CN107665736A - Method and apparatus for generating information - Google Patents
Method and apparatus for generating information Download PDFInfo
- Publication number
- CN107665736A CN107665736A CN201710914487.8A CN201710914487A CN107665736A CN 107665736 A CN107665736 A CN 107665736A CN 201710914487 A CN201710914487 A CN 201710914487A CN 107665736 A CN107665736 A CN 107665736A
- Authority
- CN
- China
- Prior art keywords
- information
- lung
- targeted customer
- tubercle
- pulmonary nodule
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Image Analysis (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
- Processing Or Creating Images (AREA)
Abstract
This application discloses the method and apparatus for generating information.One embodiment of this method includes:Receive the lung images and user record information of targeted customer;By the detection model of lung images input training in advance, the pulmonary nodule description information of the targeted customer is obtained;Based on obtained pulmonary nodule description information determine the targeted customer lung whether nodosity;Lung's nodosity in response to determining the targeted customer, the targeted customer is inputted to the forecast model of training in advance in the lung images, pulmonary nodule description information, user record information of different time, obtains the pulmonary nodule information of forecasting of the targeted customer.The embodiment is effectively utilized the lung images and user record information of targeted customer, improves the accuracy of pulmonary nodule information of forecasting.
Description
Technical field
The application is related to field of computer technology, and in particular to areas of information technology, more particularly, to generates information
Method and apparatus.
Background technology
Continuous development and maturation with information technology, computer application are also more and more extensive.It can pass through at this stage
Computer is handled medical image, to generate image information as early as possible.
Lung cancer is one of the incidence of disease and death rate highest malignant tumour in the world, can use low-dose CT at present
(Computed Tomography, CT scan) examination.But hand inspection, different stage, area doctor be present
Level difference between institute, and the difference of doctor's experience, cause the accuracy of pulmonary nodule information of forecasting that provides relatively low.
The content of the invention
The purpose of the embodiment of the present application is to propose a kind of improved method and apparatus for generating information, come solve with
The technical problem that upper background section is mentioned.
In a first aspect, the embodiment of the present application provides a kind of method for generating information, this method includes:Receive target
The lung images and user record information of user;By the detection model of above-mentioned lung images input training in advance, above-mentioned mesh is obtained
The pulmonary nodule description information of user is marked, wherein, above-mentioned detection model is used to characterize lung images and pulmonary nodule description information
Corresponding relation;Based on obtained pulmonary nodule description information determine above-mentioned targeted customer lung whether nodosity;In response to
Lung's nodosity of above-mentioned targeted customer is determined, by lung images of the above-mentioned targeted customer in different time, pulmonary nodule description
The forecast model of information, user record information input training in advance, obtains the pulmonary nodule information of forecasting of above-mentioned targeted customer, its
In, above-mentioned forecast model is used to characterize lung images, pulmonary nodule description information and note of the above-mentioned targeted customer in different time
Record information and the corresponding relation of pulmonary nodule information of forecasting.
In certain embodiments, above-mentioned detection model is trained obtain in the following manner:Utilize Three dimensional convolution nerve
Network, using lung's sample image as input, using the pulmonary nodule description information of above-mentioned lung's sample image as output, training
Obtain detection model.
In certain embodiments, the above method also includes:In response to determining that the pulmonary nodule of above-mentioned targeted customer describes letter
When the tubercle type that breath includes is part-solid tubercle, in the range of the nodule boundary of the lung images of above-mentioned targeted customer,
Image segmentation is carried out using gray threshold set in advance, the lung images for obtaining above-mentioned targeted customer include the reality of tubercle
Part and non-solid portion;Registration and analysis are carried out in the lung images of different time to above-mentioned targeted customer, it is same so as to obtain
One position tubercle size variation information and tubercle composition transfer information, and by obtained same position tubercle size variation information and
Tubercle composition transfer information is recorded as user record information, wherein, tubercle composition include reality composition and non-reality into
Point.
In certain embodiments, above-mentioned forecast model is trained obtain in the following manner:Utilize multiple linear regression
Method, pulmonary nodule description information, user record information to sample of users tubercle included in the lung images of different time
And the characteristics of image based on deep neural network extraction is analyzed and processed, and obtains forecast model.
In certain embodiments, above-mentioned forecast model is trained obtain in the following manner:Remember net using shot and long term
Network, family is mixed the sample with respectively in lung's sample image of different time, pulmonary nodule description information, user record information as defeated
Enter, training obtains at least one submodel;The output vector of above-mentioned at least one submodel is merged, after being merged
Sum vector;Using full convolutional neural networks, using above-mentioned sum vector as input, the tubercle information for mixing the sample with family is used as output,
Training obtains forecast model.
Second aspect, the embodiment of the present application provide a kind of device for being used to generate information, and the device includes:Receive single
Member, for receiving the lung images and user record information of targeted customer;First generation unit, for above-mentioned lung images are defeated
Enter the detection model of training in advance, obtain the pulmonary nodule description information of above-mentioned targeted customer, wherein, above-mentioned detection model is used for
Characterize the corresponding relation of lung images and pulmonary nodule description information;Determining unit, for based on obtained pulmonary nodule description
Information determine above-mentioned targeted customer lung whether nodosity;Second generation unit, in response to determining above-mentioned targeted customer
Lung's nodosity, by above-mentioned targeted customer different time lung images, pulmonary nodule description information, user record information
The forecast model of training in advance is inputted, obtains the pulmonary nodule information of forecasting of above-mentioned targeted customer, wherein, above-mentioned forecast model is used
In characterizing, above-mentioned targeted customer is pre- in the lung images of different time, pulmonary nodule description information and record information and pulmonary nodule
The corresponding relation of measurement information.
In certain embodiments, above-mentioned detection model is trained obtain in the following manner:Utilize Three dimensional convolution nerve
Network, using lung's sample image as input, using the pulmonary nodule description information of above-mentioned lung's sample image as output, training
Obtain detection model.
