CN112438755A - Heart parameter acquisition method, device and storage medium - Google Patents
Heart parameter acquisition method, device and storage medium Download PDFInfo
- Publication number
- CN112438755A CN112438755A CN201910813370.XA CN201910813370A CN112438755A CN 112438755 A CN112438755 A CN 112438755A CN 201910813370 A CN201910813370 A CN 201910813370A CN 112438755 A CN112438755 A CN 112438755A
- Authority
- CN
- China
- Prior art keywords
- processing
- parameters
- data
- preset
- cardiac
- 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
- 238000000034 method Methods 0.000 title claims abstract description 94
- 238000012545 processing Methods 0.000 claims abstract description 130
- 230000000747 cardiac effect Effects 0.000 claims abstract description 84
- 230000011218 segmentation Effects 0.000 claims abstract description 65
- 230000002107 myocardial effect Effects 0.000 claims abstract description 38
- 238000011156 evaluation Methods 0.000 claims abstract description 22
- 238000002604 ultrasonography Methods 0.000 claims description 73
- 238000005259 measurement Methods 0.000 claims description 33
- 238000012549 training Methods 0.000 claims description 28
- 230000008569 process Effects 0.000 claims description 17
- 230000008859 change Effects 0.000 claims description 13
- 238000006073 displacement reaction Methods 0.000 claims description 13
- 210000001174 endocardium Anatomy 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 6
- 238000010801 machine learning Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 description 28
- 210000004165 myocardium Anatomy 0.000 description 14
- 238000004422 calculation algorithm Methods 0.000 description 13
- 238000001914 filtration Methods 0.000 description 5
- 230000002861 ventricular Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 210000005240 left ventricle Anatomy 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 239000003550 marker Substances 0.000 description 3
- 230000010016 myocardial function Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000013256 Gubra-Amylin NASH model Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000010009 beating Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013184 cardiac magnetic resonance imaging Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000002592 echocardiography Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 239000012528 membrane Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
- A61B8/0883—Detecting organic movements or changes, e.g. tumours, cysts, swellings for diagnosis of the heart
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- 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
- G06T2207/10132—Ultrasound image
-
- 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
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Public Health (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Pathology (AREA)
- Epidemiology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Primary Health Care (AREA)
- Cardiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biophysics (AREA)
- Databases & Information Systems (AREA)
- Radiology & Medical Imaging (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Veterinary Medicine (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The invention provides a heart parameter acquisition method, which comprises the following steps: acquiring an ultrasonic image sequence, wherein the content corresponding to the ultrasonic image sequence comprises a heart area; acquiring preset processing parameters, wherein the processing parameters at least comprise: envelope type parameters, scanning type parameters and initial frame parameters; processing the ultrasonic image sequence at least according to the envelope type parameter and the scanning type parameter to obtain heart contour data after segmentation processing, wherein the heart contour data at least comprises: endocardial curve data, epicardial curve data, and myocardial centerline data; processing the heart contour data after the segmentation processing at least based on a preset feature searching method and the initial frame parameters to obtain a plurality of measuring points; processing the plurality of measuring points based on a preset tracking method to obtain a tracking result; based at least on the tracking results, cardiac evaluation parameters are obtained.
Description
Technical Field
The invention relates to the technical field of ultrasonic diagnosis devices, in particular to an ultrasonic analysis method for evaluating the local function of a heart.
Background
At present, there are many methods for evaluating the local function of the heart, including echocardiogram, multi-layer spiral CT, cardiac MRI, etc., but they are affected by the cost and complex operation, etc., and their clinical application is limited. Echocardiography has been widely used because of its portability, innocuity, and other characteristics. The myocardial strain and strain rate analysis is widely used for evaluating myocardial function, and the measurement and analysis result is not dependent on angle and can quantitatively evaluate local myocardial function.
Quantitative analysis of myocardial function can be obtained by means of direct measurement or M-shaped curves and other tools, and is limited by the complexity and limitation of the operation of the traditional measurement mode, so that the heart evaluation parameters obtained by calculation are low in precision and poor in robustness.
Disclosure of Invention
The invention provides a heart parameter acquisition system and an analysis method, which are used for calculating heart evaluation parameters after heart ultrasonic images are automatically segmented through an artificial intelligence model, so that the later processing (such as diagnosis) of workers (such as doctors) is facilitated.
One embodiment of the present invention provides a cardiac parameter acquisition method, including: acquiring an ultrasonic image sequence, wherein the content corresponding to the ultrasonic image sequence comprises a heart area; acquiring preset processing parameters, wherein the processing parameters at least comprise: envelope type parameters, scanning type parameters and initial frame parameters; processing the ultrasonic image sequence at least according to the envelope type parameter and the scanning type parameter to obtain heart contour data after segmentation processing, wherein the heart contour data at least comprises: endocardial curve data, epicardial curve data, and myocardial centerline data; processing the heart contour data after the segmentation processing at least based on a preset feature searching method and the initial frame parameters to obtain a plurality of measuring points; processing the plurality of measuring points based on a preset tracking method to obtain a tracking result; based at least on the tracking results, cardiac evaluation parameters are obtained.
