CN110400298A - Detection method, device, equipment and the medium of heart clinical indices - Google Patents
Detection method, device, equipment and the medium of heart clinical indices Download PDFInfo
- Publication number
- CN110400298A CN110400298A CN201910667589.3A CN201910667589A CN110400298A CN 110400298 A CN110400298 A CN 110400298A CN 201910667589 A CN201910667589 A CN 201910667589A CN 110400298 A CN110400298 A CN 110400298A
- Authority
- CN
- China
- Prior art keywords
- cardiac
- parameter
- network
- image
- images
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 25
- 230000000747 cardiac effect Effects 0.000 claims abstract description 172
- 238000013528 artificial neural network Methods 0.000 claims abstract description 44
- 238000012549 training Methods 0.000 claims description 79
- 238000012360 testing method Methods 0.000 claims description 76
- 238000000034 method Methods 0.000 claims description 39
- 238000013507 mapping Methods 0.000 claims description 21
- 230000006870 function Effects 0.000 claims description 18
- 238000003860 storage Methods 0.000 claims description 17
- 230000004913 activation Effects 0.000 claims description 12
- 210000005240 left ventricle Anatomy 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 7
- 230000002107 myocardial effect Effects 0.000 claims description 6
- 210000003205 muscle Anatomy 0.000 claims description 5
- 230000002861 ventricular Effects 0.000 claims description 4
- 210000005242 cardiac chamber Anatomy 0.000 claims description 3
- 238000012512 characterization method Methods 0.000 abstract description 3
- 238000012546 transfer Methods 0.000 abstract description 3
- 238000009826 distribution Methods 0.000 description 12
- 230000008569 process Effects 0.000 description 12
- 238000010586 diagram Methods 0.000 description 9
- 238000012545 processing Methods 0.000 description 8
- 230000005291 magnetic effect Effects 0.000 description 6
- 230000007246 mechanism Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000013527 convolutional neural network Methods 0.000 description 5
- 238000013461 design Methods 0.000 description 5
- 238000003709 image segmentation Methods 0.000 description 5
- 238000003384 imaging method Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 230000000306 recurrent effect Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 3
- 230000009977 dual effect Effects 0.000 description 3
- 238000003475 lamination Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000003042 antagnostic effect Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000002591 computed tomography Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 208000019622 heart disease Diseases 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 210000004165 myocardium Anatomy 0.000 description 2
- 210000005036 nerve Anatomy 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 208000021908 Myocardial disease Diseases 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 210000004204 blood vessel Anatomy 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000001427 coherent effect Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 210000004351 coronary vessel Anatomy 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000004064 dysfunction Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 230000004217 heart function Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000010412 perfusion Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000013102 re-test Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 210000004872 soft tissue Anatomy 0.000 description 1
- 208000019553 vascular disease Diseases 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
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
- G06T7/0014—Biomedical image inspection using an image reference approach
-
- 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
- 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
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- 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
- G06T2207/30048—Heart; Cardiac
Landscapes
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Pathology (AREA)
- Databases & Information Systems (AREA)
- Radiology & Medical Imaging (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Image Analysis (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
- Image Processing (AREA)
Abstract
This application provides detection method, device, equipment and the media of a kind of heart clinical indices, comprising: using the self-learning capability of artificial neural network, establishes the corresponding relationship between cardiac MR images and specified parameter；Wherein, specified parameter includes cardiac CT image, for the heart clinical indices of cardiac MR images, and for the heart clinical indices of cardiac CT image；Obtain the current cardiac MR image of patient；By corresponding relationship, currently assigned parameter corresponding with current cardiac MR image is determined；Specifically, it is determined that currently assigned parameter corresponding with current cardiac MR image, comprising: by specified parameter corresponding to cardiac MR images identical with current cardiac MR image in corresponding relationship, be determined as currently assigned parameter.It can be used for the assessment of polymorphic type heart clinical indices under MR image mode different from CT's.The complex relationship between characterization polymorphic type heart clinical indices is excavated, task dependencies is obtained and realizes it in the transfer of different image modes.
Description
Technical field
This application involves medical sciences, the especially detection method of heart clinical indices, device, equipment and medium.
Background technique
CT (Computed Tomography, CT scan) and MR (Magnetic Resonance, core
Magnetic resonance) to the myocardial perfusion imaging applied at present, the detection methods such as electron emitter layer scanning have carried out great perfect.
For MR with its unique Noninvasive to the chambers of the heart and big blood vessel imaging ability, the one kind for becoming clinical cardiology is important
Imaging mode.MR is widely used in assessing cardiac function and structure various aspects, and image quality and acquisition time aspect are not
It is disconnected to make substantial progress；
And CT density resolution is high, has good specificity to Lesions such as soft tissues.And it is instantaneous clear
The improvement of degree and volume scan coverage speed promotes its development, the noninvasive solution as assessment coronary artery and cardiac structure function
Imageological examination is cutd open to have a extensive future.
However, requiring cardiac imaging to check that entering derived techniques combines the further investigation of the mechanism of heart disease
Using being more advantageous to reflecting myocardium and vascular disease states in this way, significantly improve the energy for the analysis that doctor changes dysfunction
Power.
But the complex relationship between polymorphic type heart clinical indices allows study task dependencies appropriate become difficult.
These indexs have different dimensions simultaneously, so that general learning method is difficult to characterize its general character and difference.Certain indexs are also
It is influenced by different factors and has notable difference, such as: there are much relations in the myocardial wall thickness of different zones and myocardial segment direction.
In addition, the difference between different image modes also proposes across the image mode estimation of polymorphic type heart clinical indices
Challenge, such as: MR and CT shows significant difference.
Summary of the invention
In view of described problem, the application is proposed in order to provide overcoming described problem or at least being partially solved described ask
Detection method, device, equipment and the medium of the heart clinical indices of topic, comprising:
A kind of detection method of heart clinical indices, comprising:
Using the self-learning capability of artificial neural network, the corresponding relationship between cardiac MR images and specified parameter is established；
Wherein, the specified parameter includes cardiac CT image, is schemed for the heart clinical indices of cardiac MR images, and for heart CT
The heart clinical indices of picture；
Obtain the current cardiac MR image of patient；
By the corresponding relationship, currently assigned parameter corresponding with the current cardiac MR image is determined；Specifically, really
Fixed currently assigned parameter corresponding with the current cardiac MR image, comprising: by the corresponding relationship with the current cardiac
Specified parameter corresponding to the identical cardiac MR images of MR image is determined as the currently assigned parameter.
Further,
The heart clinical indices include left ventricular epicardium profile, left ventricle inner membrance profile, left compartment muscle profile, left ventricle
Outer film location, film location in left ventricle, left compartment muscle position, local chambers of the heart wall thickness WT, heart chamber volume Dim, and, chambers of the heart wall
At least one of with myocardial area Area.
Further,
Establish the corresponding relationship between cardiac MR images and specified parameter, comprising:
Obtain the sample data of the corresponding relationship for establishing between the specified parameter and the cardiac MR images；
The characteristic and its rule for analyzing the cardiac MR images determine the artificial mind according to the characteristic and its rule
Network structure and its network parameter through network；
Using the sample data, the network structure and the network parameter are trained and are tested, described in determination
The corresponding relationship of specified parameter and the cardiac MR images.
