CN113538336A - Method, device, equipment and medium for generating full-heart multi-phase positioning image - Google Patents
Method, device, equipment and medium for generating full-heart multi-phase positioning image Download PDFInfo
- Publication number
- CN113538336A CN113538336A CN202110656995.7A CN202110656995A CN113538336A CN 113538336 A CN113538336 A CN 113538336A CN 202110656995 A CN202110656995 A CN 202110656995A CN 113538336 A CN113538336 A CN 113538336A
- Authority
- CN
- China
- Prior art keywords
- heart
- image
- dynamic
- images
- phase
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- 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)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Quality & Reliability (AREA)
- Radiology & Medical Imaging (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The present application relates to the field of medical technology, and in particular, to a method, an apparatus, a device, and a medium for generating a full-cardiac multi-phase localization image. The method comprises the following steps: acquiring a heart dynamic image of a tested object, wherein the heart dynamic image comprises dynamic images of a plurality of heart sections of the tested object, each dynamic image comprises a plurality of frame images corresponding to the heart sections, and each frame image indicates a heart phase of the heart section at a corresponding moment; the heart dynamic image is input into a pre-constructed machine learning model, and a full heart multi-phase positioning image corresponding to the heart of the tested object is generated through the machine learning model. The method can improve the positioning processing efficiency of the magnetic resonance scanning.
Description
Technical Field
The present application relates to the field of medical technology, and in particular, to a method, an apparatus, a device, and a medium for generating a full-cardiac multi-phase localization image.
Background
Medical imaging examinations play a very important role as an aid in the clinical diagnosis of diseases, and when medical scanning is performed, fixed-position/planar scanning of a target region of a subject is often required to assist in the diagnosis, treatment, and the like of diseases. Taking magnetic resonance imaging as an example, for a conventional magnetic resonance scanning process, a doctor manually calibrates a reference positioning line after recognizing an anatomical position through a pre-scanning image to obtain a positioning image, and then applies a scanning sequence to perform accurate scanning based on the anatomical position of the positioning image.
In a traditional mode, when positioning line positioning is carried out and a positioning image is determined, a measured object needs to be matched with a screen for many times, time consumption is long, and positioning processing efficiency is low.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a method, an apparatus, a device, and a medium for generating a full-cardiac multi-phase positioning image, which can improve the efficiency of positioning processing.
A method of generating a full cardiac multi-phase scout image, the method comprising:
acquiring a heart dynamic image of a tested object, wherein the heart dynamic image comprises dynamic images of a plurality of heart sections of the tested object, each dynamic image comprises a plurality of frame images corresponding to the heart sections, and each frame image indicates a heart phase of the heart section at a corresponding moment;
the heart dynamic image is input into a pre-constructed machine learning model, and a full heart multi-phase positioning image corresponding to the heart of the tested object is generated through the machine learning model.
In one embodiment, the cardiac section comprises: any one of a transverse section, a coronal section, and a sagittal section.
In one embodiment, a dynamic image of the heart of the subject is acquired while the subject is in a single breath.
In one embodiment, acquiring a dynamic image of a heart of a subject includes:
and acquiring dynamic images of a plurality of heart sections of the tested object in parallel by a simultaneous multilayer parallel excitation mode to obtain the heart dynamic image of the tested object.
In one embodiment, acquiring a dynamic image of the heart of the subject includes:
and acquiring dynamic images of a plurality of heart sections of the tested object by undersampling the heart sections to obtain the heart dynamic image of the tested object.
In one embodiment, inputting the dynamic image of the heart into a pre-constructed machine learning model, and generating a full-cardiac multi-phase positioning image corresponding to the heart of the measured object by the machine learning model, includes:
performing feature extraction on the dynamic images of the heart sections through a machine learning model to obtain feature data corresponding to the dynamic images;
based on each characteristic data, carrying out image reconstruction to obtain a full heart dynamic image;
based on the full-heart dynamic image, a full-heart multi-phase positioning image of the heart of the measured object is determined.
In one embodiment, the method further includes:
receiving a myocardial image scanning request;
determining a target phase positioning image from the full-cardiac multi-phase positioning images based on the myocardial image scanning request;
and positioning the cardiac muscle part to be scanned of the detected object based on the target phase positioning image, and scanning to obtain a cardiac muscle sequence image corresponding to the cardiac muscle image scanning request.