In some embodiments, the processing the ultrasound image sequence according to at least the envelope type parameter and the scanning type parameter to obtain segmented processed cardiac contour data includes: processing the ultrasonic image sequence by adopting a corresponding preset contour identification method based on the preset envelope type parameter to obtain heart contour data; and processing the heart contour data by adopting a corresponding preset segmentation method based on the preset scanning type parameters to obtain segmented heart contour data.
In some embodiments, the processing the ultrasound image sequence by using a corresponding preset contour identification method based on the preset envelope type parameter to obtain heart contour data includes: and when the envelope type parameter is an automatic envelope type, processing the ultrasonic image sequence by using a segmentation model to obtain heart contour data.
In some embodiments, the segmentation model is a machine learning model; the segmentation model is obtained by the following method: acquiring a training set, wherein the training set comprises a sample ultrasonic image sequence and marking data, and the marking data comprises at least one of endocardium curve data, epicardium curve data and myocardial midline data; and training the initial model by using the training set to obtain a segmentation model.
In some embodiments, the processing the segmented cardiac contour data based on at least a preset feature search method and the initial frame parameter to obtain a plurality of measurement points includes: and processing each section of heart contour data of an initial frame in the ultrasonic image sequence by using a preset feature searching method to obtain a plurality of measuring points of the initial frame.
In some embodiments, the processing the plurality of measurement points based on a preset tracking method to obtain a tracking result includes: dividing the ultrasonic image sequence into an initial frame and a plurality of subsequent frames based on preset initial frame parameters; based on each measuring point of an initial frame, obtaining corresponding displacement change information of each measuring point of the initial frame in a subsequent frame by using a preset tracking method; and obtaining the speed information of each measuring point of the initial frame based on the time difference between the initial frame and the corresponding subsequent frame.
In some embodiments, said obtaining cardiac evaluation parameters based at least on said tracking results comprises: obtaining heart evaluation parameters based on corresponding displacement change information and speed information of each measuring point of the initial frame in a subsequent frame; the cardiac evaluation parameters include at least: myocardial strain information, myocardial strain rate information.
One of the embodiments of the present invention provides a cardiac parameter acquisition apparatus, including: the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring an ultrasonic image sequence, and the content corresponding to the ultrasonic image sequence comprises a heart area; a second obtaining module, configured to obtain a preset processing parameter, where the processing parameter at least includes: envelope type parameters, scanning type parameters and initial frame parameters; a first processing module, configured to process the ultrasound image data according to at least the envelope type parameter and the scanning type parameter to obtain segmented cardiac contour data, where the cardiac contour data at least includes: endocardial curve data, epicardial curve data, and myocardial centerline data; the second processing module is used for processing the heart contour data subjected to segmentation processing at least based on a preset feature searching method and the initial frame parameters to obtain a plurality of measuring points; the second processing module is further configured to process the plurality of measurement points based on a preset tracking method to obtain a tracking result; the second processing module is further configured to obtain a cardiac evaluation parameter based on at least the tracking result.
One of the embodiments of the present invention provides an ultrasonic apparatus including: a storage medium storing a computer program; a processor for executing the computer program for carrying out the steps of the method for cardiac parameter acquisition according to any one of claims 1 to 7.
One of the embodiments of the present invention provides a computer storage medium including: the computer storage medium has stored thereon a computer program which, when being executed by a processor, is adapted to carry out the steps of the cardiac parameter acquisition method according to any one of claims 1 to 7.
The embodiment of the invention has the advantages that:
1) the automatic segmentation of the outer membrane in the heart region is realized by adopting a segmentation model based on machine learning, and manual or semi-automatic segmentation processing based on a segmentation algorithm is also provided; not only the robustness of the segmentation is ensured, but also the sufficient segmentation flexibility and precision are ensured.
2) The finally obtained heart evaluation parameters are more accurate through accurate tracking of the segmented myocardial regions.
Drawings
Fig. 1 is a schematic structural component diagram of a cardiac parameter acquisition apparatus 100 according to some embodiments of the present invention.
Fig. 2 is a first flowchart of a cardiac parameter acquisition method 200 according to some embodiments of the invention.
Fig. 3 is a flow diagram of a feature search method 300 according to some embodiments of the invention.
FIG. 4 is a flow chart diagram two of a cardiac parameter acquisition method 400 according to some embodiments of the invention.
Detailed Description
The invention is further illustrated by the following specific figures and examples.
As shown in fig. 1, the cardiac parameter acquisition apparatus 100 may include: a first acquisition module 110, a second acquisition module 120, a first processing module 130, and a second processing module 140. In some embodiments, the cardiac parameter acquisition method 200 may be implemented by the cardiac parameter acquisition apparatus 100.
A first obtaining module 110, configured to obtain an ultrasound image sequence, where content corresponding to the ultrasound image sequence includes a heart region.
A second obtaining module 120, configured to obtain preset processing parameters, where the processing parameters at least include: envelope type parameter, scanning type parameter, initial frame parameter.
A first processing module 130, configured to process the ultrasound image sequence according to at least the envelope type parameter and the scanning type parameter to obtain segmented cardiac contour data, where the cardiac contour data at least includes: endocardial curve data, epicardial curve data, and myocardial centerline data.
A second processing module 140, configured to process the segmented cardiac contour data based on at least a preset feature search method and the initial frame parameter, so as to obtain a plurality of measurement points.
The second processing module 140 is further configured to process the plurality of measurement points based on a preset tracking method to obtain a tracking result;
the second processing module 140 is further configured to obtain a cardiac evaluation parameter based on at least the tracking result.
In some embodiments, the first processing module 130 is further configured to process the ultrasound image sequence by using a corresponding preset contour identification method based on the preset envelope type parameter, so as to obtain heart contour data; and processing the heart contour data by adopting a corresponding preset segmentation method based on the preset scanning type parameters to obtain segmented heart contour data.
In some embodiments, the first processing module 130 includes a first segmentation module 131, and the first segmentation module 131 is configured to process the ultrasound image sequence with a segmentation model to obtain cardiac contour data when the envelope type parameter is an automatic envelope type.
In some embodiments, the segmentation model is a machine learning model.
In some embodiments, the first processing module 130 includes a training module 132, the training module 132 configured to obtain a training set, the training set including a sequence of sample ultrasound images and marker data, the marker data including at least one of endocardial curve data, epicardial curve data, and myocardial centerline data; and training the initial model by using the training set to obtain a segmentation model.
In some embodiments, the second processing module 140 is configured to process each segment of the cardiac contour data of an initial frame in the ultrasound image sequence by using a preset feature search method, so as to obtain a plurality of measurement points of the initial frame.
In some embodiments, the second processing module 140 is configured to divide the ultrasound image sequence into an initial frame and a plurality of subsequent frames based on a preset initial frame parameter; based on each measuring point of an initial frame, obtaining corresponding displacement change information of each measuring point of the initial frame in a subsequent frame by using a preset tracking method; and obtaining the speed information of each measuring point of the initial frame based on the time difference between the initial frame and the corresponding subsequent frame.
In some embodiments, the second processing module 140 is configured to obtain a cardiac evaluation parameter based on the corresponding displacement change information and velocity information of each measurement point of the initial frame in a subsequent frame; the cardiac evaluation parameters include at least: myocardial strain information, myocardial strain rate information.
The embodiment of the cardiac analysis apparatus 100 is related to the embodiment of the cardiac analysis method 200, and details thereof are not repeated here, and refer to the related description of fig. 2 to fig. 4.
As shown in fig. 2, the cardiac parameter acquisition method 200 is implemented by the cardiac parameter acquisition apparatus 100. The method 200 may include:
step 210: a sequence of ultrasound images may be acquired, the content corresponding to the sequence of ultrasound images including a region of the heart. Acquiring preset processing parameters, wherein the processing parameters can include but are not limited to: envelope type parameter, scan type parameter, initial frame parameter, etc.
In some embodiments, the first acquisition module 110 may acquire a sequence of ultrasound images. In some embodiments, the first obtaining module 110 may obtain the sequence of ultrasound images by way of an ultrasound device (e.g., a color ultrasound device, a black and white ultrasound device, a palm ultrasound device, etc.), a database (e.g., a PACS system), or the like. The corresponding content of the sequence of ultrasound images comprises at least a cardiac region, which may be, for example, an echocardiographic sequence. In some embodiments, the content to which the sequence of ultrasound images corresponds may include at least a left ventricular region. In some embodiments, the first acquisition module 110 may acquire ultrasound video data in which the content corresponding to the sequence of images includes a cardiac region. The first obtaining module 110 may split the ultrasound video data to obtain an ultrasound image sequence. In some embodiments, the first obtaining module 110 may perform sampling processing on the ultrasound image sequence to obtain a sampled ultrasound image sequence. For example, the first acquisition module 110 performs sampling processing on the acquired 1000 frames of ultrasound image sequences by taking 5 as a unit, so as to obtain 200 frames of ultrasound image sequences. In some embodiments, the first acquisition module 110 may acquire electrocardiogram data of an ultrasound image sequence synchronously while acquiring the ultrasound image sequence.
In some embodiments, the second obtaining module 120 may obtain preset processing parameters. The processing parameters are used to set and/or indicate subsequent processing operations. In some embodiments, the envelope type parameters may include at least: an automatic envelope type, a semi-automatic envelope type, a manual envelope type, etc. When the type is an automatic envelope type, the cardiac analysis apparatus 100 automatically performs envelope processing on the ultrasound image sequence. When the envelope type is semi-automatic, the cardiac analysis apparatus 100 sets a plurality of points by a user, and performs envelope processing on an ultrasound image sequence by combining a preset segmentation algorithm. When the envelope is of the manual envelope type, the analysis apparatus 100 performs envelope processing on the ultrasound image sequence through points set by the user. In some embodiments, the scan type parameter may include at least: apical long-axis pattern type (ALAX), apical two-chamber pattern type (A2C), apical four-chamber pattern type (A4C), PSAXB, PSAXM, PSAXAP, etc. In some embodiments, the corresponding segmentation method may be determined from the scan type parameter. In some embodiments, the processing parameters may also include a segmentation method parameter. The segmentation method parameters may include at least: a first staging method, a second staging method, and a third staging method. The cardiac analysis apparatus 100 may perform segmentation processing on the envelope processed ultrasound image sequence according to different segmentation methods. In some embodiments, the initial frame parameter may be set automatically by the apparatus 100 or manually by a user. For example, the initial frame parameter may be set to an ultrasound image corresponding to an R-wave of an electrocardiogram. It should be noted that the initial frame parameter may be set according to the actual application, and the present invention is not limited to this.
Step 220: the ultrasound image sequence may be processed at least according to the envelope type parameter and the scanning type parameter to obtain segmented cardiac contour data, where the cardiac contour data at least includes: endocardial curve data, epicardial curve data, and myocardial centerline data.
In some embodiments, the first processing module 130 may envelope the sequence of ultrasound images. In some embodiments, the first processing module 130 may process the ultrasound image sequence by using a corresponding preset contour identification method based on the preset envelope type parameter, so as to obtain heart contour data. In some embodiments, the first processing module 130 may include a first segmentation module 131. In some embodiments, when the envelope type parameter is an automatic envelope type, the first processing module 131 may process the ultrasound image sequence by using a segmentation model to obtain cardiac contour data. In some embodiments, the segmentation model may be a machine learning model. The machine learning model may be a deep learning network model. The deep learning network model may include, but is not limited to: GAN model, VAE model, VGG16 model, VGG19 model, R-CNN model, Fast R-CNN model, Faster R-CNN model, FCN model, etc.
In some embodiments, the first processing module 130 may include a training module 132. In some embodiments, the segmentation model may be obtained by: the training module 132 obtains a training set comprising a sample ultrasound image sequence and marker data comprising at least endocardial curve data, epicardial curve data, and/or myocardial centerline data; the training module 132 trains the initial model using the training set to obtain a segmentation model. In some embodiments, the training module 132 may acquire the sample ultrasound image sequence by way of an ultrasound device (e.g., a color ultrasound device, a black and white ultrasound device, a palm ultrasound device, etc.), a database (e.g., a PACS system), or the like. The sequence of ultrasound images includes at least heart region data, for example, the sequence of ultrasound images may be echocardiographic data. In some embodiments, the training module 132 may acquire sample ultrasound video data in which a sequence of images includes a cardiac region. The training module 132 may perform splitting processing on the sample ultrasound video data to obtain a sample ultrasound image sequence. In some embodiments, the sample ultrasound image sequence may be subjected to preprocessing operations including, but not limited to: denoising processing, enhancing processing, normalizing processing, scaling, rotating or translating the sample and the like. In some embodiments, the marking data includes at least: endocardial curve data, epicardial curve data, and/or myocardial centerline data. The training module 132 generates a training set based on the sample ultrasound image sequence and the labeled image sequence, and trains an initial model using the training set to obtain a segmentation model.
In some embodiments, when the envelope type parameter is a manual envelope type, the first processing module 130 may generate an endocardial envelope curve by a user manually inputting a number of points (for example, inputting at least six points), and the first processing module 130 generates an epicardial curve and a myocardial midline according to the generated endocardial curve. In some embodiments, the first processing module 130 may output and display the generated endocardial curve, epicardial curve and myocardium centerline, and adjust the generated curve points by the user (including the positions of the points on the curve and the thickness of the myocardium between the epicardial curve and the epicardial curve, etc.), so as to make the generated heart contour data more accurate.
In some embodiments, when the envelope type parameter is a semi-automatic envelope type, the first processing module 130 may generate an endocardium envelope curve by a preset segmentation algorithm by inputting a number of points (e.g., inputting three points, which may include a point selected from a cardiac apex point and two cardiac base points) manually by a user, and the first processing module 130 generates an epicardium curve and a myocardial centerline according to the generated endocardium curve. The preset segmentation algorithm may include a morphology algorithm, a level set algorithm, and the like. The segmentation algorithm can evolve according to the energy function of the curve to obtain the heart contour data. In some embodiments, the first processing module 130 may output and display the generated endocardial curve, epicardial curve and myocardium centerline, and adjust the generated curve points by a user (e.g., a doctor) (including the positions of the points on the curve and the thickness of the myocardium between the epicardial curve and the like) so as to make the generated cardiac contour data more accurate.
Step 230: the segmented cardiac contour data may be processed based on at least a preset feature search method and the initial frame parameter to obtain a plurality of measurement points.
In some embodiments, the first processing module 130 may process the cardiac contour data by using a preset segmentation method based on a preset scanning type parameter, so as to obtain segmented cardiac contour data. In some embodiments, the first processing module 130 may select a corresponding segmentation type parameter according to a preset scan type parameter, and perform segmentation processing on the cardiac contour data (e.g., divide the cardiac contour data into several segments). In some embodiments, when the scan type parameter is: when ALAX, A2C, A4C is used, the second processing module 140 selects the first segment type, and can symmetrically divide the heart contour data into seven segments from the left heart base point to the right heart base point. That is, the endocardial curve, epicardial curve and the myocardium midline are symmetrically divided into seven segments for treatment. In some embodiments, the scan type parameter is: when PSAXB or PSAXM is used, the second processing module 140 selects the second segment type, and may divide the heart contour data into six segments symmetrically from the left cardinal point to the right cardinal point. In some embodiments, the scan type parameter is: in PSAXAP, the second processing module 140 selects the third segment type, and may divide the heart contour data into four segments symmetrically from the left cardinal point to the right cardinal point.
In some embodiments, the second processing module 140 may process the segmented cardiac contour data to obtain a plurality of measurement points based on at least a preset feature search method.
In some embodiments, the second processing module 140 may divide the ultrasound image sequence into an initial frame and a subsequent frame based on a preset initial frame parameter. In some embodiments, the second processing module 140 may process each segment of the cardiac contour data of the initial frame in the ultrasound image sequence by using the preset feature search method 300 to obtain a plurality of measurement points of the initial frame. For the contents of the feature search method 300, reference may be made to the related description of fig. 3, which is not described herein again.
Step 240: the plurality of measurement points may be processed based on a preset tracking method to obtain a tracking result.
In some embodiments, the second processing module 140 may obtain, based on each measurement point of an initial frame, corresponding displacement change information of each measurement point of the initial frame in a subsequent frame by using a preset tracking method; and obtaining the speed information of each measuring point of the initial frame based on the time difference between the initial frame and the corresponding subsequent frame. In some embodiments, the predetermined tracking method may include, but is not limited to, a block matching method, an optical flow tracking algorithm, and the like. In some embodiments, the second processing module 140 may track, by using an optical flow tracking algorithm, corresponding measurement points of measurement points in a subsequent frame in an initial frame, and separately track, according to the difference in the number of segments in which the measurement points are located, a measurement point pair set in different segments (i.e., a point pair between each measurement point in the initial frame and a corresponding measurement point on each subsequent frame), so as to obtain measurement point displacement variation information in each segment region; and dividing the displacement change information by the time difference between the two image frames of the subsequent frame and the initial frame to obtain the speed information of each measuring point of the initial frame based on the time difference between the initial frame and the corresponding subsequent frame.
Step 250: a cardiac evaluation parameter may be obtained based at least on the tracking results.
In some embodiments, the second processing module 140 may obtain the cardiac evaluation parameter based on the corresponding displacement change information and velocity information of each measurement point of the initial frame for the subsequent frame. In some embodiments, the cardiac evaluation parameters may include at least: myocardial strain information, myocardial strain rate information. In some embodiments, the cardiac evaluation parameters may include: myocardial strain information, myocardial strain rate information, displacement change information of a measurement point, velocity information of a measurement point, and the like. In some embodiments, the second processing module 140 may obtain the myocardial strain information by comparing the length of the whole segment on the cardiac contour data, and divide the myocardial strain information by the time interval to obtain the myocardial strain rate information.
In particular, myocardial strain information may be obtained by relatively varying the distance between different regions of the myocardium. For example, at time t0, there are two arbitrary points P1 and P2 on the heart, the distance between P1 and P2 is L0, at time t1, due to the beating of the heart, the positions of P1 and P2 change, and the positions move to P1 'and P2', respectively, and the distance between P1 'and P2' is L1, so that the corresponding strain information is (L1-L0)/L0, and the strain rate information can be obtained by dividing the strain information by the time interval, that is, the corresponding strain rate information is (L1-L0)/(L0x (t1-t 0)).
In some embodiments, after all the frames of the ultrasound image sequence are processed iteratively, the second processing module 140 performs a smoothing operation on the obtained displacement change information of the measurement point, the velocity information, the myocardial strain information, and the myocardial strain rate information to obtain corresponding curves, and calculates image frame values (corresponding to time values on the electrocardiogram) at peak time of each curve.
As shown in fig. 3, the feature search method 300 may include:
step 310: the second processing module 140 performs gaussian low-pass filtering on the ultrasound image sequence to obtain a low-pass filtered ultrasound image sequence.
In some embodiments, the second processing module 140 may perform a low-pass filtering process on the ultrasound image sequence by using a gaussian convolution kernel, and the size of the gaussian convolution kernel may be set according to the actual application, for example, the size of the gaussian convolution kernel is 5 × 5.
Step 320: the second processing module 140 performs convolution processing on the ultrasound image sequence after low-pass filtering by using a laplacian operator to obtain an ultrasound image sequence after laplacian processing.
In some embodiments, the second processing module 140 may perform filtering processing on the ultrasound image sequence by using a laplacian, and the size of the laplacian may be set according to an actual application, for example, the size of the laplacian is 5 × 5.
Step 330: the second processing module 140 traverses all positions in the laplacian-processed ultrasound image sequence where the pixel values are greater than the preset threshold, and uses the points corresponding to the obtained positions as measurement points.
In some embodiments, the preset threshold may be set according to practical applications, for example, the preset threshold is set to 30, 50 or the preset threshold is set to be in a range of 60 to 70.
As shown in fig. 4, a cardiac parameter acquisition method 400 is shown below, which cardiac parameter acquisition method 400 may be implemented by the cardiac analysis apparatus 100. The cardiac parameter acquisition method 400 may include:
In step 403, different envelope processing is performed according to the envelope type parameter. Specifically, in step 403, the cardiac analysis apparatus 100 may perform manual envelope processing, semi-automatic envelope processing, or automatic envelope processing according to different envelope type parameters.
At step 404, when the envelope type parameter is a manual envelope type, at least 6 points are input by a user for envelope processing. The heart analysis device 100 acquires at least 6 points input by the user, and generates an endocardial curve, an epicardial curve, and a myocardial centerline using a segmentation module.
In step 405, when the envelope type parameter is a semi-automatic envelope type, three points (one apex point and two cardinal points) are input by a user for envelope processing. The heart analysis device 100 acquires three points (one apex point and two cardinal points) input by the user, and generates endocardial and epicardial curves and a myocardial centerline using a segmentation module.
In the above step 404 or 405, the segmentation module employs a segmentation algorithm to generate an endocardium curve, an epicardium curve, and a myocardium midline; the segmentation algorithm includes, but is not limited to, morphology, level set, etc., and the segmentation algorithm evolves according to the energy function of the curve to obtain the final segmentation contour.
In step 406, when the envelope type parameter is an automatic envelope type, the cardiac analysis apparatus 100 performs envelope processing using the segmentation model. The heart analysis apparatus 100 generates endocardial and epicardial curves and myocardial centerlines using a segmentation model.
Specifically, the segmentation model is implemented in step 414. In step 414, the cardiac analysis apparatus 100 trains a segmentation model network for generating an endocardial curve, an epicardial curve, and a myocardial centerline of the left ventricle by collecting data. The training process of the segmentation model comprises the following steps: the method comprises the steps of obtaining an ultrasonic image scanned by the left ventricle as sample data, using a left ventricle endocardium curve, an epicardium curve and a myocardium center line which are marked with the sample data by a marking tool as marking data, and dividing the sample data with the marking data into a training set, a verification set and a test set which are respectively used for training, optimizing and testing a segmentation model. The segmentation model uses a left ventricle ultrasonic image as input, outputs a single-channel image with the same size as the input image after a series of operations of forming a network by basic artificial neurons such as convolution, maximum pooling, linear rectification filtering, transposed convolution and the like, and on the output image, the pixel value of the position of an endocardium curve can be set to be 1, the pixel value of the position of an epicardium curve can be set to be 3, the pixel value of the position of a myocardial centerline can be set to be 2, and the rest non-curve parts are 0; the output image is judged by pixel values to obtain the endocardium curve, the epicardium curve and the myocardium central line which correspond to the output image;
in step 407, the heart analysis apparatus 100 outputs heart contour data including endocardial curve, epicardial curve, and myocardium centerline.
In step 408, the cardiac analysis device 100 performs different segmentation modes according to the scan type parameters. Specifically, in step 408, the cardiac contour data may be divided into a plurality of segments according to the scan type parameter, and the specific number of segments may be determined by the scan type parameter.
In step 409, when the scanning mode is ALAX, A2C, A4C, the heart analysis device 100 divides the endocardial curve, epicardial curve, and myocardial midline into seven segments.
In step 410, when the scanning mode is PSAXB or PSAXM, the heart analysis apparatus 100 divides the endocardial curve, the epicardial curve, and the myocardium centerline into six segments.
In step 411, when the scanning mode is PSAXAP, the heart analysis apparatus 100 divides the endocardial curve, the epicardial curve, and the myocardium midline into four segments.
In step 412, the cardiac analysis apparatus 100 performs segmentation processing on the endocardium and epicardium curves and the myocardium centerline using different segmentation models, searches all the measurement points in all the segments for the initial frame using a feature search method, tracks all the corresponding measurement points for the subsequent frame using an optical flow tracking algorithm, and calculates corresponding displacement change information, velocity information, myocardium strain information, and myocardium strain rate information.
And step 413, smoothing the corresponding curves, and outputting all the curves and the heart evaluation parameters.
Finally, it should be noted that the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention has been described in detail with reference to examples, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, which should be covered by the claims of the present invention.
Claims (10)
1. A method of cardiac parameter acquisition, comprising:
acquiring an ultrasonic image sequence, wherein the content corresponding to the ultrasonic image sequence comprises a heart area;
acquiring preset processing parameters, wherein the processing parameters at least comprise: envelope type parameters, scanning type parameters and initial frame parameters;
processing the ultrasonic image sequence at least according to the envelope type parameter and the scanning type parameter to obtain heart contour data after segmentation processing, wherein the heart contour data at least comprises: endocardial curve data, epicardial curve data, and myocardial centerline data;
processing the heart contour data after the segmentation processing at least based on a preset feature searching method and the initial frame parameters to obtain a plurality of measuring points;
processing the plurality of measuring points based on a preset tracking method to obtain a tracking result;
based at least on the tracking results, cardiac evaluation parameters are obtained.
2. The method according to claim 1, wherein the processing the ultrasound image sequence according to at least the envelope type parameter and the scanning type parameter to obtain segmented processed cardiac contour data comprises:
processing the ultrasonic image sequence by adopting a corresponding preset contour identification method based on the preset envelope type parameter to obtain heart contour data;
and processing the heart contour data by adopting a corresponding preset segmentation method based on the preset scanning type parameters to obtain segmented heart contour data.
3. The method according to claim 2, wherein the processing the ultrasound image sequence by using a corresponding preset contour recognition method based on the preset envelope type parameter to obtain heart contour data comprises:
and when the envelope type parameter is an automatic envelope type, processing the ultrasonic image sequence by using a segmentation model to obtain heart contour data.
4. The method of claim 3, wherein the segmentation model is a machine learning model;
the segmentation model is obtained by the following method:
acquiring a training set, wherein the training set comprises a sample ultrasonic image sequence and marking data, and the marking data comprises at least one of endocardium curve data, epicardium curve data and myocardial midline data;
and training the initial model by using the training set to obtain a segmentation model.
5. The method of claim 1, wherein the processing the segmented cardiac contour data to obtain a plurality of measurement points based on at least a preset feature search method and the initial frame parameters comprises:
and processing each section of heart contour data of an initial frame in the ultrasonic image sequence by using a preset feature searching method to obtain a plurality of measuring points of the initial frame.
6. The method according to claim 5, wherein the processing the plurality of measurement points based on a predetermined tracking method to obtain a tracking result comprises:
dividing the ultrasonic image sequence into an initial frame and a plurality of subsequent frames based on preset initial frame parameters;
based on each measuring point of an initial frame, obtaining corresponding displacement change information of each measuring point of the initial frame in a subsequent frame by using a preset tracking method;
and obtaining the speed information of each measuring point of the initial frame based on the time difference between the initial frame and the corresponding subsequent frame.
7. The method of claim 6, wherein obtaining cardiac evaluation parameters based at least on the tracking results comprises:
obtaining heart evaluation parameters based on corresponding displacement change information and speed information of each measuring point of the initial frame in a subsequent frame;
the cardiac evaluation parameters include at least: myocardial strain information, myocardial strain rate information.
8. A cardiac parameter acquisition apparatus, comprising:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring an ultrasonic image sequence, and the content corresponding to the ultrasonic image sequence comprises a heart area;
a second obtaining module, configured to obtain a preset processing parameter, where the processing parameter at least includes: envelope type parameters, scanning type parameters and initial frame parameters;
a first processing module, configured to process the ultrasound image data according to at least the envelope type parameter and the scanning type parameter to obtain segmented cardiac contour data, where the cardiac contour data at least includes: endocardial curve data, epicardial curve data, and myocardial centerline data;
the second processing module is used for processing the heart contour data subjected to segmentation processing at least based on a preset feature searching method and the initial frame parameters to obtain a plurality of measuring points;
the second processing module is further configured to process the plurality of measurement points based on a preset tracking method to obtain a tracking result;
the second processing module is further configured to obtain a cardiac evaluation parameter based on at least the tracking result.
9. An ultrasound device, comprising:
a storage medium storing a computer program;
a processor for executing the computer program for carrying out the steps of the method for cardiac parameter acquisition according to any one of claims 1 to 7.
