CN108510495A - A kind of lung image data processing method based on artificial intelligence, apparatus and system - Google Patents
A kind of lung image data processing method based on artificial intelligence, apparatus and system Download PDFInfo
- Publication number
- CN108510495A CN108510495A CN201810311958.0A CN201810311958A CN108510495A CN 108510495 A CN108510495 A CN 108510495A CN 201810311958 A CN201810311958 A CN 201810311958A CN 108510495 A CN108510495 A CN 108510495A
- Authority
- CN
- China
- Prior art keywords
- lung
- image data
- data
- clouds
- analysis results
- 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
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
- G06T2207/30061—Lung
- G06T2207/30064—Lung nodule
Abstract
The embodiment of the present application provides a kind of lung image data processing method and device based on artificial intelligence, the method includes：Lung's case data are received, lung's case data, which are converted to lung mechanics data, to be stored in image database beyond the clouds；It trains to obtain neural network prediction model using the lung mechanics data of the high in the clouds image data library storage；Lung image data are received, the lung image data are analyzed using the neural network prediction model, obtain the predictive analysis results for including at least Lung neoplasm position and good pernicious prediction probability；Export the predictive analysis results.The embodiment of the present application can effectively improve the accuracy rate and efficiency of lung cancer early screening.
Description
Technical field
The invention relates to field of computer technology, and in particular to a kind of lung image data based on artificial intelligence
Processing method, device and system.
Background technology
Lung cancer is a kind of common malignant tumour, and great threat is constituted to human health and life.It is united according to AUTHORITATIVE DATA
Meter, China's lung cancer neopathy number is more than 730,000 within 2014, and death toll about 610,000, incidence and mortality occupies all kinds of cancers
Disease is the first.It is well known that the lung cancer of more early discovery, therapeutic effect is better.International early stage of lung cancer action plan (I-ELCAP) institute
It was found that lung cancer, I phase lung cancer accounts for 80% or more.The I phase patients to undergo surgery, 10 years survival rates 92%.And most people have found
Middle and advanced stage is in when lung cancer, therapeutic effect is generally bad, and the lung cancer five year survival rate in China is less than 20% at present.Therefore,
How to realize that the early screening of lung cancer becomes an important problem.
In the prior art, there are a kind of cloud systems can collect patient cases, the image that the doctor of each hospital uploads
Data, and store corresponding information.Doctor can carry out artificial diagosis, analysis to image data online, be sieved with carrying out lung cancer early stage
Look into work.
Applicant passes through the study found that cloud system of the existing technology only realizes the store function of image data,
The Intelligent treatment to image data is not provided, it is still desirable to rely on artificial progress diagosis, screening, low, mistake that there are efficiency
Examine the high defect of rate, rate of missed diagnosis.
Invention content
The embodiment of the present application provides a kind of lung image data processing method, device and system based on artificial intelligence,
Aim to solve the problem that the prior art manually carries out the problem that lung cancer early screening efficiency is low, accuracy rate is low.
For this purpose, the embodiment of the present application provides the following technical solutions：
The first aspect of the embodiment of the present application discloses a kind of lung image data processing method based on artificial intelligence, packet
It includes：Lung's case data are received, lung's case data are converted into lung mechanics data storage image data beyond the clouds
In library；It trains to obtain neural network prediction model using the lung mechanics data of the high in the clouds image data library storage；It receives
Lung image data are analyzed the lung image data using the neural network prediction model, are included at least
The predictive analysis results of Lung neoplasm position and good pernicious prediction probability；Export the predictive analysis results.
The second aspect of the embodiment of the present application discloses a kind of lung image data processing system based on artificial intelligence,
Including：High in the clouds image input module, for receiving lung image data and lung's case data；High in the clouds image database is used
It is stored in lung's case data are converted to lung mechanics data；Forecast analysis module, for pre- based on neural network
It surveys model to analyze the lung image data, obtains including at least the pre- of Lung neoplasm position and good pernicious prediction probability
Survey analysis result；Wherein, the neural network prediction model is according to the lung mechanics number of the high in the clouds image data library storage
It is obtained according to training；High in the clouds image output module, for exporting the predictive analysis results.
The third aspect of the embodiment of the present application discloses a kind of for the lung image data processing based on artificial intelligence
Device includes that either more than one program one of them or more than one program is stored in and deposits by memory and one
In reservoir, and it is configured to execute the one or more programs by one or more than one processor to include to be used for
Carry out the following instruction operated：Lung's case data are received, lung's case data are converted into lung mechanics data and are deposited
Storage is beyond the clouds in image database；It trains to obtain nerve net using the lung mechanics data of the high in the clouds image data library storage
Network prediction model；Lung image data are received, the lung image data are divided using the neural network prediction model
Analysis obtains the predictive analysis results for including at least Lung neoplasm position and good pernicious prediction probability；Export the forecast analysis knot
Fruit.
The fourth aspect of the embodiment of the present application discloses a kind of machine readable media, is stored thereon with instruction, when by one
Or multiple processors are when executing so that device executes the lung image data processing based on artificial intelligence as described in relation to the first aspect
Method.