Further,
The sample data of corresponding relationship of the acquisition for establishing between the specified parameter and the cardiac MR images
The step of, comprising:
Collect different patients cardiac MR images and specified parameter；
The cardiac MR images are analyzed and combined with the expertise information prestored, is chosen and the specified parameter
Relevant data are as the cardiac MR images；
The data pair that the specified parameter and the cardiac MR images chosen are constituted, as sample data.
Further,
The network structure, comprising: multi-task learning neural network, and, inverse mapping neural network；
And/or
The network parameter, comprising: the input number of plies exports the number of plies, and the convolution number of plies, intensive block number, intensive block includes the number of plies,
Initial weight, and, at least one of bias.
Further,
The network structure and the network parameter are trained, comprising:
A part of data in the sample data are chosen as training sample, by the heart in the training sample
MR image is input to the network structure, is trained, is obtained by the activation primitive and the network parameter of the network structure
To hands-on result；
Determine the hands-on error between the hands-on result specified parameter corresponding in the training sample
Whether satisfaction presets training error；
When the hands-on error meets the default training error, determine to the network structure and the network
The training of parameter is completed；
And/or
The network structure and the network parameter are tested, comprising:
Another part data in the sample data are chosen as test sample, by the heart in the test sample
Dirty MR image is input in the network structure that the training is completed, described in the activation primitive and the training completion
Network parameter is tested, and actual test result is obtained；
Determine the actual test error between the actual test result specified parameter corresponding in the test sample
Whether satisfaction sets test error；
When the actual test error meets the setting test error, determine to the network structure and the network
The test of parameter is completed.
Further,
The network structure and the network parameter are trained, further includes:
When the hands-on error is unsatisfactory for the setting training error, pass through the error energy of the network structure
Function updates the network parameter；
Re -training is carried out by the activation primitive and the updated network parameter of the network structure, until
Hands-on error after the re -training meets the setting training error；
And/or
The network structure and the network parameter are tested, further includes:
When the actual test error is unsatisfactory for the setting test error, the network structure and the network are joined
Number carries out re -trainings, until the actual test error setting test error at a slow speed after the re -training.
A kind of computing device of heart clinical indices, comprising:
Establish module, for the self-learning capability using artificial neural network, establish cardiac MR images and specified parameter it
Between corresponding relationship；Wherein, the specified parameter includes cardiac CT image, for the heart clinical indices of cardiac MR images, with
And the heart clinical indices for cardiac CT image；
Module is obtained, for obtaining the current cardiac MR image of patient；
Determining module, for by the corresponding relationship, determination to be corresponding currently assigned with the current cardiac MR image
Parameter；Specifically, it is determined that currently assigned parameter corresponding with the current cardiac MR image, comprising: will be in the corresponding relationship
Specified parameter corresponding to cardiac MR images identical with the current cardiac MR image, is determined as the currently assigned parameter.
A kind of equipment, including processor, memory and be stored on the memory and can transport on the processor
Capable computer program, the computer program realize the inspection of heart clinical indices as described above when being executed by the processor
The step of survey method.
A kind of computer readable storage medium stores computer program, the meter on the computer readable storage medium
Calculation machine program realizes the step of detection method of heart clinical indices as described above when being executed by processor.
The application has the following advantages:
In embodiments herein, by using artificial neural network self-learning capability, establish cardiac MR images with
Corresponding relationship between specified parameter；Wherein, the specified parameter includes cardiac CT image, is faced for the heart of cardiac MR images
Bed index, and the heart clinical indices for cardiac CT image；Obtain the current cardiac MR image of patient；Pass through the correspondence
Relationship determines currently assigned parameter corresponding with the current cardiac MR image；Specifically, it is determined that scheming with the current cardiac MR
As corresponding currently assigned parameter, comprising: scheme heart MR identical with the current cardiac MR image in the corresponding relationship
As corresponding specified parameter, it is determined as the currently assigned parameter, can be used for polymorphic type under MR image mode different from CT's
The assessment of heart clinical indices.It excavates the complex relationship between characterization polymorphic type heart clinical indices, obtain task dependencies
And realize its different image modes transfer, establish Knowledge-sharing Mechanism under different image modes.
Detailed description of the invention
It, below will be to attached needed in the description of the present application in order to illustrate more clearly of the technical solution of the application
Figure is briefly described, it should be apparent that, the drawings in the following description are only some examples of the present application, for this field
For those of ordinary skill, without any creative labor, it can also be obtained according to these attached drawings other attached
Figure.
Fig. 1 is a kind of step flow chart of the detection method for heart clinical indices that one embodiment of the application provides；
Fig. 2 is a kind of artificial neural network structure of the detection method for heart clinical indices that one embodiment of the application provides
Schematic diagram；
Fig. 3 is a kind of multi-task learning network knot of the detection method for heart clinical indices that one embodiment of the application provides
Structure schematic diagram；
Fig. 4 is that a kind of inverse mapping network structure of the detection method for heart clinical indices that one embodiment of the application provides is shown
It is intended to；
Fig. 5 is a kind of comparing with heterogeneous networks for the detection method for heart clinical indices that one embodiment of the application provides
Estimate the model simplification test result schematic diagram of linear index and plane index；
Fig. 6 is a kind of comparing with heterogeneous networks for the detection method for heart clinical indices that one embodiment of the application provides
The model simplification test result schematic diagram of two-dimensional cardiac image segmentation；
Fig. 7 is a kind of four depth networks of the detection method for heart clinical indices that one embodiment of the application provides not
With the transmission learning outcome under configuration, and the transmission learning outcome schematic diagram of three-dimensional DenseNet；
Fig. 8 is MR the and CT image pattern for the detection method that the application one implements a kind of heart clinical indices provided
Schematic diagram；
Fig. 9 is a kind of heterogeneous networks of the detection method for heart clinical indices that one embodiment of the application provides to three kinds of hearts
The single frames estimation error comparison schematic diagram of dirty index；
Figure 10 is a kind of structural block diagram of the detection device for heart clinical indices that one embodiment of the application provides；
Figure 11 is a kind of structural schematic diagram of computer equipment of one embodiment of the invention.
Specific embodiment
It is with reference to the accompanying drawing and specific real to keep the objects, features and advantages of the application more obvious and easy to understand
Applying mode, the present application will be further described in detail.Obviously, described embodiment is some embodiments of the present application, without
It is whole embodiments.Based on the embodiment in the application, those of ordinary skill in the art are not before making creative work
Every other embodiment obtained is put, shall fall in the protection scope of this application.
Referring to Fig.1, a kind of detection method of heart clinical indices of one embodiment of the application offer is provided, comprising:
S110, the self-learning capability using artificial neural network are established corresponding between cardiac MR images and specified parameter
Relationship；Wherein, the specified parameter includes cardiac CT image, for the heart clinical indices of cardiac MR images, and is directed to the heart
The heart clinical indices of dirty CT image；
S120, the current cardiac MR image for obtaining patient；
S130, pass through the corresponding relationship, determining currently assigned parameter corresponding with the current cardiac MR image；Specifically
Ground determines corresponding with the current cardiac MR image currently assigned parameter, comprising: by the corresponding relationship with it is described currently
Specified parameter corresponding to the identical cardiac MR images of cardiac MR images is determined as the currently assigned parameter.
