Vascular Elasticity Risk Early Warning System and Method
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-03
- Publication Date
- 2026-08-14
AI Technical Summary
[0003]现有的血管弹性检测技术采用接触式或非接触式检测，接触式检测需在体表粘贴电极或压力传感器，操作复杂且易受皮肤状态影响，长期监测可能引发皮肤过敏或不适；而超声检测依赖专业设备与操作人员，成本高昂，且无法实现连续动态监测；另外还有利用动脉硬化检测仪进行检测，通过压力袖带测量踝肱指数，仅能反映下肢动脉状态，无法全面评估全身血管弹性；
1、本发明通过动态高阶谐波分析，利用傅里叶变换或小波变换将脉搏波信号分解为不同频率成分的谐波，提取高阶谐波的相对幅度、相位变化率等反映血管弹性变化的特征参数，将连续采集到的特征参数按时间顺序排列，构建时间序列数据，用于动态分析血管弹性变化趋势，全面评估全身血管弹性，同时，引入时间序列分析，对连续采集的脉搏波信号进行动态高阶谐波分析，捕捉血管弹性随时间的变化趋势，而非单一时间点的静态评估，提高检测敏感度与评估精度。
Smart Images
Figure CN122556927A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of signal processing technology, and in particular to a vascular elasticity risk early warning system and method. Background Technology
[0002] Vascular elasticity refers to the ability of the arterial wall to deform under pressure changes and its ability to return to its original shape. It mainly reflects the compliance and elastic modulus of large and medium-sized arteries. Decreased vascular elasticity is usually manifested as increased arterial stiffness elevation, and is one of the important early markers of pathological processes such as vascular aging, atherosclerosis, hypertension, and diabetes.
[0003] Existing vascular elasticity testing technologies employ contact or non-contact methods. Contact methods require attaching electrodes or pressure sensors to the skin surface, which is complex to operate and easily affected by skin condition; long-term monitoring may cause skin allergies or discomfort. Ultrasound testing relies on specialized equipment and operators, is costly, and cannot achieve continuous dynamic monitoring. Additionally, there are methods that use arteriosclerosis detectors to measure the ankle-brachial index with a pressure cuff, which only reflects the condition of the lower limb arteries and cannot comprehensively assess the elasticity of the entire body's blood vessels. Non-contact detection, which uses single-spectrum RGB image analysis to extract pulse wave signals using only visible light, is susceptible to interference from ambient light fluctuations and human movement. Minor human movements can introduce motion artifacts, leading to incorrect skin color region segmentation and distorted pulse wave signals, which cannot meet clinical accuracy requirements.
[0004] Therefore, it is necessary to provide a new vascular elasticity risk early warning system and method to solve the above-mentioned technical problems. Summary of the Invention
[0005] To address the aforementioned technical problems, this invention provides a vascular elasticity risk early warning system and method.
[0006] The vascular elasticity risk warning system provided by this invention includes: The multispectral image acquisition module is used to continuously acquire multispectral video images of specific areas of the human body; The image processing module is used to fuse the acquired multispectral video images and extract the pulse wave signal; The dynamic higher harmonic analysis module is used to perform multi-scale harmonic analysis on the extracted pulse wave signal and extract higher harmonic characteristic parameters that reflect changes in vascular elasticity. The risk assessment module, based on extracted high-order harmonic feature parameters, combined with user basic information and historical data, uses machine learning algorithms to assess vascular elasticity and cardiovascular disease risk. The early warning and feedback module issues warning signals based on risk assessment results and provides personalized health advice.
[0007] Furthermore, the multispectral image acquisition module includes a multispectral LED light source and a high-resolution CMOS sensor, which can simultaneously emit light of multiple wavelengths such as red, green, blue and near-infrared light, and acquire reflected light to form multispectral video images.
[0008] Furthermore, the image processing module includes: The multispectral image fusion submodule uses a deep learning-based image fusion algorithm to fuse image information from different wavelengths to generate a high-quality pulse wave signal source image. The skin color segmentation submodule applies a dynamic threshold segmentation algorithm and combines human motion information to achieve accurate skin color region segmentation. The morphological optimization submodule performs opening and closing operations on the segmented skin-colored regions to remove noise and fill holes. The skin color segmentation submodule includes: The dynamic threshold calculation unit dynamically updates the grayscale threshold of skin color pixels through exponential weighted moving average or moving average method, and determines the threshold range based on the statistical distribution of historical frames. The motion adaptive adjustment unit detects the human motion region using optical flow or inter-frame difference methods, relaxes the threshold range for dynamic regions, and tightens the threshold range for static regions. The multi-channel fusion unit combines the dynamic thresholds of the chroma and luminance channels in the color space and generates the final segmentation threshold through weighted fusion.
[0009] Furthermore, the dynamic higher-order harmonic analysis module includes: The harmonic decomposition submodule uses Fourier transform or wavelet transform to decompose the pulse wave signal into harmonics of different frequency components. The feature parameter extraction submodule extracts feature parameters that reflect changes in vascular elasticity, such as the relative amplitude and phase change rate of higher-order harmonics. The time series construction submodule arranges the continuously collected feature parameters in chronological order to construct time series data for dynamic analysis of vascular elasticity change trends.
[0010] Furthermore, the risk assessment module includes: The data preprocessing submodule normalizes the extracted higher-order harmonic characteristic parameters to eliminate the influence of individual differences and measurement conditions. The machine learning model submodule uses a pre-trained deep learning model, combined with basic user information and historical data, to assess vascular elasticity and cardiovascular disease risk. The risk level classification submodule classifies vascular elasticity into different levels based on the model output and provides the corresponding cardiovascular disease risk probability.
[0011] Furthermore, the early warning and feedback module includes: The real-time early warning submodule immediately issues an early warning signal when it detects a significant decrease in vascular elasticity or an increased risk of cardiovascular disease. The personalized recommendations submodule provides users with personalized health advice and lifestyle adjustment plans based on the risk assessment results. The health report generation submodule regularly generates detailed health reports, including trends in vascular elasticity, cardiovascular disease risk assessment results, and health recommendations.
[0012] Another aspect of the present invention provides a method for early warning of vascular elasticity risks, the method comprising the following steps: S1. Use a multispectral image acquisition module to continuously acquire multispectral video images of specific areas of the human body; S2. Use the image processing module to perform fusion processing and skin color segmentation on the acquired multispectral video images, and extract the pulse wave signal; S3. The extracted pulse wave signal is subjected to multi-scale harmonic analysis through the dynamic high-order harmonic analysis module to extract high-order harmonic characteristic parameters that reflect changes in vascular elasticity. S4. Input the extracted high-order harmonic feature parameters into the risk assessment module, and combine them with user basic information and historical data to use a machine learning model to assess vascular elasticity and cardiovascular disease risk. S5. Based on the risk assessment results, issue early warning signals through the early warning and feedback module and provide personalized health advice.
[0013] Furthermore, in step S2, skin color segmentation is achieved in the following way: In the YCbCr color space, dynamic thresholds are calculated for the Cb and Cr channels respectively. The dynamic thresholds are updated based on the exponentially weighted moving average of the grayscale values of the skin pixels in the current frame, and the threshold range is determined by combining the mean and standard deviation of historical frames. By using optical flow to detect human motion areas, the threshold range for dynamic areas is expanded by 20% to 30%, while the threshold range for static areas is reduced by 10% to 20%. Perform a logical AND operation on the binarized results of the Cb and Cr channels to generate an initial skin color mask; Morphological opening operations are applied to the initial skin color mask to remove small-area noise, and closing operations are applied to fill internal holes.
[0014] Furthermore, step S3 also includes: The continuously acquired pulse wave signals are dynamically tracked to capture the trend of changes in vascular elasticity over time. By analyzing the changes in harmonic components at different time scales, multi-scale characteristic parameters can be extracted to improve the sensitivity to changes in vascular elasticity.
[0015] Furthermore, in step S4, the training process of the machine learning model includes: Collect datasets of known vascular elasticity states, including data obtained through traditional contact measurement methods; The dataset is divided into a training set and a test set. The model is trained using the training set and validated and optimized using the test set. The training process is iterated continuously until the model achieves satisfactory evaluation accuracy.