Lung image data processing method provided by the embodiments of the present application based on artificial intelligence, apparatus and system, can be with
Lung's case data are received, lung's case data are converted into lung mechanics data storage image database beyond the clouds
In；It trains to obtain neural network prediction model using the lung mechanics data of the high in the clouds image data library storage；Receive lung
Portion's image data analyzes the lung image data using the neural network prediction model, obtains including at least lung
The predictive analysis results of nodule position and good pernicious prediction probability, and the predictive analysis results can be exported.Due to this Shen
Please embodiment can collect a large amount of lung's case data, be converted into structural data structure model case database, and
It can train to obtain neural network prediction model using model case data, lung image data analyze using it pre-
It surveys, thus obtained prediction result is more accurate, and reduces the cost of artificial diagosis, improves screening efficiency.
Description of the drawings
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments described in application, for those of ordinary skill in the art, without creative efforts,
Other drawings may also be obtained based on these drawings.
Fig. 1 is one application scenarios schematic diagram of the embodiment of the present application；
Fig. 2 is the lung image data processing method flow chart based on artificial intelligence that one embodiment of the application provides；
Fig. 3 is neural network prediction model training process schematic diagram provided by the embodiments of the present application；
Fig. 4 is the lung image data processing system schematic diagram provided by the embodiments of the present application based on artificial intelligence；
Fig. 5 is the application scenarios schematic diagram that another embodiment of the application provides；
Fig. 6 is a kind of in the lung image data processing equipment of artificial intelligence shown according to an exemplary embodiment
Block diagram.
Specific implementation mode
The embodiment of the present application provides a kind of lung image data processing method based on artificial intelligence, apparatus and system,
It can effectively improve the accuracy rate and efficiency of lung cancer early screening.
The term used in the embodiment of the present application is the purpose only merely for description specific embodiment, is not intended to be limiting
The application.In the embodiment of the present application and "an" of singulative used in the attached claims, " described " and "the"
It is also intended to including most forms, unless context clearly shows that other meanings.It is also understood that term used herein
"and/or" refer to and include one or more associated list items purposes any or all may combine.
It is the exemplary application scene of the embodiment of the present application referring to Fig. 1.Wherein, method provided by the embodiments of the present application can
With applied in scene as shown in Figure 1, user such as doctor can upload lung image data and lung's case data, and
It is stored in image cloud.It can be real using the lung image data processing system provided by the embodiments of the present application based on artificial intelligence
Now to the analysis of the lung image data, predictive analysis results are obtained, user can also download the predictive analysis results.When
So, the embodiment of the present application is also applied in other scenes, is not limited herein.
It should be noted that above application scene is merely for convenience of understanding the application and showing, the embodiment party of the application
Formula is unrestricted in this regard.On the contrary, presently filed embodiment can be applied to applicable any scene.
Below in conjunction with attached drawing 2 to attached drawing 3 to lung's shadow based on artificial intelligence shown in the application exemplary embodiment
As data processing method is introduced.
Referring to Fig. 2, the lung image data processing method flow based on artificial intelligence provided for one embodiment of the application
Figure.As shown in Fig. 2, may include：
S201 receives lung's case data, lung's case data is converted to lung mechanics data and are stored in cloud
It holds in image database.
In the embodiment of the present application, lung's case data that user uploads can be received by high in the clouds image input module.
The user can be radiologist, surgeon, oncologist, case history expert, cancer expert etc..Lung's disease
Number of cases includes at least good pernicious diagnostic result according to lung image data and diagnostic data, the diagnostic data is generally comprised.
It in the embodiment of the present application, can be collected by the information of model case, realize the cloud storage of case data and shared typical disease
Example database.Specifically, lung's case data of reception can be converted to lung mechanics data storage image number beyond the clouds
According in library.The lung mechanics data may include：(1) lung image data, such as low dosage computed tomography (English
Literary full name is Computed Tomography, English abbreviation CT) lung's Digital imaging in medicine with communicate that (full name in English is
Digital Imaging and Communications in Medicine, English abbreviation DICOM) image sequence.(2) lung
Nodule position information.The Lung neoplasm location information includes at least Lung neoplasm central point information.(3) the good pernicious diagnosis of Lung neoplasm
As a result.For example, the etiologic diagnosis result of the corresponding lung image data be it is benign either it is pernicious can use True, False or
Other identifier indicates.
S202 trains to obtain neural network prediction mould using the lung mechanics data of the high in the clouds image data library storage
Type.
After structure obtains high in the clouds image database, you can trained with the lung mechanics data of utilization database purchase
To neural network prediction model.As shown in figure 3, illustrating for neural network prediction model training process provided by the embodiments of the present application
Figure.When specific implementation, it can be trained (train) using great amount of samples data as input, pass through propagated forward (Forward
Propagation), back-propagating (Backward Propagation), convolution (Convolution), pond (pooling),
The processes such as logistic regression (Soft-Max) obtain training pattern (Trained model), which includes the convolution number of plies, convolution kernel
Size, the information such as pond parameter can detect the position of doubtful Lung neoplasm and right using the training pattern i.e. prediction model
Testing result is classified (classify), and lesion localization and Lung neoplasm classification results to image are exported.In some realization sides
In formula, after obtaining enough structural datas, it can be utilized on Tensorflow or Caffe even depth learning frameworks
GPU cluster trains depth convolutional neural networks prediction model.Wherein, convolutional neural networks (full name in English Convolutional
Neural Network, English abbreviation CNN) it is a kind of feedforward neural network, its artificial neuron can respond a part
Surrounding cells in coverage area have outstanding performance for large-scale image procossing, generally comprise convolutional layer (alternating
Convolutional layer) and pond layer (pooling layer).The parameter of the model may include neural network structure, volume
Lamination parameter, connects layer parameter etc. at pond layer parameter entirely.Design parameter rule of thumb or can need to be arranged, herein without
It limits.Further, the network ginseng of the neural network prediction model can also according to new structural data, be constantly updated
Number.
