CN110400298B - Method, device, equipment and medium for detecting heart clinical index - Google Patents
Method, device, equipment and medium for detecting heart clinical index Download PDFInfo
- Publication number
- CN110400298B CN110400298B CN201910667589.3A CN201910667589A CN110400298B CN 110400298 B CN110400298 B CN 110400298B CN 201910667589 A CN201910667589 A CN 201910667589A CN 110400298 B CN110400298 B CN 110400298B
- Authority
- CN
- China
- Prior art keywords
- image
- network
- heart
- cardiac
- training
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 67
- 230000000747 cardiac effect Effects 0.000 claims abstract description 162
- 238000013528 artificial neural network Methods 0.000 claims abstract description 57
- 238000012549 training Methods 0.000 claims description 121
- 238000012360 testing method Methods 0.000 claims description 102
- 230000006870 function Effects 0.000 claims description 32
- 238000013507 mapping Methods 0.000 claims description 20
- 230000002861 ventricular Effects 0.000 claims description 18
- 230000004913 activation Effects 0.000 claims description 14
- 239000003550 marker Substances 0.000 claims description 8
- 230000002107 myocardial effect Effects 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 210000003205 muscle Anatomy 0.000 claims description 6
- 230000001419 dependent effect Effects 0.000 claims description 5
- 239000000758 substrate Substances 0.000 claims 2
- 238000003384 imaging method Methods 0.000 abstract description 19
- 238000012546 transfer Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 9
- 238000009826 distribution Methods 0.000 description 9
- 238000012545 processing Methods 0.000 description 7
- 238000013527 convolutional neural network Methods 0.000 description 6
- 238000003709 image segmentation Methods 0.000 description 6
- 230000007246 mechanism Effects 0.000 description 6
- 230000003287 optical effect Effects 0.000 description 6
- 230000014509 gene expression Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000011156 evaluation Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000008485 antagonism Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 230000002457 bidirectional effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 208000019622 heart disease Diseases 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 208000031229 Cardiomyopathies Diseases 0.000 description 1
- 208000021908 Myocardial disease Diseases 0.000 description 1
- 238000005481 NMR spectroscopy Methods 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004204 blood vessel Anatomy 0.000 description 1
- 210000005242 cardiac chamber Anatomy 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000002591 computed tomography Methods 0.000 description 1
- 210000004351 coronary vessel Anatomy 0.000 description 1
- 238000013480 data collection Methods 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
- 238000000605 extraction Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 210000004165 myocardium 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
- 230000000750 progressive effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 210000004872 soft tissue Anatomy 0.000 description 1
- 238000003325 tomography Methods 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)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Pathology (AREA)
- Epidemiology (AREA)
- Databases & Information Systems (AREA)
- Biomedical Technology (AREA)
- Quality & Reliability (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The application provides a method, a device, equipment and a medium for detecting heart clinical indexes, which comprise the following steps: establishing a corresponding relation between the heart MR image and the appointed parameter by utilizing the self-learning capability of the artificial neural network; wherein the specified parameters include a cardiac CT image, a cardiac clinical index for the cardiac MR image, and a cardiac clinical index for the cardiac CT image; acquiring a current cardiac MR image of the patient; determining a current designated parameter corresponding to the current heart MR image through the corresponding relation; specifically, determining the current specified parameters corresponding to the current cardiac MR image includes: and determining the appointed parameter corresponding to the heart MR image which is the same as the current heart MR image in the corresponding relation as the current appointed parameter. The method can be used for evaluating the multi-type heart clinical indexes under different imaging modes of MR and CT. Complex relationships between multiple types of cardiac clinical indicators are mined and characterized, task correlations are obtained, and their transfer between different imaging modalities is achieved.
Description
Technical Field
The application relates to the field of medical detection, in particular to a method, a device, equipment and a medium for detecting heart clinical indexes.
Background
CT (Computed Tomography, computerized tomography) and MR (Magnetic Resonance, nuclear magnetic resonance) have greatly perfected the currently applied detection methods of myocardial perfusion imaging, electron emitter layer scanning, etc.
MR, with its unique non-invasive imaging capability for heart chambers and large blood vessels, is an important imaging modality for clinical cardiology. MR is widely used to evaluate various aspects of cardiac structure and function, with significant advances in imaging quality and acquisition time;
and CT density resolution is high, and has good specificity for qualitative diagnosis of lesions such as soft tissues and the like. And the improvement of the instantaneous definition and the volume scanning coverage speed promotes the development of the method, and the method has wide prospect as noninvasive anatomical image examination for evaluating the structural functions of coronary arteries and hearts.
However, intensive research into the mechanisms of heart disease requires that cardiac imaging examinations enter into the joint application of multiple imaging techniques, which is more beneficial in reflecting myocardial and vascular disease states, significantly improving the physician's ability to analyze abnormal changes in function.
However, the complex relationships between the clinical indices of the multiple types of heart make it difficult to learn the proper task correlations. At the same time, the indexes have different dimensions, so that the commonality and the difference of the common learning method are hard to characterize. Some indices are also subject to significant differences due to different factors, such as: the myocardial wall thickness of the different regions has a great relationship with the myocardial segment direction.
Furthermore, the differences between different imaging modalities also present challenges for cross-imaging modality estimation of multiple types of cardiac clinical indicators, such as: MR and CT performance differ significantly.
Disclosure of Invention
In view of the problems, the present application has been made to provide a method, apparatus, device and medium for detecting a cardiac clinical index that overcomes the problems or at least partially solves the problems, including:
a method for detecting a clinical cardiac marker, comprising:
establishing a corresponding relation between the heart MR image and the appointed parameter by utilizing the self-learning capability of the artificial neural network; wherein the specified parameters include a cardiac CT image, a cardiac clinical index for the cardiac MR image, and a cardiac clinical index for the cardiac CT image;
acquiring a current cardiac MR image of the patient;
determining a current designated parameter corresponding to the current cardiac MR image through the corresponding relation; specifically, determining the current specified parameters corresponding to the current cardiac MR image comprises: and determining the appointed parameter corresponding to the heart MR image which is the same as the current heart MR image in the corresponding relation as the current appointed parameter.
