A two-dimensional speckle tracking combined with tissue Doppler OSAS cardiac mechanics analysis system

Abstract

The present invention belongs to the technical field of medical image analysis, and particularly relates to a two-dimensional speckle tracking combined with tissue Doppler OSAS cardiac mechanics analysis system. The system realizes high-precision time-axis alignment of two-dimensional ultrasonic gray-scale image sequences and myocardial tissue Doppler velocity spectra through a dual-channel synchronous acquisition module, and a parallel computing engine generates endocardial longitudinal strain, middle-layer circumferential strain, and peak velocity parameters. The dynamic correlation module uses the dynamic time warping algorithm to quantify the phase deviation between the stratified strain curve and the velocity curve, and combines the risk index model of the risk warning module to output low, medium, and high risk classifications. The visualization interface dynamically displays myocardial motion asynchrony regions through heat maps, velocity vector arrows, and abnormal flashing marks. The hardware integration components and standardized analysis processes significantly improve the detection sensitivity and diagnostic efficiency. The present invention solves the problems of multi-modal data fragmentation and early warning lag.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image analysis, and particularly to a two-dimensional speckle tracking combined with tissue Doppler OSAS cardiac mechanics analysis system. Background Art

[0002] Obstructive sleep apnea syndrome (OSAS) is a common disease characterized by nocturnal apnea and hypoxia. Long-term illness is likely to cause compensatory damage to cardiac function. In clinical diagnosis, doctors usually rely on ultrasound imaging technology to evaluate the mechanical characteristics of the left ventricular myocardium of patients. However, most of the existing ultrasound analysis methods only focus on single-modal parameters and are difficult to comprehensively capture the dynamic correlation between strain and velocity during myocardial contraction.

[0003] In the existing evaluation of the mechanical characteristics of the left ventricular myocardium of patients, in traditional methods, speckle tracking and myocardial tissue Doppler velocity spectrum are analyzed separately, and the time axes are not accurately aligned, resulting in doctors having to manually compare reports and there being a large calculation error in phase deviation. There is a lack of joint modeling of stratified strain and velocity parameters, making it difficult to detect abnormal myocardial compensation in a timely manner. At the same time, the results are presented as static numerical values, and it is impossible to visually locate high-risk areas.

[0004] In view of the above problems, an innovative design is carried out on the basis of the original OSAS cardiac mechanics analysis system. Summary of the Invention

[0005] The purpose of the present invention is to provide a two-dimensional speckle tracking combined with tissue Doppler OSAS cardiac mechanics analysis system to address the problems of fragmentation of multi-modal data, insufficient time synchronization, lag in early warning of abnormal myocardial compensation, and lack of visual interaction in the prior art.

[0006] To this end, the technical solution adopted by the present invention is as follows:

[0007] A two-dimensional speckle tracking combined with tissue Doppler OSAS cardiac mechanics analysis system includes the following modules,

[0008] M1, a dual-channel data acquisition module, which is configured to synchronously collect a sequence of two-dimensional ultrasound grayscale images of the left ventricle and myocardial tissue Doppler velocity spectra by an ultrasound probe, and align the time axes of the two types of data;

[0009] M2, a parallel computing engine module, which is connected to the dual-channel data acquisition module. The parallel computing engine module includes a first processing module and a second processing module:

[0010] The first processing module receives the sequence of two-dimensional ultrasound grayscale images and generates longitudinal strain and circumferential strain parameters of the endocardial layer, myocardial middle layer, and epicardial layer through a speckle tracking algorithm;

[0011] The second processing module receives the myocardial tissue Doppler velocity spectrum and calculates the peak myocardial motion velocity through time-domain integration. and the acceleration parameter ;

[0012] M3, the dynamic correlation module, which is connected to the parallel computing engine module. The dynamic correlation module performs time-series fusion on the endocardial longitudinal strain, mid-myocardial circumferential strain, and epicardial circumferential strain parameters of the first processing module and the peak velocity and acceleration parameters of the second processing module to generate a combined strain and velocity feature map.

