al. | Artificial intelligence in cataract: What’s new? |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181228 |