Further, the method comprises the steps of,
the cardiac clinical index includes at least one of a left ventricular epicardial contour, a left ventricular endocardial contour, a left ventricular muscle contour, a left ventricular epicardial position, a left ventricular endocardial position, a left ventricular muscle position, a local chamber wall thickness WT, a chamber volume Dim, and a chamber wall and a myocardial Area.
Further, the method comprises the steps of,
establishing a correspondence between the cardiac MR image and the specified parameter, comprising:
acquiring sample data for establishing a correspondence between the specified parameter and the cardiac MR image;
analyzing the characteristics and the rules of the heart MR image, and determining the network structure and the network parameters of the artificial neural network according to the characteristics and the rules;
training and testing the network structure and the network parameters using the sample data, determining the correspondence of the specified parameters to the cardiac MR image.
Further, the method comprises the steps of,
the step of acquiring sample data for establishing a correspondence between the specified parameter and the cardiac MR image, includes:
collecting MR images of the heart and specified parameters of different patients;
analyzing the heart MR image, and selecting data related to the specified parameters as the heart MR image by combining pre-stored expert experience information;
and taking the designated parameters and the selected data pair formed by the heart MR image as sample data.
Further, the method comprises the steps of,
the network structure comprises: a multi-task learning neural network, and, a reverse mapping neural network;
And/or the number of the groups of groups,
the network parameters include: the number of input layers, the number of output layers, the number of convolution layers, the number of dense blocks, the dense blocks comprising at least one of the number of layers, the initial weight, and the offset value.
Further, the method comprises the steps of,
training the network structure and the network parameters, including:
selecting a part of data in the sample data as a training sample, inputting the heart MR image in the training sample into the network structure, and training through an activation function of the network structure and the network parameters to obtain an actual training result;
determining whether an actual training error between the actual training result and a corresponding designated parameter in the training sample meets a preset training error;
when the actual training error meets the preset training error, determining that the training of the network structure and the network parameters is completed;
and/or the number of the groups of groups,
testing the network structure and the network parameters, including:
selecting another part of data in the sample data as a test sample, inputting the heart MR image in the test sample into the network structure with the training completed, and testing by using the activation function and the network parameters with the training completed to obtain an actual test result;
Determining whether an actual test error between the actual test result and a corresponding specified parameter in the test sample meets a set test error;
and when the actual test error meets the set test error, determining that the test on the network structure and the network parameters is completed.
Further, the method comprises the steps of,
training the network structure and the network parameters, further comprising:
when the actual training error does not meet the set training error, updating the network parameters through an error energy function of the network structure;
retraining through the activation function of the network structure and the updated network parameters until the retrained actual training error meets the set training error;
and/or the number of the groups of groups,
testing the network structure and the network parameters, further comprising:
and when the actual test error does not meet the set test error, retraining the network structure and the network parameters until the retrained actual test error is slower than the set test error.
A computing device for cardiac clinical metrics, comprising:
the building module is used for building a corresponding relation between the heart MR image and the appointed parameter by utilizing the self-learning capability of the artificial neural network; wherein the specified parameters include a cardiac CT image, a cardiac clinical index for the cardiac MR image, and a cardiac clinical index for the cardiac CT image;
An acquisition module for acquiring a current cardiac MR image of a patient;
the determining module is used for determining a current appointed parameter corresponding to the current heart MR image through the corresponding relation; specifically, determining the current specified parameters corresponding to the current cardiac MR image comprises: and determining the appointed parameter corresponding to the heart MR image which is the same as the current heart MR image in the corresponding relation as the current appointed parameter.
An apparatus comprising a processor, a memory and a computer program stored on the memory and capable of running on the processor, which when executed by the processor, performs the steps of the method of detecting a cardiac clinical marker as described above.
A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of a method of detecting a cardiac clinical marker as described above.
The application has the following advantages:
in the embodiment of the application, the self-learning capability of the artificial neural network is utilized to establish the corresponding relation between the MR image of the heart and the appointed parameter; wherein the specified parameters include a cardiac CT image, a cardiac clinical index for the cardiac MR image, and a cardiac clinical index for the cardiac CT image; acquiring a current cardiac MR image of the patient; determining a current designated parameter corresponding to the current cardiac MR image through the corresponding relation; specifically, determining the current specified parameters corresponding to the current cardiac MR image comprises: and determining the appointed parameter corresponding to the heart MR image which is the same as the current heart MR image in the corresponding relation as the current appointed parameter, and being applicable to the evaluation of multi-type heart clinical indexes under different imaging modes of MR and CT. The method comprises the steps of mining and characterizing complex relations among multi-type heart clinical indexes, acquiring task correlations, realizing transfer of the task correlations in different imaging modes, and establishing knowledge sharing mechanisms in different imaging modes.
Drawings
In order to more clearly illustrate the technical solutions of the present application, the drawings that are needed in the description of the present application will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort to a person skilled in the art.
FIG. 1 is a flowchart illustrating steps of a method for detecting a cardiac clinical index according to an embodiment of the present application;
FIG. 2 is a schematic diagram of an artificial neural network for a method for detecting cardiac clinical indicators according to an embodiment of the present application;
FIG. 3 is a schematic diagram of a multi-task learning network for a method for detecting cardiac clinical indicators according to an embodiment of the present application;
FIG. 4 is a schematic diagram of an inverse mapping network of a method for detecting cardiac clinical indicators according to an embodiment of the present application;
FIG. 5 is a simplified test result diagram of a model of estimating linear and planar indicators compared with different networks according to a method for detecting cardiac clinical indicators according to an embodiment of the present application;
FIG. 6 is a simplified test result diagram of a two-dimensional cardiac image segmentation compared with different networks for a method for detecting cardiac clinical indicators according to an embodiment of the present application;
FIG. 7 is a schematic diagram of transmission learning results of four depth networks under different configurations and transmission learning results of three-dimensional DenseNet according to a method for detecting cardiac clinical indicators according to an embodiment of the present application;
FIG. 8 is a schematic view of MR and CT image samples of a method for detecting a clinical cardiac marker according to an embodiment of the present application;
FIG. 9 is a schematic diagram showing a comparison of single frame estimation errors of three cardiac markers for different networks of a method for detecting cardiac clinical markers according to an embodiment of the present application;
FIG. 10 is a block diagram of a device for detecting clinical cardiac markers according to an embodiment of the present application;
fig. 11 is a schematic structural diagram of a computer device according to an embodiment of the present application.