[0013] M4, the OSAS risk warning module, which is connected to the dynamic correlation module. Based on the phase deviation and parameter change trend in the combined feature map, it generates a classification of the compensatory state of left ventricular systolic function and an OSAS risk index, and outputs them to the visualization and interaction interface.

[0014] Furthermore, the dynamic correlation module calculates the phase deviation between the hierarchical strain curve S(t) and the velocity curve V(t) through the dynamic time warping algorithm, specifically including:

[0015]

[0016] where is the dimensionless phase deviation metric value; π is the optimal alignment path of the two time series; is the path with the minimum cumulative distance in the alignment path π; is the time point of the hierarchical strain value, whose unit is %, including endocardial longitudinal strain and mid-myocardial circumferential strain; the time point of the myocardial motion velocity value, whose unit is cm / s, including the peak velocity ;

[0017] Before calculating the phase deviation, standardize the hierarchical strain value and the velocity value to convert them into dimensionless parameters;

[0018] When > τ, trigger the warning of abnormal myocardial compensation, where τ is the dimensionless threshold.

[0019] Furthermore, the OSAS risk warning module calculates the risk index through the following formula:

[0020]

[0021] where is the OSAS risk index, and the index range is 0-1; α and β are weight coefficients, and α + β = 1; is the change in endocardial longitudinal strain within the time window Δt, with the unit of %; is the change in peak velocity within the time window Δt, with the unit of cm / s; Δt is the length of the analysis time window, with the unit of seconds.

[0022] Furthermore, the dual-channel data acquisition module further includes a preprocessing sub-module, specifically as follows:

[0023] The preprocessing sub-module automatically segments the myocardial contour of the two-dimensional ultrasound gray-scale image sequence, and divides the endocardial layer, middle layer and epicardial layer regions;

[0024] The preprocessing sub-module filters the respiratory motion artifacts of the myocardial tissue Doppler velocity spectrum and retains the velocity signals related to the cardiac systolic phase.

[0025] Furthermore, the visual interaction interface of the OSAS risk warning module includes displaying a hierarchical strain heat map, superimposing tissue Doppler velocity vector arrows and highlighting abnormal regions:

[0026] The displayed hierarchical strain heat map is based on the hierarchical strain parameters generated by the first processing module, and maps the strain distribution of each segment of the left ventricle with a color gradient;

[0027] Red area: endocardial longitudinal strain < -20%;

[0028] Yellow area: -20% ≤ endocardial longitudinal strain ≤ -15%;

[0029] Green area: endocardial longitudinal strain > -15%;

[0030] The superimposed tissue Doppler velocity vector arrows are based on the velocity parameters generated by the second processing module , the arrow direction represents the myocardial motion direction, and the arrow length is proportional to the velocity amplitude. The velocity amplitude is defined as the absolute value of the myocardial motion velocity vector, that is, the velocity magnitude, with the unit of cm / s;

[0031] The highlighting of abnormal regions simultaneously satisfies > τ and > 0.7 for myocardial segments to be marked with a flash.

[0032] The present invention also discloses an OSAS cardiac mechanics analysis system combining two-dimensional speckle tracking and tissue Doppler, which further includes a hardware integration component. The hardware integration component consists of an inertial sensor and an edge computing device built into the ultrasound probe:

[0033] The inertial sensor built into the ultrasound probe real-time detects the displacement amount (Δx, Δy) of the ultrasound probe, and corrects the image offset through the following formula:

[0034]

[0035] Among them is the corrected two-dimensional image; is the original two-dimensional image; Δx and Δy are the displacement amounts of the ultrasonic probe in the X and Y directions, and their unit is pixel;

[0036] The edge computing device is deployed on the ultrasonic host and is used for locally calculating the layered strain parameters.