10. A computer storage medium comprising, in combination,
the computer storage medium has stored thereon a computer program which, when being executed by a processor, is adapted to carry out the steps of the cardiac parameter acquisition method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910813370.XA CN112438755A (en) | 2019-08-30 | 2019-08-30 | Heart parameter acquisition method, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910813370.XA CN112438755A (en) | 2019-08-30 | 2019-08-30 | Heart parameter acquisition method, device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112438755A true CN112438755A (en) | 2021-03-05 |
Family
ID=74742237
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910813370.XA Pending CN112438755A (en) | 2019-08-30 | 2019-08-30 | Heart parameter acquisition method, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112438755A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101066211A (en) * | 2006-05-05 | 2007-11-07 | 通用电气公司 | User interface and method for displaying information in an ultrasound system |
| US20100280355A1 (en) * | 2007-12-14 | 2010-11-04 | Grimm Richard A | System and method to characterize cardiac function |
| CN102088913A (en) * | 2008-07-10 | 2011-06-08 | 皇家飞利浦电子股份有限公司 | Ultrasonic assessment of cardiac synchronicity and viability |
| US20120078097A1 (en) * | 2010-09-27 | 2012-03-29 | Siemens Medical Solutions Usa, Inc. | Computerized characterization of cardiac motion in medical diagnostic ultrasound |
| US20150206323A1 (en) * | 2014-01-20 | 2015-07-23 | Samsung Medison Co., Ltd. | Method and apparatus for displaying medical image |
| US20160098833A1 (en) * | 2014-10-06 | 2016-04-07 | Technion Research & Development Foundation Limited | System and Method for Measurement of Myocardial Mechanical Function |
-
2019
- 2019-08-30 CN CN201910813370.XA patent/CN112438755A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101066211A (en) * | 2006-05-05 | 2007-11-07 | 通用电气公司 | User interface and method for displaying information in an ultrasound system |
| US20100280355A1 (en) * | 2007-12-14 | 2010-11-04 | Grimm Richard A | System and method to characterize cardiac function |
| CN102088913A (en) * | 2008-07-10 | 2011-06-08 | 皇家飞利浦电子股份有限公司 | Ultrasonic assessment of cardiac synchronicity and viability |
| US20120078097A1 (en) * | 2010-09-27 | 2012-03-29 | Siemens Medical Solutions Usa, Inc. | Computerized characterization of cardiac motion in medical diagnostic ultrasound |
| US20150206323A1 (en) * | 2014-01-20 | 2015-07-23 | Samsung Medison Co., Ltd. | Method and apparatus for displaying medical image |
| US20160098833A1 (en) * | 2014-10-06 | 2016-04-07 | Technion Research & Development Foundation Limited | System and Method for Measurement of Myocardial Mechanical Function |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bernard et al. | Standardized evaluation system for left ventricular segmentation algorithms in 3D echocardiography | |
| Smistad et al. | Real-time automatic ejection fraction and foreshortening detection using deep learning | |
| US8098918B2 (en) | Method and system for measuring left ventricle volume | |
| CN111493935B (en) | Artificial intelligence-based automatic prediction and identification method and system for echocardiogram | |
| US5669382A (en) | System for measuring myocardium in cardiac images | |
| JP5334692B2 (en) | Method and system for detecting 3D anatomical objects using constrained marginal space learning | |
| EP3834736A1 (en) | Method for determining cardiac cycles and ultrasonic device | |
| EP4307223A2 (en) | Detecting tooth wear using intra-oral 3d scans | |
| JP3982817B2 (en) | Image processing apparatus and image processing method | |
| US9367924B2 (en) | Method and system for segmentation of the liver in magnetic resonance images using multi-channel features | |
| CN110680399B (en) | Automatic measurement method of prenatal ultrasound image, storage medium and ultrasound equipment | |
| CN111583250B (en) | Deep learning-based ultrasonic image mitral valve positioning method and system | |
| EP2498222B1 (en) | Method and system for regression-based 4D mitral valve segmentation from 2D+T magnetic resonance imaging slices | |
| CN108597589B (en) | Model generation method, target detection method and medical imaging system | |
| CN114010227B (en) | Right ventricle characteristic information identification method and device | |
| JP3668629B2 (en) | Image diagnostic apparatus and image processing method | |
| CN112438755A (en) | Heart parameter acquisition method, device and storage medium | |
| US20240127432A1 (en) | Image sequence analysis | |
| CN115775233A (en) | Processing method and device for measuring characteristic dimension based on cardiac ultrasound video | |
| CN115880358A (en) | Construction method of positioning model, positioning method of image mark points and electronic equipment | |
| CN110598675B (en) | Ultrasonic fetal posture identification method, storage medium and electronic equipment | |
| CN113724208A (en) | Blood flow frequency spectrum signal classification method and system | |
| CN116152253B (en) | Cardiac magnetic resonance mapping quantification method, system and storage medium | |
| CN117547306B (en) | Left ventricular ejection fraction measurement method, system and device based on M-type ultrasound | |
| CN117132577B (en) | Method for non-invasively detecting myocardial tissue tension and vibration |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210305 |