CN113040825A - Rapid three-dimensional echocardiography speckle tracking method with sub-pixel precision - Google Patents

Rapid three-dimensional echocardiography speckle tracking method with sub-pixel precision Download PDF

Info

Publication number
CN113040825A
CN113040825A CN202110253373.XA CN202110253373A CN113040825A CN 113040825 A CN113040825 A CN 113040825A CN 202110253373 A CN202110253373 A CN 202110253373A CN 113040825 A CN113040825 A CN 113040825A
Authority
CN
China
Prior art keywords
sad
frame
block
tracking
image
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202110253373.XA
Other languages
Chinese (zh)
Other versions
CN113040825B (en
Inventor
孙丰荣
周晓彤
刘莹莹
边栋
邢跃林
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong University
Original Assignee
Shandong University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shandong University filed Critical Shandong University
Priority to CN202110253373.XA priority Critical patent/CN113040825B/en
Publication of CN113040825A publication Critical patent/CN113040825A/en
Application granted granted Critical
Publication of CN113040825B publication Critical patent/CN113040825B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/48—Diagnostic techniques
    • A61B8/483—Diagnostic techniques involving the acquisition of a 3D volume of data
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/52—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Pathology (AREA)
  • Radiology & Medical Imaging (AREA)
  • Biomedical Technology (AREA)
  • Veterinary Medicine (AREA)
  • Medical Informatics (AREA)
  • Physics & Mathematics (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Cardiology (AREA)
  • Image Processing (AREA)
  • Ultra Sonic Daignosis Equipment (AREA)
  • Image Analysis (AREA)

Abstract

The invention relates to a sub-pixel precision rapid three-dimensional echocardiography speckle tracking method, which comprises the following steps: selecting a tracking point by an operating doctor, and setting initial parameters such as block size, search area size and the like; performing 1/16 sub-pixel precision block matching tracking after aligning the reference block pixel values in alternate frames (odd frames); and (3) obtaining the even frame tracking point coordinates through spline interpolation calculation, and automatically marking the myocardial tissue interested region on the 3D echocardiography sequence image by a computer. The invention realizes a high-precision fast 3D echocardiography speckle tracking method, and the method has good self-adaptive characteristic and robustness. The method is suitable for the 3D STE imaging technology aiming at realizing accurate evaluation and quantitative diagnosis and treatment of the cardiac function. The 2D variant of the method of the invention is a fast 2D echocardiography speckle tracking method of sub-pixel accuracy.