[0016] Compared with related technologies, the vascular elasticity risk early warning system and method provided by the present invention have the following beneficial effects: 1. This invention utilizes dynamic higher-order harmonic analysis, employing Fourier transform or wavelet transform to decompose pulse wave signals into harmonics of different frequency components. It extracts characteristic parameters reflecting changes in vascular elasticity, such as the relative amplitude and phase change rate of the higher-order harmonics. The continuously acquired characteristic parameters are arranged chronologically to construct time-series data for dynamic analysis of vascular elasticity trends, comprehensively assessing systemic vascular elasticity. Furthermore, the introduction of time-series analysis allows for dynamic higher-order harmonic analysis of continuously acquired pulse wave signals, capturing the changing trends of vascular elasticity over time, rather than providing a static assessment at a single time point, thus improving detection sensitivity and assessment accuracy.
[0017] 2. This invention continuously acquires multispectral video images of specific areas of the human body, performs fusion processing on the acquired multispectral video images, extracts pulse wave signals, and utilizes a multispectral LED light source and a high-resolution CMOS sensor to simultaneously emit light of multiple wavelengths, including red, green, blue, and near-infrared light, and collects reflected light to form multispectral video images. A dynamic threshold segmentation algorithm, combined with human motion information, achieves precise skin color region segmentation. By combining dynamic thresholds of the chroma and luminance channels, a final segmentation threshold is generated through weighted fusion. Simultaneous illumination of the skin with multispectral light sources utilizes the differences in absorption characteristics of different wavelengths of light in blood to enhance sensitivity to changes in blood volume and improve signal quality. Through a multispectral image fusion algorithm, image information from different wavelengths is fused to extract richer and more accurate pulse wave signals. Attached Figure Description
[0018] Figure 1 This is a structural block diagram of the vascular elasticity risk early warning system provided by the present invention; Figure 2 This is a structural block diagram of the skin color segmentation submodule provided by the present invention; Figure 3 A flowchart illustrating the vascular elasticity risk warning method provided by the present invention; Figure 4 A flowchart illustrating skin color segmentation provided by this invention; Figure 5 A flowchart illustrating the training process of the machine learning model provided by this invention. Detailed Implementation
[0019] The present invention will be further described below with reference to the accompanying drawings and embodiments.
[0020] Please refer to the following: Figure 1 , Figure 2 , Figure 3 , Figure 4 as well as Figure 5 ,in, Figure 1 This is a structural block diagram of the vascular elasticity risk early warning system provided by the present invention; Figure 2 This is a structural block diagram of the skin color segmentation submodule provided by the present invention; Figure 3 A flowchart illustrating the vascular elasticity risk warning method provided by the present invention; Figure 4 A flowchart illustrating skin color segmentation provided by this invention; Figure 5 A flowchart illustrating the training process of the machine learning model provided by this invention.
[0021] Example 1 In the specific implementation process, such as Figures 1-2 As shown, the vascular elasticity risk warning system includes: The multispectral image acquisition module is used to continuously acquire multispectral video images of specific areas of the human body; The multispectral image acquisition module includes a multispectral LED light source and a high-resolution CMOS sensor, which can simultaneously emit light of multiple wavelengths such as red, green, blue and near-infrared light, and acquire reflected light to form multispectral video images; The image processing module is used to fuse the acquired multispectral video images and extract the pulse wave signal; The dynamic higher harmonic analysis module is used to perform multi-scale harmonic analysis on the extracted pulse wave signal and extract higher harmonic characteristic parameters that reflect changes in vascular elasticity. The risk assessment module, based on extracted high-order harmonic feature parameters, combined with user basic information and historical data, uses machine learning algorithms to assess vascular elasticity and cardiovascular disease risk. The early warning and feedback module issues warning signals based on risk assessment results and provides personalized health advice.
[0022] It should be noted that the image processing module includes: The multispectral image fusion submodule uses a deep learning-based image fusion algorithm to fuse image information from different wavelengths to generate a high-quality pulse wave signal source image. The skin color segmentation submodule applies a dynamic threshold segmentation algorithm and combines human motion information to achieve accurate skin color region segmentation. The morphological optimization submodule performs opening and closing operations on the segmented skin-colored regions to remove noise and fill holes. The skin color segmentation submodule includes: The dynamic threshold calculation unit dynamically updates the grayscale threshold of skin color pixels through exponential weighted moving average or moving average method, and determines the threshold range based on the statistical distribution of historical frames. The motion adaptive adjustment unit detects the human motion region using optical flow or inter-frame difference methods, relaxes the threshold range for dynamic regions, and tightens the threshold range for static regions. The multi-channel fusion unit, in the YCbCr or HSV color space, combines the dynamic thresholds of the chromaticity channel (Cb / Cr or H / S) and the luminance channel (Y or V) to generate the final segmentation threshold through weighted fusion.
[0023] It should be noted that the dynamic higher-order harmonic analysis module includes: The harmonic decomposition submodule uses Fourier transform or wavelet transform to decompose the pulse wave signal into harmonics of different frequency components. The feature parameter extraction submodule extracts feature parameters that reflect changes in vascular elasticity, such as the relative amplitude and phase change rate of higher-order harmonics. The time series construction submodule arranges the continuously collected feature parameters in chronological order to construct time series data for dynamic analysis of vascular elasticity change trends.
[0024] It should be noted that the risk assessment module includes: The data preprocessing submodule normalizes the extracted higher-order harmonic characteristic parameters to eliminate the influence of individual differences and measurement conditions. The machine learning model submodule uses a pre-trained deep learning model, combined with basic user information and historical data, to assess vascular elasticity and cardiovascular disease risk. The risk level classification submodule classifies vascular elasticity into different levels based on the model output and provides the corresponding cardiovascular disease risk probability.
[0025] It should be noted that the early warning and feedback module includes: The real-time early warning submodule immediately issues an early warning signal when it detects a significant decrease in vascular elasticity or an increased risk of cardiovascular disease. The personalized recommendations submodule provides users with personalized health advice and lifestyle adjustment plans based on the risk assessment results. The health report generation submodule regularly generates detailed health reports, including trends in vascular elasticity, cardiovascular disease risk assessment results, and health recommendations.
[0026] Example 2 In a specific implementation process, refer to Figure 3 As shown, the vascular elasticity risk early warning method includes the following steps: S1. Use a multispectral image acquisition module to continuously acquire multispectral video images of specific areas of the human body; S2. Use the image processing module to perform fusion processing and skin color segmentation on the acquired multispectral video images, and extract the pulse wave signal; S3. The extracted pulse wave signal is subjected to multi-scale harmonic analysis through the dynamic high-order harmonic analysis module to extract high-order harmonic characteristic parameters that reflect changes in vascular elasticity. S4. Input the extracted high-order harmonic feature parameters into the risk assessment module, and combine them with user basic information and historical data to use a machine learning model to assess vascular elasticity and cardiovascular disease risk. S5. Based on the risk assessment results, issue early warning signals through the early warning and feedback module and provide personalized health advice.
[0027] It should be noted that in step S2, refer to Figure 4 As shown, skin color segmentation is achieved in the following way: In the YCbCr color space, dynamic thresholds are calculated for the Cb and Cr channels respectively. The dynamic thresholds are updated based on the exponentially weighted moving average of the grayscale values of the skin pixels in the current frame, and the threshold range is determined by combining the mean and standard deviation of historical frames. By using optical flow to detect human motion areas, the threshold range for dynamic areas is expanded by 20% to 30%, while the threshold range for static areas is reduced by 10% to 20%. Perform a logical AND operation on the binarized results of the Cb and Cr channels to generate an initial skin color mask; Morphological opening operations (kernel size 5×5) are applied to the initial skin color mask to remove small-area noise, and closing operations (kernel size 7×7) are applied to fill internal holes.
[0028] It should be noted that step S3 also includes: The continuously acquired pulse wave signals are dynamically tracked to capture the trend of changes in vascular elasticity over time. By analyzing the changes in harmonic components at different time scales, multi-scale characteristic parameters can be extracted to improve the sensitivity to changes in vascular elasticity.
[0029] It should be noted that in step S4, refer to Figure 5 As shown, the training process of a machine learning model includes: Collect datasets of known vascular elasticity states, including data obtained through traditional contact measurement methods; The dataset is divided into a training set and a test set. The model is trained using the training set and validated and optimized using the test set. The training process is iterated continuously until the model achieves satisfactory evaluation accuracy.
[0030] According to embodiments of the present invention, a computing device that can be used to implement the above method includes a processor and a memory; The processor can be a multi-core processor or include multiple processors. In some embodiments, the processor may include a general-purpose main processor and one or more special coprocessors, such as a graphics processing unit (GPU), a digital signal processor (DSP), etc. In some embodiments, the processor may be implemented using custom circuitry, such as an application-specific integrated circuit (ASIC) or a field-programmable gate array (FPGA).
[0031] Memory can include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage devices. ROM can store static data or instructions required by the processor or other modules of the computer. Permanent storage devices can be read-write storage devices. Permanent storage devices can be non-volatile storage devices that retain stored instructions and data even when the computer is powered off. In some embodiments, permanent storage devices use mass storage devices (e.g., magnetic or optical disks, flash memory) as permanent storage devices. In other embodiments, permanent storage devices can be removable storage devices (e.g., floppy disks, optical drives). System memory can be read-write storage devices or volatile read-write storage devices, such as dynamic random access memory. System memory can store some or all of the instructions and data required by the processor during operation. Furthermore, memory can include any combination of computer-readable storage media, including various types of semiconductor memory chips (DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and disks and / or optical disks can also be used. In some implementations, the memory may include removable storage devices that are readable and / or writable, such as laser discs (CDs), read-only digital versatile optical discs (e.g., DVD-ROMs, dual-layer DVD-ROMs), read-only Blu-ray discs, ultra-high density optical discs, flash memory cards (e.g., SD cards, mini SD cards, Micro-SD cards, etc.), magnetic floppy disks, etc. Computer-readable storage media do not include carrier waves and transient electronic signals transmitted wirelessly or via wired connections.
[0032] It should be understood that, unless otherwise expressly stated herein, there is no strict order restriction on the execution of the above steps, and these steps may be executed in other orders. Moreover, at least some steps in the processes involved in the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but may be executed at different times. The execution order of these steps or stages is not necessarily sequential, but may be performed alternately or in turn with other steps or at least some of the steps or stages in other steps.
[0033] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or basic characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects. The scope of the invention is defined by the appended claims rather than the foregoing description. Therefore, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention, and no reference numerals in the claims should be construed as limiting the scope of the claims.
[0034] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment includes only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.
Claims
1. A vascular elasticity risk early warning system, characterized in that, include: The multispectral image acquisition module is used to continuously acquire multispectral video images of specific areas of the human body; The image processing module is used to fuse the acquired multispectral video images and extract the pulse wave signal; The dynamic higher harmonic analysis module is used to perform multi-scale harmonic analysis on the extracted pulse wave signal and extract higher harmonic characteristic parameters that reflect changes in vascular elasticity. The risk assessment module, based on extracted high-order harmonic feature parameters, combined with user basic information and historical data, uses machine learning algorithms to assess vascular elasticity and cardiovascular disease risk. The early warning and feedback module issues warning signals based on risk assessment results and provides personalized health advice.
2. The vascular elasticity risk early warning system according to claim 1, characterized in that, The multispectral image acquisition module includes a multispectral LED light source and a high-resolution CMOS sensor, which can simultaneously emit light of multiple wavelengths such as red, green, blue and near-infrared light, and acquire reflected light to form multispectral video images.
3. The vascular elasticity risk early warning system according to claim 1, characterized in that, The image processing module includes: The multispectral image fusion submodule uses a deep learning-based image fusion algorithm to fuse image information from different wavelengths to generate a high-quality pulse wave signal source image. The skin color segmentation submodule applies a dynamic threshold segmentation algorithm and combines human motion information to achieve accurate skin color region segmentation. The morphological optimization submodule performs opening and closing operations on the segmented skin-colored regions to remove noise and fill holes. The skin color segmentation submodule includes: The dynamic threshold calculation unit dynamically updates the grayscale threshold of skin color pixels through exponential weighted moving average or moving average method, and determines the threshold range based on the statistical distribution of historical frames. The motion adaptive adjustment unit detects the human motion region using optical flow or inter-frame difference methods, relaxes the threshold range for dynamic regions, and tightens the threshold range for static regions. The multi-channel fusion unit combines the dynamic thresholds of the chroma and luminance channels in the color space and generates the final segmentation threshold through weighted fusion.
4. The vascular elasticity risk early warning system according to claim 1, characterized in that, The dynamic higher-order harmonic analysis module includes: The harmonic decomposition submodule uses Fourier transform or wavelet transform to decompose the pulse wave signal into harmonics of different frequency components. The feature parameter extraction submodule extracts feature parameters that reflect changes in vascular elasticity, such as the relative amplitude and phase change rate of higher-order harmonics. The time series construction submodule arranges the continuously collected feature parameters in chronological order to construct time series data for dynamic analysis of vascular elasticity change trends.
5. The vascular elasticity risk early warning system according to claim 1, characterized in that, The risk assessment module includes: The data preprocessing submodule normalizes the extracted higher-order harmonic characteristic parameters to eliminate the influence of individual differences and measurement conditions. The machine learning model submodule uses a pre-trained deep learning model, combined with basic user information and historical data, to assess vascular elasticity and cardiovascular disease risk. The risk level classification submodule classifies vascular elasticity into different levels based on the model output and provides the corresponding cardiovascular disease risk probability.
6. The vascular elasticity risk early warning system according to claim 1, characterized in that, The early warning and feedback module includes: The real-time early warning submodule immediately issues an early warning signal when it detects a significant decrease in vascular elasticity or an increased risk of cardiovascular disease. The personalized recommendations submodule provides users with personalized health advice and lifestyle adjustment plans based on the risk assessment results. The health report generation submodule regularly generates detailed health reports, including trends in vascular elasticity, cardiovascular disease risk assessment results, and health recommendations.
7. A method for early warning of vascular elasticity risks, applicable to the vascular elasticity risk early warning system according to any one of claims 1-6, characterized in that, The method includes the following steps: S1. Use a multispectral image acquisition module to continuously acquire multispectral video images of specific areas of the human body; S2. Use the image processing module to perform fusion processing and skin color segmentation on the acquired multispectral video images, and extract the pulse wave signal; S3. The extracted pulse wave signal is subjected to multi-scale harmonic analysis through the dynamic high-order harmonic analysis module to extract high-order harmonic characteristic parameters that reflect changes in vascular elasticity. S4. Input the extracted high-order harmonic feature parameters into the risk assessment module, and combine them with user basic information and historical data to use a machine learning model to assess vascular elasticity and cardiovascular disease risk. S5. Based on the risk assessment results, issue early warning signals through the early warning and feedback module and provide personalized health advice.
8. The vascular elasticity risk warning method according to claim 7, characterized in that, In step S2, skin color segmentation is achieved in the following way: In the YCbCr color space, dynamic thresholds are calculated for the Cb and Cr channels respectively. The dynamic thresholds are updated based on the exponentially weighted moving average of the grayscale values of the skin pixels in the current frame, and the threshold range is determined by combining the mean and standard deviation of historical frames. By using optical flow to detect human motion areas, the threshold range for dynamic areas is expanded by 20% to 30%, while the threshold range for static areas is reduced by 10% to 20%. Perform a logical AND operation on the binarized results of the Cb and Cr channels to generate an initial skin color mask; Morphological opening operations are applied to the initial skin color mask to remove small-area noise, and closing operations are applied to fill internal holes.
9. The vascular elasticity risk warning method according to claim 7, characterized in that, Step S3 also includes: The continuously acquired pulse wave signals are dynamically tracked to capture the trend of changes in vascular elasticity over time. By analyzing the changes in harmonic components at different time scales, multi-scale characteristic parameters can be extracted to improve the sensitivity to changes in vascular elasticity.
10. The vascular elasticity risk warning method according to claim 7, characterized in that, In step S4, the training process of the machine learning model includes: Collect datasets of known vascular elasticity states, including data obtained through traditional contact measurement methods; The dataset is divided into a training set and a test set. The model is trained using the training set and validated and optimized using the test set. The training process is iterated continuously until the model achieves satisfactory evaluation accuracy.