A whole-heart multi-phase positioning image generation apparatus, the apparatus comprising:
the heart dynamic image acquisition module is used for acquiring a heart dynamic image of the tested object, the heart dynamic image comprises dynamic images of a plurality of heart sections of the tested object, each dynamic image comprises a plurality of frames of images corresponding to the heart sections, and each frame of image indicates the heart phase of the heart section at the corresponding moment;
and the whole-heart multi-phase positioning image generation module is used for inputting the dynamic image of the heart into a pre-constructed machine learning model and generating a whole-heart multi-phase positioning image of the heart corresponding to the tested object through the machine learning model.
A computer device comprising a memory storing a computer program and a processor implementing the steps of the method of any of the above embodiments when the processor executes the computer program.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method of any of the embodiments described above
The method, the device, the equipment and the medium for generating the full-heart multi-phase positioning image are characterized in that a heart dynamic image of a tested object is obtained, the heart dynamic image comprises dynamic images of a plurality of heart sections of the tested object, each dynamic image comprises a plurality of frames of images corresponding to the heart sections, each frame of image indicates the heart phase of the heart section at a corresponding moment, then the heart dynamic image is input into a pre-constructed machine learning model, and the full-heart multi-phase positioning image of the heart corresponding to the tested object is generated through the machine learning model. Therefore, after the heart dynamic image is acquired, the full-heart multi-phase positioning image corresponding to the measured object can be constructed and obtained based on different heart phases of different heart sections in the heart dynamic image in a machine learning system and other modes, and compared with the traditional mode that the full-heart multi-phase positioning image is acquired through holding breath for many times and is manually calibrated, the intelligent level of data processing can be improved, the data acquisition time can be shortened, and the data processing efficiency can be improved.
Drawings
FIG. 1 is a diagram illustrating an exemplary embodiment of a method for generating a full-cardiac multi-phase localization image;
FIG. 2 is a schematic flow chart diagram illustrating a method for generating a full-cardiac multi-phase scout image in one embodiment;
FIG. 3 is a schematic diagram of undersampling acquisition times in one embodiment;
FIG. 4 is a block diagram of an apparatus for generating a full-cardiac multi-phase scout image in an embodiment;
FIG. 5 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The method for generating the whole-heart multi-phase positioning image can be applied to the application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The terminal 102 may acquire a dynamic heart image of the object to be measured, and send the dynamic heart image to the server 104, where the dynamic heart image includes each sequence image corresponding to each acquisition time point of the object to be measured in a continuous time interval, and each sequence image corresponding to each acquisition time point includes a plurality of frames of single-layer images. After acquiring the dynamic heart image of the object to be measured, the server 104 may determine, from the multi-frame single-layer images corresponding to each time point, each single-layer image as a reference image, and determine each single-layer image of the remaining frames as a non-reference image of the corresponding reference image. Further, the server 104 may perform interlayer aliasing on each reference image and each corresponding non-reference image to obtain a cardiac full-phase positioning image of the object under different phase states. The terminal 102 may be a device under test in medical diagnosis and treatment, such as a magnetic resonance system, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers. The server 104 may also be a general purpose computer device or the like.
In one embodiment, as shown in fig. 2, a method for generating a full-cardiac multi-phase positioning image is provided, which is exemplified by the application of the method to the server in fig. 1, and includes the following steps:
step S202, obtaining a heart dynamic image of the tested object, wherein the heart dynamic image comprises dynamic images of a plurality of heart sections of the tested object, each dynamic image comprises a plurality of frame images corresponding to the heart sections, and each frame image indicates the heart phase of the heart section at the corresponding moment.
The dynamic heart image refers to dynamic image data of the heart in a certain time interval, and the dynamic heart image may include dynamic images of a plurality of heart slices of the measured object, such as images of a plurality of slices from the apex of the heart to the center point of the ventricle.
In this embodiment, the cardiac dynamic image may be acquired by a magnetic resonance imaging apparatus or a computed tomography imaging apparatus.
In this embodiment, the cardiac slice may include any one of a transverse slice, a coronal slice, and a sagittal slice.
In this embodiment, each dynamic image may include a plurality of frame images corresponding to the cardiac slice, each frame image indicating a cardiac phase of the cardiac slice at a corresponding time. For example, the dynamic image of a certain cardiac section may include a plurality of frames of images of the cardiac section from contraction to diastole in a continuous time, such as images 1 to 10, where each frame of image represents the cardiac phase of the corresponding cardiac section at each time.
In this embodiment, the server may perform a positioning scan on the object to obtain a heart rhythm image of the object.
Step S204, inputting the heart dynamic image into a pre-constructed machine learning model, and generating a full-heart multi-phase positioning image corresponding to the heart of the tested object through the machine learning model.
The machine learning model may be a deep learning based neural network model, and may include, but is not limited to VGGNet, AlexNet, LeNet5, and the like. It will be understood by those skilled in the art that this is by way of example only, and that in other embodiments, other models are possible, and the present application is not limited thereto.
In this embodiment, after obtaining the heart dynamic image of the whole heart, the server may input the obtained heart dynamic image into a pre-constructed machine learning model, and construct the whole-heart whole-phase image through the machine learning model to obtain the whole-heart multi-phase positioning image corresponding to the object to be measured.
In this embodiment, the full cardiac multi-phase positioning image may include slice positioning images of the heart at different cardiac phases, such as long, two, three, four or short axis and other clinically customary slice images at various cardiac motion phases of the heart, which may also include positioning slice images of the heart not limited to systolic and/or diastolic phases.
In this embodiment, the machine learning model is a pre-constructed model, and the server may obtain training data in advance and train the constructed initial machine learning model to obtain the machine learning model.
Specifically, the server may label the training data, and divide the labeled training data into a training data set and a test training set.
In this embodiment, the server may perform iterative training on the constructed initial machine learning model through a training data set, and perform testing through a test data set after the training is completed.
In this embodiment, when the server performs model training, the server may determine a model loss of the initial machine learning model based on a training result output by the model and an input training data set, and update model parameters of the initial machine learning model based on the model loss.
Further, the server may perform iterative training on the initial neural network model after the model parameters are updated until the training is completed.
In this embodiment, the server may preset a test passing rate, and indicate that the model test passes when the test passing rate reaches a preset threshold.
In this embodiment, the training data may include data in different phase states, for example, data in an expiratory state, data in an inspiratory state, or different data such as a long axis, a two-lumen, a three-lumen, a four-lumen, and a short axis.
In the method for generating the full-cardiac multi-phase positioning image, a dynamic image of the heart of the object to be measured is obtained, the dynamic image of the heart comprises dynamic images of a plurality of cardiac sections of the object to be measured, each dynamic image comprises a plurality of frames of images corresponding to the cardiac sections, each frame of image indicates the cardiac phase of the cardiac section at a corresponding moment, then the dynamic image of the heart is input into a pre-constructed machine learning model, and the full-cardiac multi-phase positioning image of the heart corresponding to the object to be measured is generated through the machine learning model. Therefore, after the heart dynamic image is acquired, the full-heart multi-phase positioning image corresponding to the measured object can be constructed and obtained based on different heart phases of different heart sections in the heart dynamic image through modes such as machine learning, compared with the traditional mode that the full-heart multi-phase positioning image is acquired through holding breath for many times and is manually calibrated, the intelligent level of data processing can be improved, the data acquisition time can be shortened, and the data processing efficiency can be improved.
In one embodiment, a dynamic image of the heart of the subject is acquired while the subject is in a single breath.
Specifically, when a heart image is acquired, the heart is located inside the thoracic cavity, and both inhalation and exhalation of the subject may interfere with the acquired image.
In this embodiment, the server may perform image acquisition when the subject is holding breath, for example, acquire a dynamic image of the heart of the subject when the subject is in an exhalation state or an inhalation state.
Specifically, after receiving the acquisition instruction, the server may send a voice prompt, such as inhalation or exhalation, to prompt the subject to inhale or exhale.
In one embodiment, acquiring a dynamic image of the heart of the subject may include: and acquiring dynamic images of a plurality of heart sections of the tested object in parallel by a simultaneous multilayer parallel excitation mode to obtain the heart dynamic image of the tested object.
In this embodiment, when the object is in an inspiration state or an expiration state, the server may simultaneously acquire images of heart muscles of the object in a plurality of slices by using a multi-layer parallel excitation method. For example, by acquiring once, a plurality of slices are excited in parallel to acquire images of the plurality of slices in parallel.
In this embodiment, the dynamic image is an image set of a plurality of cardiac phases corresponding to a plurality of acquisition times of the cardiac section in the acquisition time interval, and the server may obtain an image of each cardiac section corresponding to each acquisition time of the object to be tested in the acquisition time interval in a multi-layer parallel excitation manner for each acquisition time, so as to obtain a dynamic cardiac image of the object to be tested.