Description

Rapid three-dimensional echocardiography speckle tracking method with sub-pixel precision
Technical Field
The invention relates to a three-dimensional (3D) echocardiography Speckle tracking method, in particular to a fast 3D echocardiography Speckle tracking method with sub-pixel precision, which belongs to the technical field of medical ultrasonic image processing and aims to realize 3D STE imaging technology (3D speed tracking echo diagnosis) of accurate heart function evaluation and quantitative diagnosis and treatment.
Background
Heart failure is the last stage in the development of cardiovascular diseases such as coronary heart disease, hypertension, congenital heart disease, valvular heart disease, cardiomyopathy, etc., is the leading cause of death in patients with cardiovascular diseases and the last battlefield of clinical treatment, and is considered to be the biggest challenge facing the cardiovascular world in the 21 st century. The method has the advantages that the quantitative evaluation of the heart function is carried out in the early stage for high-risk people who are easy to suffer from heart failure, the abnormal systolic and diastolic functions are recognized as early as possible, effective intervention is carried out, the occurrence and development of the heart failure are favorably prevented and delayed, and the service life of patients suffering from cardiovascular diseases is prolonged and prognosis is improved. The Speckle tracking three-dimensional echocardiography (3D STE) is an emerging ultrasound imaging technology which aims at realizing accurate evaluation and quantitative diagnosis and treatment of cardiac function, and can accurately realize velocity vector imaging, strain and strain rate imaging of myocardial tissue and measurement and visualization of motion parameters such as left ventricular torsional angular displacement, torsional moment, torsional angular velocity and the like through tracking and motion analysis of 3D echocardiography Speckle (Speckle) or Speckle pattern (Speckle pattern), thereby providing quantitative analysis basis for a clinician to objectively evaluate the whole and local functions of the myocardium and the contraction and relaxation functions of the ventricle, and providing technical means for the clinician to quantitatively research the biological elasticity mechanical properties of the myocardial tissue.
In echocardiographic speckle tracking methods, the speckle or speckle pattern to be tracked and analyzed is, in image processing practice, a number of regions of interest (ROIs) located within the myocardial tissue, i.e., blocks or kernels of conventional Block matching processing (Block/Kernel). The echocardiography speckle tracking methods reported in the literatures mainly have two types: a speckle tracking method of block matching and a speckle tracking method of an optical flow field. The speckle tracking method (Optical flow based specific tracking) of the Optical flow field can realize high-precision myocardial tissue motion tracking, but has strict requirements on image quality and poor clinical practicability. The speckle tracking method (Block matching based speckle tracking) of Block matching has good clinical practicability and is a mainstream method in the technical field of 3D STE imaging, but the traditional Block matching method is difficult to realize high-precision myocardial tissue motion tracking. In addition, the conventional methods generally have poor clinical application performance such as computational processing efficiency, adaptive characteristics, robustness and the like.
Disclosure of Invention
Aiming at the defects that the conventional method is low in tracking precision or poor in clinical practicability and generally cannot meet the requirements of accurate cardiac function evaluation and quantitative diagnosis and treatment clinical application, the invention provides the rapid 3D echocardiogram speckle tracking method which has good self-adaptive characteristic and robustness and 1/16 sub-pixel precision and is based on the technical principles of good performance and clinical practicality.
Interpretation of terms:
1. cubic Spline Interpolation, where a Spline refers to a smooth curve passing through a series of shape points, is a mathematical method of computing such a curve.
2. The Full search algorithm, Full search, a strategy for implementing block matching operations in image processing, obtains matching blocks by evaluating all candidate points within a search area.
3. SAD, Sum of absolute differences.
4. Imaging system 3D image space, a technical term in the field of 3D medical imaging, refers in the present invention to the cartesian space where the 3D echocardiographic sequence images are located, as shown in fig. 3, three dimensions refers to X, Y, Z coordinate three dimensions of the cartesian space.
The technical scheme of the invention is as follows:
a speckle tracking method for 3D echocardiography is realized by a computer workstation configured with accurate heart function evaluation and quantitative diagnosis and treatment functions. Let 3D echocardiography sequence images consist of n frames of images, denoted as I k1,2, n, wherein IkFor the k frame image of a 3D echocardiographic sequence image, I1For an initial frame, InIs a last frame, I1,I3,...,InIs an odd number of frames anda sequence image composed of the last frame; the reference block used for block matching of the t-th frame image is denoted as RtThe reference block after pixel value calibration used for block matching of the t-th frame image is denoted as "1, 3,5
Figure BDA0002965221040000021
The best matching block obtained on the t-th frame image is denoted as MtThe position of the tracking point obtained on the t-th frame image in the 3D image space of the imaging system is recorded as rt(xt,yt,zt) Wherein (x)t,yt,zt) Representing the coordinates of the tracking point in 3D image space, the steps of the method are as follows:
s1) by the clinician at the initial frame I1Manually selecting a tracking point;
in the initial frame I of the 3D echocardiographic sequence image1The tracking point is manually selected by a clinician, and the position of the tracking point in the 3D image space of the imaging system is r1(x1,y1,z1)；
S2) setting initial parameters;
setting the size n of the myocardial tissue region of interest centered on the tracking point of each framex×ny×nzWherein n isx、ny、nzThe sizes along the directions of x, y and z axes in the 3D image space of the imaging system respectively; setting the size of a search area; setting a factor m for reference block pixel value calibration; making the current frame serial number j equal to 1; setting an initial reference block size by rj(xj,yj,zj) Forming an initial reference block R centered on a pointj(ii) a Order to
Figure BDA0002965221040000022
RjRefers to a reference block used for block matching in the jth frame image,
Figure BDA0002965221040000023
refers to a reference block, M, after pixel value calibration, used for block matching of the jth frame imagejMeans the best matching block obtained on the j frame image;
s3) reference block pixel value calibration;
j is increased by 2; let Rj＝Mj-2；
For reference block RjThe pixel value of the reference block is calibrated to obtain a reference block which is used by the jth frame image for block matching and is calibrated by the pixel value
Figure BDA0002965221040000031
S4) carrying out block matching calculation with 1/16 sub-pixel precision to obtain the coordinate r of the position of the tracking point of the current frame in the 3D image space of the imaging systemj(xj,yj,zj)；
S5) if IjIf the frame is the last frame, executing S6, otherwise executing S3;
s6) calculating the position of the even frame tracking point in the imaging system 3D image space
Respectively taking the coordinates of the odd frame and the last frame of tracking points in the 3D image space of the imaging system, and calculating the coordinates of the even frame of tracking points in the x axis, the y axis and the z axis of the imaging system through cubic spline interpolation to obtain the positions of the even frame of tracking points in the 3D image space of the imaging system and further obtain the positions of all the frame of tracking points in the 3D image space of the imaging system;
s7) forming and recording the myocardial tissue region of interest of all frames
Forming and recording the size of each frame as n by taking the tracking point of all frames as the centerx×ny×nzTo realize speckle tracking of 3D echocardiography.
According to the invention, preferably, for reference block RjIs calibrated, as shown in equation (i):
Figure BDA0002965221040000032
in the formula (I), 0<m,p,q<1，m+p+q＝1，p＝0.9(1-m)/(n-3)2·(j-3)2，q＝1-m-p。
According to the invention, preferably, in
Figure BDA0002965221040000033
For reference blocks, 1/16 sub-pixel precision block matching calculation is performed, which is implemented as follows:
s41) at IjUp to rj-2(xj-2,yj-2,zj-2) Carrying out cubic spline interpolation for the first time in the search region with the point as the center, so that 3 equally-spaced inner points among 2 adjacent nodes are interpolated to obtain 1/4 search regions with sub-pixel precision S11;
s42), adopting a full search algorithm and an SAD (sum of absolute differences) matching criterion, using an SAD iterative computation early termination strategy, carrying out primary matching in a search area S11, and recording a point corresponding to the minimum SAD value as a primary 1/4 sub-pixel precision tracking point R11;
s43) with the first 1/4 sub-pixel precision tracking point R11 as the center point, performing a second cubic spline interpolation on the surrounding 8 1/4 voxel blocks, and then interpolating 3 equally spaced interior points between 2 adjacent nodes to obtain a 1/16 sub-pixel precision search region S12.
S44) performs compression processing on the search region S12 by calculating 1/8 sub-pixel precision tracking points:
calculating SAD values of center points B1, B2, B3, B4, B5, B6, B7 and B8 of 8 1/4 voxel blocks, and dividing 1/4 voxel blocks into secondary search areas S13 by taking points (taking B5 as an example) corresponding to the minimum SAD value as centers;
s44) adopting full search algorithm and SAD matching criterion, and using SAD iterative computation to terminate strategy in advance for secondary matching, wherein the point corresponding to minimum SAD value is the tracking point r of 1/16 sub-pixel precision of current frame (j frame)j(xj,yj,zj) The block centered at this point is the best matching block M on the current frame (jth frame)j。
Preferably, in step S42) and step S44), the implementation process is as follows:
let CAmaxAs a full search candidateTotal number of blocks, LmaxFor full search candidate block z-axis direction layer number, SADminFor the current minimum SAD value, r represents the current calculated r-th candidate block, SADrsRepresenting the sum of the previous s-layer errors, SAD, of the r-th candidate blocksRepresenting the sum of the error of the s layer, the specific calculation flow is as follows:
s421) calculating an initial candidate block SAD as SADminLet r be 1;
s422) judging whether the block is the last candidate block, namely r>CAmaxIf true, then the current SADminFor minimum SAD, the calculation ends, otherwise r is increased by 1, s is equal to 1, SADrs0, and performing step S423);
s423) judging whether the candidate block is the last layer of the current candidate block, namely judging S>LmaxIf yes, then recording the current SADrsIs SADminGo to step S422), otherwise, calculate the S-th layer SAD error SADsParallel SADrs＝SADrs+SADsAnd step S424) is performed;
s424) judging the previous S-layer error and SADrsWhether it is greater than SADminIf yes, go to step S422), otherwise, S is incremented by 1, go to step S423).
The invention has the beneficial effects that:
the invention adopts a frame separation block matching processing frame; adaptively calibrating pixel values of a reference block in conjunction with medical knowledge of cardiac motion to avoid propagation and accumulation of tracking errors; 1/16 sub-pixel tracking precision is obtained by cubic spline interpolation of the search area; and the calculation efficiency of the block matching processing is improved by adopting the strategies of compressing the search area, terminating SAD iterative calculation in advance and the like. The high-precision fast 3D echocardiography speckle tracking method is realized, and the method has good self-adaptive characteristic and robustness. The method is suitable for the 3D STE imaging technology (3D speed tracking echocardiography) aiming at realizing accurate evaluation and quantitative diagnosis and treatment of the heart function. The method is beneficial to improving the clinical diagnosis and treatment level of quantitative evaluation of the cardiac function of common cardiovascular diseases such as coronary heart disease, hypertension and the like in China.
Drawings
FIG. 1(a) is a schematic diagram of the obtained 1/4 sub-pixel precision tracking point R11 and 8 1/4 voxel blocks around the tracking point;
FIG. 1(b) is a schematic diagram of 1/16 search region S12 with sub-pixel accuracy;
FIG. 1(c) is a schematic diagram of the center points B1, B2, B3, B4, B5, B6, B7 and B8 of 8 1/4 voxel blocks;
fig. 1(d) is a schematic diagram of the secondary search area S13;
FIG. 2 is a schematic flow diagram of a method of 3D echocardiography speckle tracking;