[0037] The present invention also discloses an OSAS cardiac mechanics analysis method for a two-dimensional speckle tracking combined with tissue Doppler OSAS cardiac mechanics analysis system. The OSAS cardiac mechanics analysis method includes the following steps:

[0038] Step S1: Synchronously acquire a two-dimensional ultrasonic grayscale image sequence and a myocardial tissue Doppler velocity spectrum through a dual-channel data acquisition module;

[0039] Step S2: Respectively generate layered strain parameters and velocity parameters through a parallel computing engine module;

[0040] Step S3: Calculate the phase deviation through a dynamic correlation module and the risk index ;

[0041] Step S4: Output a visualization report and a risk level through an OSAS risk warning module. The risk level includes low risk, medium risk, and high risk:

[0042] The low risk: ≤0.3;

[0043] The medium risk: 0.3 < ≤0.7;

[0044] The high risk: > 0.7.

[0045] Compared with the prior art, the advantages of the present invention are as follows:

[0046] (1) High-precision synchronization of multi-modal data. Through the dual-channel synchronous acquisition module and the preprocessing sub-module, high-precision time-axis alignment of the two-dimensional ultrasonic grayscale image sequence and the myocardial tissue Doppler velocity spectrum is realized, significantly reducing the phase deviation error caused by data fragmentation in the traditional method and providing a reliable basis for dynamic correlation analysis.

[0047] (2) Exquisitely sensitive detection of myocardial compensation: The dynamic time warping algorithm is used to quantify the phase deviation between the layered strain and velocity curves, and combined with a preset threshold to trigger an early warning, significantly improving the sensitivity of detecting myocardial motion asynchrony; through a joint model that fuses the change rate of layered strain and the decay rate of velocity, an intelligent grading early warning of OSAS-related cardiac damage is achieved, enhancing the accuracy of early anomaly recognition.

[0048] (3) Intelligent interaction and real-time diagnosis: The visualization interface displays the superposition of dynamic heat maps, velocity vector arrows, and abnormal segment markers, helping doctors quickly locate high-risk areas and greatly improving the diagnostic efficiency; the hardware integration component ensures the immediate output of the analysis results through displacement correction and local real-time calculation, meeting the clinical real-time requirements.

[0049] (4) Enhanced clinical applicability: The standardized process integrates multi-modal data acquisition, parameter calculation, risk modeling, and visualization output, enabling rapid screening and analysis, and the verification results are highly consistent with traditional diagnostic methods. Description of the Drawings

[0050] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for description in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0051] Figure 1 It is the system architecture flowchart of the present invention;

[0052] Figure 2 It is the schematic diagram of the DTW algorithm and risk assessment of the present invention;

[0053] Figure 3 It is the schematic diagram of the visualization interface of the present invention;

[0054] Figure 4 It is the hardware integration component diagram of the present invention. Detailed Embodiments

[0055] To achieve the above objectives, the present invention is realized through the following technical solutions. The present invention provides an OSAS cardiac mechanics analysis system combining two-dimensional speckle tracking and tissue Doppler, combined with the attached Figures 1 to 4 , and the system includes:

[0056] M1. A dual-channel data acquisition module, which is configured to synchronously collect the left ventricular two-dimensional ultrasound gray-scale image sequence and the myocardial tissue Doppler velocity spectrum by an ultrasound probe, and align the two types of data on the time axis.

[0057] M2. The parallel computing engine module is connected to the dual-channel data acquisition module. The parallel computing engine module includes a first processing module and a second processing module:

[0058] The first processing module receives a two-dimensional ultrasound grayscale image sequence and generates longitudinal strain and circumferential strain parameters of the endocardial layer, myocardial middle layer, and epicardial layer through the speckle tracking algorithm;

[0059] The second processing module receives the myocardial tissue Doppler velocity spectrum and calculates the peak myocardial motion velocity and acceleration parameters ;

[0060] The dual-channel data acquisition module also includes a preprocessing sub-module, specifically as follows:

[0061] The preprocessing sub-module automatically segments the myocardial contour of the two-dimensional ultrasound grayscale image sequence and divides the regions of the endocardial layer, middle layer, and epicardial layer;

[0062] The preprocessing sub-module filters the respiratory motion artifacts of the myocardial tissue Doppler velocity spectrum and retains the velocity signals related to the cardiac systolic phase;

[0063] Through automatic myocardial contour segmentation and respiratory artifact filtering, image segmentation errors and motion noise interference are eliminated, ensuring the extraction accuracy of layered strain parameters and velocity parameters and improving the data quality.