In embodiments herein, by using artificial neural network self-learning capability, establish cardiac MR images with
Corresponding relationship between specified parameter；Wherein, the specified parameter includes cardiac CT image, is faced for the heart of cardiac MR images
Bed index, and the heart clinical indices for cardiac CT image；Obtain the current cardiac MR image of patient；Pass through the correspondence
Relationship determines currently assigned parameter corresponding with the current cardiac MR image；Specifically, it is determined that scheming with the current cardiac MR
As corresponding currently assigned parameter, comprising: scheme heart MR identical with the current cardiac MR image in the corresponding relationship
As corresponding specified parameter, it is determined as the currently assigned parameter, can be used for polymorphic type under MR image mode different from CT's
The assessment of heart clinical indices.It excavates the complex relationship between characterization polymorphic type heart clinical indices, obtain task dependencies
And realize its different image modes transfer, establish Knowledge-sharing Mechanism under different image modes.
In the following, by being further described to the detection method of present exemplary embodiment cardiac clinical indices.
As described in above-mentioned steps S110, using the self-learning capability of artificial neural network, establishes cardiac MR images and specify
Corresponding relationship between parameter；Wherein, the specified parameter includes cardiac CT image, is referred to for the heart clinic of cardiac MR images
Mark, and for the heart clinical indices of cardiac CT image
Such as: display state rule of the specified parameter in cardiac MR images is analyzed using artificial neural network algorithm,
The mapping principle between patient's heart MR image and specified parameter is found by the self study of artificial neural network, adaptive characteristic.
Such as: it can use artificial neural network algorithm, by (including but not limited to following to a large amount of different volunteers
It is one or more: the age, if to have history of heart disease, gender, state of an illness etc.) cardiac MR images summarize collection, choose several aspirations
The cardiac MR images of person and specified parameter learn neural network and are trained as sample data, by adjusting network knot
Weight between structure and network node makes the relationship between neural network fitting cardiac MR images and specified parameter, finally makes nerve
Network can accurately fit the cardiac MR images of different patients and the corresponding relationship of specified parameter.
Referring to Fig. 2, it should be noted that in the embodiment of the present application, preferably by cardiac MR images and with the heart
The corresponding heart clinical indices of MR image carry out to artificial neural network (multi-task learning neural network, and, inverse mapping nerve
Network) training, it is different by two dependence parameters, the identical determining device network of structure to the input of the artificial neural network and
Output finally obtains the artificial neural network of training completion referring to dual training is carried out, in the artificial neuron completed based on the training
The network parameter of network is trained CT image corresponding with cardiac MR images and its heart clinical indices, to obtain the heart
(cardiac CT image for the heart clinical indices of cardiac MR images, and is directed to cardiac CT image for dirty MR image and specified parameter
Heart clinical indices) between corresponding relationship.
Specifically, by the parameter of each layer of multi-task learning network and each layer acquistion of inverse mapping network, scheme first in heart MR
It is trained under the data environment of picture using antagonistic training method；Then, with the cardiac CT image data with label, pass through
Loss function is finely adjusted the parameter of training acquistion, finally, by the parameter after fine tuning under the data environment of cardiac CT image
Using the training of antagonistic training method, wherein shown in the loss function such as following equation (17).
In one embodiment, the heart clinical indices include: left ventricular epicardium profile, left ventricle inner membrance profile, Zuo Xin
Room flesh profile, left ventricular epicardium position, film location in left ventricle, left compartment muscle position, local chambers of the heart wall thickness WT (Regional
Wall Thinknesses), heart chamber volume Dim (Directional Dimensions of Cavity), and, chambers of the heart wall and
At least one of myocardial area Area (Areas of Cavity and Myocardium).
In one embodiment, it can be further illustrated in step S110 in conjunction with following description and " establish cardiac MR images and refer to
Determine the corresponding relationship between parameter " detailed process.
As described in the following steps: obtaining the corresponding relationship for establishing between the specified parameter and the cardiac MR images
Sample data；
In an advanced embodiment, it can further illustrate that " acquisition can be used for establishing the operation in conjunction with following description
The detailed process of the sample data of corresponding relationship between parameter and the frosting state ".
As described in the following steps: collect different patients cardiac MR images and specified parameter；
Such as: data collection: collect the patient of different health status cardiac MR images and corresponding specified parameter；With
And collect the patient of all ages and classes cardiac MR images and corresponding specified parameter；And collect the heart of the patient of different sexes
Dirty MR image and corresponding specified parameter.
It collects operation data through a variety of ways as a result, is conducive to the amount for increasing operation data, promotes artificial neural network
Learning ability, and then promote the accuracy and reliability of determining corresponding relationship.
As described in the following steps: the cardiac MR images being analyzed and combined with the expertise information prestored, is chosen
Data relevant to the specified parameter as the cardiac MR images (such as: choose heart MR influential on specified parameter
Image is as input parameter, using specified parameter as output parameter)；
Such as: by using the cardiac MR images in the related data of the volunteer made a definite diagnosis as input parameter, by its phase
The specified parameter in data is closed as output parameter.
As described in the following steps: the data pair that the specified parameter and the cardiac MR images chosen are constituted are made
For sample data.
Such as: by obtained input, output parameter pair, a part is used as training book sample data, and a part is used as test specimens
Notebook data.
As a result, by the way that the cardiac MR images being collected into are analyzed and handled, and then sample data is obtained, operating process
Simply, operating result high reliablity.
As described in the following steps: the characteristic and its rule of the cardiac MR images are analyzed, according to the characteristic and its rule,
Determine the network structure and its network parameter of the artificial neural network.
Such as: on the influential data characteristic of cardiac image tool and its contained according to different ages, the state of an illness, gender etc.
Rule, can primarily determine the basic structure of network, the input of network, output node number, network hidden layer number, Hidden nodes, net
Network initial weight etc..
Preferably, the network structure, comprising: multi-task learning neural network, and, inverse mapping neural network.
Preferably, the network parameter, comprising: the input number of plies exports the number of plies, the convolution number of plies, intensive block number, intensive block packet
Containing the number of plies, initial weight, and, at least one of bias.
Such as: as shown in figure 3, multi-task learning network, by dense connection convolutional neural networks (Densely
Connected Convolutional Networks, DenseNet) it modifies.The multi-task learning network of foundation with
Comprising an input layer based on DenseNet, an output layer, a convolutional layer Mcnn, one has 4 layers intensive
Block (Md1), two have 8 layers of intensive block (Md2 and Md3), a warp lamination Mdcnn and a full articulamentum Mfc.Institute
Some convolutional layers have all used behavior and the superior Three dimensional convolution neural network (Convolutional of capacity modeling ability
Neural Networks, CNN), wherein different size of convolution kernel is respectively 3 containing there are three in convolutional layer3, 53And 73,
And it is 3 that the Three dimensional convolution layer in intensive block, which has then used size,3Convolution kernel.
Wherein, input layer is cardiac MR images, and after convolutional layer and three intensive blocks, network will learn to one to combine
Distribution, and Feature Mapping (feature map, fea) is obtained, two dimension is obtained using a Pixel-level classifier of warp lamination
Image segmentation result.Wherein input of the fea as full convolutional layer is referred to by the one-dimensional heart of a Recurrent networks output assessment
Mark.
