Aortic dissection risk reverse assessment method and system

Abstract

The application discloses an aortic dissection risk reverse evaluation method and system, relates to the technical field of cardiovascular disease risk evaluation, and discloses an aortic dissection risk reverse evaluation method and system, which calculates a contradiction coefficient of blood vessel stability and behavior load by integrating medical image data and daily behavior dynamic data, generates early warning information based on the difference, effectively solves the technical problems that a traditional evaluation method cannot dynamically capture mechanical changes of a blood vessel wall and identify potential risk conflicts, can dynamically monitor the contradictory relationship between the geometric stability of a blood vessel and daily behavior load in real time, generates high-risk early warning information in time, avoids the early warning loss caused by too long an examination interval in the traditional evaluation method, and improves the accuracy and timeliness of risk evaluation.

Description

Technical Field

[0001] This application relates to the field of cardiovascular disease risk assessment technology, and in particular to a method and system for reverse assessment of aortic dissection risk. Background Technology

[0002] Aortic dissection, a life-threatening vascular emergency, is characterized by its sudden onset and high mortality rate; failure to intervene promptly can lead to catastrophic consequences. Significant progress has been made in recent years in the study of the pathological mechanisms of aortic dissection, particularly in the area of ​​endothelial dysfunction. Recent research has confirmed that pro-inflammatory M1 macrophages deliver circ_0000388 via exosomes, targeting and inhibiting miR-337-3p expression, driving endothelial-mesenchymal transition (EndMT). This process leads to vascular endothelial cells acquiring a mesenchymal phenotype (upregulation of markers such as Vimentin and Snail), promoting vascular wall fibrosis and structural damage.

[0003] However, existing mechanistic studies have neglected how the transient mechanical stress on the vascular wall caused by daily activities synergistically accelerates EndMT via the circ_0000388 pathway. Furthermore, although circ_0000388 can serve as a biomarker for aortic dissection, its detection requires invasive tissue sampling or costly exosome separation techniques, making it impossible to quantify the tolerance threshold of inherent vascular wall geometric stability defects to the EndMT process. Therefore, for potentially high-risk individuals without obvious clinical symptoms, traditional assessment mechanisms often fail to identify the potential conflict between vascular geometric stability and daily behavioral load, resulting in a lack of effective early warning during critical risk windows and causing patients to miss the optimal time for early intervention. Moreover, existing technical solutions are based on analyses from single data sources, such as static assessments based solely on imaging data or isolated behavioral monitoring. The accuracy and reliability of these assessments are significantly insufficient, failing to meet the clinical need for real-time, precise risk warnings.

[0004] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main objective of this application is to provide a method and system for reverse assessment of aortic dissection risk, aiming to improve the accuracy and timeliness of aortic dissection risk assessment.

[0006] To achieve the above objectives, this application proposes a reverse risk assessment method for aortic dissection, the method comprising:

[0007] Acquire chest CT image data of the target subject using medical imaging equipment;

[0008] The chest CT image data is processed to extract the three-dimensional structural features of blood vessels to obtain vascular geometric stability feature data;

[0009] Acquire dynamic data on the daily behavior of the target object through wearable devices;

[0010] The daily behavioral dynamic data are subjected to physiological load analysis to obtain behavioral dynamic stress data;

[0011] Based on the aforementioned vascular geometric stability characteristic data and behavioral dynamic stress data, a contradiction coefficient calculation is performed to obtain the contradiction coefficient between vascular stability and behavioral load.

[0012] The discrepancy between the coefficient of contradiction between vascular stability and behavioral load and a preset discrepancy coefficient threshold is calculated, and a high-risk warning information for aortic dissection is generated based on the result of the discrepancy calculation.

[0013] In one embodiment, the step of extracting three-dimensional vascular structural features from the chest CT image data to obtain vascular geometric stability feature data includes:

[0014] The chest CT image data is processed to segment blood vessels to obtain binary image data of the blood vessel region;

[0015] The binary image data of the blood vessel region is processed to extract the centerline to obtain the blood vessel centerline data;

[0016] Based on the aforementioned vascular centerline data, calculate the curvature derivative change data;

[0017] Based on the binary image data of the blood vessel region, calculate the local cross-sectional area fluctuation data;

[0018] The data on the change in curvature derivative and the data on fluctuations in local cross-sectional area are combined to generate the data on the geometric stability characteristics of the blood vessel.

[0019] In one embodiment, the step of calculating the curvature derivative change data based on the vascular centerline data includes:

[0020] Sampling points are selected at preset intervals along the blood vessel centerline data to obtain sampling point sequence data;

[0021] Calculate the curvature angle difference between adjacent segments of the sampled point sequence data to obtain the angle difference dataset;

[0022] The standard deviation of the angle difference dataset is calculated to obtain the curvature derivative change data.

[0023] In one embodiment, the method further includes:

[0024] Obtain multi-phase historical chest CT image data of the target subject from a medical imaging database;

[0025] The step of extracting three-dimensional vascular structural features from the chest CT image data for each period of historical chest CT image data is performed to obtain vascular geometric stability feature data, thereby obtaining a historical vascular geometric stability feature dataset.

[0026] Time-series decay analysis was performed on the historical vascular geometric stability feature dataset to obtain vascular stability decay rate data;

[0027] By integrating current vascular geometric stability feature data with vascular stability decay rate data, vascular geometric stability feature data is generated.

[0028] In one embodiment, the step of performing physiological load analysis on the daily behavioral dynamic data to obtain behavioral dynamic stress data includes:

[0029] Accelerometer peak data are extracted from the aforementioned daily behavioral dynamic data;

[0030] Voice recording data is extracted from the daily behavior dynamic data and high-frequency tremor analysis is performed to obtain high-frequency tremor component data;

[0031] The accelerometer peak data and high-frequency vibration component data are weighted and fused to obtain the behavioral dynamic stress data.

[0032] In one embodiment, the steps of extracting voice recording data from the daily behavior dynamic data and performing high-frequency tremor analysis to obtain high-frequency tremor component data include:

[0033] The voice recording data is processed into frames to obtain voice frame sequence data;

[0034] The fundamental frequency fluctuation analysis is performed on the speech frame sequence data to obtain frequency oscillation amplitude data;

[0035] The proportion of frequency oscillation amplitude data exceeding a preset frequency oscillation amplitude threshold is statistically analyzed to obtain the high-frequency tremor component data.

[0036] In one embodiment, before the step of weighted fusion processing of the accelerometer peak data and high-frequency flutter component data to obtain the behavioral dynamic stress data, the method further includes:

[0037] When high-frequency tremor component data is available, emotional stress surge data is generated based on the preset mapping relationship between accelerometer peak data and aortic wall stress response.

[0038] The surge in emotional response data is overlaid onto the current accelerometer peak data to regenerate the accelerometer peak data.

[0039] In one embodiment, the step of calculating the inconsistency coefficient based on the vascular geometric stability feature data and behavioral dynamic stress data to obtain the inconsistency coefficient between vascular stability and behavioral load includes:

[0040] The vascular geometric stability feature data is mapped to stability score data;

[0041] The dynamic stress data of the behavior is mapped to load strength score data;

[0042] The ratio of the stability score data to the load intensity score data is calculated to obtain the coefficient of inconsistency between vascular stability and behavioral load.

[0043] In one embodiment, the step of calculating the difference between the inconsistency coefficient of vascular stability and behavioral load and a preset inconsistency coefficient threshold, and generating aortic dissection high-risk early warning information based on the result of the difference calculation includes:

[0044] Obtain a preset contradiction coefficient threshold from the historical case database;

[0045] The difference between the contradiction coefficient of vascular stability and behavioral load and the contradiction coefficient threshold is calculated to obtain the corresponding difference.

[0046] When the difference is lower than a preset risk margin, a high-risk warning message for aortic dissection is generated.

[0047] Furthermore, to achieve the above objectives, this application also proposes an aortic dissection risk reverse assessment system, the system comprising: a memory, a processor, and an aortic dissection risk reverse assessment program stored in the memory and executable on the processor, the aortic dissection risk reverse assessment program being configured to implement the steps of the aortic dissection risk reverse assessment method.

[0048] The aortic dissection risk reverse assessment method and system proposed in this application calculates the contradiction coefficient between vascular stability and behavioral load by integrating medical imaging data and daily behavioral dynamic data, and generates early warning information based on the difference. This effectively solves the technical problem that traditional assessment methods cannot dynamically capture changes in vascular wall mechanics and identify potential risk conflicts. It can monitor the contradictory relationship between vascular geometric stability and daily behavioral load in real time, and generate high-risk early warning information in a timely manner, avoiding the lack of early warnings caused by excessively long examination intervals in traditional assessment methods, and improving the accuracy and timeliness of risk assessment. Attached Figure Description

[0049] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0050] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a flowchart illustrating an embodiment of the reverse risk assessment method for aortic dissection in this application.

[0052] Figure 2 For this application Figure 1 A detailed flowchart of step S200;

[0053] Figure 3 For this application Figure 1 Detailed flowchart of step S400;

[0054] Figure 4 For this application Figure 1 Detailed flowchart of step S500;

[0055] Figure 5 For this application Figure 1 A detailed flowchart of step S600;

[0056] Figure 6 This is a schematic diagram of a structural embodiment of the aortic dissection risk reverse assessment system of this application.

[0057] Explanation of icon numbers:

[0058] 10. Memory; 20. Processor.

[0059] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0060] The technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of this application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0061] It should be understood that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0062] Existing mechanistic studies have neglected how the transient mechanical stress on the vascular wall caused by daily activities synergistically accelerates EndMT via the circ_0000388 pathway. Furthermore, although circ_0000388 can serve as a biomarker for aortic dissection, its detection requires invasive tissue sampling or costly exosome separation techniques, making it impossible to quantify the tolerance threshold of inherent vascular wall geometric stability defects to the EndMT process. Therefore, for potentially high-risk individuals without obvious clinical symptoms, traditional assessment mechanisms often fail to identify the potential conflict between vascular geometric stability and daily behavioral load, resulting in a lack of effective early warning during critical risk windows and causing patients to miss the optimal time for early intervention. Moreover, existing technical solutions are often based on single data sources, such as static assessments based solely on imaging data or isolated behavioral monitoring. The accuracy and reliability of these assessments are significantly insufficient, failing to meet the clinical need for real-time, precise risk warnings.

[0063] Based on this, embodiments of this application provide a reverse risk assessment method for aortic dissection, referring to... Figure 1 The aforementioned reverse risk assessment method for aortic dissection includes steps S100 to S600, wherein:

[0064] Step S100: Acquire chest CT image data of the target object using medical imaging equipment;

[0065] Step S200: Extract three-dimensional structural features of blood vessels from the chest CT image data to obtain vascular geometric stability feature data.

[0066] Step S300: Obtain dynamic data of the target object's daily behavior through wearable devices;

[0067] Step S400: Perform physiological load analysis on the daily behavioral dynamic data to obtain behavioral dynamic stress data;

[0068] Step S500: Based on the vascular geometric stability feature data and behavioral dynamic stress data, perform contradiction coefficient calculation processing to obtain the contradiction coefficient between vascular stability and behavioral load.

[0069] Step S600: The difference between the contradiction coefficient of vascular stability and behavioral load and the preset contradiction coefficient threshold is calculated, and a high-risk warning information of aortic dissection is generated based on the result of the difference calculation.

[0070] In this embodiment, vascular geometric stability feature data are quantitative indicators reflecting the stability of vascular anatomical structures, which can be achieved in various ways. For example, statistical analysis of changes in vessel wall thickness or quantitative description of the regularity of vascular branch angle distribution can yield a dataset characterizing the vascular geometric state. This data is primarily used to assess the current anatomical vulnerability of blood vessels. Daily behavioral dynamic data refers to activity-related data of the target individual in daily life, which can be obtained through various means. For example, using the accelerometer built into a smartphone to record the user's movement trajectory, or collecting user behavior pattern data through environmental monitoring devices, thus providing a basis for subsequent physiological load analysis. The calculation of the contradiction coefficient can be implemented using various mathematical models. For example, by constructing a multidimensional feature space and normalizing features of different dimensions to calculate the comprehensive distance, or by using machine learning algorithms to classify and score the input data, a quantitative indicator reflecting the degree of matching between vascular stability and behavioral load can be obtained.

[0071] This application achieves real-time capture of the dynamic evolution of vascular status and the interactive effects of behavioral load by combining multi-source data from medical imaging equipment and wearable devices. Compared with traditional assessment methods that rely solely on retrospective medical history analysis or intermittent imaging examinations, this method can cover daily activity scenarios and provide continuous monitoring, thereby solving the technical problem of not being able to provide timely warnings of potentially high-risk individuals without obvious symptoms.

[0072] This application embodiment acquires chest CT image data of the target object using medical imaging equipment. The chest CT image data provides high-resolution vascular anatomy information. Furthermore, the chest CT image data undergoes vascular three-dimensional structural feature extraction processing to obtain vascular geometric stability feature data. Specifically, this process, based on the high-resolution characteristics of CT images, accurately reflects the vulnerability of the current vascular anatomical state through quantitative analysis of geometric features such as changes in the derivative of vascular curvature and fluctuations in cross-sectional area. This avoids the shortcomings of traditional retrospective medical history analysis, which cannot reflect dynamic evolution.

[0073] Simultaneously, wearable devices acquire dynamic data on the daily behavior of the target individuals, including accelerometer peak values ​​and voice recordings, enabling continuous, non-invasive monitoring of their behavioral load. Furthermore, physiological load analysis is performed on this dynamic data to obtain dynamic behavioral stress data. For example, by weighted fusion of accelerometer peak values ​​and high-frequency tremor components in speech, the behavioral data is transformed into physiological stress indicators, effectively capturing the immediate impact of dynamic factors such as emotional stress on blood vessels and compensating for the intermittent nature of regular imaging examinations.

[0074] In this embodiment, based on the aforementioned vascular geometric stability feature data and behavioral dynamic stress data, a contradiction coefficient calculation is performed to obtain the contradiction coefficient between vascular stability and behavioral load. Specifically, this process maps vascular geometric stability feature data to stability score data and behavioral dynamic stress data to load intensity score data, calculates the ratio between the two, and directly assesses the matching degree between vascular stability and behavioral load. Thus, when the blood vessel is fragile but the behavioral load is high, the contradiction coefficient is significantly reduced, thereby accurately identifying risk contradiction points. Finally, the difference between the contradiction coefficient of vascular stability and behavioral load and a preset contradiction coefficient threshold is calculated, and a high-risk warning information for aortic dissection is generated based on the result of the difference calculation. As a preferred implementation, when the difference is lower than a preset risk margin, a high-risk warning information is generated. The threshold difference mechanism achieves risk quantification judgment, ensuring timely intervention for asymptomatic high-risk individuals and avoiding missing the optimal warning opportunity.

[0075] In one feasible implementation, refer to Figure 2 Step S200 includes steps S210 to S250, wherein:

[0076] Step S210: Perform vascular segmentation processing on the chest CT image data to obtain binary image data of the vascular region;

[0077] Step S220: Perform centerline extraction processing on the binary image data of the blood vessel region to obtain blood vessel centerline data;

[0078] Step S230: Calculate the curvature derivative change data based on the blood vessel centerline data;

[0079] Step S240: Calculate local cross-sectional area fluctuation data based on the binary image data of the blood vessel region;

[0080] Step S250: The curvature derivative change data and the local cross-sectional area fluctuation data are fused to generate the vascular geometric stability feature data.

[0081] In this embodiment, vascular segmentation refers to the precise separation of vascular structures from the background and other tissues in medical images using specific algorithms. This can be achieved using methods such as threshold-based segmentation, region growing, or deep learning models. Specifically, the binary image data of the vascular region is obtained by binarizing the segmentation results. This processing method effectively eliminates noise interference and highlights vascular morphological features, providing reliable basic data for subsequent analysis. Centerline extraction involves extracting the geometric axis of the blood vessel using topological principles and image processing techniques. This can be achieved through distance transformation, skeletonization algorithms, or graph search-based methods. The key to this step is ensuring the continuity and accuracy of the centerline, thus laying a high-precision geometric framework for subsequent curvature analysis. Curvature derivative variation data refers to an index describing the dynamic gradient of the vascular path's curvature. This can be achieved through differential geometry methods or numerical difference techniques. The core purpose of this design is to identify stress concentration areas in the blood vessel, especially points of abrupt changes in the curvature derivative, which often correspond to weak points in the vascular wall prone to tearing. Local cross-sectional area fluctuation data refers to an index reflecting the regularity of the vascular wall by quantifying changes in the vascular cross-sectional area. This can be achieved through pixel statistical analysis or geometric fitting methods. This step can accurately detect local abnormalities such as aneurysms or stenosis, because dramatic fluctuations in cross-sectional area often indicate degeneration of the blood vessel wall structure.

[0082] In this embodiment, the above-mentioned scheme systematically solves the problem of the one-sidedness of vascular geometric stability assessment through a multi-dimensional feature extraction and fusion mechanism. Vascular segmentation processing, by accurately separating the vascular structure from surrounding tissues, eliminates background noise and interference from adjacent organs in medical images, providing a clean vascular morphological basis for subsequent analysis. Centerline extraction processing utilizes the topological properties of binary images, avoiding path distortion caused by relying solely on surface contours, ensuring accurate acquisition of centerline data. Based on this, the calculation of curvature derivative variation data focuses on the dynamic gradient of vascular curvature rather than static curvature values, significantly improving sensitivity to geometric instability. Simultaneously, the calculation of local cross-sectional area fluctuation data directly utilizes the pixel distribution information of the binary image, quantifying the regularity of the vascular wall through changes in the area of ​​local regions, avoiding the loss of detail caused by global averaging. Finally, vascular geometric stability feature data is generated by fusing curvature derivative variation data and local cross-sectional area fluctuation data. This fusion mechanism organically combines the dynamic characteristics of path curvature with the spatial characteristics of cross-sectional changes, considering both the continuity of the overall vascular geometry and incorporating discrete anomalies in local structures, thereby constructing a more comprehensive and robust stability assessment index. In this way, not only is the limitation of traditional feature extraction methods relying on only a single geometric parameter overcome, but also the precise capture of subtle abnormal changes in vascular structure is achieved, providing high-confidence data support for subsequent aortic dissection risk assessment.

[0083] In one feasible implementation, the step of calculating the curvature derivative change data based on the vascular centerline data includes: selecting sampling points at preset distances along the vascular centerline data to obtain sampling point sequence data; calculating the curvature angle difference between adjacent segments of the sampling point sequence data to obtain an angle difference dataset; and performing standard deviation calculation processing on the angle difference dataset to obtain the curvature derivative change data.

[0084] In this embodiment, the sampling point sequence data refers to the set of discrete points extracted from the vessel centerline data using a uniform sampling strategy with fixed intervals. This can be achieved using an equal-interval sampling algorithm or an adaptive step-size sampling method, aiming to ensure the uniformity of the sampling point distribution and thus avoid curvature calculation bias caused by uneven sampling. The angle difference dataset can be understood as a data set generated by quantitatively analyzing the curvature angle differences between adjacent segments in the sampling point sequence data. This can be achieved through vector angle calculation or geometric trigonometric function analysis, aiming to capture the dynamic characteristics of local vessel curvature. Furthermore, the standard deviation calculation process refers to using the dispersion index in statistics to measure the degree of fluctuation in the angle difference dataset. This can be achieved using the classical standard deviation formula or an improved weighted standard deviation algorithm, aiming to objectively distinguish between gentle and drastic changes in vessel curvature.

[0085] In this embodiment, the above-described scheme ensures the uniformity and representativeness of the sampling point distribution by selecting sampling points at preset intervals along the blood vessel centerline data, providing a stable data foundation for subsequent curvature analysis. Based on this, by calculating the curvature angle difference between adjacent segments of the sampling point sequence data, the dynamic changes in local vascular curvature can be reflected more sensitively, avoiding the neglect of minute fluctuations by a single curvature value. Simultaneously, standard deviation calculation processing of the angle difference dataset further improves the accuracy and interpretability of the curvature derivative change data, enabling it to be directly correlated with the mechanical stability of the blood vessel wall. The above techniques, combined with the overall process of extracting and processing three-dimensional vascular structural features, significantly enhance the reliability of vascular geometric stability feature data, thereby providing more accurate foundational support for subsequent aortic dissection risk assessment.

[0086] In one feasible implementation, the method further includes: obtaining multi-period historical chest CT image data of the target object from a medical imaging database; performing the step of extracting three-dimensional vascular structural features from the chest CT image data to obtain vascular geometric stability feature data on each historical chest CT image data to obtain historical vascular geometric stability feature dataset; performing time-series decay analysis on the historical vascular geometric stability feature dataset to obtain vascular stability decay rate data; and fusing the current vascular geometric stability feature data and the vascular stability decay rate data to generate vascular geometric stability feature data.

[0087] In this embodiment, the medical image database refers to a dataset storing multiple medical imaging examination results of the target object. It can be implemented using the hospital's existing PACS system or cloud storage platform, with the aim of providing continuous historical data support for subsequent time-series analysis. The historical vascular geometric stability feature dataset can be understood as a feature set formed by extracting consistent features from multiple historical imaging data. It reflects the trajectory of vascular geometric features over time, ensuring the comparability of data from different periods in the feature space. The vascular stability decay rate data refers to an indicator obtained by quantitatively analyzing the rate of degradation of vascular geometric features over time. It can be implemented using mathematical modeling methods such as linear regression and exponential fitting, aiming to capture the progressive degradation trend of vascular status.

[0088] In this embodiment, the scheme achieves dynamic monitoring and assessment of vascular stability by integrating multi-period historical chest CT image data. First, multi-period historical image data of the target object is obtained from a medical imaging database. This data provides the system with continuous temporal information, enabling the tracking of changes in vascular geometric features. Next, vascular three-dimensional structural feature extraction processing is performed on each historical image data, consistent with the current image, ensuring consistency in feature extraction standards between historical and current data, thereby generating a historical vascular geometric stability feature dataset. Subsequently, time-series decay analysis is performed based on this dataset. By quantifying the vascular stability decay rate, the static vascular state assessment is transformed into dynamic degradation process monitoring. Finally, the current vascular geometric stability feature data is fused with the decay rate data to generate comprehensive vascular geometric stability feature data. This allows risk assessment to be based not only on instantaneous vascular state but also on historical deterioration rates, thus more accurately reflecting the actual risk level of the blood vessel. This solves the problem of failing to capture the long-term dynamic evolution of vascular state caused by relying solely on single chest CT image data for assessment. Especially for potentially high-risk individuals without obvious symptoms, it enables early risk warning, improving the accuracy and timeliness of aortic dissection risk assessment.

[0089] In one feasible implementation, refer to Figure 3 Step S400 includes steps S410 to S430, wherein:

[0090] Step S410: Extract accelerometer peak data from the daily behavior dynamic data;

[0091] Step S420: Extract voice recording data from the daily behavior dynamic data and perform high-frequency tremor analysis to obtain high-frequency tremor component data;

[0092] Step S430: Perform weighted fusion processing on the accelerometer peak data and high-frequency tremor component data to obtain the behavioral dynamic stress data.

[0093] In this embodiment, accelerometer peak data refers to the peak motion intensity information collected by the accelerometer sensor built into the wearable device, which can be implemented using a triaxial accelerometer or a higher-precision inertial measurement unit. High-frequency tremor component data can be understood as emotion-related parameters obtained by framing the speech signal and identifying fundamental frequency fluctuation characteristics; these can be extracted using time-frequency analysis methods such as short-time Fourier transform and wavelet transform. Weighted fusion processing refers to the process of assigning appropriate weights to different physiological signals based on their influence on aortic wall stress and then performing a comprehensive calculation. Its purpose is to integrate multi-dimensional physiological signals to ensure that the behavioral load assessment more closely reflects the actual physiological state.

[0094] In this embodiment, the scheme first monitors the target's body movement status in real time using a wearable device, extracting accelerometer peak data to quantify the contribution of physical activity intensity to the mechanical stress of the aortic wall. Simultaneously, considering the implicit impact of emotional stress on the vascular wall, voice recording data is further extracted and high-frequency tremor analysis is performed, converting the vocal cord micro-tremors caused by autonomic nervous system activation during emotional tension into quantifiable high-frequency tremor component data. Subsequently, based on the differences in the mechanisms of action of different physiological signals, the accelerometer peak data and high-frequency tremor component data are weighted and fused to generate behavioral dynamic stress data that includes explicit exercise intensity indicators and implicit emotional stress effects. This process not only overcomes the limitations of single data source analysis but also constructs a more comprehensive behavioral dynamic stress assessment framework by integrating multi-dimensional physiological signals from body movement and emotional stress, effectively solving the problem of distorted physiological load analysis results caused by ignoring emotional fluctuations in traditional methods. Thus, the system can more accurately quantify the impact of behavioral load on blood vessels, providing a reliable basis for subsequent contradiction coefficient calculation, thereby improving the accuracy and reliability of aortic dissection risk warning.

[0095] In one feasible implementation, the steps of extracting voice recording data from the daily behavior dynamic data and performing high-frequency tremor analysis to obtain high-frequency tremor component data include: performing frame-segmentation processing on the voice recording data to obtain voice frame sequence data; performing fundamental frequency fluctuation analysis on the voice frame sequence data to obtain frequency oscillation amplitude data; and calculating the proportion of the frequency oscillation amplitude data that exceeds a preset frequency oscillation amplitude threshold to obtain the high-frequency tremor component data.

[0096] In this embodiment, the speech frame sequence data refers to the data set obtained by segmenting a continuous speech signal into short time frames, which can be achieved by using a fixed time window to slide and extract speech segments. The purpose of this step is to utilize the relatively stable fundamental frequency of the speech signal within a short time window to effectively capture instantaneous frequency changes, providing localized and refined basic data for subsequent analysis. Fundamental frequency fluctuation analysis is the process of calculating the degree of fundamental frequency change for each speech frame, which can be achieved through time-frequency analysis methods such as short-time Fourier transform or wavelet transform. This step aims to quantify vocal cord vibration abnormalities caused by physiological stress, thereby accurately identifying high-frequency tremor components. Frequency oscillation amplitude data refers to a quantitative indicator reflecting the intensity of fundamental frequency fluctuations within a speech frame, which can be obtained by calculating the absolute value or variance of the fundamental frequency difference between adjacent frames. The purpose of this step is to transform the tremor characteristics in the speech signal into quantifiable numerical indicators, providing a basis for subsequent statistical analysis.

[0097] In this embodiment, by segmenting the speech recording data into frames, the feature ambiguity caused by time-varying characteristics in the overall speech signal analysis can be avoided, ensuring high stability of the fundamental frequency information within each speech frame. Based on this, fundamental frequency fluctuation analysis is performed on the segmented speech frame sequence data, directly targeting vocal cord vibration abnormalities caused by physiological stress in the speech. By quantifying the fundamental frequency oscillation amplitude, accurate identification of high-frequency tremor components is achieved. Furthermore, by setting a reasonable frequency oscillation amplitude threshold and statistically analyzing the proportion exceeding the threshold, frequency oscillations are transformed into objective quantitative indicators, eliminating subjective judgment errors and ensuring the repeatability and standardization of high-frequency tremor component data. Through the above technical solution, the systematic speech signal processing flow solves the problem of inaccurate data caused by the lack of specificity in high-frequency tremor analysis, ensuring the reliability of behavioral dynamic stress data, thereby supporting the accurate assessment of aortic dissection risk.

[0098] In one feasible implementation, before the step of weighted fusion processing of the accelerometer peak data and high-frequency tremor component data to obtain the behavioral dynamic stress data, the method further includes: when high-frequency tremor component data exists, generating emotional stress surge data according to a preset mapping relationship between accelerometer peak data and aortic wall stress response; superimposing the emotional stress surge data onto the current accelerometer peak data to regenerate the accelerometer peak data.

[0099] In this embodiment, high-frequency tremor component data refers to the proportion of frequency oscillation amplitude data exceeding a preset threshold extracted after frame-by-frame processing of voice recording data. This can be achieved using statistical analysis or machine learning classification models. In practical applications, emotional stress surge data refers to the quantitative value calculated based on a physiological correlation model between accelerometer peak data and aortic wall stress response. This value can be generated using regression models or empirical formulas validated through medical experiments. The purpose of regenerating accelerometer peak data is to embed emotional stress factors into physical motion data, thereby improving the comprehensiveness of behavioral dynamic stress assessment.

[0100] In this embodiment, the scheme first identifies the existence of high-frequency tremor component data through a conditional judgment mechanism, ensuring that subsequent processing is triggered only in emotional fluctuation scenarios, thus avoiding invalid calculations. Subsequently, emotional stress surge data is generated based on a preset mapping relationship between accelerometer peak data and aortic wall stress response. This process utilizes a medically validated physiological correlation model to ensure that the quantitative results of emotional stress accurately reflect the actual response of the aortic wall. Finally, the emotional stress surge data is superimposed on the current accelerometer peak data to form adjusted accelerometer peak data, which not only includes physical activity load but also integrates the cumulative effect of emotional stress. These techniques, through conditional triggering, precise mapping, and dynamic superposition, achieve the capture and quantitative integration of implicit emotional risks in daily behavior, significantly improving the accuracy of behavioral dynamic stress data.

[0101] In this embodiment, the scheme, combined with steps such as extracting vascular geometric stability feature data and calculating the contradiction coefficient, can more comprehensively reflect the changes in the vascular state of the target object. By dynamically integrating emotional stress factors, behavioral dynamic stress data can better serve the calculation of the contradiction coefficient between vascular stability and behavioral load, thereby providing a more reliable basis for generating high-risk early warning information.

[0102] In one feasible implementation, refer to Figure 4 Step S500 includes steps S510 to S530, wherein:

[0103] Step S510: Map the vascular geometric stability feature data to stability score data;

[0104] Step S520: Map the behavioral dynamic stress data to load strength score data;

[0105] Step S530: Calculate the ratio of the stability score data to the load intensity score data to obtain the contradiction coefficient between vascular stability and behavioral load.

[0106] In this embodiment, stability score data refers to a standardized numerical value obtained by mapping vascular geometric stability feature data. This can be achieved using linear normalization, nonlinear normalization, or piecewise function mapping. The purpose of this step is to eliminate unit differences between multidimensional parameters such as curvature derivative changes and local cross-sectional area fluctuations, thereby ensuring the accuracy of subsequent calculations. Load intensity score data refers to a standardized numerical value obtained by mapping behavioral dynamic stress data. This can be achieved using weighted average methods, fuzzy membership functions, or machine learning model prediction. The purpose of this step is to comprehensively quantify the impact of peak acceleration and high-frequency tremor components in daily behavior on physiological load, thereby improving the reliability of the assessment results.

[0107] In this embodiment, the above-mentioned scheme effectively solves the problem of distorted calculation of the inconsistency coefficient caused by different units and dimensions by mapping vascular geometric stability feature data and behavioral dynamic stress data into stability score data and load intensity score data, respectively. In the specific implementation process, the vascular geometric stability feature data is first mapped to extract the dynamic evolution information of the vascular structure contained therein and convert it into a unified scoring scale, avoiding the risk of interference from local fluctuations in the original data. Subsequently, the behavioral dynamic stress data is mapped to integrate the accelerometer peak data and high-frequency tremor component data into a comparable metric, eliminating the influence of behavioral type differences on load assessment. Finally, by calculating the ratio of stability score data to load intensity score data, rather than a simple difference, the relative imbalance between vascular stability and behavioral load can be captured more accurately. This ratio calculation method is based on standardized scoring data, avoiding weight bias caused by differences in the scale of the original data, thereby ensuring that when the behavioral load intensity exceeds the vascular stability threshold, the inconsistency coefficient can stably indicate a high-risk state.

[0108] In this embodiment, the above-described scheme is combined with the steps of acquiring vascular geometric stability characteristic data and behavioral dynamic stress data, further enhancing the scientific rigor and practicality of the overall assessment system. By introducing a standardized mapping mechanism, not only is the robustness of the inconsistency coefficient calculation enhanced, but a reliable data foundation is also provided for the subsequent generation of high-risk aortic dissection early warning information.

[0109] In one feasible implementation, refer to Figure 5 Step S600 includes steps S610 to S630, wherein:

[0110] Step S610: Obtain the preset contradiction coefficient threshold from the historical case database;

[0111] Step S620: Calculate the difference between the contradiction coefficient of vascular stability and behavioral load and the contradiction coefficient threshold to obtain the corresponding difference.

[0112] Step S630: When the difference is lower than the preset risk margin, generate the high-risk warning information for aortic dissection.

[0113] In this embodiment, the historical case database refers to an electronic database system storing a large amount of past patient clinical data. It can be implemented using a distributed or centralized database architecture, aiming to provide reliable support for setting thresholds based on real-world population risk patterns. The preset contradiction coefficient threshold is a critical value used to measure risk status, derived from statistical analysis of historical case data. It can be achieved through machine learning model training or expert-based rule setting, ensuring the threshold's dynamic adaptability. In practical applications, the difference refers to the absolute gap between the current contradiction coefficient and the threshold, which can be achieved through simple mathematical calculations or more complex weighted difference algorithms, quantifying the degree of risk deviation. Furthermore, the preset risk margin refers to a safety buffer zone set to avoid interference from minor fluctuations. It can be achieved through fixed value settings or dynamic adjustment strategies, aiming to improve the robustness of early warning triggering.

[0114] In this embodiment, the scheme first extracts a conflict coefficient threshold from a historical case database. This process utilizes real clinical data instead of traditional fixed empirical values, ensuring that the threshold dynamically reflects the characteristics of disease evolution. Subsequently, the difference between the conflict coefficient of vascular stability and behavioral load and this threshold is calculated. By quantifying the precise gap between the current risk and the safety threshold, an objective and measurable basis is provided for early warning decisions. Finally, a high-risk early warning is generated when the difference is lower than a preset risk margin. This mechanism introduces a safety buffer space, dynamically adjusting the early warning trigger point based on the risk margin, effectively filtering out minor fluctuations and activating the warning only when the risk significantly exceeds the safety range. Overall, these steps are organically combined, solving the problems of insufficient threshold reliability and lack of early warning sensitivity. Furthermore, they complement the aforementioned schemes for obtaining vascular geometric stability feature data and behavioral dynamic stress data, jointly improving the ability to identify true high-risk states and the system's stability.

[0115] In the embodiments of this application, the aortic dissection risk reverse assessment method calculates the contradiction coefficient between vascular stability and behavioral load by integrating medical imaging data and daily behavioral dynamic data, and generates early warning information based on the difference. This effectively solves the technical problem that traditional assessment methods cannot dynamically capture changes in vascular wall mechanics and identify potential risk conflicts. It can monitor the contradictory relationship between vascular geometric stability and daily behavioral load in real time, generate high-risk early warning information in a timely manner, avoid the lack of early warning caused by excessively long examination intervals in traditional assessment methods, and improve the accuracy and timeliness of risk assessment.

[0116] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the reverse assessment method for aortic dissection risk in this application. Any simple modifications based on this technical concept are within the scope of protection of this application.

[0117] This application also provides a reverse risk assessment system for aortic dissection, for reference. Figure 6 The system includes: a memory 10, a processor 20, and an aortic dissection risk reverse assessment program stored on the memory 10 and executable on the processor 20, the aortic dissection risk reverse assessment program being configured to implement the steps of the aortic dissection risk reverse assessment method.

[0118] The aortic dissection risk reverse assessment system provided in this application, employing the aortic dissection risk reverse assessment method described in the above embodiments, can improve the accuracy and timeliness of aortic dissection risk assessment. Compared with the prior art, the beneficial effects of the aortic dissection risk reverse assessment system provided in this application are the same as those of the aortic dissection risk reverse assessment method provided in the above embodiments, and other technical features of the aortic dissection risk reverse assessment system are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0119] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0120] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. All equivalent structural transformations made under the technical concept of this application using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included within the scope of patent protection of this application.

Claims

1. A reverse assessment method for the risk of aortic dissection, characterized in that, The method includes: Acquire chest CT image data of the target subject using medical imaging equipment; The chest CT image data is processed to extract the three-dimensional structural features of blood vessels to obtain vascular geometric stability feature data; Acquire dynamic data on the daily behavior of the target object through wearable devices; The daily behavioral dynamic data are subjected to physiological load analysis to obtain behavioral dynamic stress data; Based on the aforementioned vascular geometric stability characteristic data and behavioral dynamic stress data, a contradiction coefficient calculation is performed to obtain the contradiction coefficient between vascular stability and behavioral load. The discrepancy between the contradiction coefficient of vascular stability and behavioral load and the preset discrepancy coefficient threshold are calculated, and a high-risk warning information for aortic dissection is generated based on the result of the discrepancy calculation. The steps for calculating the discrepancy coefficient between vascular stability and behavioral load based on the aforementioned vascular geometric stability characteristic data and behavioral dynamic stress data include: The vascular geometric stability feature data is mapped to stability score data; The dynamic stress data of the behavior is mapped to load strength score data; Calculate the ratio of the stability score data to the load intensity score data to obtain the coefficient of inconsistency between vascular stability and behavioral load; The steps of calculating the difference between the inconsistency coefficient of vascular stability and behavioral load and a preset inconsistency coefficient threshold, and generating a high-risk early warning information for aortic dissection based on the result of the difference calculation, include: Obtain a preset contradiction coefficient threshold from the historical case database; The difference between the contradiction coefficient of vascular stability and behavioral load and the contradiction coefficient threshold is calculated to obtain the corresponding difference. When the difference is lower than a preset risk margin, a high-risk warning message for aortic dissection is generated.

2. The method for reverse risk assessment of aortic dissection as described in claim 1, characterized in that, The steps of extracting three-dimensional vascular structural features from the chest CT image data to obtain vascular geometric stability feature data include: The chest CT image data is processed to segment blood vessels to obtain binary image data of the blood vessel region; The binary image data of the blood vessel region is processed to extract the centerline to obtain the blood vessel centerline data; Based on the aforementioned vascular centerline data, calculate the curvature derivative change data; Based on the binary image data of the blood vessel region, calculate the local cross-sectional area fluctuation data; The data on the change in curvature derivative and the data on fluctuations in local cross-sectional area are combined to generate the data on the geometric stability characteristics of the blood vessel.

3. The method for reverse risk assessment of aortic dissection as described in claim 2, characterized in that, The steps for calculating the curvature derivative change data based on the aforementioned vascular centerline data include: Sampling points are selected at preset intervals along the blood vessel centerline data to obtain sampling point sequence data; Calculate the curvature angle difference between adjacent segments of the sampled point sequence data to obtain the angle difference dataset; The standard deviation of the angle difference dataset is calculated to obtain the curvature derivative change data.

4. The method for reverse risk assessment of aortic dissection as described in claim 2, characterized in that, The method further includes: Obtain multi-phase historical chest CT image data of the target subject from a medical imaging database; The step of extracting three-dimensional vascular structural features from the chest CT image data for each period of historical chest CT image data is performed to obtain vascular geometric stability feature data, thereby obtaining a historical vascular geometric stability feature dataset. Time-series decay analysis was performed on the historical vascular geometric stability feature dataset to obtain vascular stability decay rate data; By integrating current vascular geometric stability feature data with vascular stability decay rate data, vascular geometric stability feature data is generated.

5. The method for reverse risk assessment of aortic dissection as described in claim 1, characterized in that, The steps of performing physiological load analysis on the daily behavioral dynamic data to obtain behavioral dynamic stress data include: Accelerometer peak data are extracted from the aforementioned daily behavioral dynamic data; Voice recording data is extracted from the daily behavior dynamic data and high-frequency tremor analysis is performed to obtain high-frequency tremor component data; The accelerometer peak data and high-frequency vibration component data are weighted and fused to obtain the behavioral dynamic stress data.

6. The method for reverse assessment of aortic dissection risk as described in claim 5, characterized in that, The steps of extracting voice recording data from the aforementioned daily behavioral dynamic data and performing high-frequency tremor analysis to obtain high-frequency tremor component data include: The voice recording data is processed into frames to obtain voice frame sequence data; The fundamental frequency fluctuation analysis is performed on the speech frame sequence data to obtain frequency oscillation amplitude data; The proportion of frequency oscillation amplitude data exceeding a preset frequency oscillation amplitude threshold is statistically analyzed to obtain the high-frequency tremor component data.

7. The method for reverse assessment of aortic dissection risk as described in claim 5, characterized in that, Before the step of weighted fusion processing of the accelerometer peak data and high-frequency flutter component data to obtain the behavioral dynamic stress data, the method further includes: When high-frequency tremor component data is available, emotional stress surge data is generated based on the preset mapping relationship between accelerometer peak data and aortic wall stress response. The surge in emotional response data is overlaid onto the current accelerometer peak data to regenerate the accelerometer peak data.

8. A reverse risk assessment system for aortic dissection, characterized in that, The system includes: a memory, a processor, and an aortic dissection risk reverse assessment program stored in the memory and executable on the processor, the aortic dissection risk reverse assessment program being configured to implement the aortic dissection risk reverse assessment method as described in any one of claims 1 to 7.

Images