[0064] M3. The dynamic association module is connected to the parallel computing engine module. The dynamic association module performs time series fusion on the longitudinal strain of the endocardial layer, circumferential strain of the myocardial middle layer, and circumferential strain parameters of the epicardial layer of the first processing module and the peak velocity and acceleration parameters of the second processing module to generate a strain and velocity joint feature map;

[0065] The dynamic association module calculates the phase deviation between the layered strain curve S(t) and the velocity curve V(t) through the dynamic time warping algorithm, specifically including;

[0066]

[0067] where is the dimensionless phase deviation metric value; π is the optimal alignment path of the two time series; is the path with the minimum cumulative distance in the alignment path π; is the time point The layered strain value at, whose unit is %, includes the longitudinal strain of the endocardial layer and the circumferential strain of the myocardial middle layer; The time point The myocardial motion velocity value at, whose unit is cm / s, includes the peak velocity ;

[0068] Before calculating the phase deviation, the stratified strain values and velocity values are standardized and converted into dimensionless parameters;

[0069] When >τ, an early warning of abnormal myocardial compensation is triggered, where τ is a dimensionless threshold;

[0070] By aligning the stratified strain curve and the velocity curve through the dynamic time warping algorithm, the phase deviation between the two is quantified, solving the misjudgment problem caused by the asynchronous time axis in the traditional method, and significantly improving the accuracy of detecting myocardial motion asynchrony.

[0071] The M4 and OSAS risk warning module, which is connected to the dynamic association module, generates a classification of the compensatory state of left ventricular systolic function and an OSAS risk index based on the phase deviation and parameter change trend in the joint feature map, and outputs them to the visualization and interaction interface;

[0072] The OSAS risk warning module calculates the risk index through the following formula:

[0073]

[0074] where is the OSAS risk index, and the index range is 0-1; α and β are weight coefficients, and α + β = 1; is the change amount of the longitudinal strain of the endocardial layer within the time window Δt, and its unit is %; is the change amount of the peak velocity within the time window Δt, and its unit is cm / s; Δt is the length of the analysis time window, and its unit is seconds;

[0075] Based on the change rate of stratified strain and the decay rate of velocity, a risk index is constructed to achieve a three-level risk classification of OSAS-related cardiac damage, enhancing the clinical practicability of dynamic risk assessment.

[0076] The visualization and interaction interface of the OSAS risk warning module includes displaying a stratified strain heat map, superimposing tissue Doppler velocity vector arrows, and highlighting abnormal areas:

[0077] The displayed stratified strain heat map is based on the stratified strain parameters generated by the first processing module, and maps the strain distribution of each segment of the left ventricle with a color gradient;

[0078] Red area: longitudinal strain of the endocardial layer < -20%;

[0079] Yellow area: -20% ≤ longitudinal strain of the endocardial layer ≤ -15%;

[0080] Green area: longitudinal strain of the endocardial layer > -15%;

[0081] The superimposed tissue Doppler velocity vector arrows are based on the velocity parameters generated by the second processing module , where the arrow direction represents the myocardial motion direction, and the arrow length is proportional to the velocity amplitude. The velocity amplitude is defined as the absolute value of the myocardial motion velocity vector, that is, the velocity magnitude, and its unit is cm / s;

[0082] Highlight the abnormal area to flash-mark the myocardial segments that simultaneously satisfy > τ and > 0.7;

[0083] Through the dynamic superposition display of the heat map, velocity vector arrows, and abnormal markings, visually present the abnormal myocardial mechanics area, help doctors quickly locate high-risk segments, and improve the diagnosis efficiency.