It should be noted that inverse mapping network, structure and above-mentioned multi-task learning network structure are similar but contrary,
As shown in figure 4, by modifying to above-mentioned multi-task learning network structure and direction is turned.The inverse mapping network of foundation its
In, it include an input layer, an output layer, a convolutional layer Rcnn, one has 4 layers of intensive block (Rd1), and two have 8 layers
Intensive block (Rd2 and Rd3), a warp lamination Rdcnn and a full articulamentum Rfc.All convolutional layers all use
Behavior and the superior Three dimensional convolution neural network of capacity modeling ability (Convolutional Neural Networks, CNN),
Wherein, different size of convolution kernel is respectively 3 containing there are three in convolutional layer3, 53And 73, and the Three dimensional convolution layer in intensive block
Then having used size is 33Convolution kernel.It wherein, is input with two dimensional image segmentation and one-dimensional cardiac index, output result is to rebuild
Cardiac MR images.
The analysis processing for carrying out data by joint multi-task learning neural network and inverse mapping neural network as a result, can
To promote the accuracy of data processing, it is also possible that being determined to corresponding relationship between cardiac MR images and specified parameter reliable
Property improve.
As described in the following steps: using the sample data, be trained to the network structure and the network parameter
And test, determine the corresponding relationship of the specified parameter Yu the cardiac MR images.
Such as: after the completion of network design, it need to be trained with the neural network that training sample data complete design.Training
Method can be adjusted according to the problem of discovery in actual network structure and training.
As a result, by collection image data, sample data is therefrom chosen, and is trained and tests based on sample data,
It determines the corresponding relationship between cardiac MR images and specified parameter, is conducive to promote the accuracy for generating specified parameter.
Preferably, network is trained by the way of dual training, by two neural network (multi-task learning networks
With inverse mapping network) connect that (connection type includes: to connect multi-task learning network and inverse mapping network in a different order
It connects；Inverse mapping network and multi-task learning are connected to the network), and learning training is carried out respectively, respectively obtain a Joint Distribution
And it is matched.
Then, it is similar and rely on the different arbiter network of parameter that two structures are constructed, wherein the dependence parameter is following
ψ 1 and ψ 2 in formula (7) and (8), two arbiter networks respectively correspond one of them and do not repeat,.It is different for every kind
Connection type, extracts inputoutput pair from corresponding Joint Distribution, and with arbiter network to the inputoutput pair respectively into
Row is distinguished.If arbiter, which is difficult to differentiate between, outputs and inputs sample, that means that the specified parameter of assessment can be well reflected
True picture, and the image reconstructed can approaching to reality well specified parameter.
It is alternatively possible to further illustrate that step " uses the sample data, to the network structure in conjunction with following description
It is trained and tests with the network parameter, determine the corresponding relationship of the specified parameter Yu the cardiac MR images "
In detailed process that the network structure and the network parameter are trained.
As described in the following steps, a part of data in the sample data are chosen as training sample, by the training
The cardiac MR images in sample are input to the network structure, pass through the activation primitive and the network of the network structure
Parameter is trained, and obtains hands-on result；Determine that the hands-on result is corresponding in the training sample specified
Whether the hands-on error between parameter meets default training error；When the hands-on error meets the default training
When error, determine that the training to the network structure and the network parameter is completed；
Such as: the design of artificial neural network, by providing cardiac image x, further study generates the inverse mapping of x from i
G2:I → X, this helps to disclose the complex relationship between x and i.In the network of building, G1 and G2 can be indicated are as follows:
G2(i；θ)=argmaxxpθ(x|i) (2)
By G1, two networks of sequential connection of G2 obtain functionExpression formula is as follows:
G2 is pressed again, and two networks of sequential connection of G1 obtain functionExpression formula is as follows:
The cardiac index of polymorphic type describes the feature of cardiac structure from different dimensions.If understanding polymorphic type heart to refer to
The semantic relation between complicated correlation and many types of cardiac index and cardiac image in mark, can not only estimate the polymorphic type heart
Dirty index, can also therefrom reconfiguring heart image.For this purpose, considering the variation expression formula of f and g:
The design of dual training considers that the following methods of sampling goes to learn and match above-mentioned two distribution.For function f and very
Real polymorphic type cardiac index i, to (x, i), can be sentenced from distribution q (x) q (i | x) and distribution p (i) p (x | i ') extraction with full mold
Other device network Tψ1(x, i) is distinguished.It equally, can be from distribution p (i) p (x | i) and q for function g and true heart image x
(x) q (i | x ') distribution is extracted uses full mold arbiter network T to (x, i) againψ2(x, i) is distinguished, wherein 1 ψ, and ψ 2 is two differentiations
The parameter of device network acquistion.
Specifically, considering arbiter T to realize above-mentioned functionψ1So that the value of following formula maximizes:
And arbiter Tψ2So that the value of following formula maximizes:
Wherein, σ represents sigmond function, and expression formula is defined as
Section 2 in rewriting formula (7):
To which formula (7) can be write as:
When the integral of (x, i) is maximized, the integral of formula (11), about σ (Tψ2(x, i)) function, take most
Big value.Notice that function a log t+b log (1-t) obtains maximum value at t=a/ (a+b), and convolution (9) has:
Namely
Similarly
Two formulas, that is, optimal, arbiter optimized above.Finally, optimization object can be with table in conjunction with (5) (6) two formula
State into a Min-max Problems:
In addition, setting above formula realizes the corresponding parameter of processThen to have
Finally, frame is finely tuned by following loss function, in multi-task learning network, to two-dimensional cardiac image point
It cuts and loses L using DiceDice, and L is used to one-dimensional cardiac index2(i.e. ridge regression) loss.In inverse mapping network, then use
The loss function L of reconstructrecon.Finally, entire loss can write
LM&R=LDice+L2+Lrecon (17)
In above-mentioned formula:
X: source image mode image pattern；X ': reconfiguring heart image；X: source image mode image pattern collection；I: polymorphic type
Cardiac index sample；I ': estimation cardiac index；I: polymorphic type cardiac index sample set；φ: multi-task learning network acquistion distribution
Parameter；θ: the parameter of inverse mapping network acquistion distribution；qφ(i, x): multi-task learning network acquistion Joint Distribution is based on parameter
φ can write qφ(i|x)q(x)；pθ(x, i): inverse mapping network acquistion Joint Distribution is based on parameter θ, can write pθ(x|i)p
(x)。
More optionally, the network structure and the network parameter are trained, further includes:
When the hands-on error is unsatisfactory for the setting training error, pass through the error energy of the network structure
Function updates the network parameter；It is carried out by the activation primitive and the updated network parameter of the network structure
Re -training, until the hands-on error after the re -training meets the setting training error；
Such as: if test error is met the requirements, network training test is completed.
As a result, by being used to test sample obtained network structure and network parameter be trained to test, with further
Verify the reliability of network structure and network parameter.
It is alternatively possible to further illustrate that step " uses the sample data, to the network structure in conjunction with following description
It is trained and tests with the network parameter, determine the corresponding relationship of the specified parameter Yu the cardiac MR images "
In detailed process that the network structure and the network parameter are tested.
As described in the following steps, another part data in the sample data are chosen as test sample, by the survey
The cardiac MR images in sample sheet are input in the network structure that the training is completed, with the activation primitive and institute
The network parameter for stating training completion is tested, and actual test result is obtained；Determine the actual test result with it is described
Whether the actual test error between corresponding specified parameter in test sample meets setting test error；When the actual test
When error meets the setting test error, determine that the test to the network structure and the network parameter is completed.
