A method and device for evaluating vascular sclerosis and an electronic device

Abstract

The application discloses a kind of vascular sclerosis evaluation method, device and electronic equipment, the method includes: obtaining the Korotkoff signal of to-be-measured blood vessel;Scattering feature recognition is carried out to Korotkoff signal, and scattering feature set is obtained;Scattering feature set is divided according to preset proportion to obtain training set and test set;Training blood vessel sclerosis classification model based on training set;Test set is input into classification model to obtain the sclerosis evaluation result of to-be-measured blood vessel.The application extracts scattering feature to Korotkoff signal, scattering feature has higher stability, and the reliability of evaluation result can be improved;Classification based on scattering feature compared with prior art does not need a large number of data calculation, saves resource, and overall data processing efficiency is high and implementation method is simple;Meanwhile, by training classification model, classification efficiency and classification accuracy can be improved.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, specifically to a method, apparatus, and electronic device for assessing arteriosclerosis. Background Technology

[0002] The mortality and disability rates from cardiovascular diseases are rising year by year, seriously threatening the health of middle-aged and elderly people. Early screening for cardiovascular diseases is an important means to reduce morbidity, disability, and mortality. Studies have shown that arteriosclerosis is an initial and inducing factor for cardiovascular diseases. Therefore, accurate detection of arteriosclerosis is of great significance for the prevention and treatment of cardiovascular diseases.

[0003] Currently, pulse wave velocity is the most widely used non-invasive method for early detection of arteriosclerosis. However, it suffers from inaccurate measurement results, susceptibility to blood pressure fluctuations, and complex operation, limiting its clinical application. Korotkoff sounds are widely used in non-invasive blood pressure measurement, utilizing the first and fifth phases of the five-phase method to detect systolic and diastolic blood pressure, while the intermediate sound signals are often ignored. Studies have shown that the characteristics of Korotkoff sound signals are closely related to the degree of arteriosclerosis; for example, the higher the degree of arteriosclerosis, the lower the amplitude of the corresponding Korotkoff sound signal. Existing technologies mostly identify the degree of arteriosclerosis through pulse wave velocity or vascular imaging, which is complex to operate, requires a large amount of data processing, is inefficient, and consumes a significant amount of resources. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method for assessing arteriosclerosis, in order to solve the problems of complex and inefficient existing arteriosclerosis assessment processes.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] This invention provides a method for assessing arteriosclerosis, comprising:

[0007] Acquire Korotkoff sound signals from the blood vessel to be tested;

[0008] The Korotkoff tone signal is subjected to scattering feature identification to obtain a scattering feature set;

[0009] The scattering feature set is divided into a training set and a test set according to a preset ratio;

[0010] A blood vessel sclerosis classification model was trained based on the training set.

[0011] The test set is input into the classification model to obtain the hardening assessment result of the blood vessel to be tested.

[0012] Optionally, the step of performing scattering feature identification on the Korotkoff tone signal to obtain a scattering feature set includes:

[0013] Construct a wavelet scattering neural network model;

[0014] The Korotkoff tone signal is input into the wavelet scattering neural network model to extract multi-layer scattering features, thereby obtaining a scattering feature set.

[0015] Optionally, the step of inputting the Korotkoff tone signal into the wavelet scattering neural network model to extract multi-layer scattering features and obtain a scattering feature set includes:

[0016] Use the Korotkoff tone signal as the input signal for the current layer;

[0017] The input signal is subjected to continuous wavelet transform to obtain the scaling coefficients of the current layer;

[0018] The scattering characteristics of the current layer are obtained by processing the scale coefficients;

[0019] The scaling factor of the current layer is used as the input signal of the next layer, and the step of performing continuous wavelet transform on the input signal is returned to calculate the scattering feature of the next layer until the preset target number of layers is reached.

[0020] The scattering features of each layer are integrated to obtain a scattering feature set.

[0021] Optionally, processing the scale coefficients to obtain the scattering features of the current layer includes:

[0022] Perform a modulo operation on the scale coefficients to obtain the modulo operation result;

[0023] The scattering characteristics of the current layer are obtained by filtering the results of the modulus operation.

[0024] Optionally, constructing the wavelet scattering neural network model includes:

[0025] Establish a multi-resolution wavelet function ψ related to the Korotkoff tone signal function x(u). λ (u);

[0026] Based on the multi-resolution wavelet function, a wavelet scattering model function is constructed: s λ x=|x*ψ λ |*φ J , where |x*ψ λ | represents the modulus formula for wavelet transform, φ j The scaling function is the one used for the largest scale.

[0027] Optionally, the method further includes:

[0028] The Korotkoff sound signal is identified, and the first Korotkoff sound signal within a preset feature signal range is selected;

[0029] The second Korotkow tone signal is obtained by filtering out the baseline drift signal from the Korotkow tone signal;

[0030] The second Korotkoff tone signal is subjected to wavelet transform to remove high-frequency noise signals.

[0031] Optionally, training the arteriosclerosis classification model based on the training set includes:

[0032] The scattering features in the training set are classified according to a preset feature range to obtain multiple types of scattering feature sets;

[0033] A classification model for arteriosclerosis was trained using multiple types of scattering feature sets.

[0034] This invention also provides a device for assessing arteriosclerosis, comprising:

[0035] The acquisition module is used to acquire Korotkoff sound signals from the blood vessel to be tested;

[0036] The feature extraction module is used to identify the scattering features of the Korotkoff sound signal to obtain a scattering feature set;

[0037] The partitioning module is used to partition the scattering feature set into a training set and a test set according to a preset ratio;

[0038] The model building module is used to train the arteriosclerosis classification model based on the training set;

[0039] The evaluation module is used to input the test set into the classification model to obtain the hardening evaluation result of the blood vessel to be tested.

[0040] This invention also provides an electronic device, comprising:

[0041] The system includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes these computer instructions to perform the arteriosclerosis assessment method provided in this embodiment of the invention.

[0042] This invention also provides a computer-readable storage medium storing computer instructions for causing a computer to execute the arteriosclerosis assessment method provided in this invention.

[0043] The technical solution of this invention has the following advantages:

[0044] This invention provides a method for assessing arteriosclerosis. The method involves acquiring Korotkoff sound signals from the blood vessel to be tested; identifying the scattering features of the Korotkoff sound signals to obtain a scattering feature set; dividing the scattering feature set into a training set and a test set according to a preset ratio; training an arteriosclerosis classification model based on the training set; and inputting the test set into the classification model to obtain the arteriosclerosis assessment result for the blood vessel to be tested. This invention extracts scattering features from the Korotkoff sound signals, and the high stability of these features improves the reliability of the assessment results. Classification based on scattering features, compared to existing technologies, does not require extensive data computation, saving resources and resulting in high overall data processing efficiency and a simple implementation method. Furthermore, training the classification model improves both classification efficiency and accuracy. Attached Figure Description

[0045] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0046] Figure 1 This is a flowchart of the arteriosclerosis assessment method in an embodiment of the present invention;

[0047] Figure 2 This is a flowchart illustrating the scattering feature identification of Korotkoff tone signals according to an embodiment of the present invention;

[0048] Figure 3 This is a flowchart illustrating the extraction of multilayer scattering features according to an embodiment of the present invention;

[0049] Figure 4 A flowchart illustrating the scattering characteristics obtained according to an embodiment of the present invention;

[0050] Figure 5 This is a flowchart illustrating the construction of a wavelet scattering neural network model according to an embodiment of the present invention;

[0051] Figure 6 This is a schematic diagram showing the comparison before and after preprocessing of Korotkoff tone signals according to an embodiment of the present invention;

