CN115349851A - Cardiac function diagnosis method based on atrioventricular plane pump model - Google Patents
Cardiac function diagnosis method based on atrioventricular plane pump model Download PDFInfo
- Publication number
- CN115349851A CN115349851A CN202211023705.6A CN202211023705A CN115349851A CN 115349851 A CN115349851 A CN 115349851A CN 202211023705 A CN202211023705 A CN 202211023705A CN 115349851 A CN115349851 A CN 115349851A
- Authority
- CN
- China
- Prior art keywords
- atrioventricular
- left ventricle
- heart
- model
- displacement
- 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
- 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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/02028—Determining haemodynamic parameters not otherwise provided for, e.g. cardiac contractility or left ventricular ejection fraction
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/107—Measuring physical dimensions, e.g. size of the entire body or parts thereof
- A61B5/1073—Measuring volume, e.g. of limbs
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/107—Measuring physical dimensions, e.g. size of the entire body or parts thereof
- A61B5/1079—Measuring physical dimensions, e.g. size of the entire body or parts thereof using optical or photographic means
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/277—Analysis of motion involving stochastic approaches, e.g. using Kalman filters
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30048—Heart; Cardiac
Abstract
A method for cardiac function diagnosis based on an atrioventricular planar pump model, comprising: the method comprises the steps of segmenting a left ventricle area on an image through a clinically acquired cardiac nuclear magnetic resonance image, calculating left ventricle volume and flow, calculating atrioventricular plane displacement and speed through tracking of an atrioventricular characteristic point motion track, realizing model construction of the left ventricle based on cardiac dynamics, carrying out system identification through an unscented Kalman filter, optimizing model parameters through a nonlinear optimization method, and realizing cardiac function diagnosis by utilizing cardiac parameters. The method can realize accurate segmentation of the cardiac magnetic resonance short axis image, and effectively solves the problems of time and labor waste caused by manually tracing the outline; the problem that the tracking is inaccurate by the traditional normalized cross-correlation matching method is solved, and the measured atrioventricular plane displacement curve is more consistent with physiological performance; the novel heart model is constructed, the dynamic characteristics of the heart can be accurately described, the reliability of identification parameters is ensured through the unscented Kalman filter, the method is simple, the identification accuracy is high, the left ventricle model is optimized through nonlinear optimization, the obtained parameters are stable and accurate, and a doctor can be assisted in diagnosing heart diseases.
Description
Technical Field
The invention relates to a medical image processing technology, a heart dynamics model and a system identification technology, in particular to a cardiac function diagnosis method based on an atrioventricular plane pump model.
Background
In recent years, the mortality rate of cardiovascular diseases is high, and the improvement of the detection rate of heart diseases is urgent. The clinical findings show that the traditional measuring method of the interventional cardiac function parameters is not only complex, but also has great harm to human body; in addition, the discovery of heart diseases with preserved ejection fraction is also explaining the limitation of the traditional heart disease judgment index. Therefore, a comprehensive evaluation index with easily measurable parameters, non-intrusive performance, quantifiable performance and universality is urgently needed.
The cardiac magnetic resonance imaging has the characteristic of non-invasive, and can clearly display the physical structures of internal organs, muscles, bones, blood, fat and the like of a human body. Based on the measurement result of the cardiac function parameters of the patient, a doctor can quickly determine the position of the pathological changes, the type of inflammation, the cause of abnormal substances and the like, so that the diagnosis of the cardiac diseases is realized as early and accurate as possible, and the precision and the efficiency of the diagnosis of the cardiac diseases are improved.
In order to realize quantitative analysis of cardiac function parameters, modeling analysis of the heart needs to be realized. In the existing heart model, a large number of nonlinear equations often cause difficult parameter identification. In this context, we propose a modeling method based on the atrioventricular plane displacement, which uses the unscented kalman filter to identify parameters and perform nonlinear optimization to identify cardiac function parameters, thereby assisting doctors in diagnosing heart diseases.
Disclosure of Invention
The invention aims to provide a simple and high-accuracy cardiac function diagnosis method, which comprises the steps of calculating the volume and flow of a left ventricle after segmenting a left ventricle area on an image through a clinically obtained cardiac nuclear magnetic resonance image, calculating the plane displacement of an atrioventricular chamber through tracking of a motion trail of an atrioventricular characteristic point, realizing model construction of the left ventricle based on cardiac dynamics, carrying out system identification through an unscented Kalman filter, and finally optimizing the model through a nonlinear optimization method to realize cardiac function diagnosis.
In order to achieve the above object, the present invention provides a cardiac function diagnosis method based on an atrioventricular plane pump model, comprising the steps of:
s1: the heart nuclear magnetic resonance image is obtained clinically, the segmentation of the left ventricle is completed, and the calculation of the ventricle volume is realized;
s2: selecting characteristic points of the left ventricle in a nuclear magnetic resonance image of the left ventricle, and tracking the characteristic points by utilizing a tracking algorithm to obtain the plane displacement of the atrioventricular organ;
s3: based on the principle of cardiac dynamics, a left ventricle model is established:
s4: identifying physiological parameters of the heart through an unscented Kalman filter;
s5: and optimizing the model by using a nonlinear programming method.
Further, in step S1, the ventricular volume calculation step is as follows:
constructing a left ventricle segmentation training set based on a clinically collected left ventricle nuclear magnetic resonance image and a manually labeled left ventricle contour of a medical image expert; performing data enhancement on the training set to obtain a sufficient training set and a sufficient test set for training the U-NET network model; segmenting the left ventricle by using the trained U-NET to obtain the area of each slice in the meaning of the left ventricle pixel; and combining the data such as the area of the left ventricle in each slice, the physical distance between adjacent slices and the like to realize the calculation of the volume and the flow of the left ventricle of the heart individual.
Further, in step S2, the atrioventricular plane displacement calculation step is as follows:
making a atrioventricular feature point data set through a clinically measured long-axis image of a ventricle, marking, and dividing a training set and a verification set; constructing a model network structure, selecting a tracking target image and determining a search area, respectively sending feature graphs of different scales into a SimRPN network after extracting features, directly adopting a weighted sum to the output of the SimRPN network, and outputting a combined SimRPN including a classification branch and a regression branch; and calculating the offset of the anchor frame by using the L1 loss, and calculating the loss of confidence by using a cross entropy loss, thereby realizing the automatic tracking of the characteristic points of the ventricle and calculating the plane displacement of the atrioventricular organ. Compared with the situation that a plurality of abnormal values are arranged at key points of the atrioventricular plane tracked by the traditional normalized cross-correlation matching method, the measured atrioventricular plane displacement curve is more in line with physiological performance.
Further, in step S3, the heart model construction principle is as follows:
the outer wall of the left ventricle is approximated to a cylinder and a hemisphere, and an axial function model is established by analyzing by a infinitesimal method, however, the model is established only by considering the factors of the longitudinal motion of the heart, and according to the display of the cardiac radiography, the longitudinal displacement of the heart is actually accompanied by a small amount of circumferential displacement.
In order to be able to describe the heart's motion behavior more accurately, in the atrioventricular planar model, the lateral displacement is equivalent to a coefficient k RAD . The established heart model is as follows:
wherein, V AVP The speed of the plane displacement of the chamber is shown,
representing the rate of change of plane displacement of the atrioventricular cavity, A LV Is the equivalent piston area on the ventricular side, A LA Is the equivalent piston area on the atrial side, F C Is the myocardial force, Q LV Is the left heart flow, P LV The pressure of the left ventricle is shown,
indicates the rate of change of left ventricular pressure, P LA Indicating the pressure of the left atrium, C LV Indicating left ventricular compliance, k RAD Representing the radial function coefficients.
Further, in step S4, the method for identifying parameters based on the unscented kalman filter is as follows:
based on the relationship between the tension and contraction of the myocardium, the ventricular force is equivalent to a trapezoid, and the selected parameters to be identified are as follows:
wherein A is 1 And A 2 Respectively representing ventricular contraction force F VC And atrial contractile force F AC Amplitude of (D), R AVP Denotes the damping of AVP, L AVP Representing the inertia of AVP, taking the flow of the left ventricle as an input quantity, taking the plane displacement speed of the atrioventricular cavity as an observed quantity, and realizing the identification of the model parameters of the atrioventricular cavity plane pump by utilizing an unscented Kalman filter; based on the measured parameter information of damping, inertia, responsiveness, myocardial force, and the geometric and circumferential expansion of the ventricles, assessment of the heart state can be achieved.
Further, in step S5, the model is optimized based on a nonlinear optimization method. The method is characterized in that trapezoidal myocardial force is used as an initial value, each point in a myocardial force curve is used as a variable to be optimized based on a discretization thought, a kinetic equation is used as constraint, the minimum of the atrioventricular plane displacement speed estimation error is used as a judgment index, nonlinear optimization is carried out on the minimum, an interior point method is used for solving, and further optimization on heart parameters and the shape of myocardial force is achieved. According to clinical findings, the occurrence of heart diseases is often accompanied by large changes of relevant heart parameters, so that a doctor can be assisted in cardiac function diagnosis through measurement of the heart parameters.
Drawings
FIG. 1 is a flow diagram of the overall scheme of the present invention;
fig. 2 is a partitioning effect based on a U-NET network;
FIG. 3 is a schematic view of a characteristic point of a chamber;
FIG. 4 is a SiamRPN network framework;
FIG. 5 is a heart model of the atrioventricular plane;
FIG. 6 is a graph of maximum area/longitudinal displacement versus ejection fraction;
FIG. 7 shows the myocardial force F C Schematic representation.
The specific implementation mode is as follows:
the technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings and specific embodiments. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The method comprises the steps of segmenting a left ventricle area on an image through a clinically acquired cardiac nuclear magnetic resonance image, then calculating left ventricle volume and flow, calculating atrioventricular plane displacement through tracking of motion tracks of atrioventricular feature points, achieving model construction of the left ventricle based on cardiac dynamics, performing system identification through an unscented Kalman filter, and finally optimizing the model through a nonlinear optimization method to achieve cardiac function diagnosis.
Referring to fig. 1, the present invention provides a cardiac function diagnosis method based on an atrioventricular plane pump model, comprising the following steps:
s1: the heart nuclear magnetic resonance image is obtained clinically, the segmentation of the left ventricle is completed, and the calculation of the ventricle volume is realized;
s2: selecting characteristic points of the left ventricle in a nuclear magnetic resonance image of the left ventricle, and tracking the characteristic points by utilizing a tracking algorithm to obtain the plane displacement of the atrioventricular organ;
s3: based on the principle of cardiac dynamics, a left ventricle model is established:
s4: identifying physiological parameters of the heart through an unscented Kalman filter;
s5: and optimizing the model by using a nonlinear programming method.
The following describes a method for constructing a three-dimensional heart model based on ultrasound imaging according to embodiment 1:
example 1
A cardiac function diagnosis method based on an atrioventricular plane pump model comprises the following steps:
fig. 2 is a segmentation effect graph based on the U-NET network, as shown in fig. 2, in step S1, the left ventricle is segmented by clinically acquiring a cardiac magnetic resonance image, and the calculation of the ventricle volume is realized, the steps are as follows:
(1) Manually marking the contour line of the left ventricle of the heart in the image by an expert based on a clinically acquired cardiac nuclear magnetic resonance image; extracting corresponding contour lines and constructing corresponding binary segmentation images, wherein the contour of the left ventricle and the internal pixel value thereof are both 1, and the external pixel value of the contour is both 0;
(2) The obtained data set is subjected to operations such as scaling, shearing, turning, brightness adjustment and the like to realize data enhancement, so that the training data volume is increased, and the generalization capability of the model is improved; noise data are added, and the robustness of the model is improved;
(3) Training by adopting a U-NET network model, respectively embedding an attention module into each cascade layer in a decoding stage in order to improve the accuracy, and performing weight distribution on space attention by utilizing a shallow layer and an upsampling splicing result and combining the attention module to improve the segmentation performance of a target area;
(4) The area in terms of left ventricle pixels in each slice can be obtained from the left ventricle segmentation result, and the actual area of the left ventricle in each heart slice can be converted. About 25 slices are provided for each individual heart in the data set, and the volume of the left ventricle of the individual heart can be calculated according to the area of the left ventricle in each slice, the physical distance between adjacent slices and the like.
Fig. 3 is a schematic diagram of atrioventricular feature points, and as shown in fig. 3, in step S2, in the mri image of the left ventricle, feature points of the left ventricle are selected, and a tracking algorithm is used to track the feature points to obtain atrioventricular plane displacement;
through a clinically measured long-axis image of the ventricle, a data set of the atrioventricular characteristic points is made and labeled, and a training set and a verification set are divided;
constructing a model network structure, selecting a tracking target image and determining searchSearching areas, respectively sending the feature graphs with different scales into a SiamRPN network after extracting features, wherein the SiamRPN network comprises a Siamese sub-network extracted by the features and a candidate area generation network, and a network structure diagram is shown in FIG. 4; in the feature extraction subnetwork, x is belonged to R 255*255*3 Representing the detected frame image, z ∈ R 127*127*3 Representing template frame images, and obtaining corresponding features after performing feature extraction through AlexNet
And
candidate area Generation network includes Classification A cls And regression A reg And the two branches respectively carry out convolution operation on the characteristics of the template frame and the detection frame:
where, denotes the convolution operation, k denotes the number of anchor boxes, and w and h denote the width and height of the anchor boxes, respectively.
The loss function of the candidate region generation network consists of two parts, namely a classification branch loss function and a regression branch loss function, wherein the classification loss is a fork entropy loss, and the regression loss adopts an L1 smooth loss.
Wherein A is x ,A y ,A w ,A h Respectively representing the center point and length and width information of the anchor frame, T x ,T y ,T w ,T h Respectively representing the real central point and length and width information of the corresponding frame, and the standardized distance is as follows:
through the L1 smoothing loss calculation of the normalized coordinates, the available loss is:
loss＝L cls +λL reg
therefore, the automatic tracking of the characteristic points of the ventricles can be realized, the plane displacement of the ventricles can be calculated, and compared with the situation that a plurality of abnormal values exist in the key points of the ventricles tracked by the traditional normalized cross-correlation matching method, the measured plane displacement curve of the ventricles is more in line with physiological performance.
Fig. 5 is a schematic diagram of a heart model in an atrioventricular plane, as shown in fig. 5, and in step S3, the heart model is constructed according to the following principle:
the atrioventricular plane AVP is formed by the soft tissue surrounding the valves (aortic and mitral valves), which is located between the atria and ventricles, and the myocardial forces F C By ventricular contractile force F VC And atrial contractile force F AC And (4) forming. During ventricular systole (VC), AVP is controlled by F VC Pulling towards the apex of the heart and expelling blood through the aortic valve into the artery. In ventricular diastole (VR), F C Equal to 0,avp moves in reverse under the pressure differential across the piston, causing blood to flow from the left atrium, through the mitral valve, and into the left ventricle. During Atrial Contractions (AC), F C Vertically upwards, the ventricles continue to relax until the next ventricular systole.
The outer wall of the left ventricle is approximated to a cylinder and a hemisphere, and an axial function model is established by analyzing by a infinitesimal method, however, the model is established only by considering the factors of the longitudinal motion of the heart, and according to the display of the cardiac radiography, the longitudinal displacement of the heart is actually accompanied by a small amount of circumferential displacement.
In order to describe the motion characteristics of the heart more accurately, in the atrioventricular plane model, the lateral displacement is equivalent to a coefficient k RAD . The established heart model is as follows:
wherein, V AVP The speed of the plane displacement of the chamber is shown,
represents the rate of change of the plane displacement of the chamber, A LV Is the equivalent piston area on the ventricular side, A LA Is the equivalent piston area on the atrial side, F C Is the myocardial force, Q LV Is the left heart flow, P LV The pressure of the left ventricle is shown,
representing the rate of change of left ventricular pressure, P LA Indicating the pressure in the left atrium, C LV Indicating left ventricular compliance, k RAD Representing the radial function coefficients.
Fig. 6 is a schematic diagram showing a relationship between maximum area/longitudinal displacement and ejection fraction, as shown in fig. 6, in step S4, the method for identifying the physiological parameters of the heart based on the unscented kalman filter is as follows:
based on the relationship between the tension and contraction of the myocardium, the ventricular force is equivalent to a trapezoid, and the selected parameters to be identified are as follows:
wherein A is 1 And A 2 Respectively representing ventricular contraction force F VC And atrial contractionForce F AC Amplitude of (3), R AVP Denotes the damping of AVP, L AVP Representing the inertia of AVP, taking the flow of the left ventricle as an input quantity, taking the displacement speed of the plane of the atrioventricular as an observed quantity, and realizing the identification of the parameters of the model of the plane pump of the atrioventricular by utilizing an unscented Kalman filter; based on the measured parameter information of damping, inertia, responsiveness, myocardial force, and the geometric and circumferential expansion of the ventricles, assessment of the heart state can be achieved.
FIG. 7 shows the myocardial force F C Schematically, as shown in fig. 7, in step S5, the model is optimized by using a nonlinear programming method. The method is characterized in that trapezoidal myocardial force is used as an initial value, each point in a myocardial force curve is used as a variable to be optimized based on a discretization thought, a kinetic equation is used as constraint, the minimum of the atrioventricular plane displacement speed estimation error is used as a judgment index, nonlinear optimization is carried out on the minimum, an interior point method is used for solving, and further optimization on heart parameters and the shape of myocardial force is achieved. Clinical findings show that the occurrence of heart diseases is often accompanied by large changes of relevant heart parameters, so that a doctor can be assisted in diagnosing the heart function by measuring the heart parameters.
Claims (6)
1. A method for diagnosing cardiac function based on an atrioventricular plane pump model, comprising:
s1: the method comprises the steps of obtaining a heart nuclear magnetic resonance image clinically, completing segmentation of a left ventricle, and realizing calculation of the volume of the left ventricle;
s2: selecting characteristic points of the left ventricle in a nuclear magnetic resonance image of the left ventricle, and tracking the characteristic points by utilizing a tracking algorithm to obtain the plane displacement of the atrioventricular organ;
s3: establishing a left ventricle model based on the heart dynamics principle;
s4: identifying physiological parameters of the heart through an unscented Kalman filter;
s5: and optimizing the model by using a nonlinear optimization method.
2. The method for diagnosing cardiac function based on the atrioventricular plane pump model as claimed in claim 1, wherein the step of calculating the volume of the left ventricle in step S1 comprises:
constructing a left ventricle segmentation training set based on a clinically collected left ventricle nuclear magnetic resonance image and a manually labeled left ventricle contour of a medical image expert; performing data enhancement on the training set to obtain a sufficient training set and a sufficient test set for training the U-NET network model; segmenting the left ventricle by adopting the trained U-NET to obtain the area of each slice in the meaning of the left ventricle pixel; and combining the data such as the area of the left ventricle in each slice, the physical distance between adjacent slices and the like to realize the calculation of the volume and the flow of the left ventricle of the heart individual.
3. The method for diagnosing cardiac function based on the planar atrioventricular pump model as claimed in claim 1, wherein in step S2, the step of calculating the displacement of the planar atrioventricular pump comprises:
making a atrioventricular feature point data set through a clinically measured long-axis image of the left ventricle, marking, and dividing a training set and a verification set; constructing a model network structure, selecting a tracking target image and determining a search area, respectively sending feature graphs of different scales into a SimRPN network after extracting features, directly adopting weighted sum to the output of the SimRPN network, and outputting a classification branch and a regression branch by the combined SimRPN network; and calculating the offset of the anchor frame by using the L1 loss, and calculating the loss of confidence by using a cross entropy loss, thereby realizing the automatic tracking of the characteristic points of the ventricle and calculating the plane displacement of the atrioventricular organ. Compared with the situation that a plurality of abnormal values are arranged at key points of the atrioventricular plane tracked by the traditional normalized cross-correlation matching method, the measured atrioventricular plane displacement curve is more in line with physiological performance.
4. The method for diagnosing cardiac function based on the planar atrioventricular pump model as claimed in claim 1, wherein in step S3, the step of calculating the displacement of the planar atrioventricular pump is as follows:
1) The outer wall of the left ventricle is approximate to a cylinder and a hemisphere, and an axial function model is established by analyzing by a infinitesimal method, however, the model is established only by considering the factors of the longitudinal motion of the heart, and according to the display of the cardiac radiography, the longitudinal displacement of the heart is accompanied with a small amount of circumferential displacement actually.
2) In order to describe the motion characteristics of the heart more accurately, in the atrioventricular plane model, the lateral displacement is equivalent to a coefficient k RAD . The established heart model is as follows:
wherein, V AVP The speed of displacement of the plane of the chamber is shown,
represents the rate of change of the plane displacement of the chamber, A LV Is the equivalent piston area on the ventricular side, A LA Is the equivalent piston area on the atrial side, F C Is the myocardial force, Q LV Is the left heart flow, P LV The pressure of the left ventricle is shown,
representing the rate of change of left ventricular pressure, P LA Indicating the pressure in the left atrium, C LV Indicating left ventricular compliance, k RAD Representing the radial function coefficients.
5. The method as claimed in claim 1, wherein in step S4, based on the relationship between the tension and contraction of the myocardium, the ventricular force is equivalent to a trapezoid, and the parameters to be identified are selected as follows:
wherein, A 1 And A 2 Respectively, the ventricular contractility F VC And atrial contractile force F AC Amplitude of (3), R AVP Denotes the damping of AVP, L AVP Representing the inertia of AVP, taking the flow of the left ventricle as an input quantity, taking the plane displacement speed of the atrioventricular cavity as an observed quantity, and realizing the identification of the model parameters of the atrioventricular cavity plane pump by utilizing an unscented Kalman filter; based on the measured parameter information such as damping, inertia, responsiveness, myocardial force and the like, the evaluation of the heart state can be realized by combining the geometric property and the circumferential expansion coefficient of the heart chamber.
6. The method for diagnosing cardiac function based on the atrioventricular planar pump model as claimed in claim 1, wherein in step S5, the trapezoidal myocardial force is used as an initial value, each point in the myocardial force curve is used as a variable to be optimized based on a discretization idea, a kinetic equation is used as a constraint, the minimum atrioventricular planar displacement speed estimation error is used as a judgment index, and the method is used for performing nonlinear optimization and solving by using an interior point method to further optimize the cardiac parameters and the shape of the myocardial force. According to clinical findings, the occurrence of heart diseases is often accompanied by large changes of relevant heart parameters, so that a doctor can be assisted in cardiac function diagnosis through measurement of the heart parameters.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211023705.6A CN115349851A (en) | 2022-08-29 | 2022-08-29 | Cardiac function diagnosis method based on atrioventricular plane pump model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211023705.6A CN115349851A (en) | 2022-08-29 | 2022-08-29 | Cardiac function diagnosis method based on atrioventricular plane pump model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115349851A true CN115349851A (en) | 2022-11-18 |
Family
ID=84004178
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211023705.6A Pending CN115349851A (en) | 2022-08-29 | 2022-08-29 | Cardiac function diagnosis method based on atrioventricular plane pump model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115349851A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117132577A (en) * | 2023-09-07 | 2023-11-28 | 湖北大学 | Method for non-invasively detecting myocardial tissue tension and vibration |
-
2022
- 2022-08-29 CN CN202211023705.6A patent/CN115349851A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117132577A (en) * | 2023-09-07 | 2023-11-28 | 湖北大学 | Method for non-invasively detecting myocardial tissue tension and vibration |
| CN117132577B (en) * | 2023-09-07 | 2024-02-23 | 湖北大学 | Method for non-invasively detecting myocardial tissue tension and vibration |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11024025B2 (en) | Automatic quantification of cardiac MRI for hypertrophic cardiomyopathy | |
| JP7395142B2 (en) | Systems and methods for ultrasound analysis | |
| US8009887B2 (en) | Method and system for automatic quantification of aortic valve function from 4D computed tomography data using a physiological model | |
| CN101084528B (en) | A method and a system for integration of medical diagnostic information and a geometric model of a movable body | |
| Schneider et al. | Mitral annulus segmentation from 3D ultrasound using graph cuts | |
| CN110197713B (en) | Medical image processing method, device, equipment and medium | |
| CN109192305B (en) | Heart function automatic analysis method based on deep circulation neural network | |
| EP1949857B1 (en) | Ultrasonographic device | |
| CN105518684A (en) | Systems and methods for predicting location, onset, and/or change of coronary lesions | |
| CN110599499B (en) | MRI image heart structure segmentation method based on multipath convolutional neural network | |
| CN110363772B (en) | Heart MRI segmentation method and system based on antagonistic network | |
| CN110335236B (en) | Method, apparatus, system and medium for processing cardiac medical image | |
| CN109925002A (en) | Artificial intelligence echocardiogram data collection system and its collecting method | |
| CN110335235A (en) | Processing unit, processing system and the medium of cardiologic medical image | |
| JP2005514997A (en) | Observation system having means for processing an ultrasound image sequence for performing a quantitative calculation of flow in a body organ | |
| US20130013278A1 (en) | Non-invasive cardiovascular image matching method | |
| CN115349851A (en) | Cardiac function diagnosis method based on atrioventricular plane pump model | |
| Yang et al. | Lagrangian-averaged vorticity deviation of spiraling blood flow in the heart during isovolumic contraction and ejection phases | |
| Yong et al. | Automatic ventricular nuclear magnetic resonance image processing with deep learning | |
| US20110317897A1 (en) | Method and apparatus for automated localization of a moving structure | |
| CN114511564B (en) | Image analysis method for breast cancer residual tumor load based on DCE-MRI | |
| Mehta et al. | Heart Disease Diagnosis using Deep Learning | |
| Rahmatikaregar et al. | A Review of Automatic Cardiac Segmentation using Deep Learning and Deformable Models | |
| KR20220141218A (en) | An assistant diagnosis decision system for dilated cardiomyopathy based on automatic algorithm | |
| CN113902738A (en) | Heart MRI segmentation method and 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 |