[0084] The OSAS cardiac mechanics analysis system combining two-dimensional speckle tracking and tissue Doppler also includes a hardware integration component, which consists of an inertial sensor built into the ultrasound probe and an edge computing device:

[0085] The inertial sensor built into the ultrasound probe real-time detects the displacement of the ultrasound probe (Δx, Δy), and corrects the image offset through the following formula:

[0086]

[0087] where is the corrected two-dimensional image; is the original two-dimensional image; Δx, Δy are the displacements of the ultrasound probe in the X and Y directions, and their unit is pixel;

[0088] The edge computing device is deployed on the ultrasound host and is used for localizing the calculation of the layered strain parameters;

[0089] The inertial sensor built into the ultrasound probe corrects the image displacement, and the edge computing device realizes local real-time calculation, eliminates the probe movement artifacts and reduces the analysis delay, ensuring the immediacy and reliability of the result output.

[0090] The OSAS cardiac mechanics analysis method of the OSAS cardiac mechanics analysis system combining two-dimensional speckle tracking and tissue Doppler includes the following steps:

[0091] Step S1, synchronously acquire the two-dimensional ultrasound gray-scale image sequence and the myocardial tissue Doppler velocity spectrum through the dual-channel data acquisition module;

[0092] Step S2, respectively generate the layered strain parameters and velocity parameters through the parallel computing engine module;

[0093] Step S3, calculate the phase deviation through the dynamic correlation module and risk index ;

[0094] Step S4: Output a visualization report and a risk level through the OSAS risk warning module. The risk levels include low risk, medium risk, and high risk:

[0095] Low risk: ≤0.3;

[0096] Medium risk: 0.3 < ≤0.7;

[0097] High risk: >0.7;

[0098] Integrate data collection, parameter calculation, and risk assessment through a standardized process, quickly complete the full-process analysis, meet the needs of rapid clinical screening, and the verification results are highly consistent with traditional diagnostic methods.

[0099] The working processes of the system in 4 different scenarios are given below.

[0100] Such as Figure 1 , Figure 2 and Figure 4 shown, Example 1: Real-time processing and hardware optimization type

[0101] First, the ultrasonic probe collects a sequence of left ventricular two-dimensional ultrasonic gray-scale images at 50 frames per second, and simultaneously obtains the myocardial tissue Doppler velocity spectrum at a sampling rate of 1 kHz. Add time stamps to the two types of data through the hardware clock synchronization technology to ensure that the time axis alignment error < 5 ms. The probe is built with an inertial sensor to detect displacement in real time, and the displacement correction formula is applied to eliminate image offset.

[0102] Next, automatically segment the myocardial contour, divide the myocardium into the endocardial layer, middle layer, and epicardial layer, and perform a 0.5 Hz high-pass filter on the myocardial tissue Doppler velocity spectrum to remove respiratory motion noise.

[0103] Secondly, calculate the longitudinal strain and circumferential strain in the endocardial layer and middle layer regions, integrate the filtered spectrum, and extract the velocity peak value and maximum acceleration.

[0104] Again, use the dynamic time warping algorithm to align the layered strain curve and velocity curve, calculate the phase deviation, and trigger a myocardial motion asynchrony warning when the phase deviation exceeds the preset threshold.

[0105] Then, generate a risk index based on the strain change rate and velocity decay rate within the sliding time window, output the low, medium, and high risk classification results, and display the heat map, velocity vector arrow, and abnormal segment flashing mark in real time.

[0106] Finally, the edge computing device runs the localization algorithm to ensure that the processing delay is less than 100 ms, and the diagnostic results are transmitted to the hospital information system through the standard protocol.

[0107] As Figure 2 and Figure 3 shown, Example 2: High-sensitivity scientific research analysis type

[0108] First, the ultrasonic probe collects a two-dimensional ultrasonic gray-scale image sequence at a high frame rate of 100 frames per second, samples the myocardial tissue Doppler velocity spectrum at 2 kHz, optimizes the time synchronization error to 3 ms, and the probe displacement correction method is the same as that in Example 1.

[0109] Next, the optimized automatic segmentation algorithm is used to improve the myocardial stratification accuracy, and the filtering parameters are adjusted to retain more physiological signals.

[0110] Secondly, the deformation correction technology is introduced to improve the measurement accuracy of strain parameters, expand the Doppler analysis range, and increase the calculation of acceleration parameters.

[0111] Again, the constrained dynamic time warping algorithm is used to limit the curve alignment path deviation, and the phase deviation threshold is adjusted to enhance the detection sensitivity.