Detailed Description
In order that the manner in which the above recited objects, features and advantages of the present application are obtained will become more readily apparent, a more particular description of the application briefly described above will be rendered by reference to the appended drawings. It will be apparent that the described embodiments are some, but not all, embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Referring to fig. 1, a method for detecting a clinical cardiac marker according to an embodiment of the present application includes:
s110, establishing a corresponding relation between a heart MR image and a specified parameter by utilizing the self-learning capability of the artificial neural network; wherein the specified parameters include a cardiac CT image, a cardiac clinical index for the cardiac MR image, and a cardiac clinical index for the cardiac CT image;
s120, acquiring a current heart MR image of a patient;
s130, determining a current appointed parameter corresponding to the current heart MR image through the corresponding relation; specifically, determining the current specified parameters corresponding to the current cardiac MR image comprises: and determining the appointed parameter corresponding to the heart MR image which is the same as the current heart MR image in the corresponding relation as the current appointed parameter.
In the embodiment of the application, the self-learning capability of the artificial neural network is utilized to establish the corresponding relation between the MR image of the heart and the appointed parameter; wherein the specified parameters include a cardiac CT image, a cardiac clinical index for the cardiac MR image, and a cardiac clinical index for the cardiac CT image; acquiring a current cardiac MR image of the patient; determining a current designated parameter corresponding to the current cardiac MR image through the corresponding relation; specifically, determining the current specified parameters corresponding to the current cardiac MR image comprises: and determining the appointed parameter corresponding to the heart MR image which is the same as the current heart MR image in the corresponding relation as the current appointed parameter, and being applicable to the evaluation of multi-type heart clinical indexes under different imaging modes of MR and CT. The method comprises the steps of mining and characterizing complex relations among multi-type heart clinical indexes, acquiring task correlations, realizing transfer of the task correlations in different imaging modes, and establishing knowledge sharing mechanisms in different imaging modes.
Next, a method of detecting a heart clinical index according to the present exemplary embodiment will be further described.
As described in the above step S110, the self-learning ability of the artificial neural network is utilized to establish the correspondence between the MR image of the heart and the specified parameters; wherein the specified parameters include a cardiac CT image, a cardiac clinical index for a cardiac MR image, and a cardiac clinical index for a cardiac CT image
For example: the display state rule of the specified parameters in the heart MR image is analyzed by utilizing an artificial neural network algorithm, and the mapping rule between the heart MR image of the patient and the specified parameters is found through the self-learning and self-adapting characteristics of the artificial neural network.
For example: the artificial neural network algorithm can be utilized, the heart MR images of a plurality of different volunteers (including but not limited to one or more of age, whether heart disease history, gender, illness state and the like) are collected together, heart MR images of a plurality of volunteers and specified parameters are selected as sample data, the neural network is learned and trained, the relation between the heart MR images and the specified parameters is fitted by the neural network through adjusting the weight between the network structure and the network nodes, and finally the neural network can accurately fit the corresponding relation between the heart MR images and the specified parameters of different patients.
Referring to fig. 2, in the embodiment of the present application, training an artificial neural network (a multi-task learning neural network and an inverse mapping neural network) is preferably performed by using a cardiac MR image and a cardiac clinical index corresponding to the cardiac MR image, the trained artificial neural network is finally obtained by performing countermeasure training on input and output references of the artificial neural network by using two determiner networks with different dependent parameters and the same structure, and then, based on network parameters of the trained artificial neural network, a CT image corresponding to the cardiac MR image and a cardiac clinical index thereof are trained, thereby obtaining a correspondence between the cardiac MR image and a specified parameter (cardiac CT image, cardiac clinical index for the cardiac MR image, and cardiac clinical index for the cardiac CT image).
Specifically, the parameters learned by each layer of the multi-task learning network and each layer of the inverse mapping network are firstly trained by adopting an antagonism training method in the data environment of the heart MR image; and then, carrying out fine adjustment on the parameters obtained by training by using the heart CT image data with the labels through a loss function, and finally, training the parameters after fine adjustment by adopting an antagonism training method in the data environment of the heart CT image, wherein the loss function is shown in the following formula (17).
In one embodiment, the cardiac clinical index comprises: left ventricular epicardium contour, left ventricular endocardial contour, left ventricular muscle contour, left ventricular epicardium position, left ventricular endocardial position, left ventricular muscle position, local chamber wall thickness WT (Regional Wall Thinknesses), chamber volume Dim (Directional Dimensions of Cavity), and at least one of chamber wall and myocardial area Area (Areas of Cavity and Myocardium).
In an embodiment, the specific procedure of "establishing correspondence between the cardiac MR image and the specified parameters" in step S110 may be further described in conjunction with the following description.
As described in the following steps: acquiring sample data for establishing a correspondence between the specified parameter and the cardiac MR image;
in a further embodiment, the specific procedure of "obtaining sample data that can be used to establish a correspondence between the operating parameters and the frosting status" can be further described in connection with the following description.
As described in the following steps: collecting MR images of the heart and specified parameters of different patients;
for example: data collection: collecting heart MR images of patients with different health conditions and corresponding specified parameters; collecting heart MR images of patients of different ages and corresponding specified parameters; and collecting the heart MR images of the patients with different sexes and corresponding specified parameters.
Therefore, the operation data are collected through various ways, the quantity of the operation data is increased, the learning capacity of the artificial neural network is improved, and the accuracy and the reliability of the determined corresponding relation are further improved.
As described in the following steps: analyzing the heart MR image, and combining pre-stored expert experience information, and selecting data related to the specified parameters as the heart MR image (for example, selecting the heart MR image which has influence on the specified parameters as input parameters and the specified parameters as output parameters);
for example: by taking the MR image of the heart in the relevant data of the diagnosed volunteer as input parameter, the specified parameter in its relevant data is taken as output parameter.
As described in the following steps: and taking the designated parameters and the selected data pair formed by the heart MR image as sample data.
For example: the obtained input and output parameter pairs are used as training sample data, and are used as test sample data.
Therefore, the collected heart MR image is analyzed and processed, so that sample data is obtained, the operation process is simple, and the reliability of an operation result is high.
As described in the following steps: and analyzing the characteristics and the rules of the heart MR image, and determining the network structure and the network parameters of the artificial neural network according to the characteristics and the rules.
For example: according to the data characteristics of different ages, illness states, sexes and the like which have influences on the heart image and the rules contained in the data characteristics, the basic structure of the network, the number of input nodes and output nodes of the network, the number of hidden nodes of the network, the initial weight of the network and the like can be preliminarily determined.
Preferably, the network structure comprises: a multitasking learning neural network, and an inverse mapping neural network.
Preferably, the network parameters include: the number of input layers, the number of output layers, the number of convolution layers, the number of dense blocks, the dense blocks comprising at least one of the number of layers, the initial weight, and the offset value.
For example: as shown in fig. 3, the multi-task learning network is modified by modifying a dense link convolutional neural network (Densely Connected Convolutional Networks, denseNet). The built multi-task learning network is mainly composed of a DenseNet, which comprises an input layer, an output layer, a convolution layer Mcnn, a 4-layer dense block (Md 1), two 8-layer dense blocks (Md 2 and Md 3), a deconvolution layer Mdcnn and a full connection layer Mfc. All the convolution layers use three-dimensional convolution neural networks (Convolutional Neural Networks, CNN) with excellent behavior and capacity modeling capability, wherein the convolution layers contain three convolution kernels with different sizes of 33, 53 and 73 respectively, and the three-dimensional convolution layers in the dense block use convolution kernels with the size of 33.
The input layer is a heart MR image, after passing through the convolution layer and three dense blocks, the network learns a joint distribution, and obtains feature map (fea), and then a pixel-level classifier of the deconvolution layer obtains a two-dimensional image segmentation result. Wherein fea is used as the input of the full convolution layer, and the estimated one-dimensional heart index is output through a regression network.
The inverse mapping network has a structure similar to the structure of the above-mentioned multi-task learning network but opposite to the structure, and is modified and turned in direction as shown in fig. 4. The inverse mapping network is built up by using an input layer, an output layer, a convolution layer Rcnn, a 4-layer dense block (Rd 1), two 8-layer dense blocks (Rd 2 and Rd 3), a deconvolution layer Rdcnn and a full connection layer Rfc. All the convolution layers use three-dimensional convolution neural networks (Convolutional Neural Networks, CNN) with excellent behavior and capacity modeling capability, wherein the convolution layers contain three convolution kernels with different sizes of 33, 53 and 73 respectively, and the three-dimensional convolution layers in the dense block use convolution kernels with the size of 33. Wherein, two-dimensional image segmentation and one-dimensional heart index are taken as input, and the output result is a reconstructed heart MR image.
Therefore, the data is analyzed and processed through the joint multi-task learning neural network and the inverse mapping neural network, so that the accuracy of data processing can be improved, and the reliability of determining the corresponding relation between the heart MR image and the specified parameter can be improved.
As described in the following steps: training and testing the network structure and the network parameters using the sample data, determining the correspondence of the specified parameters to the cardiac MR image.
For example: after the network design is completed, training sample data is needed to train the designed neural network. The training method can be adjusted according to the actual network structure and problems found in training.
Therefore, through collecting image data, selecting sample data from the image data, training and testing based on the sample data, the corresponding relation between the MR image of the heart and the specified parameters is determined, and the accuracy of generating the specified parameters is improved.
Preferably, the network is trained by adopting an countermeasure training mode, two neural networks (a multi-task learning network and an inverse mapping network) are connected in different sequences (the connecting mode comprises the steps of connecting the multi-task learning network with the inverse mapping network, connecting the inverse mapping network with the multi-task learning network), learning training is respectively carried out, and a joint distribution is respectively obtained and matched.
Then, two structurally similar discriminator networks with different dependent parameters are constructed, wherein the dependent parameters are psi 1 and psi 2 in the following formulas (7) and (8), and the two discriminator networks respectively correspond to one of the two discriminator networks and are not repeated. For each different connection mode, the input-output pairs are extracted from the corresponding joint distribution, and the input-output pairs are respectively distinguished by a discriminator network. If the input and output samples are difficult to distinguish by the arbiter, this means that the estimated specified parameters can well reflect the real image, and the reconstructed image can well approximate the real specified parameters.
Optionally, the specific procedure of training the network structure and the network parameters in the step of training and testing the network structure and the network parameters using the sample data, determining the correspondence of the specified parameters to the cardiac MR image may be further described in connection with the following description.
Selecting a part of data in the sample data as a training sample, inputting the heart MR image in the training sample into the network structure, and training through an activation function of the network structure and the network parameters to obtain an actual training result; determining whether an actual training error between the actual training result and a corresponding designated parameter in the training sample meets a preset training error; when the actual training error meets the preset training error, determining that the training of the network structure and the network parameters is completed;
For example: the design of the artificial neural network further learns to generate the inverse mapping of X, g2:i→x, from I by giving a cardiac image X, which helps reveal the complex relationship between X and I. In the constructed network, G1 and G2 can be expressed as:
(1)
(2)
connecting two networks according to the sequence of G1 and G2 to obtain a functionThe expression is as follows:
(3)
then connecting two networks according to the sequence of G2 and G1 to obtain a functionThe expression is as follows:
(4)
multiple types of cardiac markers describe features of cardiac structures from different dimensions. If the complex correlations in the multi-type cardiac markers and the semantic relationships between the multi-type cardiac markers and the cardiac images are known, not only can the multi-type cardiac markers be estimated, but also the cardiac images can be reconstructed therefrom. To this end, consider the variational expressions for f and g:
)(5)
(6)
the challenge training is designed by taking into account the following sampling method to learn and match the two distributions. For the function f and the real multi-type cardiac index i, the pair (x, i) can be extracted from the distribution q (x) q (i|x) and the distribution p (i) p (x|i'), distinguished by the real-type discriminator network tψ1 (x, i). Likewise, for the function g and the real heart image x, the pairs (x, i) can be extracted from the distributions p (i) p (x|i) and q (x) q (i|x') and distinguished again by the real type of arbiter network tψ2 (x, i), where ψ1, ψ2 are parameters learned by both arbiter networks.
Specifically, to achieve the above function, consider that the arbiter tψ1 maximizes the value of the following equation:
(7)
and the arbiter tψ2 maximizes the value of the following equation:
(8)
wherein σ represents a sigmond function whose expression is defined as
(9)
The second term in equation (7) is rewritten:
(10)
thus equation (7) can be written as:
(11)
the integral of formula (11) if and only if the integral of (x, i) takes a maximum value, with respect toIs a function of (c), taking the maximum value. Note that the function->At->The maximum value is obtained, and the combination formula (9) is as follows:
(12)
i.e. the
(13)
Is of the same kind
(14)
The two above formulas are the most ideal and optimized discriminants. Finally, in combination with the formulas (5) (6), the optimization object can be expressed as a maximum and minimum problem:
(15)
in addition, the parameters corresponding to the implementation process are setFor, there is
(16)
Finally, the framework is trimmed by a loss function using the Dice loss for two-dimensional cardiac image segmentation in a multi-task learning networkWhile using +.>(i.e., ridge regression) loss. In the inverse mapping network, a reconstructed loss function is used>. Eventually, the entire loss can be written as
(17)
In the above formula:
x: a source imaging modality image sample; x': reconstructing a heart image; x: a source imaging modality image sample set; i: a multi-type cardiac marker sample; i': estimating heart indexes; i: a multi-type cardiac marker sample set; phi: the multitask learning network learns distributed parameters; θ: inverse mapping network learning distributed parameters; q phi (i, x): the multi-task learning network learns the joint distribution, and based on the parameter phi, q phi (i|x) q (x) can be written; pθ (x, i): the inverse mapping network learns a joint distribution based on the parameter θ, which can be written as pθ (x|i) p (x).
More optionally, training the network structure and the network parameters further includes:
when the actual training error does not meet the set training error, updating the network parameters through an error energy function of the network structure; retraining through the activation function of the network structure and the updated network parameters until the retrained actual training error meets the set training error;
for example: if the test error meets the requirement, the network training test is completed.
Therefore, the test samples are used for testing the network structure and the network parameters obtained through training, so that the reliability of the network structure and the network parameters is further verified.
Optionally, the specific procedure of testing the network structure and the network parameters in the step of training and testing the network structure and the network parameters using the sample data, determining the correspondence of the specified parameters to the cardiac MR image may be further described in connection with the following description.
Selecting another part of the sample data as a test sample, inputting the heart MR image in the test sample into the trained network structure, and testing by using the activation function and the trained network parameters to obtain an actual test result; determining whether an actual test error between the actual test result and a corresponding specified parameter in the test sample meets a set test error; and when the actual test error meets the set test error, determining that the test on the network structure and the network parameters is completed.
And (3) making a judgment standard:
(1) and analyzing the result of the two-dimensional heart image segmentation, wherein the evaluation is mainly performed by using a Dice coefficient, and the expression is as follows:
where PT denotes all pixels of the manually segmented contour region, PE denotes all pixels of the automatically segmented contour region, and PTE denotes overlapping pixels between PT and PE. The higher the Dice value, the higher the consistency between automatic and manual segmentation.
(2) For the one-dimensional linear index and the plane index, the average absolute error (Mean Absolute Error, MAE) is calculated, expressed as:
wherein the method comprises the steps of，/>The two vectors contain the actual index and the estimated index, respectively. The smaller the MAE value, the higher the accuracy of the estimation.
More optionally, the testing the network structure and the network parameters further includes:
and when the actual test error does not meet the set test error, retraining the network structure and the network parameters until the retrained actual test error is slower than the set test error.
For example: and when the test error does not meet the requirement, repeating the steps and retraining the network.
Therefore, the network structure is retrained to be retested when the test error is larger, so that more accurate and reliable network structure is obtained, and the accuracy of determining the frosting state is improved.
In the specific implementation of the method of the present application,
referring to fig. 5, (1) evaluation aspect of the index: through comparative research, the method of the application has better performance on MR data than the existing unidirectional method. Such as fullvnet and DMTRL have the advantage of a combination of CNN and recurrent neural networks (Recurrent Neural Network, RNN), but they all rely very much on a particular constraint on task dependency learning and do not have an inverse mapping network.
The method of the application is still more advantageous for other learning frameworks with inverse mapping networks.
Referring to fig. 7, (2) bidirectional parameter sharing aspect: first, a plurality of multi-task learning frameworks are established. The pixel-level classifier, regression network, and joint representation network of these multitasking frameworks are structurally identical, but the feature extraction hierarchy is different. The frame is directly trained on target data without parameter sharing (No sharing); (ii) Adopting a traditional unidirectional parameter sharing mechanism (One-way-para); (iii) employing a Bi-directional parameter sharing mechanism (Bi-para). The comparison result shows that the bidirectional parameter sharing mechanism is effective for index estimation across imaging modes in the multi-task learning framework.
Referring to fig. 8, in general, the method of the present application is capable of accurately locating cardiac structures not only in terms of image segmentation, but also in terms of index estimation, wherein "Our" in fig. 9 is a method of an embodiment of the present application.
For the device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments for relevant points.