Judgment criteria is formulated:
1. the interpretation of result to two-dimensional cardiac image segmentation is mainly judged with Dice coefficient, expression formula are as follows:
Wherein, PTIndicate all pixels for the contour area manually divided, PEIndicate all of the contour area divided automatically
Pixel, PTEIndicate PTWith PEBetween overlaid pixel.Dice value is higher, shows the consistency between dividing automatically and dividing by hand
It is higher.
2. one-dimensional linear index and plane index are calculated mean absolute error (Mean Absolute Error, MAE),
Expression formula are as follows:
Wherein y ∈ RN,Two vectors separately include the index of actual index and estimation.The value of MAE is smaller,
Show that the accuracy for estimating result is higher.
More optionally, the network structure and the network parameter are tested, further includes:
When the actual test error is unsatisfactory for the setting test error, the network structure and the network are joined
Number carries out re -trainings, until the actual test error setting test error at a slow speed after the re -training.
Such as: when test error is unsatisfactory for requiring, then repeatedly above step, re -training network.
As a result, by carrying out re -training to network structure when test error is larger to retest, be conducive to
More accurate and reliable network structure is obtained, and then promotes the accuracy determined to frosting state.
In the concrete realization,
Referring to Fig. 5-5,1. in terms of the assessment of index: passing through comparative studies, the present processes are in MR data than existing
One-way method have better performance.As FullLVNet and DMTRL has CNN and recurrent neural network (Recurrent
Neural Network, RNN) the advantages of combining, but they be all highly dependent on to task dependencies study some is specific
Constraint condition and do not have inverse mapping network.
And other learning frameworks for having inverse mapping network, the present processes still have more superiority.
Referring to Fig. 7,2. in terms of two-way parameter sharing: firstly, establishing multiple multi-task learning frames.These multitasks
Pixel-level classifier, Recurrent networks and the joint for practising frame indicate that network structure is identical, but feature extraction level is different.Then divide
Three kinds of modes distinguish test performance: (i) frame is trained directly against target data, and printenv shares (No sharing)；
(ii) traditional unidirectional parameter sharing mechanism (One-way-para) is used；(iii) two-way parameter sharing mechanism (Bi- is used
para).Comparison result shows in the multi-task learning frame of the application that two-way parameter sharing mechanism is to the index across image mode
It is effective to estimate.
Referring to Fig. 8-8, in general, not only heart knot can be accurately positioned in the present processes in terms of image segmentation
Structure, and accuracy is high in terms of index estimation, and wherein " Our " in Fig. 9 is the method for the embodiment of the present application.
For device embodiment, since it is basically similar to the method embodiment, related so being described relatively simple
Place illustrates referring to the part of embodiment of the method.
Referring to Fig.1 0, a kind of computing device of heart clinical indices of one embodiment of the application offer is provided, comprising:
Module 110 is established, for the self-learning capability using artificial neural network, establishes cardiac MR images and specified parameter
Between corresponding relationship；Wherein, the specified parameter includes cardiac CT image, for the heart clinical indices of cardiac MR images,
And the heart clinical indices for cardiac CT image；
Module 120 is obtained, for obtaining the current cardiac MR image of patient；
Determining module 130, for determining current finger corresponding with the current cardiac MR image by the corresponding relationship
Determine parameter；Specifically, it is determined that currently assigned parameter corresponding with the current cardiac MR image, comprising: by the corresponding relationship
In specified parameter corresponding to cardiac MR images identical with the current cardiac MR image, be determined as the currently assigned ginseng
Number.
In one embodiment, described image feature includes: two-dimensional cardiac image and one-dimensional heart clinical indices.
It is in one embodiment, described to establish module 310, comprising:
Acquisition submodule, for obtaining the corresponding relationship for establishing between the specified parameter and the cardiac MR images
Sample data；
Submodule is analyzed, for analyzing the characteristic and its rule of the cardiac MR images, according to the characteristic and its rule,
Determine the network structure and its network parameter of the artificial neural network；
Training submodule is trained the network structure and the network parameter for using the sample data
And test, determine the corresponding relationship of the specified parameter Yu the cardiac MR images.
In one embodiment, the acquisition submodule, comprising:
Collect submodule, for collect different patients cardiac MR images and specified parameter；
Submodule is analyzed, for the cardiac MR images to be analyzed and combined with the expertise information prestored, is chosen
Data relevant to the specified parameter are as the cardiac MR images；
Sample data generates submodule, for constitute the specified parameter and the cardiac MR images chosen
Data pair, as sample data.
In one embodiment,
The network structure, comprising: multi-task learning neural network, and, inverse mapping neural network；
And/or
The network parameter, comprising: the input number of plies exports the number of plies, and the convolution number of plies, intensive block number, intensive block includes the number of plies,
Initial weight, and, at least one of bias.
In one embodiment,
The trained submodule, comprising:
Training result generates submodule, for choosing a part of data in the sample data as training sample, incites somebody to action
The cardiac MR images in the training sample are input to the network structure, by the activation primitive of the network structure and
The network parameter is trained, and obtains hands-on result；
Training result error judgment submodule, for determining that the hands-on result is corresponding in the training sample
Whether the hands-on error between specified parameter meets default training error；
Decision sub-module is completed in training, for determining when the hands-on error meets the default training error
The training of the network structure and the network parameter is completed；
And/or
Submodule is tested, for testing the network structure and the network parameter, the test submodule, packet
It includes:
Test result generates submodule, for choosing another part data in the sample data as test sample,
The cardiac MR images in the test sample are input in the network structure that the training is completed, with the activation
The network parameter that function and the training are completed is tested, and actual test result is obtained；
Test result error judgment submodule, for determining that the actual test result is corresponding in the test sample
Whether the actual test error between specified parameter meets setting test error；
Decision sub-module is completed in test, for determining when the actual test error meets the setting test error
The test of the network structure and the network parameter is completed.
In one embodiment,
The trained submodule, further includes:
Network parameter updates submodule, for leading to when the hands-on error is unsatisfactory for the setting training error
The error energy function for crossing the network structure updates the network parameter；
First retraining submodule, for the activation primitive and the updated network by the network structure
Parameter carries out re -training, until the hands-on error after the re -training meets the setting training error；
And/or
The test submodule, further includes:
Second retraining submodule, for when the actual test error is unsatisfactory for the setting test error, to institute
It states network structure and the network parameter carries out re -training, until the actual test error after the re -training is described at a slow speed
Set test error.
Referring to Fig.1 1, a kind of computer equipment of the detection method of heart clinical indices of the invention is shown, specifically may be used
To include the following:
Above-mentioned computer equipment 12 is showed in the form of universal computing device, the component of computer equipment 12 may include but
Be not limited to: one or more processor or processing unit 16, system storage 28, connecting different system components (including is
Unite memory 28 and processing unit 16) bus 18.
Bus 18 indicates one of a few 18 structures of class bus or a variety of, including memory bus 18 or memory control
Device, peripheral bus 18, graphics acceleration port, processor or the office using 18 structure of any bus in a variety of 18 structures of bus
Domain bus 18.For example, these architectures include but is not limited to industry standard architecture (ISA) bus 18, microchannel
Architecture (MAC) bus 18, enhanced isa bus 18, audio-video frequency electronic standard association (VESA) local bus 18 and outer
Enclose component interconnection (PCI) bus 18.