[0112] Then, the risk index formula is expanded, and the acceleration parameter is fused to improve the early warning ability. The acceleration curve and multi-level heat map grading are superimposed on the visualization interface.

[0113] Finally, a high-performance computing device is used to accelerate the algorithm operation and support the export of raw data for scientific research analysis.

[0114] As Figure 1 and Figure 4 shown, Example 3: Portable primary medical adaptation type

[0115] First, the wireless ultrasonic probe collects a two-dimensional ultrasonic gray-scale image sequence at 30 frames per second, samples the myocardial tissue Doppler velocity spectrum at 500 Hz, expands the time synchronization error to 10 ms, meets the primary diagnosis requirements, and simplifies the probe displacement correction function.

[0116] Next, a lightweight segmentation algorithm is used to replace the deep learning model, and the filtering algorithm runs on the terminal device to reduce resource occupancy.

[0117] Secondly, only the longitudinal strain and velocity peak parameters of the endocardial layer are calculated, and the analysis of the middle layer and acceleration parameters is omitted to simplify the process.

[0118] Again, the dynamic time warping algorithm runs on the cloud server and returns the results, and the default phase deviation threshold is maintained to ensure consistency.

[0119] Then, the risk grading is simplified into two levels: high risk and non-high risk, and the heat map only shows two colors, red and green, to reduce the interaction complexity.

[0120] Finally, the tablet computer runs lightweight analysis software, supporting offline data caching and batch processing.

[0121] As Figures 1 - 3 shown, Example 4: Multi-center collaboration and data standardization type

[0122] First, the data is collected uniformly using standardized ultrasonic probe parameters, and the time synchronization error is strictly controlled within 5 ms.

[0123] Next, a collaborative optimization segmentation algorithm is adopted to improve the multi-center data consistency, and the filtering process is the same as that in Example 1.

[0124] Secondly, the endocardial layer and middle layer strain parameters are calculated, and the peak velocity and acceleration parameters are extracted.

[0125] Thirdly, the cloud server cluster processes the dynamic time warping algorithm in parallel, and keeps the default phase deviation threshold to unify the evaluation standard.

[0126] Then, a standardized risk index model is adopted to generate the grading results, and the 3D model on the Web side dynamically displays the abnormal mechanical distribution of each segment.

[0127] Finally, all data is encapsulated and stored according to the medical imaging standard, and the analysis results are output in a cross-platform format, supporting multi-center statistics and analysis.

[0128] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A two-dimensional speckle tracking combined with tissue Doppler OSAS cardiac mechanics analysis system, characterized in that, It includes the following modules: M1, a dual-channel data acquisition module, which is configured to synchronously acquire the left ventricular two-dimensional ultrasonic gray-scale image sequence and the myocardial tissue Doppler velocity spectrum by an ultrasonic probe, and align the two types of data on the time axis; M2, a parallel computing engine module, which is connected to the dual-channel data acquisition module. The parallel computing engine module includes 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 the endocardial layer, the middle myocardial layer and the epicardial layer through a speckle tracking algorithm; The second processing module receives the myocardial tissue Doppler velocity spectrum and calculates the peak myocardial motion velocity through time-domain integration and the acceleration parameter ; M3, a dynamic association module, which is connected to the parallel computing engine module. The dynamic association module performs time series fusion on the longitudinal strain of the endocardial layer, the circumferential strain of the middle myocardial layer and the circumferential strain parameters of the epicardial layer of the first processing module and the peak velocity and acceleration parameters of the second processing module to generate a strain and velocity joint feature map; The dynamic association module calculates the phase deviation between the hierarchical strain curve S(t) and the velocity curve V(t) through a dynamic time warping algorithm, specifically including: wherein is a dimensionless phase deviation metric value; π is the optimal alignment path of two time series; is the path with the minimum cumulative distance in the alignment path π; is the time point of the stratified strain value, whose unit is %, including the longitudinal strain of the endocardial layer and the circumferential strain of the middle myocardial layer; is the time point of the myocardial motion velocity value, whose unit is cm / s, including the peak velocity ; Before calculating the phase deviation, the stratified strain values and the velocity values are standardized and converted into dimensionless parameters; When > τ, trigger the warning of abnormal myocardial compensation, where τ is a dimensionless threshold; M4, an OSAS risk warning module, which is connected to the dynamic association module. Based on the phase deviation and parameter change trend in the joint feature map, it generates a classification of the left ventricular systolic function compensation state and an OSAS risk index, and outputs them to a visualization interaction interface; The OSAS risk warning module calculates the risk index through the following formula: where is the OSAS risk index, with the index range being 0 - 1; α and β are weight coefficients, and α + β = 1; is the change amount of the longitudinal strain of the endocardium within the time window Δt, and its unit is %; is the change amount of the peak velocity within the time window Δt, and its unit is cm / s; Δt is the length of the analysis time window, and its unit is seconds.

2. The OSAS cardiac mechanics analysis system combining two-dimensional speckle tracking and tissue Doppler according to claim 1, wherein The dual-channel data acquisition module further includes a preprocessing sub-module, specifically as follows: The preprocessing sub-module automatically segments the myocardial contour of the two-dimensional ultrasonic gray-scale image sequence and divides the endocardial layer, the middle layer and the epicardial layer regions; The preprocessing sub-module filters the respiratory motion artifacts of the myocardial tissue Doppler velocity spectrum and retains the velocity signals related to the cardiac systolic phase.

3. The OSAS cardiac mechanics analysis system combining two-dimensional speckle tracking and tissue Doppler according to claim 1, wherein The visualization interaction interface of the OSAS risk warning module includes displaying a hierarchical strain heat map, superimposing tissue Doppler velocity vector arrows and highlighting abnormal regions: The displayed hierarchical strain heat map is based on the hierarchical strain parameters generated by the first processing module and maps the strain distribution of each segment of the left ventricle with a color gradient; Red area: the longitudinal strain of the endocardial layer < -20%; Yellow area: -20% ≤ the longitudinal strain of the endocardial layer ≤ -15%; Green area: the longitudinal strain of the endocardial layer > -15%; The superimposed tissue Doppler velocity vector arrows are based on the velocity parameters generated by the second processing module , with the arrow direction indicating the myocardial motion direction, the arrow length being proportional to the velocity amplitude, and the velocity amplitude being defined as the absolute value of the myocardial motion velocity vector, i.e., the velocity magnitude, with its unit being cm / s; The abnormal area is highlighted and prompted for myocardial segments that simultaneously satisfy > τ and > 0.7 with flashing markers.

4. The OSAS cardiac mechanics analysis system combining two-dimensional speckle tracking and tissue Doppler according to claim 3, wherein It also includes a hardware integration component, which is composed of an inertial sensor and an edge computing device built into the ultrasonic probe: The inertial sensor built into the ultrasonic probe real-time detects the displacement amount (Δx, Δy) of the ultrasonic probe, and corrects the image offset through the following formula: wherein is the corrected two-dimensional image; is the original two-dimensional image; Δx and Δy are the displacement amounts of the ultrasonic probe in the X and Y directions, and their unit is pixel; The edge computing device is deployed on the ultrasonic host and is used for localizing the calculation of the hierarchical strain parameters.

5. The OSAS cardiac mechanics analysis method of the OSAS cardiac mechanics analysis system combining two-dimensional speckle tracking and tissue Doppler according to claim 1, characterized in that, The OSAS cardiac mechanics analysis method includes the following steps: Step S1, synchronously obtain the two-dimensional ultrasonic gray-scale image sequence and the myocardial tissue Doppler velocity spectrum through the dual-channel data acquisition module; Step S2, respectively generate hierarchical strain parameters and velocity parameters through the parallel computing engine module; Step S3: Calculate the phase deviation and the risk index through the dynamic association module and the risk index ; Step S4: Output a visualization report and a risk level through the OSAS risk warning module, where the risk level includes low risk, medium risk, and high risk: The low risk: ≤0.3; The medium risk: 0.3 < ≤ 0.7; The high risk: > 0.7.

Images