Referring to fig. 10, a cardiac clinical index calculating apparatus according to an embodiment of the present application includes:
the establishing module 110 is configured to establish a correspondence between the MR image of the heart and the specified parameter by using the self-learning capability of the artificial neural network; wherein the specified parameters include a cardiac CT image, a cardiac clinical index for the cardiac MR image, and a cardiac clinical index for the cardiac CT image;
an acquisition module 120 for acquiring a current cardiac MR image of the patient;
a determining module 130, configured to determine, according to the correspondence, a current specified parameter corresponding to the current cardiac MR image; specifically, determining the current specified parameters corresponding to the current cardiac MR image comprises: and determining the appointed parameter corresponding to the heart MR image which is the same as the current heart MR image in the corresponding relation as the current appointed parameter.
In an embodiment, the image features include: two-dimensional cardiac images and one-dimensional cardiac clinical indices.
In one embodiment, the establishing module 310 includes:
an acquisition sub-module for acquiring sample data for establishing a correspondence between the specified parameter and the cardiac MR image;
the analysis submodule is used for analyzing the characteristics and the rules of the heart MR image and determining the network structure and the network parameters of the artificial neural network according to the characteristics and the rules;
and the training sub-module is used for training and testing the network structure and the network parameters by using the sample data, and determining the corresponding relation between the specified parameters and the heart MR image.
In an embodiment, the acquiring sub-module includes:
a collection sub-module for collecting cardiac MR images and specified parameters of different patients;
the analysis sub-module is used for analyzing the heart MR image and combining pre-stored expert experience information to select data related to the specified parameters as the heart MR image;
and the sample data generation submodule is used for taking the specified parameters and the data pairs formed by the selected heart MR images as sample data.
In one embodiment of the present invention, in one embodiment,
the network structure comprises: a multi-task learning neural network, and, a reverse mapping neural network;
and/or the number of the groups of groups,
the network parameters include: the number of input layers, the number of output layers, the number of convolution layers, the number of dense blocks, the dense blocks comprising at least one of the number of layers, the initial weight, and the offset value.
In one embodiment of the present invention, in one embodiment,
the training sub-module comprises:
the training result generation sub-module is used for selecting a part of data in the sample data as a training sample, inputting the heart MR image in the training sample into the network structure, and training through an activation function of the network structure and the network parameters to obtain an actual training result;
the training result error judging sub-module is used for determining whether the actual training error between the actual training result and the corresponding appointed parameter in the training sample meets the preset training error;
the training completion judging sub-module is used for determining that the training of the network structure and the network parameters is completed when the actual training error meets the preset training error;
and/or the number of the groups of groups,
a testing sub-module, configured to test the network structure and the network parameter, where the testing sub-module includes:
The test result generation sub-module is used for selecting another part of data in the sample data as a test sample, inputting the heart MR image in the test sample into the network structure after the training is finished, and testing by the activation function and the network parameters after the training is finished to obtain an actual test result;
the test result error judging sub-module is used for determining whether the actual test error between the actual test result and the corresponding specified parameter in the test sample meets the set test error;
and the test completion judging sub-module is used for determining that the test on the network structure and the network parameters is completed when the actual test error meets the set test error.
In one embodiment of the present invention, in one embodiment,
the training submodule further includes:
a network parameter updating sub-module, configured to update the network parameter through an error energy function of the network structure when the actual training error does not meet the set training error;
the first retraining sub-module is used for retraining through the activation function of the network structure and the updated network parameters until the retrained actual training error meets the set training error;
And/or the number of the groups of groups,
the test sub-module further comprises:
and the second retraining sub-module is used for retraining the network structure and the network parameters when the actual test error does not meet the set test error until the retrained actual test error is slower than the set test error.
Referring to fig. 11, a computer device for a method for detecting a cardiac clinical index according to the present invention may specifically include the following:
the computer device 12 described above is embodied in the form of a general purpose computing device, and the components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, a bus 18 that connects the various system components, including the system memory 28 and the processing units 16.
Bus 18 represents one or more of several types of bus 18 structures, including a memory bus 18 or memory controller, a peripheral bus 18, an accelerated graphics port, a processor, or a local bus 18 using any of a variety of bus 18 architectures. By way of example, and not limitation, such architectures include Industry Standard Architecture (ISA) bus 18, micro channel architecture (MAC) bus 18, enhanced ISA bus 18, video Electronics Standards Association (VESA) local bus 18, and Peripheral Component Interconnect (PCI) bus 18.
Computer device 12 typically includes a variety of computer system readable media. Such media can be any available media that is accessible by computer device 12 and includes both volatile and nonvolatile media, removable and non-removable media.
The system memory 28 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM) 30 and/or cache memory 32. The computer device 12 may further include other removable/non-removable, volatile/nonvolatile computer storage media. By way of example only, storage system 34 may be used to read from or write to non-removable, nonvolatile magnetic media (commonly referred to as a "hard disk drive"). Although not shown in fig. 11, a magnetic disk drive for reading from and writing to a removable non-volatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading from or writing to a removable non-volatile optical disk such as a CD-ROM, DVD-ROM, or other optical media may be provided. In such cases, each drive may be coupled to bus 18 through one or more data medium interfaces. The memory may include at least one program product having a set (e.g., at least one) of program modules 42, the program modules 42 being configured to carry out the functions of embodiments of the invention.
A program/utility 40 having a set (at least one) of program modules 42 may be stored in, for example, a memory, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules 42, and program data, each or some combination of which may include an implementation of a network environment. Program modules 42 generally perform the functions and/or methods of the embodiments described herein.
The computer device 12 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, camera, etc.), one or more devices that enable a user to interact with the computer device 12, and/or any devices (e.g., network card, modem, etc.) that enable the computer device 12 to communicate with one or more other computing devices. Such communication may occur through an input/output (I/O) interface 22. Moreover, computer device 12 may also communicate with one or more networks such as a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network such as the Internet, through network adapter 20. As shown, network adapter 20 communicates with other modules of computer device 12 via bus 18. It should be appreciated that although not shown in fig. 11, other hardware and/or software modules may be used in connection with computer device 12, including, but not limited to: microcode, device drivers, redundant processing units 16, external disk drive arrays, RAID systems, tape drives, data backup storage systems 34, and the like.
The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, for example, to implement the method for detecting cardiac clinical indicators provided by the embodiment of the present application.
That is, the processing unit 16 realizes when executing the program: establishing a corresponding relation between the heart MR image and the appointed parameter by utilizing the self-learning capability of the artificial neural network; wherein the specified parameters include a cardiac CT image, a cardiac clinical index for the cardiac MR image, and a cardiac clinical index for the cardiac CT image; acquiring a current cardiac MR image of the patient; determining a current designated parameter corresponding to the current cardiac MR image through the corresponding relation; specifically, determining the current specified parameters corresponding to the current cardiac MR image comprises: and determining the appointed parameter corresponding to the heart MR image which is the same as the current heart MR image in the corresponding relation as the current appointed parameter.
In an embodiment of the present application, the present application further provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method for detecting a cardiac clinical index as provided in all embodiments of the present application:
That is, the program is implemented when executed by a processor: establishing a corresponding relation between the heart MR image and the appointed parameter by utilizing the self-learning capability of the artificial neural network; wherein the specified parameters include a cardiac CT image, a cardiac clinical index for the cardiac MR image, and a cardiac clinical index for the cardiac CT image; acquiring a current cardiac MR image of the patient; determining a current designated parameter corresponding to the current cardiac MR image through the corresponding relation; specifically, determining the current specified parameters corresponding to the current cardiac MR image comprises: and determining the appointed parameter corresponding to the heart MR image which is the same as the current heart MR image in the corresponding relation as the current appointed parameter.
Any combination of one or more computer readable media may be employed. The computer readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPOM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Computer program code for carrying out operations of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider). In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described by differences from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other.
While preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the scope of the embodiments of the application.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.
The above detailed description of the method, device, equipment and medium for detecting heart clinical index provided by the application applies specific examples to illustrate the principle and implementation of the application, and the above examples are only used to help understand the method and core idea of the application; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present application, the present description should not be construed as limiting the present application in view of the above.
Claims (6)
1. A method for detecting a clinical cardiac marker, comprising:
establishing a corresponding relation between a heart MR image and a specified parameter by utilizing the self-learning capability of an artificial neural network, wherein the artificial neural network comprises a network structure and a network parameter, and the network structure comprises a multi-task learning neural network and an inverse mapping neural network; training the artificial neural network through the heart MR image and the appointed parameters corresponding to the heart MR image, performing countermeasure training on input and output references of the artificial neural network through a first judging device network and a second judging device network to obtain the trained artificial neural network, wherein the two dependent parameters of the first judging device network and the second judging device network are different and have the same structure, and training the appointed parameters corresponding to the heart MR image based on the trained network parameters of the artificial neural network, so that the corresponding relation between the heart MR image and the appointed parameters is obtained; wherein the specified parameters include a cardiac CT image, a cardiac clinical index for the cardiac MR image, and a cardiac clinical index for the cardiac CT image; the cardiac clinical index includes: a left ventricular epicardium contour, a left ventricular endocardial contour, a left ventricular muscle contour, a left ventricular epicardium position, a left ventricular endocardial position, a left ventricular muscle position, a local chamber wall thickness WT, a chamber volume Dim, and at least one of a chamber wall and a myocardial Area; the establishing a correspondence between the cardiac MR image and the specified parameter includes: acquiring sample data for establishing a correspondence between the specified parameter and the cardiac MR image; analyzing the characteristics and the rules of the heart MR image, and determining the network structure and the network parameters of the artificial neural network according to the characteristics and the rules; training and testing the network structure and the network parameters by using the sample data, and determining the corresponding relation between the specified parameters and the heart MR image; the step of acquiring sample data for establishing a correspondence between the specified parameter and the cardiac MR image, includes: collecting MR images of the heart and specified parameters of different patients; analyzing the heart MR image, and selecting data related to the specified parameters as the heart MR image by combining pre-stored expert experience information; taking the specified parameters and the selected data pairs formed by the heart MR images as sample data; training the network structure and the network parameters, including: selecting a part of data in the sample data as a training sample, inputting the heart MR image in the training sample into the network structure, and training through an activation function of the network structure and the network parameters to obtain an actual training result; determining whether an actual training error between the actual training result and a corresponding designated parameter in the training sample meets a preset training error; when the actual training error meets the preset training error, determining that the training of the network structure and the network parameters is completed; and/or testing the network structure and the network parameters, including: selecting another part of data in the sample data as a test sample, inputting the heart MR image in the test sample into the network structure with the training completed, and testing by using the activation function and the network parameters with the training completed to obtain an actual test result; determining whether an actual test error between the actual test result and a corresponding specified parameter in the test sample meets a set test error; when the actual test error meets the set test error, determining that the test on the network structure and the network parameters is completed;
Acquiring a current cardiac MR image of the patient;
determining a current designated parameter corresponding to the current cardiac MR image through the corresponding relation; determining a current specified parameter corresponding to the current cardiac MR image, comprising: and determining the appointed parameter corresponding to the heart MR image which is the same as the current heart MR image in the corresponding relation as the current appointed parameter.
2. The method of claim 1, wherein the step of determining the position of the substrate comprises,
the network parameters include: the number of input layers, the number of output layers, the number of convolution layers, the number of dense blocks, the dense blocks comprising at least one of the number of layers, the initial weight, and the offset value.
3. The method of claim 1, wherein the step of determining the position of the substrate comprises,
training the network structure and the network parameters, further comprising:
when the actual training error does not meet the set training error, updating the network parameters through an error energy function of the network structure;
retraining through the activation function of the network structure and the updated network parameters until the retrained actual training error meets the set training error;
and/or the number of the groups of groups,
Testing the network structure and the network parameters, further comprising:
and when the actual test error does not meet the set test error, retraining the network structure and the network parameters until the retrained actual test error is slower than the set test error.
4. A computing device for cardiac clinical metrics, comprising:
the building module is used for building a corresponding relation between the heart MR image and the appointed parameter by utilizing the self-learning capability of the artificial neural network; training the artificial neural network through the heart MR image and the appointed parameters corresponding to the heart MR image, performing countermeasure training on input and output references of the artificial neural network through a first judging device network and a second judging device network to obtain the trained artificial neural network, wherein the two dependent parameters of the first judging device network and the second judging device network are different and have the same structure, and training the appointed parameters corresponding to the heart MR image based on the trained network parameters of the artificial neural network, so that the corresponding relation between the heart MR image and the appointed parameters is obtained; wherein the specified parameters include a cardiac CT image, a cardiac clinical index for the cardiac MR image, and a cardiac clinical index for the cardiac CT image; the establishing a correspondence between the cardiac MR image and the specified parameter includes: acquiring sample data for establishing a correspondence between the specified parameter and the cardiac MR image; analyzing the characteristics and the rules of the heart MR image, and determining the network structure and the network parameters of the artificial neural network according to the characteristics and the rules; training and testing the network structure and the network parameters by using the sample data, and determining the corresponding relation between the specified parameters and the heart MR image; the step of acquiring sample data for establishing a correspondence between the specified parameter and the cardiac MR image, includes: collecting MR images of the heart and specified parameters of different patients; analyzing the heart MR image, and selecting data related to the specified parameters as the heart MR image by combining pre-stored expert experience information; taking the specified parameters and the selected data pairs formed by the heart MR images as sample data; training the network structure and the network parameters, including: selecting a part of data in the sample data as a training sample, inputting the heart MR image in the training sample into the network structure, and training through an activation function of the network structure and the network parameters to obtain an actual training result; determining whether an actual training error between the actual training result and a corresponding designated parameter in the training sample meets a preset training error; when the actual training error meets the preset training error, determining that the training of the network structure and the network parameters is completed; and/or testing the network structure and the network parameters, including: selecting another part of data in the sample data as a test sample, inputting the heart MR image in the test sample into the network structure with the training completed, and testing by using the activation function and the network parameters with the training completed to obtain an actual test result; determining whether an actual test error between the actual test result and a corresponding specified parameter in the test sample meets a set test error; when the actual test error meets the set test error, determining that the test on the network structure and the network parameters is completed;
An acquisition module for acquiring a current cardiac MR image of a patient;
the determining module is used for determining a current appointed parameter corresponding to the current heart MR image through the corresponding relation; specifically, determining the current specified parameters corresponding to the current cardiac MR image comprises: and determining the appointed parameter corresponding to the heart MR image which is the same as the current heart MR image in the corresponding relation as the current appointed parameter.
5. An apparatus comprising a processor, a memory, and a computer program stored on the memory and capable of running on the processor, which when executed by the processor, implements the method of any one of claims 1 to 3.
6. A computer readable storage medium, characterized in that it has stored thereon a computer program which, when executed by a processor, implements the method according to any of claims 1 to 3.
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 CN110400298A (en) | 2019-11-01 |
| CN110400298B true 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) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110853012B (en) * | 2019-11-11 | 2022-09-06 | 苏州锐一仪器科技有限公司 | Method, apparatus and computer storage medium for obtaining cardiac parameters |
| 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 |
| 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 |
| 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 |
Citations (7)
| 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 |
| 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 |
| 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 |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030035573A1 (en) * | 1999-12-22 | 2003-02-20 | Nicolae Duta | Method for learning-based object detection in cardiac magnetic resonance images |
| EP3336571A1 (en) * | 2017-04-05 | 2018-06-20 | Siemens Healthcare GmbH | Assignment of mr fingerprints on the basis of neural networks |
| US10984905B2 (en) * | 2017-11-03 | 2021-04-20 | 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 (7)
| 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 |
| 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 |
| 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 |
|---|
| 无创性MR在老年心功能评价中的临床应用;叶大春等;《医学影像学杂志》;20040531;第14卷(第5期);第376-378页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110400298A (en) | 2019-11-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110400298B (en) | Method, device, equipment and medium for detecting heart clinical index | |
| Ouyang et al. | Video-based AI for beat-to-beat assessment of cardiac function | |
| US11664125B2 (en) | System and method for deep learning based cardiac electrophysiology model personalization | |
| US20210151187A1 (en) | Data-Driven Estimation of Predictive Digital Twin Models from Medical Data | |
| 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 | |
| US20230342918A1 (en) | Image-driven brain atlas construction method, apparatus, device and storage medium | |
| CN111476757A (en) | Coronary artery patch data detection method, system, storage medium and terminal | |
| JP6885517B1 (en) | Diagnostic support device and model generation device | |
| CN111612756B (en) | Coronary artery specificity calcification detection method and device | |
| Freiman et al. | Improving CCTA‐based lesions' hemodynamic significance assessment by accounting for partial volume modeling in automatic coronary lumen segmentation | |
| Li et al. | Recurrent aggregation learning for multi-view echocardiographic sequences segmentation | |
| CN111340794B (en) | Quantification method and device for coronary artery stenosis | |
| de Siqueira et al. | Artificial intelligence applied to support medical decisions for the automatic analysis of echocardiogram images: A systematic review | |
| Reddy et al. | Video-based deep learning for automated assessment of left ventricular ejection fraction in pediatric patients | |
| Loncaric et al. | Integration of artificial intelligence into clinical patient management: focus on cardiac imaging | |
| Kagiyama et al. | Machine learning in cardiovascular imaging | |
| Badano et al. | Artificial intelligence and cardiovascular imaging: A win-win combination. | |
| Seetharam et al. | Clinical inference from cardiovascular imaging: paradigm shift towards machine-based intelligent platform | |
| Wehbe et al. | Deep learning for cardiovascular imaging: A review | |
| 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 | |
| Laumer et al. | Weakly supervised inference of personalized heart meshes based on echocardiography videos | |
| CN115330669A (en) | Computer-implemented method, system, and storage medium for predicting disease quantification parameters of an anatomical structure | |
| CN111598966B (en) | Magnetic resonance imaging method and device based on generation countermeasure network | |
| CN112164447B (en) | Image processing method, device, equipment and storage medium |
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 |