In certain embodiments, said apparatus also includes:Cutting unit, for the lung in response to determining above-mentioned targeted customer
When the tubercle type that portion's tubercle description information includes is part-solid tubercle, in the tubercle of the lung images of above-mentioned targeted customer
In bounds, image segmentation is carried out using gray threshold set in advance, the lung images for obtaining above-mentioned targeted customer are wrapped
Solid portion and non-solid portion containing tubercle;Analytic unit, for above-mentioned targeted customer different time lung images
Registration and analysis are carried out, so as to obtaining same position tubercle size variation information and tubercle composition transfer information, and will be obtained
Same position tubercle size variation information and tubercle composition transfer information are recorded as user record information, wherein, tubercle
Composition includes reality composition and non-reality composition.
In certain embodiments, above-mentioned forecast model is trained obtain in the following manner:Utilize multiple linear regression
Method, pulmonary nodule description information, user record information to sample of users tubercle included in the lung images of different time
And the characteristics of image based on deep neural network extraction is analyzed and processed, and obtains forecast model.
In certain embodiments, above-mentioned forecast model is trained obtain in the following manner:Remember net using shot and long term
Network, family is mixed the sample with respectively in lung's sample image of different time, pulmonary nodule description information, user record information as defeated
Enter, training obtains at least one submodel;The output vector of above-mentioned at least one submodel is merged, after being merged
Sum vector;Using full convolutional neural networks, using above-mentioned sum vector as input, the tubercle information for mixing the sample with family is used as output,
Training obtains forecast model.
The third aspect, the embodiment of the present application provide a kind of terminal, and the terminal includes:One or more processors;Storage
Device, for storing one or more programs, when said one or multiple programs are by said one or multiple computing devices,
So that said one or multiple processors realize the method as described in any implementation in first aspect.
Fourth aspect, the embodiment of the present application provide a kind of computer-readable recording medium, are stored thereon with computer journey
Sequence, it is characterised in that the side as described in any implementation in first aspect is realized when the computer program is executed by processor
Method.
The method and apparatus for generating information that the embodiment of the present application provides, the lung images of targeted customer are received first
With user record information, above-mentioned lung images are inputted into detection model afterwards, obtain the pulmonary nodule description of above-mentioned targeted customer
Information, the pulmonary nodule description information for being then based on obtaining determine above-mentioned targeted customer lung whether nodosity, finally respond
In it is determined that lung's nodosity of above-mentioned targeted customer, above-mentioned targeted customer is retouched in the lung images of different time, pulmonary nodule
Information, user record information input prediction model are stated, obtains the pulmonary nodule information of forecasting of above-mentioned targeted customer, so as to effectively sharp
With the lung images and user record information of targeted customer, pulmonary nodule information of forecasting is automatically generated, improves pulmonary nodule
The accuracy of information of forecasting.
Brief description of the drawings
By reading the detailed description made to non-limiting example made with reference to the following drawings, the application's is other
Feature, objects and advantages will become more apparent upon:
Fig. 1 is that the application can apply to exemplary system architecture figure therein;
Fig. 2 is the flow chart for being used to generate one embodiment of the method for information according to the application;
Fig. 3 is the schematic diagram for being used to generate an application scenarios of the method for information according to the application;
Fig. 4 is the structural representation for being used to generate one embodiment of the device of information according to the application;
Fig. 5 is adapted for the structural representation of the computer system of the terminal device for realizing the embodiment of the present application.
Embodiment
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to
Be easy to describe, illustrate only in accompanying drawing to about the related part of invention.
It should be noted that in the case where not conflicting, the feature in embodiment and embodiment in the application can phase
Mutually combination.Describe the application in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 shows the method for being used to generate information that can apply the application or the implementation of the device for generating information
The exemplary system architecture 100 of example.
As shown in figure 1, system architecture 100 can include terminal device 101,102,103, network 104 and server 105.
Network 104 between terminal device 101,102,103 and server 105 provide communication link medium.Network 104 can be with
Including various connection types, such as wired, wireless communication link or fiber optic cables etc..
User can be interacted with using terminal equipment 101,102,103 by network 104 with server 105, to receive or send out
Send message etc..Various client applications, such as web browser applications, doctor can be installed on terminal device 101,102,103
Treat class application etc..
Terminal device 101,102,103 can be the various electronic equipments for having display screen and supporting information to input, and wrap
Include but be not limited to smart mobile phone, tablet personal computer, pocket computer on knee and desktop computer etc..
Server 105 can be to provide the server of various services, such as to being shown on terminal device 101,102,103
Webpage provides the background server supported.Background server can be to data such as the lung images received and user record information
Carry out the processing such as analyzing, and result (such as pulmonary nodule information of forecasting) is fed back into terminal device.
It should be noted that the method for being used to generate information that the embodiment of the present application is provided can pass through terminal device
101st, 102,103 perform, can also be performed by server 105, can also by server 105 and terminal device 101,102,
103 common execution.Correspondingly, it can be arranged in terminal device 101,102,103, can also set for generating the device of information
Be placed in server 105, can be arranged at unit in server 105 and by other units be arranged at terminal device 101,
102nd, in 103.The application is not limited this.
It should be understood that the number of the terminal device, network and server in Fig. 1 is only schematical.According to realizing need
Will, can have any number of terminal device, network and server.
With continued reference to Fig. 2, it illustrates the flow for being used to generate one embodiment of the method for information according to the application
200.This is used for the method for generating information, comprises the following steps:
Step 201, the lung images and user record information of targeted customer are received.
In the present embodiment, electronic equipment (such as the terminal shown in Fig. 1 thereon is run for generating the method for information
Equipment 101,102,103 or server 105) can directly or indirectly receive targeted customer lung images and user record letter
Breath, wherein, above-mentioned targeted customer can refer to Lung neoplasm patient or doubtful Lung neoplasm patient.The lung images of targeted customer can
To refer at least lung images that targeted customer is obtained by CT scan.User record information can refer to and targeted customer
Related various information, for example, the diagnostic message of doctor, pathological information, tubercle size variation information, tubercle composition transfer are believed
Breath, lung cancer gene information, medical record information (for example, the age, sex, history of disease, whether smoking, family's medical history etc.) etc..
Here, an information aggregate can be pre-established for above-mentioned targeted customer, the information aggregate can be used for storing above-mentioned target
The lung images and user record information of user's difference follow up time.
Step 202, lung images are inputted to the detection model of training in advance, obtain the pulmonary nodule description letter of targeted customer
Breath.
In the present embodiment, above-mentioned electronic equipment can input lung images the detection model of training in advance, obtain
The pulmonary nodule description information of targeted customer is stated, wherein, above-mentioned detection model can be used for characterizing lung images and pulmonary nodule
The corresponding relation of description information, wherein, above-mentioned pulmonary nodule description information can include at least one of following:Nodule position, knot
Save size, nodule boundary and tubercle type.Above-mentioned tubercle type can include solid nodules, non-solid nodules, part-solid knot
Section, ground glass nodule, benign protuberance, Malignant Nodules, calcium scoring etc..As an example, above-mentioned detection model can be ground
Personnel are studied carefully based on being pre-established to the statistics of substantial amounts of lung images and pulmonary nodule description information, be stored with multiple lungs
The mapping table of image and the corresponding relation of pulmonary nodule description information.
In some optional implementations of the present embodiment, the above-mentioned method for being used to generate information can also include:It is first
First, when the tubercle type included in response to determining the pulmonary nodule description information of above-mentioned targeted customer is part-solid tubercle,
Above-mentioned electronic equipment can use gray scale threshold set in advance in the range of the nodule boundary of the lung images of above-mentioned targeted customer
Value progress image segmentation, the solid portion and non-solid portion that the lung images for obtaining above-mentioned targeted customer include tubercle,
Here, above-mentioned gray threshold can be obtained by various modes, for example, it may be researcher is based on to largely including portion
The research of the lung images of point solid nodules and pre-establish;Then, above-mentioned electronic equipment can exist to above-mentioned targeted customer
The lung images of different time carry out registration and analysis, become so as to obtain same position tubercle size variation information and tubercle composition
Change information, and obtained same position tubercle size variation information and tubercle composition transfer information are entered as user record information
Row record, wherein, tubercle composition can include reality composition and non-reality composition.As an example, above-mentioned electronic equipment can be adopted
With existing method for registering (for example, based on half-tone information method, transpositions domain and feature based method etc.) to above-mentioned targeted customer
Registration is carried out in the lung images of different time, and contrast decile is carried out in the change of different time to the tubercle of same position
Analysis, so as to obtain same position tubercle size variation information (for example, volume doubling time etc.) and tubercle composition transfer information.
In some optional implementations of the present embodiment, above-mentioned electronic equipment or other be used for train above-mentioned detection
The electronic equipment of model can train to obtain above-mentioned detection model in the following manner:Using Three dimensional convolution neutral net, by lung
Portion's sample image is detected as input using the pulmonary nodule description information of above-mentioned lung's sample image as output, training
Model, wherein, the pulmonary nodule description information of above-mentioned lung's sample image can be including at least one of following:Nodule position, knot
Save size, nodule boundary and tubercle type.Above-mentioned lung's sample image can be from existing lung cancer data storehouse, such as LIDC
(Lung Image Database Consortium, lung images database alliance) database, or the CT that hospital obtains
Scan image, the information such as nodule position, tubercle size, nodule boundary, tubercle type are marked through image department or clinical department expert,
The lung images description information of above-mentioned lung's sample image can be obtained.As an example, above-mentioned electronic equipment can use comprising
The Three dimensional convolution neutral net of 10 convolutional layers (wherein, 5 downward convolutional layers and 5 upward convolutional layers), Three dimensional convolution god
Loss function through network can include following several classes:The tubercle type of pulmonary nodule, nodule position, tubercle radius size, knot
Save border etc..Using lung's sample image as input, in lung's sample image that image department or clinical department expert are marked
Nodule position, tubercle size, nodule boundary, the tubercle type information of tubercle are used as output.During training, lung's sample image is made
For input, the tubercle position for determining the output result of the Three dimensional convolution neutral net and being marked through image department or clinical department expert
Put, the penalty values between tubercle size, nodule boundary, tubercle type, and Three dimensional convolution nerve is optimized and revised based on penalty values
The network parameter of network, so as to the Three dimensional convolution neutral net after being trained, i.e. detection model.
Step 203, based on obtained pulmonary nodule description information determine targeted customer lung whether nodosity.
In the present embodiment, above-mentioned electronic equipment can determine according to the pulmonary nodule description information that step 202 obtains
State targeted customer lung whether nodosity.
Step 204, in response to lung's nodosity of determination targeted customer, targeted customer is schemed in the lung of different time
The forecast model of picture, pulmonary nodule description information, user record information input training in advance, obtains the pulmonary nodule of targeted customer
Information of forecasting.
In the present embodiment, in response to lung's nodosity of the above-mentioned targeted customer of determination, above-mentioned electronic equipment can be by mesh
Mark the prediction mould that user inputs training in advance in the lung images, pulmonary nodule description information, user record information of different time
Type, so as to obtain the pulmonary nodule information of forecasting of above-mentioned targeted customer.Wherein, above-mentioned forecast model is used to characterize above-mentioned target use
Family is in lung images, pulmonary nodule description information and the record information pass corresponding with pulmonary nodule information of forecasting of different time
System.Above-mentioned pulmonary nodule information of forecasting can include the change of tubercle internal structure, the change of tubercle composition, the change of tubercle size
Change etc..After obtaining the pulmonary nodule information of forecasting of above-mentioned targeted customer, above-mentioned electronic equipment can also be according to pulmonary nodule
Information of forecasting exports prompt message, for example, above-mentioned electronic equipment can be based on pulmonary nodule information of forecasting, determines to believe as follows
Breath:Baseline diameter≤15mm (millimeter) tubercle, diameter increases 2mm compared with baseline;Baseline diameter>15mm tubercle, with base
Line is compared to diameter increase more than 15%;Former pure ground glass shadow density increase wherein goes out actuality composition, or former mixed density knot
Reality composition increases in section;Emerging pulmonary nodule;It was found that tubercle in trachea-bronchial epithelial cell wall thickening, luminal stenosis or tube chamber
Person.
Above-mentioned electronic equipment can also provide lung images feature and corresponding report information using attention model, pay attention to
Power model can also produce one " notice scope " and represent to pay close attention to when next output when output is produced
Which of input part, output is then produced according to the region of concern.For example it is to be noted that power model can be by above-mentioned three-dimensional volume
The characteristics of image that each layer of product neutral net extracts to obtain obtains each tubercle class in lung images by coding-decoded model
The weight vectors of type.When the pulmonary nodule for detecting above-mentioned targeted customer is that a certain tubercle type is to extract the tubercle type
Weight vectors, and gained attention by the process of deconvolution and tried hard to or thermodynamic chart, the attention is tried hard to or thermodynamic chart can embody disease
Become corresponding positional information.