Computer equipment 12 typically comprises a variety of computer system readable media.These media can be it is any can be by
The usable medium that computer equipment 12 accesses, including volatile and non-volatile media, moveable and immovable medium.
System storage 28 may include the computer system readable media of form of volatile memory, such as arbitrary access
Memory (RAM) 30 and/or cache memory 32.Computer equipment 12 may further include other movement/it is not removable
Dynamic, volatile/non-volatile computer decorum storage medium.Only as an example, storage system 34 can be used for read and write can not
Mobile, non-volatile magnetic media (commonly referred to as " hard disk drive ").Although being not shown in Figure 11, can provide for can
The disc driver of mobile non-volatile magnetic disk (such as " floppy disk ") read-write, and to removable anonvolatile optical disk (such as CD-
ROM, DVD-ROM or other optical mediums) read-write CD drive.In these cases, each driver can pass through one
A or multiple data mediums interface is connected with bus 18.Memory may include at least one program product, the program product
With one group of (for example, at least one) program module 42, these program modules 42 are configured to perform the function of various embodiments of the present invention
Energy.
Program/utility 40 with one group of (at least one) program module 42, can store in memory, for example,
Such program module 42 includes --- but being not limited to --- operating system, one or more application program, other program moulds
It may include the realization of network environment in block 42 and program data, each of these examples or certain combination.Program mould
Block 42 usually executes function and/or method in embodiment described in the invention.
Computer equipment 12 can also with one or more external equipments 14 (such as keyboard, sensing equipment, display 24,
Camera etc.) communication, the equipment interacted with the computer equipment 12 can be also enabled a user to one or more to be communicated, and/
Or with enable the computer equipment 12 and one or more other calculate any equipment that equipment are communicated (such as network interface card,
Modem etc.) communication.This communication can be carried out by interface input/output (I/O) 22.Also, computer equipment
12 can also by network adapter 20 and one or more network (such as local area network (LAN)), wide area network (WAN) and/or
Public network (such as internet) communication.As shown, network adapter 20 passes through other of bus 18 and computer equipment 12
Module communication.It should be understood that although being not shown in Figure 11 other hardware and/or software can be used in conjunction with computer equipment 12
Module, including but not limited to: microcode, device driver, redundant processing unit 16, external disk drive array, RAID system,
Tape drive and data backup storage system 34 etc..
Processing unit 16 by the program that is stored in system storage 28 of operation, thereby executing various function application and
Data processing, such as realize the detection method of heart clinical indices provided by the embodiment of the present invention.
That is, above-mentioned processing unit 16 is realized when executing above procedure: using the self-learning capability of artificial neural network, building
Corresponding relationship between vertical cardiac MR images and specified parameter；Wherein, the specified parameter includes cardiac CT image, for heart
The heart clinical indices of MR image, and the heart clinical indices for cardiac CT image；Obtain the current cardiac MR figure of patient
Picture；By the corresponding relationship, currently assigned parameter corresponding with the current cardiac MR image is determined；Specifically, it is determined that with
The corresponding currently assigned parameter of the current cardiac MR image, comprising: will scheme in the corresponding relationship with the current cardiac MR
The specified parameter as corresponding to identical cardiac MR images is determined as the currently assigned parameter.
In embodiments of the present invention, the present invention also provides a kind of computer readable storage medium, it is stored thereon with computer
Program realizes the detection method of the heart clinical indices provided such as all embodiments of the application when the program is executed by processor:
That is, realization when being executed by processor to program: using the self-learning capability of artificial neural network, establishing heart MR
Corresponding relationship between image and specified parameter；Wherein, the specified parameter includes cardiac CT image, for cardiac MR images
Heart clinical indices, and the heart clinical indices for cardiac CT image；Obtain the current cardiac MR image of patient；Pass through institute
Corresponding relationship is stated, determines currently assigned parameter corresponding with the current cardiac MR image；Specifically, it is determined that working as front center with described
The corresponding currently assigned parameter of dirty MR image, comprising: by the heart identical with the current cardiac MR image in the corresponding relationship
Specified parameter corresponding to dirty MR image is determined as the currently assigned parameter.
It can be using any combination of one or more computer-readable media.Computer-readable medium can be calculating
Machine gram signal media or computer readable storage medium.Computer readable storage medium for example can be --- but it is unlimited
In system, device or the device of --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, or any above combination.Computer
The more specific example (non exhaustive list) of readable storage medium storing program for executing includes: electrical connection with one or more conducting wires, portable
Formula computer disk, hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory
(EPOM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device or
Above-mentioned any appropriate combination.In this document, computer readable storage medium can be it is any include or storage program
Tangible medium, the program can be commanded execution system, device or device use or in connection.
Computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including --- but
It is not limited to --- electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be
Any computer-readable medium other than computer readable storage medium, which can send, propagate or
Transmission is for by the use of instruction execution system, device or device or program in connection.
The computer for executing operation of the present invention can be write with one or more programming languages or combinations thereof
Program code, above procedure design language include object oriented program language --- such as Java, Smalltalk, C+
+, further include conventional procedural programming language --- such as " C " language or similar programming language.Program code
It can fully execute on the user computer, partly execute, held as an independent software package on the user computer
Part executes on the remote computer or holds on a remote computer or server completely on the user computer for row, part
Row.In situations involving remote computers, remote computer can pass through the network of any kind --- including local area network
(LAN) or wide area network (WAN) --- it is connected to subscriber computer, or, it may be connected to outer computer (such as using because of spy
Service provider is netted to connect by internet).All the embodiments in this specification are described in a progressive manner, each
What embodiment stressed is the difference from other embodiments, the mutual coherent in same and similar part between each embodiment
See.
Although preferred embodiments of the embodiments of the present application have been described, once a person skilled in the art knows bases
This creative concept, then additional changes and modifications can be made to these embodiments.So the following claims are intended to be interpreted as
Including preferred embodiment and all change and modification within the scope of the embodiments of the present application.
Finally, it is to be noted that, herein, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning
Covering non-exclusive inclusion, so that process, method, article or terminal device including a series of elements not only wrap
Those elements are included, but also including other elements that are not explicitly listed, or further includes for this process, method, article
Or the element that terminal device is intrinsic.In the absence of more restrictions, being wanted by what sentence "including a ..." limited
Element, it is not excluded that there is also other identical elements in process, method, article or the terminal device for including the element.
Above to detection method, device, equipment and the medium of heart clinical indices provided herein, carry out in detail
It introduces, specific examples are used herein to illustrate the principle and implementation manner of the present application, the explanation of above embodiments
It is merely used to help understand the present processes and its core concept；At the same time, for those skilled in the art, according to this
The thought of application, there will be changes in the specific implementation manner and application range, in conclusion the content of the present specification is not answered
It is interpreted as the limitation to the application.
Claims (10)
1. a kind of detection method of heart clinical indices characterized by comprising
Using the self-learning capability of artificial neural network, the corresponding relationship between cardiac MR images and specified parameter is established；Wherein,
The specified parameter includes cardiac CT image, for the heart clinical indices of cardiac MR images, and for cardiac CT image
Heart clinical indices；
Obtain the current cardiac MR image of patient；
By the corresponding relationship, currently assigned parameter corresponding with the current cardiac MR image is determined；Specifically, it is determined that with
The corresponding currently assigned parameter of the current cardiac MR image, comprising: will scheme in the corresponding relationship with the current cardiac MR
The specified parameter as corresponding to identical cardiac MR images is determined as the currently assigned parameter.
2. the method according to claim 1, wherein
The heart clinical indices include: left ventricular epicardium profile, left ventricle inner membrance profile, left compartment muscle profile, outside left ventricle
Film location, film location in left ventricle, left compartment muscle position, local chambers of the heart wall thickness WT, heart chamber volume Dim, and, chambers of the heart wall and
At least one of myocardial area Area.
3. the method according to claim 1, wherein establishing the corresponding pass between cardiac MR images and specified parameter
System, comprising:
Obtain the sample data of the corresponding relationship for establishing between the specified parameter and the cardiac MR images；
The characteristic and its rule for analyzing the cardiac MR images determine the artificial neural network according to the characteristic and its rule
The network structure and its network parameter of network；
Using the sample data, the network structure and the network parameter are trained and are tested, determined described specified
The corresponding relationship of parameter and the cardiac MR images.
4. according to the method described in claim 3, it is characterized in that, the acquisition is for establishing the specified parameter and the heart
The step of sample data of corresponding relationship between dirty MR image, comprising:
Collect different patients cardiac MR images and specified parameter；
The cardiac MR images are analyzed and combined with the expertise information prestored, is chosen related to the specified parameter
Data as the cardiac MR images；
The data pair that the specified parameter and the cardiac MR images chosen are constituted, as sample data.
5. according to the method described in claim 4, it is characterized in that,
The network structure, comprising: multi-task learning neural network, and, inverse mapping neural network；
And/or
The network parameter, comprising: the input number of plies exports the number of plies, and the convolution number of plies, intensive block number, intensive block includes the number of plies, initially
Weight, and, at least one of bias.
6. according to the described in any item methods of claim 3-5, which is characterized in that
The network structure and the network parameter are trained, comprising:
A part of data in the sample data are chosen as training sample, the heart MR in the training sample is schemed
As being input to the network structure, it is trained by the activation primitive and the network parameter of the network structure, obtains reality
Border training result；
Whether determine the hands-on result hands-on error specified between parameter corresponding in the training sample
Meet default training error；
When the hands-on error meets the default training error, determine to the network structure and the network parameter
It is described training complete；
And/or
The network structure and the network parameter are tested, comprising:
Another part data in the sample data are chosen as test sample, by the heart MR in the test sample
Image is input in the network structure that the training is completed, the network completed with the activation primitive and the training
Parameter is tested, and actual test result is obtained；
Whether determine the actual test result actual test error specified between parameter corresponding in the test sample
Meet setting test error；
When the actual test error meets the setting test error, determine to the network structure and the network parameter
The test complete.
7. according to the method described in claim 6, it is characterized in that,
The network structure and the network parameter are trained, further includes:
When the hands-on error is unsatisfactory for the setting training error, pass through the error energy function of the network structure
Update the network parameter；
Re -training is carried out by the activation primitive and the updated network parameter of the network structure, until described
Hands-on error after re -training meets the setting training error；
And/or
The network structure and the network parameter are tested, further includes:
When the actual test error is unsatisfactory for the setting test error, to the network structure and the network parameter into
Row re -training, until the actual test error setting test error at a slow speed after the re -training.
8. a kind of computing device of heart clinical indices characterized by comprising
Module is established, for the self-learning capability using artificial neural network, is established between cardiac MR images and specified parameter
Corresponding relationship；Wherein, the specified parameter includes cardiac CT image, for the heart clinical indices of cardiac MR images, Yi Jizhen
To the heart clinical indices of cardiac CT image；
Module is obtained, for obtaining the current cardiac MR image of patient；
Determining module, for determining currently assigned parameter corresponding with the current cardiac MR image by the corresponding relationship；
Specifically, it is determined that currently assigned parameter corresponding with the current cardiac MR image, comprising: by the corresponding relationship with it is described
Specified parameter corresponding to the identical cardiac MR images of current cardiac MR image is determined as the currently assigned parameter.
9. a kind of equipment, which is characterized in that including processor, memory and be stored on the memory and can be at the place
The computer program run on reason device is realized when the computer program is executed by the processor as appointed in claim 1 to 7
Method described in one.
10. a kind of computer readable storage medium, which is characterized in that store computer journey on the computer readable storage medium
Sequence realizes the method as described in any one of claims 1 to 7 when the computer program is executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910667589.3A CN110400298B (en) | 2019-07-23 | 2019-07-23 | Method, device, equipment and medium for detecting heart clinical index |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910667589.3A CN110400298B (en) | 2019-07-23 | 2019-07-23 | Method, device, equipment and medium for detecting heart clinical index |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110400298A true CN110400298A (en) | 2019-11-01 |
| CN110400298B CN110400298B (en) | 2023-10-31 |
Family
ID=68325889
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910667589.3A Active CN110400298B (en) | 2019-07-23 | 2019-07-23 | Method, device, equipment and medium for detecting heart clinical index |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110400298B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110786839A (en) * | 2019-11-22 | 2020-02-14 | 中山大学 | Method, device, equipment and medium for generating instantaneous waveform-free ratio |
| CN110853012A (en) * | 2019-11-11 | 2020-02-28 | 苏州锐一仪器科技有限公司 | Method, apparatus and computer storage medium for obtaining cardiac parameters |
| CN111062948A (en) * | 2019-11-18 | 2020-04-24 | 北京航空航天大学合肥创新研究院 | Multi-tissue segmentation method based on fetal four-chamber cardiac section image |
| CN111126274A (en) * | 2019-12-24 | 2020-05-08 | 深圳市检验检疫科学研究院 | Method, device, equipment and medium for detecting inbound target population |
| CN111127400A (en) * | 2019-11-29 | 2020-05-08 | 深圳蓝韵医学影像有限公司 | Method and device for detecting breast lesions |
| CN111192255A (en) * | 2019-12-30 | 2020-05-22 | 上海联影智能医疗科技有限公司 | Index detection method, computer device, and storage medium |
| CN111275686A (en) * | 2020-01-20 | 2020-06-12 | 中山大学 | Method and device for generating medical image data for artificial neural network training |