In the above-mentioned embodiment, through the mode of the parallel excitation of multilayer simultaneously, the dynamic image of a plurality of heart tangent planes of measurand is gathered in parallel, obtains the heart dynamic image of measurand to, thereby, make can gather the dynamic image of a plurality of heart tangent planes of whole heart through once gathering, compare and gather in the conventional mode tangent plane one by one, this application scheme can reduce the acquisition time, promotes collection efficiency. Moreover, the breath holding time of the tested object can be reduced, and the user experience is improved.
In one embodiment, acquiring a dynamic image of the heart of the subject may include: and acquiring dynamic images of a plurality of heart sections of the tested object by undersampling the heart sections to obtain the heart dynamic image of the tested object.
In this embodiment, the server may perform undersampling when performing sampling of the heart slice, for example, for a whole heart, only 10 dynamic images of the heart slice may be acquired, and then the dynamic image of the heart of the measured object may be obtained by way of compressed sensing.
Similarly, for the dynamic images of each heart section, when the server collects the dynamic images, the server can collect part of the frame images at the collection time in an undersampling mode, and the dynamic images corresponding to the heart sections are obtained based on the multiframe images obtained by undersampling.
For example, referring to fig. 3, when a certain cardiac slice is sampled, the server may acquire only frame images at 20 acquisition times in an undersampling manner, such as acquiring a frame image at the whole time, that is, time 1 and time 2 …, then construct a full-phase image based on the undersampled frame images in a compressed sensing manner, for example, construct an image at time 1.5 between time 1 and time 2, or an image at time 1.75, and obtain a dynamic image of each cardiac slice corresponding to the full-phase based on the obtained frame images.
In the above embodiment, the dynamic images of the plurality of cardiac slices of the object to be measured are acquired in an undersampling manner, and the dynamic images of the heart of the object to be measured are obtained, so that the number of acquisitions can be reduced, and the acquisition efficiency can be improved.
In one embodiment, the server acquires dynamic images of a plurality of heart slices of the measured object in an undersampling mode of the heart slices, and performs aliasing removal processing on images among different heart slices in an interlayer aliasing mode to obtain a heart dynamic tone image of the whole heart of the measured object.
In one embodiment, inputting the dynamic image of the heart into a pre-constructed machine learning model, and generating a full-cardiac multi-phase positioning image corresponding to the heart of the measured object by the machine learning model may include: performing feature extraction on the dynamic images of the heart sections through a machine learning model to obtain feature data corresponding to the dynamic images; based on each characteristic data, carrying out image reconstruction to obtain a full heart dynamic image; based on the full-heart dynamic image, a full-heart multi-phase positioning image of the heart of the measured object is determined.
Specifically, the server may perform convolution and down-sampling processing on each input moving image by using a machine learning model to obtain a feature map corresponding to each moving image, i.e., obtain feature data.
Further, the server may perform image reconstruction processing on the obtained feature numbers to obtain a full heart dynamic image corresponding to the full heart.
Further, the server can identify positioning images corresponding to each cardiac phase from the full-cardiac dynamic image, such as positioning images corresponding to a long axis, two-chamber, three-chamber, four-chamber and a short axis, or positioning images corresponding to an expiration state and an inspiration state, so as to obtain a full-cardiac multi-phase positioning image.
In one embodiment, the method may further include: receiving a myocardial image scanning request; determining a target phase positioning image from the full-cardiac multi-phase positioning images based on the myocardial image scanning request; and positioning the cardiac muscle part to be scanned of the detected object based on the target phase positioning image, and scanning to obtain a cardiac muscle sequence image corresponding to the cardiac muscle image scanning request.
As mentioned above, in the magnetic resonance scanning process, the physician usually first identifies the anatomical location from the pre-scan image and then manually marks the reference location line, and then applies the scan sequence to the anatomical location of the location image to perform the precise scan.
In this embodiment, after acquiring the full-cardiac multi-phase positioning image of the object to be measured, the server may determine a target-phase positioning image from the full-cardiac multi-phase positioning image according to a myocardial image scanning request of the user, and perform accurate scanning using the determined target-phase positioning image as a positioning image to obtain a myocardial sequence image corresponding to the myocardial image scanning request.
In this embodiment, the heart full-phase positioning image includes slice positioning images of the heart in different heart phases, and the server may determine the corresponding target phase positioning image based on the requirements of the practical application, and perform subsequent scanning processing without acquiring and positioning each of the different heart phases, thereby reducing the data processing time and improving the data processing efficiency.
In this embodiment, the whole-heart multi-phase localization image generating method may be applied to Imaging sequences such as Gradient Echo (GRE), Balanced Steady-State Free Precession (bSSFP) sequence, Echo Planar Echo Imaging (EPI) sequence, and the like. The phase modulation may be implemented by radio frequency pulses and/or gradient fields, and the server may also be implemented by other means, which is not limited in this application.