In some optional implementations of the present embodiment, above-mentioned electronic equipment or other be used for train above-mentioned prediction mould
The electronic equipment of type can train to obtain above-mentioned forecast model in the following manner:Using multiple linear regression method, to above-mentioned sample
The pulmonary nodule description information of this user tubercle included in the lung images of different time, user record information and it is based on
The characteristics of image of deep neural network extraction is analyzed and processed, and obtains forecast model.Figure based on deep neural network extraction
As feature can refer to that each layer based on the Three dimensional convolution neutral net after above-mentioned training is extracted obtained characteristics of image.As
Example, above-mentioned electronic equipment can be extracted with pulmonary nodule description information, user record information and based on deep neural network
Characteristics of image etc. be independent variable, become with tubercle and turn to dependent variable, establish the temporal characteristics curve of each tubercle, it is big by analyzing
Measure sample of users tubercle curve variation tendency and pulmonary nodule description information, user record information and based on depth nerve
The respective change of the characteristics of image of network extraction, the forecast model that tubercle changes over time is established using multiple linear regression method.
In some optional implementations of the present embodiment, above-mentioned electronic equipment or other be used for train above-mentioned prediction mould
The electronic equipment of type can train to obtain above-mentioned forecast model in the following manner:It is possible, firstly, to utilize shot and long term memory network
(Long-Short Term Memory, LSTM), lung sample image, pulmonary nodule of the family in different time are mixed the sample with respectively
Description information, user record information obtain at least one submodel as input, training;Then, by above-mentioned at least one submodule
The output vector of type merges, the sum vector after being merged;Finally, using full convolutional neural networks, by above-mentioned sum vector
As input, sample of users can be referred to as output, the tubercle information of above-mentioned sample of users by mixing the sample with the tubercle information at family
The various information of Lung neoplasm, for example, the position for the Lung neoplasm that diagnosis obtains, size, border, type etc., training is predicted
Model.As an example, above-mentioned electronic equipment can mix the sample with lung sample image of the family in different time as input, utilize
Shot and long term memory network, training obtain a submodel, can also mix the sample with family in the pathological information of different time as defeated
Enter, using shot and long term memory network, training obtains a submodel, multiple submodels can be obtained according to aforesaid way.Will
To multiple submodels output characteristic vector merge, so as to obtain a sum vector.
With continued reference to Fig. 3, Fig. 3 is the signal for being used to generate the application scenarios of the method for information according to the present embodiment
Figure.In Fig. 3 application scenarios, terminal receives the lung images and user record information of targeted customer first;Afterwards, above-mentioned end
Lung images are inputted detection model by end, obtain the pulmonary nodule description information of targeted customer;Then, above-mentioned terminal is based on obtaining
Pulmonary nodule description information determine lung's nodosity of targeted customer;Finally, above-mentioned terminal by targeted customer in different time
Lung images, pulmonary nodule description information, user record information input training in advance forecast model, obtain targeted customer's
Pulmonary nodule information of forecasting, and prompt message " tubercle diameter≤15mm (millimeter), with base is exported according to pulmonary nodule information of forecasting
Line is compared to diameter increase 2mm;Emerging pulmonary nodule;Need to pay special attention to!", will be as shown in Figure 3.It should be noted that
Fig. 3 is used only for exemplary explanation terminal-pair prompt message output, rather than the restriction to exported content, in use, can be with
Various information are exported according to being actually needed.
The method that above-described embodiment of the application provides is effectively utilized the lung images and user record letter of targeted customer
Breath, automatically generates pulmonary nodule information of forecasting, improves the accuracy of pulmonary nodule information of forecasting, and meets targeted customer
The demand of property.
With further reference to Fig. 4, as the realization to method shown in above-mentioned each figure, it is used to generate letter this application provides one kind
One embodiment of the device of breath, the device embodiment is corresponding with the embodiment of the method shown in Fig. 2, and the device can specifically answer
For in various electronic equipments.
As shown in figure 4, the device 400 for being used to generate information of the present embodiment includes:The generation of receiving unit 401, first is single
Member 402, the generation unit 404 of determining unit 403 and second.Wherein, receiving unit 401 is used for the lung images for receiving targeted customer
With user record information;First generation unit 402 is used for the detection model that above-mentioned lung images are inputted to training in advance, obtains
The pulmonary nodule description information of targeted customer is stated, wherein, above-mentioned detection model is used to characterize lung images and pulmonary nodule description
The corresponding relation of information, wherein, above-mentioned pulmonary nodule description information includes at least one of following:Nodule position, tubercle size, knot
Save border and tubercle type;Determining unit 403 is used to determine above-mentioned targeted customer's based on obtained pulmonary nodule description information
Lung whether nodosity;Second generation unit 404 is used for lung's nodosity in response to determining above-mentioned targeted customer, by above-mentioned mesh
Mark the prediction mould that user inputs training in advance in the lung images, pulmonary nodule description information, user record information of different time
Type, the pulmonary nodule information of forecasting of above-mentioned targeted customer is obtained, wherein, above-mentioned forecast model exists for characterizing above-mentioned targeted customer
Lung images, pulmonary nodule description information and the record information of different time and the corresponding relation of pulmonary nodule information of forecasting.
In the present embodiment, for the receiving unit 401 of the device 400 that generates information, the first generation unit 402, determine
The specific processing of the generation unit 404 of unit 403 and second and its caused technique effect can be respectively with reference to 2 corresponding embodiments of figure
Middle step 201, step 202, the related description of step 203 and step 204, will not be repeated here.
In some optional implementations of the present embodiment, above-mentioned detection model can be trained in the following manner
Arrive:Using Three dimensional convolution neutral net, using lung's sample image as input, by the pulmonary nodule of above-mentioned lung's sample image
Description information obtains detection model as output, training, wherein, the pulmonary nodule description information of above-mentioned lung's sample image includes
At least one of below:Nodule position, tubercle size, nodule boundary and tubercle type.
In some optional implementations of the present embodiment, said apparatus can also include:Cutting unit (does not show in figure
Go out), the tubercle type for including in response to determining the pulmonary nodule description information of above-mentioned targeted customer is part-solid knot
During section, in the range of the nodule boundary of the lung images of above-mentioned targeted customer, image is carried out using gray threshold set in advance
Segmentation, the solid portion and non-solid portion that the lung images for obtaining above-mentioned targeted customer include tubercle;Analytic unit is (in figure
It is not shown), for carrying out registration and analysis in the lung images of different time to above-mentioned targeted customer, so as to obtain same position
Tubercle size variation information and tubercle composition transfer information, and by obtained same position tubercle size variation information and tubercle into
Change information is divided to be recorded as user record information, wherein, tubercle composition includes reality composition and non-reality composition.
In some optional implementations of the present embodiment, above-mentioned forecast model can be trained in the following manner
Arrive:Using multiple linear regression method, to the pulmonary nodule of sample of users tubercle included in the lung images of different time
Description information, user record information and the characteristics of image based on deep neural network extraction are analyzed and processed, and are predicted
Model.
In some optional implementations of the present embodiment, above-mentioned forecast model can be trained in the following manner
Arrive:Using shot and long term memory network, lung's sample image, the pulmonary nodule for mixing the sample with family respectively in different time describe letter
Breath, user record information obtain at least one submodel as input, training;By the output vector of above-mentioned at least one submodel
Merge, the sum vector after being merged;Using full convolutional neural networks, using above-mentioned sum vector as input, mix the sample with
The tubercle information at family obtains forecast model as output, training.
Below with reference to Fig. 5, it illustrates suitable for for realizing the computer system 500 of the terminal device of the embodiment of the present application
Structural representation.Terminal device shown in Fig. 5 is only an example, to the function of the embodiment of the present application and should not use model
Shroud carrys out any restrictions.
As shown in figure 5, computer system 500 includes CPU (CPU, Central Processing Unit)
501, its can according to the program being stored in read-only storage (ROM, Read Only Memory) 502 or from storage part
508 programs being loaded into random access storage device (RAM, Random Access Memory) 503 and perform it is various appropriate
Action and processing.In RAM 503, also it is stored with system 500 and operates required various programs and data.CPU 501、ROM
502 and RAM 503 is connected with each other by bus 504.Input/output (I/O, Input/Output) interface 505 is also connected to
Bus 504.
I/O interfaces 505 are connected to lower component:Importation 506 including keyboard, mouse etc.;Penetrated including such as negative electrode
Spool (CRT, Cathode Ray Tube), liquid crystal display (LCD, Liquid Crystal Display) etc. and loudspeaker
Deng output par, c 507;Storage part 508 including hard disk etc.;And including such as LAN (LAN, Local Area
Network) the communications portion 509 of the NIC of card, modem etc..Communications portion 509 is via such as internet
Network performs communication process.Driver 510 is also according to needing to be connected to I/O interfaces 505.Detachable media 511, such as disk,
CD, magneto-optic disk, semiconductor memory etc., it is arranged on as needed on driver 510, in order to the calculating read from it
Machine program is mounted into storage part 508 as needed.
Especially, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiment of the disclosure includes a kind of computer program product, it includes being carried on computer-readable medium
On computer program, the computer program include be used for execution flow chart shown in method program code.In such reality
To apply in example, the computer program can be downloaded and installed by communications portion 509 from network, and/or from detachable media
511 are mounted.When the computer program is performed by CPU (CPU) 501, perform what is limited in the present processes
Above-mentioned function.It should be noted that the above-mentioned computer-readable medium of the application can be computer-readable signal media or
Computer-readable recording medium either the two any combination.Computer-readable recording medium for example can be --- but
Be not limited to --- electricity, magnetic, optical, electromagnetic, system, device or the device of infrared ray or semiconductor, or it is any more than combination.
The more specifically example of computer-readable recording medium can include but is not limited to:Electrical connection with one or more wires,
Portable computer diskette, hard disk, random access storage device (RAM), read-only storage (ROM), erasable type may be programmed read-only deposit
Reservoir (EPROM or flash memory), optical fiber, portable compact disc read-only storage (CD-ROM), light storage device, magnetic memory
Part or above-mentioned any appropriate combination.In this application, computer-readable recording medium can any be included or store
The tangible medium of program, the program can be commanded the either device use or in connection of execution system, device.And
In the application, computer-readable signal media can include believing in a base band or as the data that a carrier wave part is propagated
Number, wherein carrying computer-readable program code.The data-signal of this propagation can take various forms, including but not
It is limited to electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be computer
Any computer-readable medium beyond readable storage medium storing program for executing, the computer-readable medium can send, propagate or transmit use
In by instruction execution system, device either device use or program in connection.Included on computer-readable medium
Program code any appropriate medium can be used to transmit, include but is not limited to:Wirelessly, electric wire, optical cable, RF etc., Huo Zheshang
Any appropriate combination stated.
Flow chart and block diagram in accompanying drawing, it is illustrated that according to the system of the various embodiments of the application, method and computer journey
Architectural framework in the cards, function and the operation of sequence product.At this point, each square frame in flow chart or block diagram can generation
The part of one module of table, program segment or code, the part of the module, program segment or code include one or more use
In the executable instruction of logic function as defined in realization.It should also be noted that marked at some as in the realization replaced in square frame
The function of note can also be with different from the order marked in accompanying drawing generation.For example, two square frames succeedingly represented are actually
It can perform substantially in parallel, they can also be performed in the opposite order sometimes, and this is depending on involved function.Also to note
Meaning, the combination of each square frame and block diagram in block diagram and/or flow chart and/or the square frame in flow chart can be with holding
Function as defined in row or the special hardware based system of operation are realized, or can use specialized hardware and computer instruction
Combination realize.
Being described in unit involved in the embodiment of the present application can be realized by way of software, can also be by hard
The mode of part is realized.Described unit can also be set within a processor, for example, can be described as:A kind of processor bag
Include receiving unit, the first generation unit, determining unit and the second generation unit.Wherein, the title of these units is in certain situation
Under do not form restriction to the unit in itself, for example, receiving unit is also described as " receiving lung's figure of targeted customer
The unit of picture and user record information ".
As on the other hand, present invention also provides a kind of computer-readable medium, the computer-readable medium can be
Included in device described in above-described embodiment;Can also be individualism, and without be incorporated the device in.Above-mentioned calculating
Machine computer-readable recording medium carries one or more program, when said one or multiple programs are performed by the device so that should
Device:Receive the lung images and user record information of targeted customer;By the detection mould of above-mentioned lung images input training in advance
Type, the pulmonary nodule description information of above-mentioned targeted customer is obtained, wherein, above-mentioned detection model is used to characterize lung images and lung
The corresponding relation of tubercle description information, wherein, above-mentioned pulmonary nodule description information includes at least one of following:Nodule position, knot
Save size, nodule boundary and tubercle type;Determine that the lung of above-mentioned targeted customer is based on obtained pulmonary nodule description information
No nodosity;Lung's nodosity in response to determining above-mentioned targeted customer, above-mentioned targeted customer is schemed in the lung of different time
The forecast model of picture, pulmonary nodule description information, user record information input training in advance, obtains the lung of above-mentioned targeted customer
Tubercle information of forecasting, wherein, above-mentioned forecast model is used to characterizing above-mentioned targeted customer to be tied in the lung images of different time, lung
Save description information and record information and the corresponding relation of pulmonary nodule information of forecasting.
Above description is only the preferred embodiment of the application and the explanation to institute's application technology principle.People in the art
Member should be appreciated that invention scope involved in the application, however it is not limited to the technology that the particular combination of above-mentioned technical characteristic forms
Scheme, while should also cover in the case where not departing from foregoing invention design, carried out by above-mentioned technical characteristic or its equivalent feature
The other technical schemes for being combined and being formed.Such as features described above has similar work(with (but not limited to) disclosed herein
The technical scheme that the technical characteristic of energy is replaced mutually and formed.
Claims (12)
- A kind of 1. method for generating information, it is characterised in that methods described includes:Receive the lung images and user record information of targeted customer;By the detection model of lung images input training in advance, the pulmonary nodule description information of the targeted customer is obtained, Wherein, the detection model is used for the corresponding relation for characterizing lung images and pulmonary nodule description information;Based on obtained pulmonary nodule description information determine the targeted customer lung whether nodosity;Lung's nodosity in response to determining the targeted customer, by lung images of the targeted customer in different time, lung The forecast model of portion's tubercle description information, user record information input training in advance, obtains the pulmonary nodule of the targeted customer Information of forecasting, wherein, the forecast model is used to characterizing the targeted customer to be retouched in the lung images of different time, pulmonary nodule State information and record information and the corresponding relation of pulmonary nodule information of forecasting.
- 2. according to the method for claim 1, it is characterised in that the detection model is that training obtains in the following manner 's:Using Three dimensional convolution neutral net, using lung's sample image as input, by the pulmonary nodule of lung's sample image Description information obtains detection model as output, training.
- 3. according to the method for claim 1, it is characterised in that methods described also includes:When the tubercle type that pulmonary nodule description information in response to determining the targeted customer includes is part-solid tubercle, In the range of the nodule boundary of the lung images of the targeted customer, image segmentation is carried out using gray threshold set in advance, The solid portion and non-solid portion that the lung images for obtaining the targeted customer include tubercle;Registration and analysis are carried out in the lung images of different time to the targeted customer, so as to obtain same position tubercle size Change information and tubercle composition transfer information, and obtained same position tubercle size variation information and tubercle composition transfer are believed Breath is recorded as user record information, wherein, tubercle composition includes reality composition and non-reality composition.
- 4. according to the method for claim 1, it is characterised in that the forecast model is that training obtains in the following manner 's:Using multiple linear regression method, the pulmonary nodule of sample of users tubercle included in the lung images of different time is retouched State information, user record information and the characteristics of image based on deep neural network extraction to be analyzed and processed, obtain predicting mould Type.
- 5. according to the method for claim 1, it is characterised in that the forecast model is that training obtains in the following manner 's:Using shot and long term memory network, lung's sample image, the pulmonary nodule for mixing the sample with family respectively in different time describe letter Breath, user record information obtain at least one submodel as input, training;The output vector of at least one submodel is merged, the sum vector after being merged;Using full convolutional neural networks, using the sum vector as input, the tubercle information at family is mixed the sample with as output, training Obtain forecast model.
- 6. a kind of device for being used to generate information, it is characterised in that described device includes:Receiving unit, for receiving the lung images and user record information of targeted customer;First generation unit, for the lung images to be inputted to the detection model of training in advance, obtain the targeted customer's Pulmonary nodule description information, wherein, the detection model is used to characterize lung images pass corresponding with pulmonary nodule description information System;Determining unit, for determined based on obtained pulmonary nodule description information the targeted customer lung whether nodosity;Second generation unit, for lung's nodosity in response to determining the targeted customer, by the targeted customer in difference The lung images of time, pulmonary nodule description information, the forecast model of user record information input training in advance, obtain the mesh The pulmonary nodule information of forecasting of user is marked, wherein, the forecast model is used to characterize lung of the targeted customer in different time Portion's image, pulmonary nodule description information and record information and the corresponding relation of pulmonary nodule information of forecasting.
- 7. device according to claim 6, it is characterised in that the detection model is that training obtains in the following manner 's:Using Three dimensional convolution neutral net, using lung's sample image as input, by the pulmonary nodule of lung's sample image Description information obtains detection model as output, training.
- 8. device according to claim 6, it is characterised in that described device also includes:Cutting unit, the tubercle type for including in response to determining the pulmonary nodule description information of the targeted customer is portion When dividing solid nodules, in the range of the nodule boundary of the lung images of the targeted customer, gray threshold set in advance is used Carry out image segmentation, the solid portion and non-solid portion that the lung images for obtaining the targeted customer include tubercle;Analytic unit, it is same so as to obtain for carrying out registration and analysis in the lung images of different time to the targeted customer One position tubercle size variation information and tubercle composition transfer information, and by obtained same position tubercle size variation information and Tubercle composition transfer information is recorded as user record information, wherein, tubercle composition include reality composition and non-reality into Point.
- 9. device according to claim 6, it is characterised in that the forecast model is that training obtains in the following manner 's:Using multiple linear regression method, the pulmonary nodule of sample of users tubercle included in the lung images of different time is retouched State information, user record information and the characteristics of image based on deep neural network extraction to be analyzed and processed, obtain predicting mould Type.
- 10. according to the method for claim 6, it is characterised in that the forecast model is that training obtains in the following manner 's:Using shot and long term memory network, lung's sample image, the pulmonary nodule for mixing the sample with family respectively in different time describe letter Breath, user record information obtain at least one submodel as input, training;The output vector of at least one submodel is merged, the sum vector after being merged;Using full convolutional neural networks, using the sum vector as input, the tubercle information at family is mixed the sample with as output, training Obtain forecast model.
- 11. a kind of terminal, including:One or more processors;Storage device, for storing one or more programs,When one or more of programs are by one or more of computing devices so that one or more of processors Realize the method as described in any in claim 1-5.
- 12. a kind of computer-readable recording medium, is stored thereon with computer program, it is characterised in that the computer program quilt The method as described in any in claim 1-5 is realized during computing device.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710914487.8A CN107665736B (en) | 2017-09-30 | 2017-09-30 | Method and apparatus for generating information |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710914487.8A CN107665736B (en) | 2017-09-30 | 2017-09-30 | Method and apparatus for generating information |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107665736A true CN107665736A (en) | 2018-02-06 |
CN107665736B CN107665736B (en) | 2021-05-25 |
Family
ID=61097596
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710914487.8A Active CN107665736B (en) | 2017-09-30 | 2017-09-30 | Method and apparatus for generating information |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107665736B (en) |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108549850A (en) * | 2018-03-27 | 2018-09-18 | 联想(北京)有限公司 | A kind of image-recognizing method and electronic equipment |
CN108899087A (en) * | 2018-06-22 | 2018-11-27 | 中山仰视科技有限公司 | X-ray intelligent diagnosing method based on deep learning |
CN109411080A (en) * | 2018-09-18 | 2019-03-01 | 北京大学 | A kind of man-machine coordination tubercle risk rating system based on ultrasound data |
CN109636819A (en) * | 2018-12-17 | 2019-04-16 | 郑州云海信息技术有限公司 | A kind of Lung neoplasm growth rate prediction meanss based on deep learning |
CN109948671A (en) * | 2019-03-04 | 2019-06-28 | 腾讯科技(深圳)有限公司 | Image classification method, device, storage medium and endoscopic imaging arrangement |
CN110033019A (en) * | 2019-03-06 | 2019-07-19 | 腾讯科技(深圳)有限公司 | Method for detecting abnormality, device and the storage medium of human body |
CN110097580A (en) * | 2019-03-18 | 2019-08-06 | 山东师范大学 | A kind of ultrasound image marker motion tracking method |
CN110232686A (en) * | 2019-06-19 | 2019-09-13 | 东软医疗***股份有限公司 | Acquisition methods, device, CT equipment and the storage medium of Lung neoplasm follow-up image |
CN110858399A (en) * | 2018-08-24 | 2020-03-03 | 西门子医疗有限公司 | Method and apparatus for providing virtual tomographic stroke follow-up detection images |
CN111062876A (en) * | 2018-10-17 | 2020-04-24 | 北京地平线机器人技术研发有限公司 | Method and device for correcting model training and image correction and electronic equipment |
CN111832581A (en) * | 2020-09-21 | 2020-10-27 | 平安科技(深圳)有限公司 | Lung feature recognition method and device, computer equipment and storage medium |
CN111860076A (en) * | 2019-04-30 | 2020-10-30 | 百度时代网络技术(北京)有限公司 | Method and apparatus for generating information |
CN111932482A (en) * | 2020-09-25 | 2020-11-13 | 平安科技(深圳)有限公司 | Method and device for detecting target object in image, electronic equipment and storage medium |
CN112581522A (en) * | 2020-11-30 | 2021-03-30 | 平安科技(深圳)有限公司 | Method and device for detecting position of target object in image, electronic equipment and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE10255525A1 (en) * | 2001-11-29 | 2003-07-17 | Siemens Corp Res Inc | Automated lung node segmentation using dynamic programming and EM-based classification |
CN104700118A (en) * | 2015-03-18 | 2015-06-10 | 中国科学院自动化研究所 | Pulmonary nodule benignity and malignancy predicting method based on convolutional neural networks |
CN106156484A (en) * | 2016-06-08 | 2016-11-23 | 中国科学院自动化研究所 | Disease of brain individuation Forecasting Methodology based on nuclear magnetic resonance image and system |
CN106780460A (en) * | 2016-12-13 | 2017-05-31 | 杭州健培科技有限公司 | A kind of Lung neoplasm automatic checkout system for chest CT image |
CN107016665A (en) * | 2017-02-16 | 2017-08-04 | 浙江大学 | A kind of CT pulmonary nodule detection methods based on depth convolutional neural networks |
CN107103187A (en) * | 2017-04-10 | 2017-08-29 | 四川省肿瘤医院 | The method and system of Lung neoplasm detection classification and management based on deep learning |
-
2017
- 2017-09-30 CN CN201710914487.8A patent/CN107665736B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE10255525A1 (en) * | 2001-11-29 | 2003-07-17 | Siemens Corp Res Inc | Automated lung node segmentation using dynamic programming and EM-based classification |
CN104700118A (en) * | 2015-03-18 | 2015-06-10 | 中国科学院自动化研究所 | Pulmonary nodule benignity and malignancy predicting method based on convolutional neural networks |
CN106156484A (en) * | 2016-06-08 | 2016-11-23 | 中国科学院自动化研究所 | Disease of brain individuation Forecasting Methodology based on nuclear magnetic resonance image and system |
CN106780460A (en) * | 2016-12-13 | 2017-05-31 | 杭州健培科技有限公司 | A kind of Lung neoplasm automatic checkout system for chest CT image |
CN107016665A (en) * | 2017-02-16 | 2017-08-04 | 浙江大学 | A kind of CT pulmonary nodule detection methods based on depth convolutional neural networks |
CN107103187A (en) * | 2017-04-10 | 2017-08-29 | 四川省肿瘤医院 | The method and system of Lung neoplasm detection classification and management based on deep learning |
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108549850A (en) * | 2018-03-27 | 2018-09-18 | 联想(北京)有限公司 | A kind of image-recognizing method and electronic equipment |
CN108549850B (en) * | 2018-03-27 | 2021-07-16 | 联想(北京)有限公司 | Image identification method and electronic equipment |
CN108899087A (en) * | 2018-06-22 | 2018-11-27 | 中山仰视科技有限公司 | X-ray intelligent diagnosing method based on deep learning |
CN110858399A (en) * | 2018-08-24 | 2020-03-03 | 西门子医疗有限公司 | Method and apparatus for providing virtual tomographic stroke follow-up detection images |
CN110858399B (en) * | 2018-08-24 | 2023-11-17 | 西门子医疗有限公司 | Method and apparatus for providing post-examination images of a virtual tomographic stroke |
CN109411080A (en) * | 2018-09-18 | 2019-03-01 | 北京大学 | A kind of man-machine coordination tubercle risk rating system based on ultrasound data |
CN109411080B (en) * | 2018-09-18 | 2022-05-20 | 北京大学 | Man-machine cooperation nodule risk rating system based on ultrasonic data |
CN111062876A (en) * | 2018-10-17 | 2020-04-24 | 北京地平线机器人技术研发有限公司 | Method and device for correcting model training and image correction and electronic equipment |
CN111062876B (en) * | 2018-10-17 | 2023-08-08 | 北京地平线机器人技术研发有限公司 | Method and device for correcting model training and image correction and electronic equipment |
CN109636819A (en) * | 2018-12-17 | 2019-04-16 | 郑州云海信息技术有限公司 | A kind of Lung neoplasm growth rate prediction meanss based on deep learning |
CN109948671A (en) * | 2019-03-04 | 2019-06-28 | 腾讯科技(深圳)有限公司 | Image classification method, device, storage medium and endoscopic imaging arrangement |
CN109948671B (en) * | 2019-03-04 | 2021-11-30 | 腾讯医疗健康(深圳)有限公司 | Image classification method, device, storage medium and endoscopic imaging equipment |
CN110033019A (en) * | 2019-03-06 | 2019-07-19 | 腾讯科技(深圳)有限公司 | Method for detecting abnormality, device and the storage medium of human body |
CN110033019B (en) * | 2019-03-06 | 2021-07-27 | 腾讯科技(深圳)有限公司 | Method and device for detecting abnormality of human body part and storage medium |
CN110097580B (en) * | 2019-03-18 | 2021-10-19 | 山东师范大学 | Ultrasonic image marker motion tracking method |
CN110097580A (en) * | 2019-03-18 | 2019-08-06 | 山东师范大学 | A kind of ultrasound image marker motion tracking method |
CN111860076A (en) * | 2019-04-30 | 2020-10-30 | 百度时代网络技术(北京)有限公司 | Method and apparatus for generating information |
CN110232686A (en) * | 2019-06-19 | 2019-09-13 | 东软医疗***股份有限公司 | Acquisition methods, device, CT equipment and the storage medium of Lung neoplasm follow-up image |
CN111832581B (en) * | 2020-09-21 | 2021-01-29 | 平安科技(深圳)有限公司 | Lung feature recognition method and device, computer equipment and storage medium |
CN111832581A (en) * | 2020-09-21 | 2020-10-27 | 平安科技(深圳)有限公司 | Lung feature recognition method and device, computer equipment and storage medium |
WO2021189912A1 (en) * | 2020-09-25 | 2021-09-30 | 平安科技(深圳)有限公司 | Method and apparatus for detecting target object in image, and electronic device and storage medium |
CN111932482A (en) * | 2020-09-25 | 2020-11-13 | 平安科技(深圳)有限公司 | Method and device for detecting target object in image, electronic equipment and storage medium |
CN112581522A (en) * | 2020-11-30 | 2021-03-30 | 平安科技(深圳)有限公司 | Method and device for detecting position of target object in image, electronic equipment and storage medium |
CN112581522B (en) * | 2020-11-30 | 2024-05-07 | 平安科技(深圳)有限公司 | Method and device for detecting position of target in image, electronic equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN107665736B (en) | 2021-05-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107665736A (en) | Method and apparatus for generating information | |
US20190096060A1 (en) | Method and apparatus for annotating medical image | |
US11937962B2 (en) | Systems and methods for automated and interactive analysis of bone scan images for detection of metastases | |
CN111161275B (en) | Method and device for segmenting target object in medical image and electronic equipment | |
CN107680088A (en) | Method and apparatus for analyzing medical image | |
Narayan et al. | [Retracted] Enhance‐Net: An Approach to Boost the Performance of Deep Learning Model Based on Real‐Time Medical Images | |
Gatta et al. | Towards a modular decision support system for radiomics: A case study on rectal cancer | |
US9031306B2 (en) | Diagnostic and prognostic histopathology system using morphometric indices | |
CN107658023A (en) | Disease forecasting method, apparatus, medium and electronic equipment | |
CN109308469A (en) | Method and apparatus for generating information | |
CN107910060A (en) | Method and apparatus for generating information | |
US20220130525A1 (en) | Artificial intelligence orchestration engine for medical studies | |
KR20130029576A (en) | Method for serving health related information and poi information based on cloud and system therefor | |
He et al. | Deep metric attention learning for skin lesion classification in dermoscopy images | |
CN108171167B (en) | Method and apparatus for exporting image | |
CN113240661A (en) | Deep learning-based lumbar vertebra analysis method, device, equipment and storage medium | |
Bogowicz et al. | Perfusion CT radiomics as potential prognostic biomarker in head and neck squamous cell carcinoma | |
CN114820861A (en) | MR synthetic CT method, equipment and computer readable storage medium based on cycleGAN | |
Kuker et al. | A deep learning-aided automated method for calculating metabolic tumor volume in diffuse large B-cell lymphoma | |
Li et al. | An Integrated Clinical‐MR Radiomics Model to Estimate Survival Time in Patients With Endometrial Cancer | |
CN107708076A (en) | Method and apparatus for pushing access information | |
CN106530386B (en) | Method and system for volume rendering of medical images | |
CN113393445B (en) | Breast cancer image determination method and system | |
CN112037886B (en) | Radiotherapy plan making device, method and storage medium | |
Cadrin-Chênevert | Unleashing the Power of Deep Learning for Breast Cancer Detection through Open Mammography Datasets |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |