CN119924891A - A two-dimensional speckle tracking combined with tissue Doppler analysis system for OSAS cardiac mechanics - Google Patents
A two-dimensional speckle tracking combined with tissue Doppler analysis system for OSAS cardiac mechanics Download PDFInfo
- Publication number
- CN119924891A CN119924891A CN202510429222.3A CN202510429222A CN119924891A CN 119924891 A CN119924891 A CN 119924891A CN 202510429222 A CN202510429222 A CN 202510429222A CN 119924891 A CN119924891 A CN 119924891A
- Authority
- CN
- China
- Prior art keywords
- strain
- osas
- module
- risk
- velocity
- 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
Links
Landscapes
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The invention belongs to the technical field of medical image analysis, and particularly relates to an OSAS heart mechanics analysis system combining two-dimensional speckle tracking and tissue Doppler, which realizes high-precision time axis alignment of a two-dimensional ultrasonic gray-scale image sequence and a myocardial tissue Doppler velocity spectrum through a two-channel synchronous acquisition module and generates longitudinal strain of an endocardial layer, circumferential strain of a middle layer and velocity peak value parameters by a parallel calculation engine. The dynamic association module adopts a dynamic time warping algorithm to quantify the phase deviation of the layered strain curve and the speed curve, and combines the risk index model of the risk early warning module to output low, medium and high three-level risk classification. The visual interface dynamically displays the myocardial motion asynchronous area through thermodynamic diagrams, speed vector arrows and abnormal flickering marks, and the hardware integrated component and the standardized analysis flow obviously improve the detection sensitivity and the diagnosis efficiency. The invention solves the problem of multi-mode data fracturing and early warning hysteresis.
Description
Technical Field
The invention relates to the technical field of medical image analysis, in particular to an OSAS heart mechanics analysis system combining two-dimensional speckle tracking and tissue Doppler.
Background
Obstructive Sleep Apnea Syndrome (OSAS) is a common disorder characterized by nocturnal apnea and hypoxia, and long-term illness is prone to compensatory impairment of cardiac function. In clinical diagnosis, doctors usually rely on ultrasonic imaging technology to evaluate the mechanical characteristics of the heart muscle of the left ventricle of a patient, however, most of the existing ultrasonic analysis methods only pay attention to single-mode parameters, and are difficult to comprehensively capture the dynamic relevance of the strain and the speed of the heart muscle in the contraction process.
In the traditional method, the spot tracking and the Doppler velocity spectrum separation analysis of myocardial tissue are carried out, the time axis is not aligned accurately, so that doctors need to compare and report manually, the phase deviation calculation error is large, the joint modeling of layered strain and velocity parameters is lacking, myocardial compensation abnormality is difficult to discover timely, and meanwhile, the result is presented in a static numerical value, so that a high risk area cannot be positioned in a visual way.
Aiming at the problems, the novel design is carried out on the basis of the original OSAS heart mechanics analysis system.
Disclosure of Invention
The invention aims to provide an OSAS heart mechanical analysis system combining two-dimensional speckle tracking and tissue Doppler, which aims at solving the problems of multi-mode data fracture, insufficient time synchronism, myocardial compensation abnormal early warning lag and visual interaction deficiency in the prior art.
For this purpose, the invention adopts the following technical scheme:
An OSAS heart mechanics analysis system combining two-dimensional speckle tracking and tissue Doppler comprises the following modules,
M1, a double-channel data acquisition module, wherein the module is configured to synchronously acquire a left ventricle two-dimensional ultrasonic gray scale image sequence and a myocardial tissue Doppler velocity spectrum by an ultrasonic probe, and realize time axis alignment of the two types of data;
m2, parallel computing engine module, this module is connected with binary channels data acquisition module, parallel computing engine module includes first processing module and second processing module:
The first processing module receives a two-dimensional ultrasonic gray-scale image sequence, and generates longitudinal strain and circumferential strain parameters of an endocardial layer, a myocardial middle layer and an epicardial layer through a speckle tracking algorithm;
the second processing module receives Doppler velocity spectrum of myocardial tissue, calculates myocardial motion velocity peak value through time domain integration Acceleration parameter;
M3, a dynamic association module which is connected with the parallel computing engine module, wherein the dynamic association module carries out time series fusion on the parameters of the longitudinal strain of the endocardium layer, the circumferential strain of the middle layer and the circumferential strain of the epicardium layer of the first processing module and the speed peak value and the acceleration parameter of the second processing module to generate a strain and speed combined characteristic map;
And M4, an OSAS risk early warning module, which is connected with the dynamic association module, generates left ventricular contraction function compensation state grading and an OSAS risk index based on the phase deviation and the parameter change trend in the combined characteristic map, and outputs the left ventricular contraction function compensation state grading and the OSAS risk index to the visual interaction interface.
Further, the dynamic correlation module calculates the phase deviation of the layered strain curve S (t) and the speed curve V (t) through a dynamic time warping algorithm, and specifically comprises the following steps of;
Wherein the method comprises the steps of Pi is the optimal alignment path of two time sequences; a path that is the smallest cumulative distance among the aligned paths pi; For the point in time The unit of the layered strain value is that the unit comprises the longitudinal strain of an endocardial layer and the circumferential strain of a middle layer of cardiac muscle; Time point Myocardial motion velocity values in cm/s, including velocity peaks;
Prior to calculating the phase deviation, the layered strain valuesAnd velocity valueCarrying out standardization processing and converting into dimensionless parameters;
When (when) Triggering myocardial decompensation abnormality early warning when tau is larger than tau, wherein tau is a dimensionless threshold value.
Further, the OSAS risk early warning module calculates a risk index according to the following formula:
Wherein the method comprises the steps of The risk index is OSAS risk index, the index range is 0-1, alpha and beta are weight coefficients, and alpha+beta=1; The variable quantity of the longitudinal strain of the endocardial layer in the time window delta t is expressed as a unit; the change amount of the speed peak value in the time window delta t is expressed in cm/s, delta t is the analysis time window length and the unit is seconds.
Further, the dual-channel data acquisition module further comprises a preprocessing sub-module, which is specifically as follows:
the preprocessing submodule carries out automatic segmentation of myocardial contours on the two-dimensional ultrasonic gray scale image sequence and divides areas of an endocardial layer, a middle layer and an epicardial layer;
and the preprocessing submodule carries out respiratory motion artifact filtering on the Doppler velocity spectrum of the myocardial tissue and retains a relevant velocity signal of the systole.
Further, the visual interaction interface of the OSAS risk early warning module includes displaying a layered strain thermodynamic diagram, overlaying a tissue doppler velocity vector arrow, and an abnormal region highlighting prompt:
the display layered strain thermodynamic diagram maps strain distribution of each segment of the left ventricle in a color gradient based on the layered strain parameters generated by the first processing module;
red area, i.e. longitudinal strain of endocardial layer is less than-20%;
yellow region, wherein the longitudinal strain of endocardial layer is less than or equal to-20 percent and less than or equal to-15 percent;
green area, longitudinal strain of endocardial layer > 15%;
the superimposed tissue Doppler velocity vector arrow is based on the velocity parameter generated by the second processing module The direction of the myocardial motion is represented by an arrow direction, the length of the arrow is in direct proportion to the velocity amplitude, and the velocity amplitude is defined as the absolute value of a myocardial motion velocity vector, namely the velocity size, and the unit is cm/s;
the highlight prompt pair of the abnormal region simultaneously meets the following conditions T>0.7 Myocardial segments were scintillation labeled.
The invention also discloses an OSAS heart mechanics analysis system of the two-dimensional speckle tracking joint tissue Doppler, which further comprises a hardware integrated component, wherein the hardware integrated component is composed of an inertial sensor and an edge computing device which are arranged in an ultrasonic probe:
the ultrasonic probe built-in inertial sensor detects the displacement amount (deltax, deltay) of the ultrasonic probe in real time, and corrects the image offset by the following formula:
Wherein the method comprises the steps of Is a corrected two-dimensional image; The displacement of the ultrasonic probe in the X, Y direction is shown as delta x and delta y, and the unit is a pixel;
The edge computing device is deployed on an ultrasonic host and is used for locally computing the layered strain parameters.
The invention also discloses an OSAS heart mechanics analysis method of the OSAS heart mechanics analysis system of the two-dimensional speckle tracking combined tissue Doppler, which comprises the following steps:
step S1, synchronously acquiring a two-dimensional ultrasonic gray-scale image sequence and a myocardial tissue Doppler velocity spectrum through a two-channel data acquisition module;
s2, respectively generating layered strain parameters and speed parameters through a parallel computing engine module;
S3, calculating phase deviation through a dynamic correlation module Risk index;
Step S4, outputting a visual report and risk levels through an OSAS risk early warning module, wherein the risk levels comprise low risk, medium risk and high risk:
the low risk: ≤0.3;
the stroke risk is 0.3< > ≤0.7;
The high risk:>0.7。
compared with the prior art, the invention has the advantages that:
(1) The multi-mode data is synchronized with high precision, and the two-dimensional ultrasonic gray-scale image sequence is aligned with the high-precision time axis of the Doppler velocity spectrum of the myocardial tissue through the two-channel synchronous acquisition module and the preprocessing sub-module, so that the phase deviation error caused by data splitting in the traditional method is remarkably reduced, and a reliable basis is provided for dynamic association analysis.
(2) The intelligent grading early warning of the OSAS related heart injury is realized by combining a combined model of a layered strain change rate and a speed attenuation rate, and the accuracy of early abnormal recognition is improved.
(3) The intelligent interaction and real-time diagnosis are realized, the visual interface is displayed by superposition of dynamic thermodynamic diagrams, speed vector arrows and abnormal segment marks, a doctor is helped to quickly locate a high-risk area, diagnosis efficiency is greatly improved, and the hardware integration assembly ensures the instant output of analysis results through displacement correction and localized real-time calculation, so that the clinical real-time requirement is met.
(4) The clinical applicability is enhanced, the standardized flow integrates multi-mode data acquisition, parameter calculation, risk modeling and visual output, rapid screening and analysis are realized, and the verification result is highly consistent with the traditional diagnosis method.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a system architecture of the present invention;
FIG. 2 is a schematic diagram of a DTW algorithm and risk assessment according to the present invention;
FIG. 3 is a schematic diagram of a visual interface of the present invention;
Fig. 4 is a diagram of a hardware integrated component of the present invention.
Detailed Description
To achieve the above objective, the present invention provides an OSAS heart mechanical analysis system of two-dimensional speckle tracking combined tissue doppler, and the system comprises, in combination with fig. 1 to 4:
m1, a double-channel data acquisition module, wherein the module is configured to synchronously acquire a left ventricle two-dimensional ultrasonic gray scale image sequence and a myocardial tissue Doppler velocity spectrum by an ultrasonic probe, and realize time axis alignment of the two types of data.
M2, a parallel computing engine module, wherein the module is connected with the two-channel data acquisition module, and the parallel computing engine module comprises a first processing module and a second processing module:
The first processing module receives the two-dimensional ultrasonic gray-scale image sequence and generates longitudinal strain and circumferential strain parameters of an endocardial layer, a myocardial middle layer and an epicardial layer through a speckle tracking algorithm;
The second processing module receives Doppler velocity spectrum of myocardial tissue and calculates myocardial motion velocity peak value through time domain integration Acceleration parameter;
The dual-channel data acquisition module also comprises a preprocessing sub-module, and the preprocessing sub-module is specifically as follows:
The preprocessing submodule carries out automatic segmentation of myocardial contours on the two-dimensional ultrasonic gray scale image sequence and divides the areas of an endocardial layer, a middle layer and an epicardial layer;
the preprocessing sub-module carries out respiratory motion artifact filtering on the Doppler velocity spectrum of the myocardial tissue and retains the related velocity signals of the systole;
By means of automatic segmentation of myocardial contours and respiratory artifact filtering, image segmentation errors and motion noise interference are eliminated, extraction accuracy of layered strain parameters and speed parameters is ensured, and data quality is improved.
M3, a dynamic association module, which is connected with the parallel computing engine module, and is used for carrying out time sequence fusion on the parameters of the longitudinal strain of the endocardium layer, the circumferential strain of the middle layer of the cardiac muscle and the circumferential strain of the epicardium layer of the first processing module and the speed peak value and the acceleration parameter of the second processing module to generate a strain and speed combined characteristic map;
The dynamic association module calculates the phase deviation of the layered strain curve S (t) and the speed curve V (t) through a dynamic time warping algorithm, and specifically comprises the following steps of;
Wherein the method comprises the steps of Pi is the optimal alignment path of two time sequences; a path that is the smallest cumulative distance among the aligned paths pi; For the point in time The unit of the layered strain value is that the unit comprises the longitudinal strain of an endocardial layer and the circumferential strain of a middle layer of cardiac muscle; Time point Myocardial motion velocity values in cm/s, including velocity peaks;
Prior to calculating the phase deviation, the layered strain valuesAnd velocity valueCarrying out standardization processing and converting into dimensionless parameters;
When (when) Triggering myocardial decompensation abnormality early warning when tau is equal to a dimensionless threshold;
The layered strain curve and the speed curve are aligned through a dynamic time warping algorithm, the phase deviation of the layered strain curve and the speed curve is quantized, the problem of misjudgment caused by time axis dyssynchrony in the traditional method is solved, and the accuracy of myocardial motion dyssynchrony detection is remarkably improved.
M4, an OSAS risk early warning module, which is connected with the dynamic association module, generates left ventricular contraction function compensation state grading and an OSAS risk index based on phase deviation and parameter change trend in the combined feature map, and outputs the left ventricular contraction function compensation state grading and the OSAS risk index to the visual interaction interface;
the OSAS risk early warning module calculates a risk index by the following formula:
Wherein the method comprises the steps of The risk index is OSAS risk index, the index range is 0-1, alpha and beta are weight coefficients, and alpha+beta=1; The variable quantity of the longitudinal strain of the endocardial layer in the time window delta t is expressed as a unit; the change amount of the speed peak value in the time window delta t is expressed in cm/s, delta t is the length of the analysis time window, and the unit is seconds;
and constructing a risk index based on the layered strain change rate and the speed attenuation rate, realizing three-level risk classification of OSAS-related heart injury, and enhancing the clinical practicability of dynamic risk assessment.
The visual interaction interface of the OSAS risk early warning module includes displaying a layered strain thermodynamic diagram, overlaying tissue doppler velocity vector arrows and an abnormal region highlighting prompt:
Displaying a layered strain thermodynamic diagram, and mapping strain distribution of each segment of the left ventricle in a color gradient based on the layered strain parameters generated by the first processing module;
red area, i.e. longitudinal strain of endocardial layer is less than-20%;
yellow region, wherein the longitudinal strain of endocardial layer is less than or equal to-20 percent and less than or equal to-15 percent;
green area, longitudinal strain of endocardial layer > 15%;
the superimposed tissue doppler velocity vector arrow is based on the velocity parameter generated by the second processing module The direction of the myocardial motion is represented by an arrow direction, the length of the arrow is in direct proportion to the velocity amplitude, and the velocity amplitude is defined as the absolute value of a myocardial motion velocity vector, namely the velocity size, and the unit is cm/s;
the highlight prompt pair of the abnormal region simultaneously meets the following conditions TA myocardial segment of >0.7 is scintillant-labeled;
Through dynamic superposition display of thermodynamic diagrams, speed vector arrows and abnormal marks, myocardial mechanics abnormal areas are visually presented, so that doctors are helped to quickly locate high-risk segments, and diagnosis efficiency is improved.
The OSAS heart mechanics analysis system of the two-dimensional speckle tracking combined tissue Doppler further comprises a hardware integration component, wherein the hardware integration component is composed of an inertial sensor and an edge computing device which are arranged in an ultrasonic probe:
the ultrasonic probe built-in inertial sensor detects the displacement amount (deltax, deltay) of the ultrasonic probe in real time, and corrects the image offset by the following formula:
Wherein the method comprises the steps of Is a corrected two-dimensional image; The displacement of the ultrasonic probe in the X, Y direction is shown as delta x and delta y, and the unit is a pixel;
The edge computing equipment is deployed on the ultrasonic host and is used for locally computing the layered strain parameters;
The ultrasonic probe is internally provided with an inertial sensor to correct image displacement, the edge computing equipment realizes localized real-time computation, eliminates probe movement artifacts, reduces analysis delay, and ensures the instantaneity and reliability of result output.
An OSAS heart mechanics analysis method of an OSAS heart mechanics analysis system combining two-dimensional speckle tracking and tissue doppler, the OSAS heart mechanics analysis method comprises the following steps:
step S1, synchronously acquiring a two-dimensional ultrasonic gray-scale image sequence and a myocardial tissue Doppler velocity spectrum through a two-channel data acquisition module;
s2, respectively generating layered strain parameters and speed parameters through a parallel computing engine module;
S3, calculating phase deviation through a dynamic correlation module Risk index;
Step S4, outputting a visual report and risk levels through an OSAS risk early warning module, wherein the risk levels comprise low risk, medium risk and high risk:
low risk: ≤0.3;
risk of the middle part is 0.3< > ≤0.7;
High risk:>0.7;
The data acquisition, parameter calculation and risk assessment are integrated through a standardized flow, the whole flow analysis is completed rapidly, the clinical rapid screening requirement is met, and the verification result is highly consistent with the traditional diagnosis method.
The operation of the system in 4 different scenarios is given below.
As shown in fig. 1,2 and 4, embodiment 1: real-time processing and hardware optimization
Firstly, an ultrasonic probe acquires a left ventricle two-dimensional ultrasonic gray scale image sequence at 50 frames/second, acquires a myocardial tissue Doppler velocity spectrum at a 1kHz sampling rate, adds a time stamp for two types of data through a hardware clock synchronization technology, ensures that a time axis alignment error is less than 5ms, detects displacement in real time by a built-in inertial sensor of the probe, and eliminates image offset by applying a displacement correction formula.
Then, the myocardial contours are automatically segmented, the myocardium is divided into an endocardial layer, a middle layer and an epicardial layer, and 0.5Hz high-pass filtering is performed on the Doppler velocity spectrum of the myocardial tissue to remove respiratory motion noise.
And secondly, calculating longitudinal strain and circumferential strain in the endocardial layer and middle layer areas, integrating the filtered frequency spectrum, and extracting a speed peak value and maximum acceleration.
And thirdly, aligning the layered strain curve with the speed curve by adopting a dynamic time warping algorithm, calculating phase deviation, and triggering the asynchronous early warning of myocardial motion when the phase deviation exceeds a preset threshold value.
And then, generating a risk index according to the strain change rate and the speed attenuation rate in the sliding time window, outputting a low, medium and high three-level risk classification result, and displaying a thermodynamic diagram, a speed vector arrow and an abnormal segment flickering mark in real time.
Finally, the edge computing device locally runs an algorithm, ensures that the processing delay is less than 100ms, and transmits the diagnosis result to the hospital information system through a standard protocol.
As shown in FIGS. 2 and 3, example 2 high sensitivity scientific research analysis type
Firstly, an ultrasonic probe acquires a two-dimensional ultrasonic gray-scale image sequence at a high frame rate of 100 frames/second, a myocardial tissue Doppler velocity spectrum is sampled at a high frequency of 2kHz, a time synchronization error is optimized to 3ms, and a probe displacement correction method is consistent with embodiment 1.
And then, improving the layering precision of the cardiac muscle by adopting an optimized automatic segmentation algorithm, and adjusting filtering parameters to keep more physiological signals.
Secondly, a deformation correction technology is introduced to improve the measurement precision of strain parameters, the Doppler analysis range is expanded, and the calculation of acceleration parameters is increased.
And thirdly, limiting curve alignment path offset by adopting a constraint type dynamic time warping algorithm, and adjusting a phase deviation threshold value to enhance detection sensitivity.
Then, a risk index formula is expanded, acceleration parameters are fused to improve early warning capability, and a visual interface is overlaid with an acceleration curve and multi-level thermodynamic diagram classification is carried out.
And finally, using high-performance computing equipment to accelerate algorithm operation, and supporting the derivation of the original data for scientific research analysis.
Example 3 Portable substrate medical Adaptation type as shown in FIGS. 1 and 4
Firstly, the wireless ultrasonic probe collects a two-dimensional ultrasonic gray-scale image sequence at 30 frames/second, the myocardial tissue Doppler velocity spectrum is sampled at 500Hz, the time synchronization error is expanded to 10ms, the basic diagnosis requirement is met, and the probe displacement correction function is simplified.
And then, adopting a lightweight segmentation algorithm to replace a deep learning model, and running a filtering algorithm on the terminal equipment to reduce the resource occupation.
Secondly, only longitudinal strain and speed peak parameters of the endocardial layer are calculated, and middle layer and acceleration parameter analysis is omitted so as to simplify the flow.
Again, the cloud server runs a dynamic time warping algorithm and returns results, maintaining a default phase deviation threshold to ensure consistency.
Then, the risk classification is simplified into two stages of high risk and non-high risk, and the thermodynamic diagram only displays red and green colors so as to reduce interaction complexity.
And finally, the tablet personal computer runs light analysis software and supports offline data caching and batch processing.
Embodiment 4 Multi-center collaboration and data normalization as illustrated in FIGS. 1-3
Firstly, uniformly acquiring data by using parameters of a standardized ultrasonic probe, wherein the time synchronization error is strictly controlled within 5 ms.
Then, the multi-center data consistency is improved by adopting a collaborative optimization segmentation algorithm, and the filtering processing flow is consistent with embodiment 1.
And secondly, calculating the strain parameters of the endocardium and the middle layer, and extracting the speed peak value and the acceleration parameter.
And thirdly, the cloud server cluster processes the dynamic time warping algorithm in parallel, and a default phase deviation threshold value is kept to unify evaluation standards.
And then, generating a grading result by adopting a standardized risk index model, and dynamically displaying mechanical abnormal distribution of each segment by a Web end three-dimensional model.
Finally, all data are packaged and stored according to medical image standards, analysis results are output in a cross-platform format, and multi-center statistics and analysis are supported.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (7)
1. An OSAS heart mechanics analysis system combining two-dimensional speckle tracking and tissue Doppler is characterized by comprising the following modules,
M1, a double-channel data acquisition module, wherein the module is configured to synchronously acquire a left ventricle two-dimensional ultrasonic gray scale image sequence and a myocardial tissue Doppler velocity spectrum by an ultrasonic probe, and realize time axis alignment of the two types of data;
m2, parallel computing engine module, this module is connected with binary channels data acquisition module, parallel computing engine module includes first processing module and second processing module:
The first processing module receives a two-dimensional ultrasonic gray-scale image sequence, and generates longitudinal strain and circumferential strain parameters of an endocardial layer, a myocardial middle layer and an epicardial layer through a speckle tracking algorithm;
the second processing module receives Doppler velocity spectrum of myocardial tissue, calculates myocardial motion velocity peak value through time domain integration Acceleration parameter;
M3, a dynamic association module which is connected with the parallel computing engine module, wherein the dynamic association module carries out time series fusion on the parameters of the longitudinal strain of the endocardium layer, the circumferential strain of the middle layer and the circumferential strain of the epicardium layer of the first processing module and the speed peak value and the acceleration parameter of the second processing module to generate a strain and speed combined characteristic map;
And M4, an OSAS risk early warning module, which is connected with the dynamic association module, generates left ventricular contraction function compensation state grading and an OSAS risk index based on the phase deviation and the parameter change trend in the combined characteristic map, and outputs the left ventricular contraction function compensation state grading and the OSAS risk index to the visual interaction interface.
2. The OSAS heart mechanical analysis system of two-dimensional speckle tracking joint tissue doppler according to claim 1, wherein the dynamic correlation module calculates the phase deviation of the layered strain curve S (t) from the velocity curve V (t) by a dynamic time warping algorithm, comprising in particular;
Wherein the method comprises the steps of Pi is the optimal alignment path of two time sequences; a path that is the smallest cumulative distance among the aligned paths pi; For the point in time The unit of the layered strain value is that the unit comprises the longitudinal strain of an endocardial layer and the circumferential strain of a middle layer of cardiac muscle; For the point in time Myocardial motion velocity values in cm/s, including velocity peaks;
Prior to calculating the phase deviation, the layered strain valuesAnd velocity valueCarrying out standardization processing and converting into dimensionless parameters;
When (when) Triggering myocardial decompensation abnormality early warning when tau is larger than tau, wherein tau is a dimensionless threshold value.
3. The OSAS cardiac mechanical analysis system of two-dimensional speckle-tracking joint tissue doppler of claim 1, wherein the OSAS risk early warning module calculates a risk index by the following formula:
Wherein the method comprises the steps of The risk index is OSAS risk index, the index range is 0-1, alpha and beta are weight coefficients, and alpha+beta=1; The variable quantity of the longitudinal strain of the endocardial layer in the time window delta t is expressed as a unit; the change amount of the speed peak value in the time window delta t is expressed in cm/s, delta t is the analysis time window length and the unit is seconds.
4. The OSAS cardiac mechanical analysis system of two-dimensional speckle-tracking joint tissue doppler of claim 1, wherein the dual-channel data acquisition module further comprises a preprocessing sub-module, in particular:
the preprocessing submodule carries out automatic segmentation of myocardial contours on the two-dimensional ultrasonic gray scale image sequence and divides areas of an endocardial layer, a middle layer and an epicardial layer;
and the preprocessing submodule carries out respiratory motion artifact filtering on the Doppler velocity spectrum of the myocardial tissue and retains a relevant velocity signal of the systole.
5. The OSAS cardiac mechanical analysis system of two-dimensional speckle tracking joint tissue doppler of claim 1, wherein the visual interaction interface of the OSAS risk early warning module comprises displaying a layered strain thermodynamic diagram, superimposed tissue doppler velocity vector arrows, and abnormal region highlighting cues:
the display layered strain thermodynamic diagram maps strain distribution of each segment of the left ventricle in a color gradient based on the layered strain parameters generated by the first processing module;
red area, i.e. longitudinal strain of endocardial layer is less than-20%;
yellow region, wherein the longitudinal strain of endocardial layer is less than or equal to-20 percent and less than or equal to-15 percent;
green area, longitudinal strain of endocardial layer > 15%;
the superimposed tissue Doppler velocity vector arrow is based on the velocity parameter generated by the second processing module The direction of the myocardial motion is represented by an arrow direction, the length of the arrow is in direct proportion to the velocity amplitude, and the velocity amplitude is defined as the absolute value of a myocardial motion velocity vector, namely the velocity size, and the unit is cm/s;
the highlight prompt pair of the abnormal region simultaneously meets the following conditions T>0.7 Myocardial segments were scintillation labeled.
6. The OSAS cardiac mechanical analysis system of two-dimensional speckle-tracking joint tissue doppler of claim 5, further comprising a hardware integration component composed of an ultrasound probe built-in inertial sensor and an edge computing device:
the ultrasonic probe built-in inertial sensor detects the displacement amount (deltax, deltay) of the ultrasonic probe in real time, and corrects the image offset by the following formula:
Wherein the method comprises the steps of Is a corrected two-dimensional image; The displacement of the ultrasonic probe in the X, Y direction is shown as delta x and delta y, and the unit is a pixel;
The edge computing device is deployed on an ultrasonic host and is used for locally computing the layered strain parameters.
7. The OSAS heart mechanical analysis method of two-dimensional speckle tracking joint tissue doppler OSAS heart mechanical analysis system of claim 1, wherein said OSAS heart mechanical analysis method comprises the steps of:
step S1, synchronously acquiring a two-dimensional ultrasonic gray-scale image sequence and a myocardial tissue Doppler velocity spectrum through a two-channel data acquisition module;
s2, respectively generating layered strain parameters and speed parameters through a parallel computing engine module;
S3, calculating phase deviation through a dynamic correlation module Risk index;
Step S4, outputting a visual report and risk levels through an OSAS risk early warning module, wherein the risk levels comprise low risk, medium risk and high risk:
the low risk: ≤0.3;
the stroke risk is 0.3< > ≤0.7;
The high risk:>0.7。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510429222.3A CN119924891B (en) | 2025-04-08 | 2025-04-08 | OSAS heart mechanical analysis system combining two-dimensional speckle tracking and tissue Doppler |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510429222.3A CN119924891B (en) | 2025-04-08 | 2025-04-08 | OSAS heart mechanical analysis system combining two-dimensional speckle tracking and tissue Doppler |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119924891A true CN119924891A (en) | 2025-05-06 |
| CN119924891B CN119924891B (en) | 2025-07-04 |
Family
ID=95537187
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202510429222.3A Active CN119924891B (en) | 2025-04-08 | 2025-04-08 | OSAS heart mechanical analysis system combining two-dimensional speckle tracking and tissue Doppler |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119924891B (en) |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020072674A1 (en) * | 2000-12-07 | 2002-06-13 | Criton Aline Laure | Strain rate analysis in ultrasonic diagnostic images |
| CN101926657A (en) * | 2009-06-18 | 2010-12-29 | 深圳迈瑞生物医疗电子股份有限公司 | Method for tracking features of ultrasound pattern and system thereof |
| 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 |
| US9629615B1 (en) * | 2013-09-06 | 2017-04-25 | University Of Louisville Research Foundation, Inc. | Combined B-mode / tissue doppler approach for improved cardiac motion estimation in echocardiographic images |
| CN107072531A (en) * | 2014-05-06 | 2017-08-18 | 塞克利心血管成像股份有限公司 | Method and system for the dynamic (dynamical) analysis of myocardial wall |
| CN108042154A (en) * | 2017-12-08 | 2018-05-18 | 浙江中医药大学 | Two dimensional echocardiogram formation center flesh shape, movement and deformation analysis method |
| CN111080618A (en) * | 2019-12-13 | 2020-04-28 | 浙江大学 | GTV-based heart left ventricle motion analysis method |
| CN111508004A (en) * | 2020-04-29 | 2020-08-07 | 中国人民解放军总医院 | Wall motion abnormity ultrasonic processing method, system and equipment based on deep learning |
| CN113384293A (en) * | 2021-06-12 | 2021-09-14 | 北京医院 | Integrated machine learning method for coronary heart disease screening based on two-dimensional spot tracking technology |
| CN114515169A (en) * | 2021-12-31 | 2022-05-20 | 西安交通大学 | Ultrasonic myocardial tissue multi-parameter imaging method and system |
| US20230128956A1 (en) * | 2021-12-31 | 2023-04-27 | Xi'an Jiaotong University | Myocardial multiparametric ultrasound imaging method and system |
| CN116889425A (en) * | 2022-04-01 | 2023-10-17 | 通用电气精准医疗有限责任公司 | Method and system for excluding pericardium from cardiac strain calculations |
| CN116977373A (en) * | 2023-08-03 | 2023-10-31 | 浙江大学 | Myocardial motion tracking method and system based on bidirectional joint learning |
| CN117745672A (en) * | 2023-12-19 | 2024-03-22 | 汕头大学 | A left ventricular motion estimation method, electronic device and storage medium |
| CN118750027A (en) * | 2024-08-20 | 2024-10-11 | 深圳市宝安区妇幼保健院 | An AI-based quantitative analysis method for echocardiography |
-
2025
- 2025-04-08 CN CN202510429222.3A patent/CN119924891B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020072674A1 (en) * | 2000-12-07 | 2002-06-13 | Criton Aline Laure | Strain rate analysis in ultrasonic diagnostic images |
| 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 |
| CN101926657A (en) * | 2009-06-18 | 2010-12-29 | 深圳迈瑞生物医疗电子股份有限公司 | Method for tracking features of ultrasound pattern and system thereof |
| US9629615B1 (en) * | 2013-09-06 | 2017-04-25 | University Of Louisville Research Foundation, Inc. | Combined B-mode / tissue doppler approach for improved cardiac motion estimation in echocardiographic images |
| CN107072531A (en) * | 2014-05-06 | 2017-08-18 | 塞克利心血管成像股份有限公司 | Method and system for the dynamic (dynamical) analysis of myocardial wall |
| CN108042154A (en) * | 2017-12-08 | 2018-05-18 | 浙江中医药大学 | Two dimensional echocardiogram formation center flesh shape, movement and deformation analysis method |
| CN111080618A (en) * | 2019-12-13 | 2020-04-28 | 浙江大学 | GTV-based heart left ventricle motion analysis method |
| CN111508004A (en) * | 2020-04-29 | 2020-08-07 | 中国人民解放军总医院 | Wall motion abnormity ultrasonic processing method, system and equipment based on deep learning |
| CN113384293A (en) * | 2021-06-12 | 2021-09-14 | 北京医院 | Integrated machine learning method for coronary heart disease screening based on two-dimensional spot tracking technology |
| CN114515169A (en) * | 2021-12-31 | 2022-05-20 | 西安交通大学 | Ultrasonic myocardial tissue multi-parameter imaging method and system |
| US20230128956A1 (en) * | 2021-12-31 | 2023-04-27 | Xi'an Jiaotong University | Myocardial multiparametric ultrasound imaging method and system |
| CN116889425A (en) * | 2022-04-01 | 2023-10-17 | 通用电气精准医疗有限责任公司 | Method and system for excluding pericardium from cardiac strain calculations |
| CN116977373A (en) * | 2023-08-03 | 2023-10-31 | 浙江大学 | Myocardial motion tracking method and system based on bidirectional joint learning |
| CN117745672A (en) * | 2023-12-19 | 2024-03-22 | 汕头大学 | A left ventricular motion estimation method, electronic device and storage medium |
| CN118750027A (en) * | 2024-08-20 | 2024-10-11 | 深圳市宝安区妇幼保健院 | An AI-based quantitative analysis method for echocardiography |
Non-Patent Citations (2)
| Title |
|---|
| 贺丽;李沿江;: "二维斑点追踪技术评价右心室功能的研究进展", 中国医学影像学杂志, no. 11, 27 November 2013 (2013-11-27) * |
| 陈雪;刘燕娜;郭良云;盛旅德;丁勇民;: "二维斑点追踪成像评估阻塞性睡眠呼吸暂停综合征左心室整体及分层心肌应变", 中国医学影像学杂志, no. 12, 25 December 2017 (2017-12-25) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119924891B (en) | 2025-07-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107529646B (en) | Non-contact heart rate measurement method and device based on Euler image amplification | |
| Tasli et al. | Remote PPG based vital sign measurement using adaptive facial regions | |
| CN111281367A (en) | Anti-interference non-contact heart rate detection method based on face video | |
| CN117357093B (en) | Non-contact respiratory rate monitoring method and system based on dual-spectrum facial video | |
| KR101652641B1 (en) | Method for Image registration Using ECG signal and Apparatus Thereof | |
| CN114332278B (en) | A deep learning-based OCTA image motion correction method | |
| JP6452236B2 (en) | Eyeball identification device and eyeball identification method | |
| CN118982492B (en) | A jitter distortion correction image processing system for ophthalmic OCT | |
| CN117192458B (en) | Magnetic resonance image generation method for eliminating motion artifact based on dynamic tracking technology | |
| CN111127511A (en) | Non-contact heart rate monitoring method | |
| US20250252562A1 (en) | Method for estimating heart rate on basis of corrected image, and device therefor | |
| CN120392025A (en) | Multimodal optical non-contact health monitoring system and method | |
| US20240298906A1 (en) | Pulse wave detector | |
| CN112155511A (en) | Method for compensating human eye shake in OCT (optical coherence tomography) acquisition process based on deep learning | |
| US20250157042A1 (en) | Measurement Method, Recording Medium, and Measurement System | |
| US20250025052A1 (en) | Vision-Based Cardiorespiratory Monitoring | |
| CN117745808B (en) | Electrocardiogram image positioning comparison method based on photogrammetry | |
| CN119924891A (en) | A two-dimensional speckle tracking combined with tissue Doppler analysis system for OSAS cardiac mechanics | |
| CN115482209B (en) | Pulse data processing method, device, equipment and computer readable storage medium | |
| US20220160260A1 (en) | System and method for measuring biomedical signal | |
| Al-Fahdawi et al. | An automatic corneal subbasal nerve registration system using FFT and phase correlation techniques for an accurate DPN diagnosis | |
| CN109009020A (en) | A kind of health monitoring systems and health monitor method | |
| TW202224622A (en) | Method and system for dynamic physiological characteristic region capturing | |
| Qiu et al. | Low-cost gaze and pulse analysis using realsense | |
| US20250078563A1 (en) | Pulse wave detection device and pulse wave detection method |
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 |