In the embodiment of the invention, the method for generating the full-cardiac multi-phase positioning image is suitable for the magnetic resonance cardiac scanning rapid positioning scanning. The novel method provides that: under the state of one breath holding or free breathing, a high-resolution and larger-coverage-range full-center (Volumeric) 2D cine is obtained based on a self-calibration dynamic simultaneous multi-layer parallel excitation technology and a high-speed single shot (cine imaging), under the condition that a multi-layer parallel simultaneous excitation element (MB factor) is equal to 2, 30-40 layers or more are acquired, and a higher parallel effect can be achieved by utilizing more channel coils, so that more slices can be scanned in the same time, the coverage rate of scanning can be increased, or the scanning time can be greatly reduced in the same coverage range, and the breath holding time of a detected object is reduced.
In the present embodiment, the reconstruction processing is performed by using Machine Learning (ML), so as to obtain the slice locations of multiple cardiac phases, such as the long axis, two-chamber, three-chamber, four-chamber, or short axis, and the images of other clinically-used slices in each cardiac motion phase of the heart, that is, cine imaging, including or not limited to the location slices in the systolic phase and/or the diastolic phase. The method has very significant clinical significance, provides direct positioning assistance for other cardiac magnetic resonance imaging applications, greatly increases the positioning coverage rate, for example, in the clinical application, positioning images based on the systolic phase are required to be positioned or acquired, and in some clinical applications, the positioning images acquired in a specific phase can be selected and applied as the positioning basis of subsequent scanning according to specific cardiac applications.
In one of the good application scenarios, such as the SENC (Strain-encoded) scan, since the SENC scan is a sequence for measuring the myocardial Strain, and the most important location in the measurement is the myocardial location during systole, the general procedure includes the above-mentioned two-chamber and four-chamber cardiac location techniques, and the two-chamber and four-chamber, and short-axis cine imaging are combined to accurately locate the respective locations during systole. According to the scheme, through Easy Scan with CINE (film imaging) function, the conventional multiple breath-holding scanning can be shortened to the CINE Easy Scan with 2D multilayer full coverage of only one breath-holding in the whole positioning process, the positioning process and the scanning time are greatly simplified, and the scanning of the SENC technology is quicker and easier.
Meanwhile, 2D tMB CINE Easy Scan (2D dynamic multilayer parallel simultaneous excitation CINE imaging fast Scan) has good practical significance for analysis of cardiac function in practice.
It should be understood that, although the steps in the flowchart of fig. 2 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a portion of the steps in fig. 2 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
In one embodiment, as shown in fig. 4, there is provided a full-cardiac multi-phase positioning image generation apparatus, including: a heart dynamic image acquisition module 100 and a full-heart multi-phase positioning image generation module 200, wherein:
the heart dynamic image acquiring module 100 is configured to acquire a heart dynamic image of the measured object, where the heart dynamic image includes dynamic images of a plurality of heart slices of the measured object, each dynamic image includes a plurality of frame images corresponding to the heart slices, and each frame image indicates a heart phase of the heart slice at a corresponding time.
The whole-heart multi-phase positioning image generation module 200 is configured to input the dynamic image of the heart into a pre-constructed machine learning model, and generate a whole-heart multi-phase positioning image of the heart corresponding to the object to be measured through the machine learning model.
In one embodiment, the cardiac section may comprise: any one of a transverse section, a coronal section, and a sagittal section.
In one embodiment, a dynamic image of the heart of the subject is acquired while the subject is in a single breath.
In one embodiment, the heart dynamic image acquiring module 100 is configured to acquire dynamic images of a plurality of heart slices of the object to be tested in parallel by simultaneous multi-layer parallel excitation, so as to obtain a heart dynamic image of the object to be tested.
In one embodiment, the heart dynamic image acquiring module 100 acquires dynamic images of a plurality of heart slices of the measured object by undersampling the heart slices to obtain a heart dynamic image of the measured object.
In one embodiment, the whole-heart multi-phase positioning image generation module 200 may include:
and the feature extraction submodule is used for extracting features of the dynamic images of the heart sections through the machine learning model to obtain feature data corresponding to the dynamic images.
And the reconstruction submodule is used for carrying out image reconstruction based on the characteristic data to obtain a full heart dynamic image.
And the whole-heart multi-phase positioning image generation sub-module is used for determining a whole-heart multi-phase positioning image of the heart of the measured object based on the whole-heart dynamic image.
In one embodiment, the apparatus may further include:
a request receiving module, configured to receive a myocardial image scan request.
And the target phase positioning image determining module is used for determining a target phase positioning image from the full-heart multi-phase positioning images based on the myocardial image scanning request.
And the positioning and scanning module is used for positioning the myocardial part to be scanned of the detected object based on the target phase positioning image and scanning to obtain a myocardial sequence image corresponding to the myocardial image scanning request.
For the specific definition of the whole-heart multi-phase positioning image generation apparatus, reference may be made to the above definition of the whole-heart multi-phase positioning image generation method, which is not described herein again. The modules in the whole-heart multi-phase positioning image generation device can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 5. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing data such as dynamic heart images and full-heart multi-phase positioning images. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of generating a full cardiac multi-phase scout image.
Those skilled in the art will appreciate that the architecture shown in fig. 5 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, there is provided a computer device comprising a memory storing a computer program and a processor implementing the following steps when the processor executes the computer program: acquiring a heart dynamic image of a tested object, wherein the heart dynamic image comprises dynamic images of a plurality of heart sections of the tested object, each dynamic image comprises a plurality of frame images corresponding to the heart sections, and each frame image indicates a heart phase of the heart section at a corresponding moment; the heart dynamic image is input into a pre-constructed machine learning model, and a full heart multi-phase positioning image corresponding to the heart of the tested object is generated through the machine learning model.
In one embodiment, the cardiac section may comprise: any one of a transverse section, a coronal section, and a sagittal section.
In one embodiment, the dynamic image of the heart of the measured object is acquired when the measured object is in a single breath state.
In one embodiment, the processor, when executing the computer program, is configured to acquire a dynamic image of a heart of a subject, and may include: and acquiring dynamic images of a plurality of heart sections of the tested object in parallel by a simultaneous multilayer parallel excitation mode to obtain the heart dynamic image of the tested object.
In one embodiment, the processor, when executing the computer program, is configured to acquire a dynamic image of a heart of the subject, and may include: and acquiring dynamic images of a plurality of heart sections of the tested object by undersampling the heart sections to obtain the heart dynamic image of the tested object.
In one embodiment, the processor, when executing the computer program, implements inputting the dynamic images of the heart into a pre-constructed machine learning model, and generating full-cardiac multi-phase positioning images of the heart corresponding to the measured object by the machine learning model, may include: performing feature extraction on the dynamic images of the heart sections through a machine learning model to obtain feature data corresponding to the dynamic images; based on each characteristic data, carrying out image reconstruction to obtain a full heart dynamic image; based on the full-heart dynamic image, a full-heart multi-phase positioning image of the heart of the measured object is determined.
In one embodiment, the processor, when executing the computer program, may further implement the following steps: receiving a myocardial image scanning request; determining a target phase positioning image from the full-cardiac multi-phase positioning images based on the myocardial image scanning request; and positioning the cardiac muscle part to be scanned of the detected object based on the target phase positioning image, and scanning to obtain a cardiac muscle sequence image corresponding to the cardiac muscle image scanning request.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of: acquiring a heart dynamic image of a tested object, wherein the heart dynamic image comprises dynamic images of a plurality of heart sections of the tested object, each dynamic image comprises a plurality of frame images corresponding to the heart sections, and each frame image indicates a heart phase of the heart section at a corresponding moment; the heart dynamic image is input into a pre-constructed machine learning model, and a full heart multi-phase positioning image corresponding to the heart of the tested object is generated through the machine learning model.
In one embodiment, the cardiac section may comprise: any one of a transverse section, a coronal section, and a sagittal section.
In one embodiment, the dynamic image of the heart of the measured object is acquired when the measured object is in a single breath state.
In one embodiment, the computer program, when executed by the processor, for implementing acquiring a dynamic image of a heart of a subject, may include: and acquiring dynamic images of a plurality of heart sections of the tested object in parallel by a simultaneous multilayer parallel excitation mode to obtain the heart dynamic image of the tested object.
In one embodiment, the computer program, when executed by the processor, for implementing acquiring a dynamic image of a heart of a subject, may include: and acquiring dynamic images of a plurality of heart sections of the tested object by undersampling the heart sections to obtain the heart dynamic image of the tested object.
In one embodiment, the computer program when executed by the processor for implementing inputting the dynamic images of the heart into a pre-constructed machine learning model, and generating the full-cardiac multi-phase positioning images of the heart corresponding to the measured object by the machine learning model may include: performing feature extraction on the dynamic images of the heart sections through a machine learning model to obtain feature data corresponding to the dynamic images; based on each characteristic data, carrying out image reconstruction to obtain a full heart dynamic image; based on the full-heart dynamic image, a full-heart multi-phase positioning image of the heart of the measured object is determined.
In one embodiment, the computer program when executed by the processor may further implement the steps of: receiving a myocardial image scanning request; determining a target phase positioning image from the full-cardiac multi-phase positioning images based on the myocardial image scanning request; and positioning the cardiac muscle part to be scanned of the detected object based on the target phase positioning image, and scanning to obtain a cardiac muscle sequence image corresponding to the cardiac muscle image scanning request.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A method for generating a full cardiac multi-phase scout image, the method comprising:
acquiring a heart dynamic image of a measured object, wherein the heart dynamic image comprises dynamic images of a plurality of heart sections of the measured object, each dynamic image comprises a plurality of frame images corresponding to the heart sections, and each frame image indicates a heart phase of the heart section at a corresponding moment;
and inputting the heart dynamic image into a pre-constructed machine learning model, and generating a full-heart multi-phase positioning image corresponding to the heart of the measured object through the machine learning model.
2. The method of claim 1, wherein the cardiac section comprises: any one of a transverse section, a coronal section, and a sagittal section.
3. The method of claim 1, wherein the acquiring of the dynamic image of the subject's heart is acquired while the subject is in a single breath.
4. The method of claim 1, wherein the acquiring a dynamic image of the subject's heart comprises:
and acquiring dynamic images of a plurality of heart sections of the tested object in parallel by a simultaneous multilayer parallel excitation mode to obtain the heart dynamic image of the tested object.
5. The method of claim 1, wherein the acquiring a dynamic image of the subject's heart comprises:
and acquiring dynamic images of a plurality of heart sections of the tested object by undersampling the heart sections to obtain the heart dynamic image of the tested object.
6. The method of claim 1, wherein the inputting the dynamic images of the heart into a pre-constructed machine learning model, and generating full-cardiac multi-phase positioning images of the heart corresponding to the subject by the machine learning model, comprises:
performing feature extraction on the dynamic images of the heart sections through the machine learning model to obtain feature data corresponding to the dynamic images;
based on each feature data, carrying out image reconstruction to obtain a full heart dynamic image;
based on the full-cardiac dynamic image, a full-cardiac multi-phase positioning image of the heart of the subject is determined.
7. The method of claim 1, further comprising:
receiving a myocardial image scanning request;
determining a target phase positioning image from the full-heart multi-phase positioning images based on the myocardial image scanning request;
and positioning the cardiac muscle part to be scanned of the object to be detected based on the target period phase positioning image, and scanning to obtain a cardiac muscle sequence image corresponding to the cardiac muscle image scanning request.
8. A full cardiac multi-phase scout image generating apparatus, the apparatus comprising:
the heart dynamic image acquisition module is used for acquiring a heart dynamic image of a measured object, wherein the heart dynamic image comprises dynamic images of a plurality of heart sections of the measured object, each dynamic image comprises a plurality of frame images corresponding to the heart sections, and each frame image indicates a heart phase of the heart section at a corresponding moment;
and the whole-heart multi-phase positioning image generation module is used for inputting the heart dynamic image into a pre-constructed machine learning model and generating a whole-heart multi-phase positioning image corresponding to the heart of the measured object through the machine learning model.
9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110656995.7A CN113538336A (en) | 2021-06-11 | 2021-06-11 | Method, device, equipment and medium for generating full-heart multi-phase positioning image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110656995.7A CN113538336A (en) | 2021-06-11 | 2021-06-11 | Method, device, equipment and medium for generating full-heart multi-phase positioning image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113538336A true CN113538336A (en) | 2021-10-22 |
Family
ID=78095906
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110656995.7A Pending CN113538336A (en) | 2021-06-11 | 2021-06-11 | Method, device, equipment and medium for generating full-heart multi-phase positioning image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113538336A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115662579A (en) * | 2022-10-27 | 2023-01-31 | 拓微摹心数据科技(北京)有限公司 | Method, device, storage medium and computer equipment for splitting and dynamically displaying multi-phase sequence |
| CN116671974A (en) * | 2023-06-06 | 2023-09-01 | 河北大学 | Magnetic positioning system for ultrasonic inspection |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005046480A1 (en) * | 2003-11-05 | 2005-05-26 | General Electric Company | Methods and apparatus for assisting cardiac resynchronization therapy |
| CN104200500A (en) * | 2014-07-29 | 2014-12-10 | 沈阳东软医疗系统有限公司 | Method and device for reconstructing heart image |
| CN109507622A (en) * | 2018-11-16 | 2019-03-22 | 上海联影医疗科技有限公司 | MR imaging method and system |
| CN109745062A (en) * | 2019-01-30 | 2019-05-14 | 腾讯科技（深圳）有限公司 | Generation method, device, equipment and the storage medium of CT image |
-
2021
- 2021-06-11 CN CN202110656995.7A patent/CN113538336A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005046480A1 (en) * | 2003-11-05 | 2005-05-26 | General Electric Company | Methods and apparatus for assisting cardiac resynchronization therapy |
| CN104200500A (en) * | 2014-07-29 | 2014-12-10 | 沈阳东软医疗系统有限公司 | Method and device for reconstructing heart image |
| CN109507622A (en) * | 2018-11-16 | 2019-03-22 | 上海联影医疗科技有限公司 | MR imaging method and system |
| CN109745062A (en) * | 2019-01-30 | 2019-05-14 | 腾讯科技（深圳）有限公司 | Generation method, device, equipment and the storage medium of CT image |
Non-Patent Citations (2)
| Title |
|---|
| 张文兰 等: ""组织多普勒超声心脏图像的动态三维重建方法"", 《中国医学装备》 * |
| 黄小倩 等: ""基于同时多层激发和并行成像的心脏磁共振电影成像"" * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115662579A (en) * | 2022-10-27 | 2023-01-31 | 拓微摹心数据科技(北京)有限公司 | Method, device, storage medium and computer equipment for splitting and dynamically displaying multi-phase sequence |
| CN116671974A (en) * | 2023-06-06 | 2023-09-01 | 河北大学 | Magnetic positioning system for ultrasonic inspection |
| CN116671974B (en) * | 2023-06-06 | 2024-02-06 | 河北大学 | Magnetic positioning system for ultrasonic inspection |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5523718B2 (en) | Medical imaging device | |
| US9684979B2 (en) | MRI 3D cine imaging based on intersecting source and anchor slice data | |
| CN110031786B (en) | Magnetic resonance image reconstruction method, magnetic resonance imaging apparatus, and medium | |
| JP2013223724A (en) | Method for spatially resolving and determining magnetic resonance relaxation parameters in examination area rapidly | |
| JP2013102889A (en) | Magnetic resonance imaging apparatus | |
| US20150238149A1 (en) | Magnetic resonance imaging apparatus | |
| KR101809213B1 (en) | Method and apparatus for multi-slice imaging of t2-relaxation time | |
| CN107505584B (en) | Magnetic resonance data acquisition triggering method and device | |
| JP2007319689A (en) | Method for displaying heart and magnetic resonance system | |
| US20130338486A1 (en) | Diffusion tensor magnetic resonance imaging method | |
| CN113538336A (en) | Method, device, equipment and medium for generating full-heart multi-phase positioning image | |
| US11269036B2 (en) | System and method for phase unwrapping for automatic cine DENSE strain analysis using phase predictions and region growing | |
| EP3979898B1 (en) | Mri-derived strain-based measurements and related image data acquisitions, image data processing, patient evaluations and monitoring methods and systems | |
| US8660634B2 (en) | System and method for improved lung fourier decomposition magnetic resonance imaging | |
| US20160061923A1 (en) | Sheet tractography using diffusion tensor mri | |
| CN109077728B (en) | Quantitative myocardial magnetic resonance imaging method, apparatus and storage medium | |
| CN109247938B (en) | Quantitative myocardial magnetic resonance imaging method, apparatus and storage medium | |
| Seemann et al. | Dynamic lung water MRI during exercise stress | |
| WO2020034676A1 (en) | Myocardial quantitative magnetic resonance imaging method and device, and storage medium | |
| KR20220030896A (en) | Method for correcting magnetic resonance imaging error using heart rate interval | |
| KR20220092685A (en) | Method for generating ECV map by predicting hematocrit using Neural network | |
| WO2016127053A2 (en) | Continuous and rapid quantification of stroke volume from magnetohydrodynamic voltages in magnetic resonance imaging | |
| CN110073233A (en) | Method for the diagnostic measurement data by magnetic resonance equipment record heart | |
| US20170224259A1 (en) | Magnetic resonance imaging method to non-invasively measure blood oxygen saturation | |
| CN112986878B (en) | Coronary artery imaging method and magnetic resonance imaging system |
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 |