S203 is received lung image data, is carried out to the lung image data using the neural network prediction model
Analysis obtains the predictive analysis results for including at least Lung neoplasm position and good pernicious prediction probability.
After receiving lung image data, it can be confirmed whether the image data meets preset condition, it, can be with if meeting
The lung image data are analyzed using neural network prediction model.Preset condition can be whether as thin layer scanning number
According to or other conditions.After training neural network prediction model, you can using by lung image data as the prediction model
Input, export to the qualitative analysis of the lung image data.The qualitative analysis may include Lung neoplasm position
And the classification results of lung image data, the classification results for example can be the classification knots for including good pernicious prediction probability
Fruit.For example, it is benign or malignant tumour that the qualitative analysis, which may include Lung neoplasm center point coordinate, the Lung neoplasm,
Probabilistic information.
S204 exports the predictive analysis results.
In some embodiments, the method can also include：Quantitative analysis is carried out to the lung image data, it is defeated
Go out quantitative analysis results；The quantitative analysis results include Lung neoplasm maximum diameter, most path, volume, maximal density or minimum close
Degree.When specific implementation, image analysis can be carried out to the lung image data using computer-assisted analysis module, output is fixed
Measure analysis result.
In some embodiments, the method can also include：Morphological analysis is carried out to the lung image data,
Determine that sign of lobulation, spicule sign or blood vessel close on relationship.Specifically, can utilize morphological analysis module to lung image data into
Row morphological analysis.For example, the recess of sign of lobulation is likely to occur between two adjacent salient points, is extracted and tied with three-dimensional Lung neoplasm
The outer surface of fruit is input, can calculate leaflet index.For another example, spicule sign can pass through the ladder of calculating Lung neoplasm 2 dimensional region
Degree-normal orthogonal index, to distinguish the smooth of the edge, undercoat thorn and lance thorn.For another example, multiplanar reconstruction, curved surface weight can also be passed through
It builds and shows Lung neoplasm and blood vessel and bronchial morphological relationship with Shaded surface display.
In some embodiments, the download request that can receive predictive analysis results, exports the predictive analysis results.
User can be based on the predictive analysis results such as doctor, make diagnosis and treatment report file quantitative analysis results, described in upload
Diagnosis and treatment report file.The diagnosis and treatment report file may include analysis of cases report, clinic diagnosis report etc..Based on artificial intelligence
Lung image data processing system receive and store the diagnosis and treatment report file obtained based on the predictive analysis results.More into one
Step ground can also generate new lung and tie according to the lung image data, the predictive analysis results, the diagnosis and treatment report file
Structure data store the new lung mechanics data.In this way, can enrich constantly, update case database, can also utilize
The newer data update neural network prediction model of case database.
Correspondingly, the method further includes：More according to the new lung mechanics data of the high in the clouds image data library storage
The network parameter of the new neural network prediction model.The network parameter includes but not limited to convolution layer parameter, pond layer ginseng
Number connects layer parameter etc. entirely.Specifically the neural network prediction model can periodically be iterated update on backstage, using continuous
Newly-increased case data training neural network prediction model updates and optimizes the information such as network parameter, the self study of implementation model,
From raising, the accuracy of neural network prediction model is continuously improved and enhances the stability of forecast analysis.
The corresponding equipment of method provided by the embodiments of the present application is introduced below.
Referring to Fig. 4, the lung image data processing system signal based on artificial intelligence provided for one embodiment of the application
Figure.
A kind of lung image data processing system 400 based on artificial intelligence, including：
High in the clouds image input module 401, for receiving lung image data and lung's case data.Wherein, the cloud
The specific implementation of end image input module 401 is referred to the S201 of embodiment illustrated in fig. 2 and realizes.
High in the clouds image database 402, for lung's case data to be converted to the storage of lung mechanics data.Its
In, the specific implementation of the high in the clouds image database 402 is referred to the S202 of embodiment illustrated in fig. 2 and realizes.
Forecast analysis module 403 is obtained for being analyzed the lung image data in neural network prediction model
Including at least Lung neoplasm position and the predictive analysis results of good pernicious prediction probability；Wherein, the neural network prediction model
It trains to obtain according to the lung mechanics data of the high in the clouds image data library storage.Wherein, the forecast analysis module 403
Specific implementation is referred to the S203 of embodiment illustrated in fig. 2 and realizes.
High in the clouds image output module 404, for exporting the predictive analysis results.Wherein, the high in the clouds image output mould
The specific implementation of block 404 is referred to the S204 of embodiment illustrated in fig. 2 and realizes.
In some embodiments, the system also includes：Computer-assisted analysis module, for the lung image
Data carry out image analysis, export quantitative analysis results；The quantitative analysis results include Lung neoplasm maximum diameter, most path, body
Product, maximal density or minimum density.
In some embodiments, the system also includes：Morphological analysis module, for the lung image data
Morphological analysis is carried out, determines that sign of lobulation, spicule sign or blood vessel close on relationship.