[0052] Figure 7 This is a flowchart illustrating the training of a blood vessel sclerosis classification model according to an embodiment of the present invention;

[0053] Figure 8 This is a schematic diagram of the arteriosclerosis assessment device in an embodiment of the present invention;

[0054] Figure 9 This is a schematic diagram of the structure of an electronic device in an embodiment of the present invention. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] According to an embodiment of the present invention, an embodiment of a method for evaluating arteriosclerosis is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0057] This embodiment provides a method for assessing arteriosclerosis, which can be used for early screening of arteriosclerosis, such as... Figure 1 As shown, this method for assessing arteriosclerosis includes the following steps:

[0058] Step S1: Acquire Korotkoff sound signals from the blood vessel to be tested. Specifically, the Korotkoff sound signal is obtained by applying pressure to the blood vessel and then depressurizing it, collecting the friction and impact sound signals that are synchronized with the pulse after the blood flow reopens the blood vessel.

[0059] Step S2: Identify the scattering features of the Korotkoff sound signal to obtain a scattering feature set. Specifically, the wavelet scattering neural network model identifies and extracts the scattering features of the Korotkoff sound signal. Compared to other deep neural networks (such as CNN) and existing vascular image processing and recognition processes, it does not require large datasets and extensive computing resources for training and evaluation, saving resources while improving efficiency and accuracy. Furthermore, the scattering features extracted by the wavelet scattering neural network have high stability, which can improve the reliability of the evaluation results.

[0060] Step S3: Divide the scattering feature set into a training set and a test set according to a preset ratio. Specifically, the scattering feature set of the Korotkoff tone signal is randomly divided into a training set and a test set according to a preset ratio. During the training process, the convergence curve tends to stabilize after multiple iterations.

[0061] Step S4: Train the arteriosclerosis classification model based on the training set. Specifically, by training the arteriosclerosis classification model, the scattering features are classified more accurately. Through the classification model, the condition of arteriosclerosis can be quickly and accurately determined.

[0062] Step S5: Input the test set into the classification model to obtain the hardening evaluation results of the blood vessels to be tested.

[0063] Through steps S1 to S5 above, the arteriosclerosis assessment method provided by this embodiment of the invention extracts scattering features from Korotkoff sound signals. The scattering features have high stability, which can improve the reliability of the assessment results. Classification based on scattering features saves resources compared to existing technologies by not requiring a large amount of data computation. The overall data processing efficiency is high and the implementation method is simple. At the same time, by training the classification model, the classification efficiency and classification accuracy can be improved.

[0064] Specifically, in one embodiment, step S2 described above is as follows: Figure 2 As shown, the specific steps include the following:

[0065] Step S21: Construct a wavelet scattering neural network model.

[0066] Step S22: Input the Korotkoff tone signal into the wavelet scattering neural network model to extract multi-layer scattering features and obtain the scattering feature set.

[0067] Specifically, by establishing a wavelet scattering neural network model to process Korotkoff tone signals, the amount of data processing is greatly reduced, and the extracted scattering features have higher stability, thereby improving the reliability of subsequent classification based on scattering features.

[0068] Specifically, in one embodiment, step S21 described above is as follows: Figure 3 As shown, the specific steps include the following:

[0069] Step S211: Establish a multi-resolution wavelet function ψ related to the Korotkoff tone signal function x(u). λ (u). Specifically, the resolution wavelet function ψ λ (u) is obtained from a two-dimensional wavelet through a rotation and scaling filter function ψ, i.e.: s J [λ1]x, where λ=2 j γ∈^=G×z, j∈z, γ∈G. Here, j determines ψ. λ The scale of (u), γ determines ψ λ The direction of (u).

[0070] Step S212: Construct wavelet scattering model function based on multi-resolution wavelet function: s λ x=|x*ψ λ |*φ J , where |x*ψ λ | represents the modulus formula for wavelet transform, φ J Let be the scaling function for the largest scale. Specifically, let the scattering path P = {λ1, λ2, λ3, ..., λ...} m}, where m is the maximum path length, the scattering characteristic coefficients {S} of the window from order 1 to m can be obtained.J (φ)f,S J (λ)f,...,S J (λ1, ..., λ) m In a scattering network, the first-order coefficients are equivalent to the size-invariant eigentransform (SIFT) coefficients.

[0071] Specifically, by establishing a wavelet scattering neural network model, the amount of data processing is greatly reduced, and the extracted scattering features have higher stability.

[0072] Specifically, in one embodiment, step S22 described above is as follows: Figure 4 As shown, the specific steps include the following:

[0073] Step S221: Use the Korotkoff tone signal as the input signal for the current layer.

[0074] Step S222: Perform a continuous wavelet transform on the input signal to obtain the scaling coefficients of the current layer. Specifically, the scaling coefficients of the current layer can be obtained by averaging the input signal using a wavelet low-pass filter.

[0075] Step S223: Process the scale coefficients to obtain the scattering characteristics of the current layer. Specifically, the scattering characteristics obtained after modulo filtering the scale coefficients have higher stability.

[0076] Step S224: Use the scaling coefficients of the current layer as the input signal for the next layer, and return to the step of performing continuous wavelet transform on the input signal to calculate the scattering features of the next layer, until the preset target number of layers is reached. Specifically, by further extracting features from the scaling coefficients and iterating through the above steps, the feature output of each layer of the wavelet scattering neural network model can be obtained.

[0077] Step S225: Integrate the scattering features of each layer to obtain a scattering feature set. Specifically, combine the scattering features output from all layers to obtain the final scattering feature set.

[0078] Specifically, wavelet scattering scaling extracts the optimal features without adjusting weights, significantly reducing computational load. Existing network models, on the other hand, require learning and feedback structure calculations, which are computationally intensive and resource-intensive.

[0079] Specifically, in one embodiment, step S223 described above is as follows: Figure 5 As shown, the specific steps include the following:

[0080] Step S2231: Perform modulo operation on the scaling coefficients to obtain the modulo operation result.

[0081] Step S2232: Filter the modulus operation result to obtain the scattering characteristics of the current layer.

[0082] Specifically, by performing modulo operations and then filtering with a wavelet low-pass filter, the extracted scattering features can be made more stable.

[0083] Specifically, in one embodiment, before step S2 described above, the following steps are further included:

[0084] Step S201: Identify the Korotkoff sound signal and select the first Korotkoff sound signal within a preset characteristic signal range. Specifically, for example... Figure 6 As shown, signals within the range of A and B are captured. By excluding signals outside the preset signal range, unnecessary data processing is reduced, and resources are saved.

[0085] Step S202: Filter out the baseline drift signal from the Korotkoff tone signal to obtain the second Korotkoff tone signal. Specifically, filtering out the baseline drift signal can eliminate baseline drift interference in the Korotkoff tone signal.

[0086] Step S203: Perform wavelet transform on the second Korotkow tone signal to remove high-frequency noise. Specifically, wavelet transform can remove high-frequency noise above the conventional frequency range of the Korotkow tone, thereby improving the signal-to-noise ratio.

[0087] Specifically, in one embodiment, step S4 described above is as follows: Figure 7 As shown, the specific steps include the following:

[0088] Step S41: Classify the scattering features in the training set according to the preset feature range to obtain multiple types of scattering feature sets.

[0089] Step S42: Train the arteriosclerosis classification model using multiple types of scattering feature sets.

[0090] Specifically, training a blood vessel sclerosis classification model can improve classification speed and accuracy, thus increasing efficiency. Long Short-Term Memory (LSTM) neural networks can also be used for classifying waveforms, including:

[0091] The input layer is used to input the Korotkoff sound scattering features;

[0092] LSTM layers are used to record the long-term and short-term characteristics of the scattering coefficients;

[0093] The fully connected layer takes as input all the outputs of the LSTM layer and is used to classify the long and short time features recorded by the LSTM.

[0094] Dropout layers, as hidden layers in classification models, randomly ignore certain hidden nodes during training and testing to reduce the number of features and prevent overfitting.

[0095] The Softmax layer outputs the classification results of Korotkoff tone signals. The network's prediction results select the class with the highest posterior probability as the discrimination result.

[0096] The output layer outputs Korotkoff sound classification labels.

[0097] This embodiment also provides a device for assessing arteriosclerosis, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0098] This embodiment provides a device for assessing arteriosclerosis, such as... Figure 8 As shown, it includes:

[0099] The acquisition module 101 is used to acquire the Korotkoff sound signal of the blood vessel to be tested. For details, please refer to the relevant description of step S1 in the above method embodiment, which will not be repeated here.

[0100] The feature extraction module 102 is used to identify the scattering features of Korotkoff sound signals to obtain a scattering feature set. For details, please refer to the relevant description of step S2 in the above method embodiment, which will not be repeated here.

[0101] The partitioning module 103 is used to partition the scattering feature set into a training set and a test set according to a preset ratio. For details, please refer to the relevant description of step S3 in the above method embodiment, which will not be repeated here.

[0102] The model building module 104 is used to train the arteriosclerosis classification model based on the training set. For details, please refer to the relevant description of step S4 in the above method embodiment, which will not be repeated here.

[0103] The evaluation module 105 is used to input the test set into the classification model to obtain the hardening evaluation result of the blood vessel to be tested. For details, please refer to the relevant description of step S5 in the above method embodiment, which will not be repeated here.

[0104] In this embodiment, the arteriosclerosis assessment device is presented in the form of a functional unit. Here, a unit refers to an ASIC circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0105] Further functional descriptions of the above modules are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0106] According to embodiments of the present invention, an electronic device is also provided, such as... Figure 9 As shown, the electronic device may include a processor 901 and a memory 902, wherein the processor 901 and the memory 902 may be connected via a bus or other means. Figure 9 Taking the example of a connection between China and Israel via a bus.

[0107] Processor 901 can be a Central Processing Unit (CPU). Processor 901 can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations of the above types of chips.

[0108] The memory 902, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the methods in the method embodiments of the present invention. The processor 901 executes various functional applications and data processing of the processor by running the non-transitory software programs, instructions, and modules stored in the memory 902, thereby implementing the methods in the above method embodiments.

[0109] The memory 902 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the processor 901, etc. Furthermore, the memory 902 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 902 may optionally include memory remotely located relative to the processor 901, and these remote memories may be connected to the processor 901 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0110] One or more modules are stored in memory 902, and when executed by processor 901, they perform the methods described in the above method embodiments.

[0111] The specific details of the aforementioned electronic device can be understood by referring to the relevant descriptions and effects in the above method embodiments, and will not be repeated here.

[0112] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk drive (HDD), or solid-state drive (SSD), etc.; the storage medium can also include combinations of the above types of memory.

[0113] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and all such modifications and variations fall within the scope defined by the appended claims.

Claims

1. A method for assessing arteriosclerosis, characterized in that, include: Acquire Korotkoff sound signals from the blood vessel to be tested; The Korotkoff tone signal is subjected to scattering feature identification to obtain a scattering feature set; The scattering feature set is divided into a training set and a test set according to a preset ratio; A blood vessel sclerosis classification model was trained based on the training set. The test set is input into the classification model to obtain the hardening assessment result of the blood vessel to be tested; The step of performing scattering feature identification on the Korotkoff tone signal to obtain a scattering feature set includes: Construct a wavelet scattering neural network model; The Korotkoff tone signal is input into the wavelet scattering neural network model to extract multi-layer scattering features, thereby obtaining a scattering feature set; The construction of the wavelet scattering neural network model includes: Establish a multi-resolution wavelet function related to the Korotkoff tone signal function x(u) The multi-resolution wavelet function Wavelets in two dimensions are filtered by rotation and scaling functions To obtain, that is: ,in ,in, Determining multi-resolution wavelet functions The scale, Determining multi-resolution wavelet functions The direction; Construct a wavelet scattering model function based on the multi-resolution wavelet function: ,in, The modulus formula for wavelet transform is... The scaling function is the one used for the largest scale.

2. The method for assessing arteriosclerosis according to claim 1, characterized in that, The process involves inputting the Korotkoff tone signal into the wavelet scattering neural network model to extract multi-layer scattering features, resulting in a scattering feature set, including: Use the Korotkoff tone signal as the input signal for the current layer; The input signal is subjected to continuous wavelet transform to obtain the scaling coefficients of the current layer; The scattering characteristics of the current layer are obtained by processing the scale coefficients; The scaling factor of the current layer is used as the input signal of the next layer, and the step of performing continuous wavelet transform on the input signal is returned to calculate the scattering feature of the next layer until the preset target number of layers is reached. The scattering features of each layer are integrated to obtain a scattering feature set.

3. The method for evaluating arteriosclerosis according to claim 2, characterized in that, The process of processing the scale coefficients to obtain the scattering features of the current layer includes: Perform a modulo operation on the scale coefficients to obtain the modulo operation result; The scattering characteristics of the current layer are obtained by filtering the results of the modulus operation.

4. The method for assessing arteriosclerosis according to claim 1, characterized in that, Before performing scattering feature identification on the Korotkoff tone signal to obtain a scattering feature set, the method further includes: The Korotkoff sound signal is identified, and the first Korotkoff sound signal within a preset feature signal range is selected; The second Korotkow tone signal is obtained by filtering out the baseline drift signal from the Korotkow tone signal; The second Korotkoff tone signal is subjected to wavelet transform to remove high-frequency noise signals.

5. The method for assessing arteriosclerosis according to claim 1, characterized in that, The training of the arteriosclerosis classification model based on the training set includes: The scattering features in the training set are classified according to a preset feature range to obtain multiple types of scattering feature sets; A classification model for arteriosclerosis was trained using multiple types of scattering feature sets.

6. A device for assessing arteriosclerosis, characterized in that, include: The acquisition module is used to acquire Korotkoff sound signals from the blood vessel to be tested; The feature extraction module is used to perform scattering feature recognition on the Korotkoff sound signal to obtain a scattering feature set; wherein, the step of performing scattering feature recognition on the Korotkoff sound signal to obtain a scattering feature set includes: A wavelet scattering neural network model is constructed; the Korotkoff tone signal is input into the wavelet scattering neural network model to extract multi-layer scattering features, thereby obtaining a scattering feature set; wherein, the construction of the wavelet scattering neural network model includes: establishing a multi-resolution wavelet function related to the Korotkoff tone signal function x(u). The multi-resolution wavelet function Wavelets in two dimensions are filtered by rotation and scaling functions To obtain, that is: ,in ,in, Determining multi-resolution wavelet functions The scale, Determining multi-resolution wavelet functions The direction; constructing a wavelet scattering model function based on the multi-resolution wavelet function: ,in, This is the modulus formula for wavelet transform. The scaling function is for the largest scale. The partitioning module is used to partition the scattering feature set into a training set and a test set according to a preset ratio; The model building module is used to train the arteriosclerosis classification model based on the training set; The evaluation module is used to input the test set into the classification model to obtain the hardening evaluation result of the blood vessel to be tested.

7. An electronic device, characterized in that, include: A memory and a processor are communicatively connected, the memory stores computer instructions, and the processor executes the computer instructions to perform the arteriosclerosis assessment method according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a computer to perform the arteriosclerosis assessment method according to any one of claims 1-5.

Images