| CN112819789A (en) * | 2020-02-28 | 2021-05-18 | 上海联影智能医疗科技有限公司 | Apparatus and method for cardiac assessment |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5435310A (en) * | 1993-06-23 | 1995-07-25 | University Of Washington | Determining cardiac wall thickness and motion by imaging and three-dimensional modeling |
| US20030035573A1 (en) * | 1999-12-22 | 2003-02-20 | Nicolae Duta | Method for learning-based object detection in cardiac magnetic resonance images |
| CN1839392A (en) * | 2003-06-25 | 2006-09-27 | 美国西门子医疗解决公司 | Automated regional myocardial assessment for cardiac imaging |
| CN1914617A (en) * | 2004-02-03 | 2007-02-14 | 美国西门子医疗解决公司 | Systems and methods for automated diagnosis and decision support for heart related diseases and conditions |
| WO2015109254A2 (en) * | 2014-01-17 | 2015-07-23 | Morpheus Medical, Inc. | Apparatus, methods and articles for four dimensional (4d) flow magnetic resonance imaging |
| CN106096632A (en) * | 2016-06-02 | 2016-11-09 | 哈尔滨工业大学 | Based on degree of depth study and the ventricular function index prediction method of MRI image |
| US20180292484A1 (en) * | 2017-04-05 | 2018-10-11 | Siemens Healthcare Gmbh | Method, neural network, and magnetic resonance apparatus for assigning magnetic resonance fingerprints |
| CN108898622A (en) * | 2018-07-05 | 2018-11-27 | 深圳大学 | A kind of the representation of athletic method, apparatus and computer readable storage medium of heart |
| CN109215014A (en) * | 2018-07-02 | 2019-01-15 | 中国科学院深圳先进技术研究院 | Training method, device, equipment and the storage medium of CT image prediction model |
| US20190139641A1 (en) * | 2017-11-03 | 2019-05-09 | Siemens Healthcare Gmbh | Artificial intelligence for physiological quantification in medical imaging |
-
2019
- 2019-07-23 CN CN201910667589.3A patent/CN110400298B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5435310A (en) * | 1993-06-23 | 1995-07-25 | University Of Washington | Determining cardiac wall thickness and motion by imaging and three-dimensional modeling |
| US20030035573A1 (en) * | 1999-12-22 | 2003-02-20 | Nicolae Duta | Method for learning-based object detection in cardiac magnetic resonance images |
| CN1839392A (en) * | 2003-06-25 | 2006-09-27 | 美国西门子医疗解决公司 | Automated regional myocardial assessment for cardiac imaging |
| CN1914617A (en) * | 2004-02-03 | 2007-02-14 | 美国西门子医疗解决公司 | Systems and methods for automated diagnosis and decision support for heart related diseases and conditions |
| WO2015109254A2 (en) * | 2014-01-17 | 2015-07-23 | Morpheus Medical, Inc. | Apparatus, methods and articles for four dimensional (4d) flow magnetic resonance imaging |
| CN106096632A (en) * | 2016-06-02 | 2016-11-09 | 哈尔滨工业大学 | Based on degree of depth study and the ventricular function index prediction method of MRI image |
| US20180292484A1 (en) * | 2017-04-05 | 2018-10-11 | Siemens Healthcare Gmbh | Method, neural network, and magnetic resonance apparatus for assigning magnetic resonance fingerprints |
| US20190139641A1 (en) * | 2017-11-03 | 2019-05-09 | Siemens Healthcare Gmbh | Artificial intelligence for physiological quantification in medical imaging |
| CN109215014A (en) * | 2018-07-02 | 2019-01-15 | 中国科学院深圳先进技术研究院 | Training method, device, equipment and the storage medium of CT image prediction model |
| CN108898622A (en) * | 2018-07-05 | 2018-11-27 | 深圳大学 | A kind of the representation of athletic method, apparatus and computer readable storage medium of heart |
Non-Patent Citations (1)
| Title |
|---|
| 叶大春等: "无创性MR在老年心功能评价中的临床应用", 《医学影像学杂志》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110853012A (en) * | 2019-11-11 | 2020-02-28 | 苏州锐一仪器科技有限公司 | Method, apparatus and computer storage medium for obtaining cardiac parameters |
| CN111062948A (en) * | 2019-11-18 | 2020-04-24 | 北京航空航天大学合肥创新研究院 | Multi-tissue segmentation method based on fetal four-chamber cardiac section image |
| CN111062948B (en) * | 2019-11-18 | 2022-09-13 | 北京航空航天大学合肥创新研究院 | Multi-tissue segmentation method based on fetal four-chamber cardiac section image |
| CN110786839A (en) * | 2019-11-22 | 2020-02-14 | 中山大学 | Method, device, equipment and medium for generating instantaneous waveform-free ratio |
| CN111127400A (en) * | 2019-11-29 | 2020-05-08 | 深圳蓝韵医学影像有限公司 | Method and device for detecting breast lesions |
| CN111126274A (en) * | 2019-12-24 | 2020-05-08 | 深圳市检验检疫科学研究院 | Method, device, equipment and medium for detecting inbound target population |
| CN111192255A (en) * | 2019-12-30 | 2020-05-22 | 上海联影智能医疗科技有限公司 | Index detection method, computer device, and storage medium |
| CN111192255B (en) * | 2019-12-30 | 2024-04-26 | 上海联影智能医疗科技有限公司 | Index detection method, computer device, and storage medium |
| CN111275686A (en) * | 2020-01-20 | 2020-06-12 | 中山大学 | Method and device for generating medical image data for artificial neural network training |
| CN111275686B (en) * | 2020-01-20 | 2023-05-26 | 中山大学 | Method and device for generating medical image data for artificial neural network training |
| CN112819789A (en) * | 2020-02-28 | 2021-05-18 | 上海联影智能医疗科技有限公司 | Apparatus and method for cardiac assessment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110400298B (en) | 2023-10-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110400298A (en) | Detection method, device, equipment and the medium of heart clinical indices | |
| US11960571B2 (en) | Method and apparatus for training image recognition model, and image recognition method and apparatus | |
| Medvedofsky et al. | 2D and 3D echocardiography-derived indices of left ventricular function and shape: relationship with mortality | |
| Young et al. | Computational cardiac atlases: from patient to population and back | |
| JP2018185856A (en) | Evolving contextual clinical data engine for medical information | |
| CN103814384B (en) | Tracking based on image | |
| CN109192305B (en) | Heart function automatic analysis method based on deep circulation neural network | |
| Banerjee et al. | A completely automated pipeline for 3D reconstruction of human heart from 2D cine magnetic resonance slices | |
| Yang et al. | 3D motion modeling and reconstruction of left ventricle wall in cardiac MRI | |
| Liu et al. | Deep learning-based automated left ventricular ejection fraction assessment using 2-D echocardiography | |
| CN110414607A (en) | Classification method, device, equipment and the medium of capsule endoscope image | |
| Van Der Geest et al. | Evaluation of a new method for automated detection of left ventricular boundaries in time series of magnetic resonance images using an active appearance motion model | |
| Ahn et al. | Unsupervised motion tracking of left ventricle in echocardiography | |
| CN111340794B (en) | Quantification method and device for coronary artery stenosis | |
| CN109620293A (en) | A kind of image-recognizing method, device and storage medium | |
| Miller et al. | An implementation of patient-specific biventricular mechanics simulations with a deep learning and computational pipeline | |
| Loncaric et al. | Integration of artificial intelligence into clinical patient management: focus on cardiac imaging | |
| Wehbe et al. | Deep learning for cardiovascular imaging: A review | |
| JP7369437B2 (en) | Evaluation system, evaluation method, learning method, trained model, program | |
| Kagiyama et al. | Machine learning in cardiovascular imaging | |
| JP7325411B2 (en) | Method and apparatus for analyzing echocardiogram | |
| Badano et al. | Artificial intelligence and cardiovascular imaging: A win-win combination. | |
| Gonzales et al. | TVnet: Automated time-resolved tracking of the tricuspid valve plane in MRI long-axis cine images with a dual-stage deep learning pipeline | |
| Wong et al. | Velocity-based cardiac contractility personalization from images using derivative-free optimization | |
| Nedadur et al. | The cardiac surgeon's guide to artificial intelligence |
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 |