In some embodiments, the high in the clouds image input module is additionally operable to：It receives and is based on the predictive analysis results
Obtained diagnosis and treatment report file；The high in the clouds image database is additionally operable to：According to the predictive analysis results, diagnosis and treatment report text
Part, lung image data generate new lung mechanics data, store the lung mechanics data.
In some embodiments, the system also includes：Neural network prediction model update module, for according to
The network parameter of neural network prediction model described in the new lung mechanics data update of high in the clouds image data library storage.
It is more clearly understood that presently filed embodiment for the ease of those skilled in the art, below with an application scenarios
Specific example the application embodiment is introduced.It should be noted that the specific example is only so that art technology
Personnel more clearly understand the application, but presently filed embodiment is not limited to the specific example.
Referring to Fig. 5, the application scenarios schematic diagram provided for another embodiment of the application.As shown in figure 5, image cloud can be with
Lung neoplasm computer-assisted analysis CAD modules, the forecast analysis model module based on big data analysis, LDCT (Low dose
Computed Tomography, low dosage computed tomography) equipment, doctor, patient realize communication connection.The image
Cloud can specifically include high in the clouds image input module, high in the clouds image input module, high in the clouds image database.Wherein, Lung neoplasm meter
Calculation machine assistant analysis CAD modules can be provided commonly for lung cancer early stage intelligent diagnostics (English with the prediction model based on big data analysis
Literary full name is Lung cancer Early Diagnosis LCED, English abbreviation LCED).
As shown in figure 5, in a typical application scenarios, the high in the clouds image input module for including by image cloud, shadow
As equipment technician, the spiral CT lung images of image department doctor upload DICOM format, while online submission patient assessment's is basic
Information：Whether gender the age, familial inheritance medical history (whether suffer from lung cancer), works and the length of smoking in air pollution high-risk environment
Etc. information.After high in the clouds image input module receives lung CT image, the output of forecast analysis module can be called to include at least lung
The predictive analysis results of nodule position and good pernicious prediction probability.Utilize Lung neoplasm computer-assisted analysis CAD modules, output
Quantitative analysis results including Lung neoplasm maximum diameter, most path, volume, maximal density or minimum density.Utilize form credit
It analyses module and morphological analysis is carried out to the lung image data, determine that sign of lobulation, spicule sign or blood vessel close on relationship etc..This
Outside, the high in the clouds diagosis function of being provided by the system, image department doctor can use tool browsing lung images (turn over layer, put
Contracting, window width and window level etc.), it is reported in conjunction with the automated analysis of image, completes the Artificial Diagnosis of Lung neoplasm and report work(using high in the clouds
Diagnostic imaging can be submitted to report online；Clinician's comprehensive imaging diagnosis reports that Lung neoplasm qualitatively and quantitatively analysis report also may be used
Artificial diagosis is carried out again using high in the clouds diagosis function and intelligent auxiliary tool, completes final clinic diagnosis report.Image is examined
Disconnected report and clinic diagnosis report are eventually sent to uploader by cloud community；The pathological analysis report etc. uploaded by doctor
Final clinic diagnosis report, realizes that the result of Lung neoplasm forecast analysis confirms and be added into case database, and final realize is finished
The construction of the cloud storage and model case database of structure data.Neural network prediction model is periodically being iterated more from the background
Newly, neural network prediction model is trained using constantly newly-increased case data, updates and optimize the information such as network parameter, realize mould
The self study of type is continuously improved the accuracy of neural network prediction model and enhances the stability of forecast analysis from improving.
As shown in figure 5, in a typical application scenarios, each user can pass through lung provided by the embodiments of the present application
Portion's video data processing system realizes that the processing of lung image data, each user are collectively formed with lung image data processing system
High in the clouds community.Wherein, each user may include expert team, image department doctor team, clinician team, image technician group
Team etc..As shown in table 1, different user roles has different system permissions, realizes different responsibility functions.For example, expert
Team can be based on the forecast analysis for the Lung neoplasm position and good pernicious prediction probability that lung image data processing system exports
As a result, generating and uploading diagnosis report.Image department doctor team can upload image data, utilize lung image data processing system
System obtains the predictive analysis results etc. of Lung neoplasm position and good pernicious prediction probability.By Fig. 5 and table 1 it is found that each user can
To realize data transmission, processing and interaction by lung image data processing system provided by the embodiments of the present application.The application is real
The lung image data processing system for applying example offer provides intercommunion platform for each user, improves lung image data processing
Efficiency.
1 high in the clouds community role's explanation of table
In conclusion the embodiment of the present application can reach following advantageous effect：
1, intellectual analysis, prediction are realized.It by the collection of big data, storage and shares, constructs model case database,
And constantly increase.Artificial intelligence neural networks prediction model based on deep learning can realize self based on this big data platform
Training and self promotion realize the automatic intelligent analysis and prediction of the early screening of diagnosing, save a large amount of manpower objects
Power.
2, intercommunion platform is provided.By the system, image department doctor team can share exchange diagosis experience, complete shadow
As diagnosis report；The qualitatively and quantitatively analysis report that clinician can provide according to the diagnosis report of image doctor, system, meeting
It examines and completes diagnosis and treatment scheme and report.Simultaneously image department doctor and clinician, such as Neurology doctor, can it is same specially
Exchange medical diagnosis on disease experience is shared in the communities Ke Hua, lung image data processing cloud community intersects learning and communication between providing subject
Facility environment and tool.
3, diagnosis efficiency is improved, misdiagnosis rate is reduced.The quantitative, the qualitative analysis provided by this system, image department doctor
Life can check that Lung neoplasm detection is reported with quantitative analysis, reduce wrong diagnosis and escape rate；Pass through Lung neoplasm CAD modules, forecast analysis
The detection efficiency of Lung neoplasm can be improved in the auxiliary of module, reduces image department doctor's scoring time；Clinician can check quantitative
Analysis report and qualitative analysis report, improve the working efficiency of pernicious diagnosis good to Lung neoplasm, greatly reduce rate of missed diagnosis.
It is the device for the lung image data processing based on artificial intelligence that another embodiment of the application provides referring to Fig. 6
Block diagram.Including：At least one processor 601 (such as CPU), memory 602 and at least one communication bus 603, for real
Connection communication between these existing devices.Processor 601 is used to execute the executable module stored in memory 602, such as counts
Calculation machine program.Memory 602 may include high-speed random access memory (RAM：Random Access Memory), it is also possible to
Further include non-labile memory (non-volatile memory), for example, at least a magnetic disk storage.One or one
A procedure above is stored in memory, and be configured to by one or more than one processor 601 execute it is one or
More than one program of person includes the instruction for being operated below：Lung's case data are received, by lung's case data
Lung mechanics data are converted to store in image database beyond the clouds；It is tied using the lung of the high in the clouds image data library storage
Structure data train to obtain neural network prediction model；Lung image data are received, the neural network prediction model pair is utilized
The lung image data are analyzed, and the forecast analysis knot for including at least Lung neoplasm position and good pernicious prediction probability is obtained
Fruit；Export the predictive analysis results.
In some embodiments, the specific execution the one or more programs that are additionally operable to of processor 601 include
Instruction for being operated below：Quantitative analysis is carried out to the lung image data, exports quantitative analysis results；It is described fixed
It includes Lung neoplasm maximum diameter, most path, volume, maximal density or minimum density to measure analysis result.
In some embodiments, the specific execution the one or more programs that are additionally operable to of processor 601 include
Instruction for being operated below：Morphological analysis is carried out to the lung image data, determines sign of lobulation, spicule sign or blood
Pipe closes on relationship.
In some embodiments, the specific execution the one or more programs that are additionally operable to of processor 601 include
Instruction for being operated below：Receive the diagnosis and treatment report file obtained based on the predictive analysis results；According to the lung
Portion's image data, the predictive analysis results, the diagnosis and treatment report file generate new lung mechanics data, store the new lung
Portion's structural data.
In some embodiments, the specific execution the one or more programs that are additionally operable to of processor 601 include
Instruction for being operated below：It is refreshing described in new lung mechanics data update according to the high in the clouds image data library storage
Network parameter through Network Prediction Model.
In the exemplary embodiment, it includes the non-transitorycomputer readable storage medium instructed, example to additionally provide a kind of
Such as include the memory of instruction, above-metioned instruction can be executed by the processor of device to complete the above method.For example, described non-provisional
Property computer readable storage medium can be that ROM, random access memory (RAM), CD-ROM, tape, floppy disk and light data are deposited
Store up equipment etc..
A kind of machine readable media, such as the machine readable media can be non-transitorycomputer readable storage medium,
When the instruction in the medium is executed by the processor of device (terminal or server) so that device is able to carry out a kind of base
In the lung image data processing method of artificial intelligence, the method includes：Lung's case data are received, by lung's case
Data are converted to lung mechanics data and store in image database beyond the clouds；Utilize the lung of the high in the clouds image data library storage
Portion's structural data trains to obtain neural network prediction model；Lung image data are received, the neural network prediction mould is utilized
Type analyzes the lung image data, obtains the prediction point for including at least Lung neoplasm position and good pernicious prediction probability
Analyse result；Export the predictive analysis results.
Wherein, the setting of the application device each unit or module is referred to Fig. 2 and is realized to method shown in Fig. 3,
This is not repeated.
Those skilled in the art after considering the specification and implementing the invention disclosed here, will readily occur to its of the application
Its embodiment.This application is intended to cover any variations, uses, or adaptations of the application, these modifications, purposes or
Person's adaptive change follows the general principle of the application and includes the undocumented common knowledge in the art of the disclosure
Or conventional techniques.The description and examples are only to be considered as illustrative, and the true scope and spirit of the application are by following
Claim is pointed out.
It should be understood that the application is not limited to the precision architecture for being described above and being shown in the accompanying drawings, and
And various modifications and changes may be made without departing from the scope thereof.Scope of the present application is only limited by the accompanying claims
It should be noted that herein, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to
Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also include other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.The application can be by calculating
Described in the general context for the computer executable instructions that machine executes, such as program module.Usually, program module includes holding
The routine of row particular task or realization particular abstract data type, program, object, component, data structure etc..It can also divide
The application is put into practice in cloth computing environment, in these distributed computing environments, by connected long-range by communication network
Processing equipment executes task.In a distributed computing environment, program module can be located at the local including storage device
In remote computer storage medium.
Each embodiment in this specification is described in a progressive manner, identical similar portion between each embodiment
Point just to refer each other, and each embodiment focuses on the differences from other embodiments.Especially for device reality
For applying example, since it is substantially similar to the method embodiment, so describing fairly simple, related place is referring to embodiment of the method
Part explanation.The apparatus embodiments described above are merely exemplary, wherein described be used as separating component explanation
Unit may or may not be physically separated, the component shown as unit may or may not be
Physical unit, you can be located at a place, or may be distributed over multiple network units.It can be according to the actual needs
Some or all of module therein is selected to achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying
In the case of creative work, you can to understand and implement.The above is only the specific implementation mode of the application, should be referred to
Go out, for those skilled in the art, under the premise of not departing from the application principle, can also make several
Improvements and modifications, these improvements and modifications also should be regarded as the protection domain of the application.
Claims (12)
1. a kind of lung image data processing method based on artificial intelligence, which is characterized in that including：
Lung's case data are received, lung's case data are converted into lung mechanics data storage image data beyond the clouds
In library；
It trains to obtain neural network prediction model using the lung mechanics data of the high in the clouds image data library storage；
Lung image data are received, the lung image data are analyzed using the neural network prediction model, are obtained
Including at least Lung neoplasm position and the predictive analysis results of good pernicious prediction probability；
Export the predictive analysis results.
2. according to the method described in claim 1, it is characterized in that, the method further includes：
Quantitative analysis is carried out to the lung image data, exports quantitative analysis results；The quantitative analysis results include lung knot
Save maximum diameter, most path, volume, maximal density or minimum density.
3. according to the method described in claim 1, it is characterized in that, the method further includes：
Morphological analysis is carried out to the lung image data, determines that sign of lobulation, spicule sign or blood vessel close on relationship.
4. according to the method described in claim 1, it is characterized in that, the method further includes：
Receive the diagnosis and treatment report file obtained based on the predictive analysis results；
New lung mechanics number is generated according to the lung image data, the predictive analysis results, the diagnosis and treatment report file
According to the storage new lung mechanics data.
5. according to the method described in claim 4, it is characterized in that, the method further includes：
The net of neural network prediction model described in new lung mechanics data update according to the high in the clouds image data library storage
Network parameter.
6. a kind of lung image data processing system based on artificial intelligence, which is characterized in that including：
High in the clouds image input module, for receiving lung image data and lung's case data；
High in the clouds image database, for lung's case data to be converted to the storage of lung mechanics data；
Forecast analysis module is analyzed the lung image data for being based on neural network prediction model, is obtained at least
Predictive analysis results including Lung neoplasm position and good pernicious prediction probability；Wherein, the neural network prediction model according to
The lung mechanics data of the high in the clouds image data library storage train to obtain；
High in the clouds image output module, for exporting the predictive analysis results.
7. system according to claim 6, which is characterized in that further include：
Computer-assisted analysis module exports quantitative analysis results for carrying out image analysis to the lung image data；Institute
It includes Lung neoplasm maximum diameter, most path, volume, maximal density or minimum density to state quantitative analysis results.
8. system according to claim 6, which is characterized in that further include：
Morphological analysis module determines sign of lobulation, spicule sign or blood for carrying out morphological analysis to the lung image data
Pipe closes on relationship.
9. system according to claim 6, which is characterized in that the high in the clouds image input module is additionally operable to：
Receive the diagnosis and treatment report file obtained based on the predictive analysis results；
The high in the clouds image database is additionally operable to：
New lung mechanics data are generated according to the predictive analysis results, diagnosis and treatment report file, lung image data, store institute
State lung mechanics data.
10. system according to claim 9, which is characterized in that further include：
Neural network prediction model update module, for the new lung mechanics data according to the high in the clouds image data library storage
Update the network parameter of the neural network prediction model.
11. a kind of device for the lung image data processing based on artificial intelligence, which is characterized in that include memory,
And one either more than one program one of them or more than one program be stored in memory, and be configured to
It includes the finger for being operated below to execute the one or more programs by one or more than one processor
It enables：
Lung's case data are received, lung's case data are converted into lung mechanics data storage image data beyond the clouds
In library；
It trains to obtain neural network prediction model using the lung mechanics data of the high in the clouds image data library storage；
Lung image data are received, the lung image data are analyzed using the neural network prediction model, are obtained
Including at least Lung neoplasm position and the predictive analysis results of good pernicious prediction probability；
Export the predictive analysis results.
12. a kind of machine readable media is stored thereon with instruction, when executed by one or more processors so that device is held
Lung image data processing method based on artificial intelligence of the row as described in one or more in Claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810311958.0A CN108510495A (en) | 2018-04-09 | 2018-04-09 | A kind of lung image data processing method based on artificial intelligence, apparatus and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810311958.0A CN108510495A (en) | 2018-04-09 | 2018-04-09 | A kind of lung image data processing method based on artificial intelligence, apparatus and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108510495A true CN108510495A (en) | 2018-09-07 |
Family
ID=63380904
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810311958.0A Pending CN108510495A (en) | 2018-04-09 | 2018-04-09 | A kind of lung image data processing method based on artificial intelligence, apparatus and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108510495A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109410170A (en) * | 2018-09-14 | 2019-03-01 | 沈阳东软医疗系统有限公司 | Image processing method, device and equipment |
| CN109509541A (en) * | 2018-10-26 | 2019-03-22 | 广州金域医学检验中心有限公司 | Method and device is signed and issued in report based on medical image |
| CN110097969A (en) * | 2019-05-10 | 2019-08-06 | 安徽科大讯飞医疗信息技术有限公司 | A kind of analysis method of diagnosis report, device and equipment |
| CN110223743A (en) * | 2019-06-14 | 2019-09-10 | 吉林大学第一医院 | A kind of structuring processing method and system for pulmonary cancer diagnosis record |
| CN110335256A (en) * | 2019-06-18 | 2019-10-15 | 广州智睿医疗科技有限公司 | A kind of pathology aided diagnosis method |
| CN111223579A (en) * | 2019-12-16 | 2020-06-02 | 郑州大学第一附属医院 | Remote medical image enhancement system and method based on artificial intelligence |
| CN112530580A (en) * | 2020-12-03 | 2021-03-19 | 上海联影智能医疗科技有限公司 | Medical image picture processing method and computer readable storage medium |
| CN112530578A (en) * | 2020-12-02 | 2021-03-19 | 中国科学院大学宁波华美医院 | Viral pneumonia intelligent diagnosis system based on multi-mode information fusion |
| CN112820399A (en) * | 2021-01-26 | 2021-05-18 | 北京小白世纪网络科技有限公司 | Method and device for automatically diagnosing benign and malignant thyroid nodules |
| CN113130067A (en) * | 2021-04-01 | 2021-07-16 | 上海市第一人民医院 | Intelligent reminding method for ultrasonic examination based on artificial intelligence |
| US11094411B2 (en) | 2018-10-26 | 2021-08-17 | Guangzhou Kingmed Center for Clinical Laboratory Co., Ltd. | Methods and devices for pathologically labeling medical images, methods and devices for issuing reports based on medical images, and computer-readable storage media |
| CN113763353A (en) * | 2021-09-06 | 2021-12-07 | 杭州类脑科技有限公司 | Lung ultrasonic image detection system |
| CN113903452A (en) * | 2021-09-18 | 2022-01-07 | 宁波大学医学院附属医院 | Auxiliary diagnosis system for pulmonary nodules of doctors |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040010481A1 (en) * | 2001-12-07 | 2004-01-15 | Whitehead Institute For Biomedical Research | Time-dependent outcome prediction using neural networks |
| CN105139390A (en) * | 2015-08-14 | 2015-12-09 | 四川大学 | Image processing method for detecting pulmonary tuberculosis focus in chest X-ray DR film |
| CN106372390A (en) * | 2016-08-25 | 2017-02-01 | 姹ゅ钩 | Deep convolutional neural network-based lung cancer preventing self-service health cloud service system |
| CN106599549A (en) * | 2016-11-25 | 2017-04-26 | 上海联影医疗科技有限公司 | Computer-aided diagnosis system and method, and medical system |
| CN107451609A (en) * | 2017-07-24 | 2017-12-08 | 上海交通大学 | Lung neoplasm image identification system based on depth convolutional neural networks |
| CN107680678A (en) * | 2017-10-18 | 2018-02-09 | 北京航空航天大学 | Based on multiple dimensioned convolutional neural networks Thyroid ultrasound image tubercle auto-check system |
-
2018
- 2018-04-09 CN CN201810311958.0A patent/CN108510495A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040010481A1 (en) * | 2001-12-07 | 2004-01-15 | Whitehead Institute For Biomedical Research | Time-dependent outcome prediction using neural networks |
| CN105139390A (en) * | 2015-08-14 | 2015-12-09 | 四川大学 | Image processing method for detecting pulmonary tuberculosis focus in chest X-ray DR film |
| CN106372390A (en) * | 2016-08-25 | 2017-02-01 | 姹ゅ钩 | Deep convolutional neural network-based lung cancer preventing self-service health cloud service system |
| CN106599549A (en) * | 2016-11-25 | 2017-04-26 | 上海联影医疗科技有限公司 | Computer-aided diagnosis system and method, and medical system |
| CN107451609A (en) * | 2017-07-24 | 2017-12-08 | 上海交通大学 | Lung neoplasm image identification system based on depth convolutional neural networks |
| CN107680678A (en) * | 2017-10-18 | 2018-02-09 | 北京航空航天大学 | Based on multiple dimensioned convolutional neural networks Thyroid ultrasound image tubercle auto-check system |
Non-Patent Citations (2)
| Title |
|---|
| 张鸿: "《基于人工智能的多媒体数据挖掘和应用实例》", 31 January 2018 * |
| 罗风莲 等: "最大径 2 cm 以下肺小结节诊断中超高分辨率薄层 CT 的应用", 《实用医学杂志》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109410170A (en) * | 2018-09-14 | 2019-03-01 | 沈阳东软医疗系统有限公司 | Image processing method, device and equipment |
| CN109509541B (en) * | 2018-10-26 | 2020-07-07 | 广州金域医学检验中心有限公司 | Report issuing method and device based on medical image |
| CN109509541A (en) * | 2018-10-26 | 2019-03-22 | 广州金域医学检验中心有限公司 | Method and device is signed and issued in report based on medical image |
| US11094411B2 (en) | 2018-10-26 | 2021-08-17 | Guangzhou Kingmed Center for Clinical Laboratory Co., Ltd. | Methods and devices for pathologically labeling medical images, methods and devices for issuing reports based on medical images, and computer-readable storage media |
| CN110097969A (en) * | 2019-05-10 | 2019-08-06 | 安徽科大讯飞医疗信息技术有限公司 | A kind of analysis method of diagnosis report, device and equipment |
| CN110223743A (en) * | 2019-06-14 | 2019-09-10 | 吉林大学第一医院 | A kind of structuring processing method and system for pulmonary cancer diagnosis record |
| CN110335256A (en) * | 2019-06-18 | 2019-10-15 | 广州智睿医疗科技有限公司 | A kind of pathology aided diagnosis method |
| WO2020253886A1 (en) * | 2019-06-18 | 2020-12-24 | 广州智睿医疗科技有限公司 | Supplementary diagnosis method for pathology |
| CN111223579A (en) * | 2019-12-16 | 2020-06-02 | 郑州大学第一附属医院 | Remote medical image enhancement system and method based on artificial intelligence |
| CN112530578A (en) * | 2020-12-02 | 2021-03-19 | 中国科学院大学宁波华美医院 | Viral pneumonia intelligent diagnosis system based on multi-mode information fusion |
| CN112530580A (en) * | 2020-12-03 | 2021-03-19 | 上海联影智能医疗科技有限公司 | Medical image picture processing method and computer readable storage medium |
| CN112820399A (en) * | 2021-01-26 | 2021-05-18 | 北京小白世纪网络科技有限公司 | Method and device for automatically diagnosing benign and malignant thyroid nodules |
| CN113130067A (en) * | 2021-04-01 | 2021-07-16 | 上海市第一人民医院 | Intelligent reminding method for ultrasonic examination based on artificial intelligence |
| CN113130067B (en) * | 2021-04-01 | 2022-11-08 | 上海市第一人民医院 | Intelligent reminding method for ultrasonic examination based on artificial intelligence |
| CN113763353A (en) * | 2021-09-06 | 2021-12-07 | 杭州类脑科技有限公司 | Lung ultrasonic image detection system |
| CN113903452A (en) * | 2021-09-18 | 2022-01-07 | 宁波大学医学院附属医院 | Auxiliary diagnosis system for pulmonary nodules of doctors |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108510495A (en) | A kind of lung image data processing method based on artificial intelligence, apparatus and system | |
| Xu et al. | Deep learning predicts lung cancer treatment response from serial medical imaging | |
| Kaur et al. | A systematic review on metaheuristic optimization techniques for feature selections in disease diagnosis: open issues and challenges | |
| Chakraborty et al. | SuFMoFPA: A superpixel and meta-heuristic based fuzzy image segmentation approach to explicate COVID-19 radiological images | |
| Kasinathan et al. | Cloud-based lung tumor detection and stage classification using deep learning techniques | |
| CN107622492A (en) | Lung splits dividing method and system | |
| CN108615237A (en) | A kind of method for processing lung images and image processing equipment | |
| WO2021059607A1 (en) | Machine learning system and method, integration server, information processing device, program, and inference model generation method | |
| Osman et al. | SOM-LWL method for identification of COVID-19 on chest X-rays | |
| CN106599549A (en) | Computer-aided diagnosis system and method, and medical system | |
| JP7346553B2 (en) | Determining the growth rate of objects in a 3D dataset using deep learning | |
| CN112614133B (en) | Three-dimensional pulmonary nodule detection model training method and device without anchor point frame | |
| Dodia et al. | Recent advancements in deep learning based lung cancer detection: A systematic review | |
| Huang et al. | Optimal deep learning neural network using ISSA for diagnosing the oral cancer | |
| CN109003651A (en) | Lesion report preparing apparatus and method | |
| Suman et al. | Attention based CNN-LSTM network for pulmonary embolism prediction on chest computed tomography pulmonary angiograms | |
| Zhang | Computer-aided diagnosis for pneumoconiosis staging based on multi-scale feature mapping | |
| Wang et al. | Deep learning for automatic identification of nodule morphology features and prediction of lung cancer | |
| Li et al. | An improved random forests approach for interactive lobar segmentation on emphysema detection | |
| Kaluva et al. | An automated workflow for lung nodule follow-up recommendation using deep learning | |
| Lian et al. | Imaging-based deep graph neural networks for survival analysis in early stage lung cancer using ct: A multicenter study | |
| Yang et al. | 3D multi‐view squeeze‐and‐excitation convolutional neural network for lung nodule classification | |
| JP7317136B2 (en) | Machine learning system and method, integrated server, information processing device, program, and inference model creation method | |
| Bong et al. | Adaptive multi-objective archive-based hybrid scatter search for segmentation in lung computed tomography imaging | |
| Liu et al. | Ratio of injured lung volume fraction in prognosis evaluation of acute PQ poisoning |
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 | ||
| CB02 | Change of applicant information |
Address after: 110179 No. 177-1 Innovation Road, Hunnan District, Shenyang City, Liaoning Province Applicant after: DongSoft Medical System Co., Ltd. Address before: 110179 No. 177-1 Innovation Road, Hunnan District, Shenyang City, Liaoning Province Applicant before: Dongruan Medical Systems Co., Ltd., Shenyang |
|
| CB02 | Change of applicant information | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180907 |
|
| RJ01 | Rejection of invention patent application after publication |