fig. 3 is a schematic diagram of a 3D image space coordinate system of the imaging system.
Detailed Description
The invention is further defined in the following, but not limited to, the figures and examples in the description.
Example 1
In the embodiment, a home-made heart ultrasonic diagnostic apparatus and a computer workstation are configured in an ultrasonic examination room of a cardiology department of a certain hospital. The main function of the workstation is to implement a 3D STE imaging technique based on the method of the invention. Namely, the method of the invention is used for carrying out speckle tracking on 3D echocardiography image data, and realizing the speed vector imaging, strain and strain rate imaging of myocardial tissues, the measurement and the imaging of motion parameters such as left ventricle torsional angular displacement, torsional moment, torsional angular velocity and the like. Through a USB flash disk and other mobile storage media, a clinician transfers and stores 3D echocardiogram image data acquired by an ultrasonic diagnostic apparatus to a workstation, and 3D echocardiogram sequence images are set to be composed of n frames of images and are respectively marked as Ik1,2, n, wherein IkFor the k frame image of a 3D echocardiographic sequence image, I1For an initial frame, InIs a last frame, I1,I3,...,InThe sequence image is formed by odd frames and last frames; the reference block used for block matching of the t-th frame image is denoted as RtThe reference block after pixel value calibration used for block matching of the t-th frame image is denoted as "1, 3,5
Figure BDA0002965221040000051
The best matching block obtained on the t-th frame image is denoted as MtThe position of the tracking point obtained on the t-th frame image in the 3D image space of the imaging system is recorded as rt(xt,yt,zt) Wherein (x)t,yt,zt) Representing the coordinates of the tracking point in 3D image space (as shown in fig. 3), as shown in fig. 2, the steps of the method are as follows:
s1) by the clinician at the initial frame I1Manually selecting a tracking point;
in the initial frame I of the 3D echocardiographic sequence image1The tracking point is manually selected by a clinician, and the position of the tracking point in the 3D image space of the imaging system is r1(x1,y1,z1)；
S2) setting initial parameters;
setting the size n of the myocardial tissue region of interest centered on the tracking point of each framex×ny×nzWherein n isx、ny、nzThe sizes along the directions of x, y and z axes in the 3D image space of the imaging system respectively; setting the size of a search area; setting a factor m for reference block pixel value calibration; making the current frame serial number j equal to 1; setting an initial reference block size by rj(xj,yj,zj) Forming an initial reference block R centered on a pointj(ii) a Order to
Figure BDA0002965221040000061
RjRefers to a reference block used for block matching in the jth frame image,
Figure BDA0002965221040000062
refers to a reference block, M, after pixel value calibration, used for block matching of the jth frame imagejMeans the best matching block obtained on the j frame image;
s3) reference block pixel value calibration;
j is increased by 2; let Rj＝Mj-2；
For reference blockRjThe pixel value of the reference block is calibrated to obtain a reference block which is used by the jth frame image for block matching and is calibrated by the pixel value
Figure BDA0002965221040000063
S4) carrying out block matching calculation with 1/16 sub-pixel precision to obtain the coordinate r of the position of the tracking point of the current frame in the 3D image space of the imaging systemj(xj,yj,zj)；
S5) if IjIf the frame is the last frame, executing S6, otherwise executing S3;
s6) calculating the position of the even frame tracking point in the imaging system 3D image space
Respectively taking the coordinates of the odd frame and the last frame of tracking points in the 3D image space of the imaging system, and calculating the coordinates of the even frame of tracking points in the x axis, the y axis and the z axis of the imaging system through cubic spline interpolation to obtain the positions of the even frame of tracking points in the 3D image space of the imaging system and further obtain the positions of all the frame of tracking points in the 3D image space of the imaging system;
s7) forming and recording the myocardial tissue region of interest of all frames
Forming and recording the size of each frame as n by taking the tracking point of all frames as the centerx×ny×nzTo realize speckle tracking of 3D echocardiography.
By using the recorded region of interest, the workstation software reliably, conveniently and efficiently assists clinicians in accurately evaluating the cardiac function of a patient and quantitatively diagnosing and treating the cardiac function of the patient through modes or means such as velocity vector imaging, strain and strain rate imaging of myocardial tissues, left ventricle torsional motion parameter measurement and imaging and the like.
Example 2
A method of 3D echocardiographic speckle tracking according to embodiment 1, which differs in that:
for reference block RjIs calibrated, as shown in equation (i):
Figure BDA0002965221040000064
in the formula (I), 0<m,p,q<1，m+p+q＝1，p＝0.9(1-m)/(n-3)2·(j-3)2，q＝1-m-p。
Example 3
A method of 3D echocardiographic speckle tracking according to embodiment 1, which differs in that:
to be provided with
Figure BDA0002965221040000071
For reference blocks, 1/16 sub-pixel precision block matching calculation is performed, which is implemented as follows:
s41) at IjUp to rj-2(xj-2,yj-2,zj-2) Carrying out cubic spline interpolation for the first time in the search region with the point as the center, so that 3 equally-spaced inner points among 2 adjacent nodes are interpolated to obtain 1/4 search regions with sub-pixel precision S11;
s42), adopting a full search algorithm and an SAD (sum of absolute differences) matching criterion, using an SAD iterative computation early termination strategy, carrying out primary matching in a search area S11, and recording a point corresponding to the minimum SAD value as a primary 1/4 sub-pixel precision tracking point R11;
s43) with the first 1/4 sub-pixel precision tracking point R11 as the center point, performing a second cubic spline interpolation on the surrounding 8 1/4 voxel blocks (such as the region with a1-a8 as the vertex in fig. 1 (a)), and then interpolating 3 equally spaced interior points between 2 adjacent nodes to obtain a 1/16 sub-pixel precision search region S12 (such as fig. 1 (b)).
S44) performs compression processing on the search region S12 by calculating 1/8 sub-pixel precision tracking points:
calculating SAD values of center points B1, B2, B3, B4, B5, B6, B7 and B8 (as shown in fig. 1(C)) of 8 1/4 voxel blocks, and taking a point (for example, B5) corresponding to the minimum SAD value as a center to define 1/4 voxel blocks as a secondary search region S13 (as shown in fig. 1(d), regions with C1-C7 and R11 as vertexes);
s44) adopting a full search algorithm and SAD matching criteria, anUsing SAD iterative computation early termination strategy to carry out secondary matching, wherein the point corresponding to the minimum SAD value is the tracking point r of the current frame (jth frame) 1/16 with sub-pixel precisionj(xj,yj,zj) The block centered at this point is the best matching block M on the current frame (jth frame)j。
The 2D variant of the method is a fast 2D echocardiogram speckle tracking method with sub-pixel precision, namely the method is also suitable for carrying out speckle tracking on the 2D echocardiogram.
Example 4
A method of 3D echocardiographic speckle tracking according to embodiment 1, which differs in that:
step S42) and step S44), the specific implementation process is as follows:
let CAmaxTotal number of candidate blocks for full search, LmaxFor full search candidate block z-axis direction layer number, SADminFor the current minimum SAD value, r represents the current calculated r-th candidate block, SADrsRepresenting the sum of the previous s-layer errors, SAD, of the r-th candidate blocksRepresenting the sum of the error of the s layer, the specific calculation flow is as follows:
s421) calculating an initial candidate block SAD as SADminLet r be 1;
s422) judging whether the block is the last candidate block, namely r>CAmaxIf true, then the current SADminFor minimum SAD, the calculation ends, otherwise r is increased by 1, s is equal to 1, SADrs0, and performing step S423);
s423) judging whether the candidate block is the last layer of the current candidate block, namely judging S>LmaxIf yes, then recording the current SADrsIs SADminGo to step S422), otherwise, calculate the S-th layer SAD error SADsParallel SADrs＝SADrs+SADsAnd step S424) is performed;
s424) judging the previous S-layer error and SADrsWhether it is greater than SADminIf yes, go to step S422), otherwise, S is incremented by 1, go to step S423).

Claims (4)

1. A3D echocardiogram speckle tracking method is characterized in that a 3D echocardiogram sequence image is composed of n frames of images which are respectively marked as Ik1,2, n, wherein IkFor the k frame image of a 3D echocardiographic sequence image, I1For an initial frame, InIs a last frame, I1,I3,...,InThe sequence image is formed by odd frames and last frames; the reference block used for block matching of the t-th frame image is denoted as RtThe reference block after pixel value calibration used for block matching of the t-th frame image is denoted as "1, 3,5
Figure FDA0002965221030000011
The best matching block obtained on the t-th frame image is denoted as MtThe position of the tracking point obtained on the t-th frame image in the 3D image space of the imaging system is recorded as rt(xt,yt,zt) Wherein (x)t,yt,zt) Representing the coordinates of the tracking point in 3D image space, the steps of the method are as follows:
s1) in the initial frame I1Manually selecting a tracking point;
in the initial frame I of the 3D echocardiographic sequence image1The tracking point is selected manually, and the position of the tracking point in the 3D image space of the imaging system is r1(x1,y1,z1)；
S2) setting initial parameters;
setting the size n of the myocardial tissue region of interest centered on the tracking point of each framex×ny×nzWherein n isx、ny、nzThe sizes along the directions of x, y and z axes in the 3D image space of the imaging system respectively; setting the size of a search area; setting a factor m for reference block pixel value calibration; making the current frame serial number j equal to 1; setting an initial reference block size by rj(xj,yj,zj) Forming an initial reference block R centered on a pointj(ii) a Order to
Figure FDA0002965221030000012
RjRefers to a reference block used for block matching in the jth frame image,
Figure FDA0002965221030000013
refers to a reference block, M, after pixel value calibration, used for block matching of the jth frame imagejMeans the best matching block obtained on the j frame image;
s3) reference block pixel value calibration;
j is increased by 2; let Rj＝Mj-2；
For reference block RjThe pixel value of the reference block is calibrated to obtain a reference block which is used by the jth frame image for block matching and is calibrated by the pixel value
Figure FDA0002965221030000014
S4) carrying out block matching calculation with 1/16 sub-pixel precision to obtain the coordinate r of the position of the tracking point of the current frame in the 3D image space of the imaging systemj(xj,yj,zj)；
S5) if IjIf the frame is the last frame, executing S6, otherwise executing S3;
s6) calculating the position of the even frame tracking point in the imaging system 3D image space
Respectively taking the coordinates of the odd frame and the last frame of tracking points in the 3D image space of the imaging system, and calculating the coordinates of the even frame of tracking points in the x axis, the y axis and the z axis of the imaging system through cubic spline interpolation to obtain the positions of the even frame of tracking points in the 3D image space of the imaging system and further obtain the positions of all the frame of tracking points in the 3D image space of the imaging system;
s7) forming and recording the myocardial tissue region of interest of all frames
Forming and recording the size of each frame as n by taking the tracking point of all frames as the centerx×ny×nzTo realize speckle tracking of 3D echocardiography.
2. The method for 3D echocardiography speckle tracking according to claim 1, wherein R is the reference blockjIs calibrated, as shown in equation (i):
Figure FDA0002965221030000021
in the formula (I), 0<m,p,q<1，m+p+q＝1，p＝0.9(1-m)/(n-3)2·(j-3)2，q＝1-m-p。
3. The method of 3D echocardiography speckle tracking according to claim 1, further comprising performing a speckle reduction on the image to obtain a speckle reduction image
Figure FDA0002965221030000022
For reference blocks, 1/16 sub-pixel precision block matching calculation is performed, which is implemented as follows:
s41) at IjUp to rj-2(xj-2,yj-2,zj-2) Carrying out first cubic spline interpolation on the search region with the point as the center to obtain 1/4 search regions with sub-pixel precision S11;
s42) adopting a full search algorithm and an SAD matching criterion, using an SAD iterative computation early termination strategy, carrying out primary matching in a search region S11, and recording a point corresponding to the minimum SAD value as a primary 1/4 sub-pixel precision tracking point R11;
s43) taking the first 1/4 sub-pixel precision tracking point R11 as a central point, carrying out second cubic spline interpolation on 8 surrounding 1/4 pixel blocks to obtain a search area S12 with 1/16 sub-pixel precision;
s44) performs compression processing on the search region S12 by calculating 1/8 sub-pixel precision tracking points:
calculating SAD values of center points B1, B2, B3, B4, B5, B6, B7 and B8 of 8 1/4 voxel blocks, and taking a point corresponding to the minimum SAD value as a center to demarcate 1/4 voxel blocks as a secondary search area S13;
s44) using the full search algorithm and the SAD matching criterion, and using SAD iterationCalculating a termination strategy in advance, performing secondary matching, wherein the point corresponding to the minimum SAD value is the tracking point r of the current frame 1/16 with sub-pixel precisionj(xj,yj,zj) The block centered at this point is the best matching block M on the current framej。
4. The method for speckle tracking of 3D echocardiography according to claim 3, wherein the steps S42) and S44) are implemented as follows:
let CAmaxTotal number of candidate blocks for full search, LmaxFor full search candidate block z-axis direction layer number, SADminFor the current minimum SAD value, r represents the current calculated r-th candidate block, SADrsRepresenting the sum of the previous s-layer errors, SAD, of the r-th candidate blocksRepresenting the sum of the error of the s layer, the specific calculation flow is as follows:
s421) calculating an initial candidate block SAD as SADminLet r be 1;
s422) judging whether the block is the last candidate block, namely r>CAmaxIf true, then the current SADminFor minimum SAD, the calculation ends, otherwise r is increased by 1, s is equal to 1, SADrs0, and performing step S423);
s423) judging whether the candidate block is the last layer of the current candidate block, namely judging S>LmaxIf yes, then recording the current SADrsIs SADminGo to step S422), otherwise, calculate the S-th layer SAD error SADsParallel SADrs＝SADrs+SADsAnd step S424) is performed;
s424) judging the previous S-layer error and SADrsWhether it is greater than SADminIf yes, go to step S422), otherwise, S is incremented by 1, go to step S423).
CN202110253373.XA 2021-03-08 2021-03-08 Rapid three-dimensional echocardiography speckle tracking method with sub-pixel precision Active CN113040825B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110253373.XA CN113040825B (en) 2021-03-08 2021-03-08 Rapid three-dimensional echocardiography speckle tracking method with sub-pixel precision

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110253373.XA CN113040825B (en) 2021-03-08 2021-03-08 Rapid three-dimensional echocardiography speckle tracking method with sub-pixel precision

Publications (2)

Publication Number Publication Date
CN113040825A true CN113040825A (en) 2021-06-29
CN113040825B CN113040825B (en) 2022-02-11

Family

ID=76510602

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110253373.XA Active CN113040825B (en) 2021-03-08 2021-03-08 Rapid three-dimensional echocardiography speckle tracking method with sub-pixel precision

Country Status (1)

Country Link
CN (1) CN113040825B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101617947A (en) * 2009-07-17 2010-01-06 重庆医科大学 Cardiac muscle texture automatic analyzing method based on spot tracking technology
CN103429164A (en) * 2012-03-30 2013-12-04 东芝医疗系统株式会社 Ultrasonic diagnostic device, image processing device, and image processing method
CN103810708A (en) * 2014-02-13 2014-05-21 西安交通大学 Method and device for perceiving depth of laser speckle image
CN103873875A (en) * 2014-03-25 2014-06-18 山东大学 Layering sub pixel motion estimation method for image super resolution
US20150327838A1 (en) * 2012-11-15 2015-11-19 Imperial Innovations Ltd Echocardiography

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101617947A (en) * 2009-07-17 2010-01-06 重庆医科大学 Cardiac muscle texture automatic analyzing method based on spot tracking technology
CN103429164A (en) * 2012-03-30 2013-12-04 东芝医疗系统株式会社 Ultrasonic diagnostic device, image processing device, and image processing method
US20150327838A1 (en) * 2012-11-15 2015-11-19 Imperial Innovations Ltd Echocardiography
CN103810708A (en) * 2014-02-13 2014-05-21 西安交通大学 Method and device for perceiving depth of laser speckle image
CN103873875A (en) * 2014-03-25 2014-06-18 山东大学 Layering sub pixel motion estimation method for image super resolution

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
金培培: "超声心动图三维散斑跟踪技术及其应用研究", 《中国优秀硕士学位论文全文数据库（电子期刊） 信息科技辑》 *

Also Published As

Publication number Publication date
CN113040825B (en) 2022-02-11

Similar Documents

Publication Publication Date Title
Hashemi et al. Global time-delay estimation in ultrasound elastography
US6716175B2 (en) Autonomous boundary detection system for echocardiographic images
Reant et al. Evaluation of global left ventricular systolic function using three-dimensional echocardiography speckle-tracking strain parameters
Kühl et al. High-resolution transthoracic real-time three-dimensional echocardiography: quantitation of cardiac volumes and function using semi-automatic border detection and comparison with cardiac magnetic resonance imaging
US10321892B2 (en) Computerized characterization of cardiac motion in medical diagnostic ultrasound
JP4427548B2 (en) Apparatus and method for analyzing locally deformable motion
Dwivedi et al. Reference values for mitral and tricuspid annular dimensions using two-dimensional echocardiography
WO2010113998A1 (en) Medical image diagnosis device and volume calculating method
US20230326034A1 (en) Automated right ventricle medical imaging and computation of clinical parameters
Caselli et al. Three-dimensional echocardiographic characterization of patients with left ventricular noncompaction
JP2023502449A (en) Intelligent measurement aids for ultrasound imaging and related devices, systems and methods
JP6039876B2 (en) System and method for measuring left ventricular torsion
US20180192987A1 (en) Ultrasound systems and methods for automatic determination of heart chamber characteristics
JPH06503483A (en) Automatic digital image quantification system
CN111815597A (en) Left ventricle long and short axis tangent plane extraction method and device based on CT image, computer equipment and storage medium
TWI445520B (en) Methods of comparison of non - invasive cardiovascular status
Beyls et al. Usefulness of right ventricular longitudinal shortening fraction to detect right ventricular dysfunction in acute cor pulmonale related to COVID-19
US20110317897A1 (en) Method and apparatus for automated localization of a moving structure
Yue et al. Speckle tracking in intracardiac echocardiography for the assessment of myocardial deformation
WO2024126468A1 (en) Echocardiogram classification with machine learning
CN113040825B (en) Rapid three-dimensional echocardiography speckle tracking method with sub-pixel precision
Perperidis et al. Temporal compounding: a novel implementation and its impact on quality and diagnostic value in echocardiography
CN113592802B (en) Mitral valve annular displacement automatic detection system based on ultrasonic image
Bouchez et al. In-vivo validation of a new clinical tool to quantify three-dimensional myocardial strain using ultrasound
CN108289655B (en) Ultrasound cardiac assessment of a heart with medial axis flexion and lateral eccentricity

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant