CN111481187A

CN111481187A - Artificial intelligence arrhythmia detection method based on arterial pressure wave characteristics

Info

Publication number: CN111481187A
Application number: CN202010460777.1A
Authority: CN
Inventors: 张闻涛; 郑颖; 李祥
Original assignee: Tongxintang Technology Shenzhen Co ltd
Current assignee: Tongxintang Health Technology Beijing Co ltd
Priority date: 2020-05-27
Filing date: 2020-05-27
Publication date: 2020-08-04
Anticipated expiration: 2040-05-27
Also published as: CN111481187B

Abstract

The application relates to a method for detecting arrhythmia based on artificial intelligence of arterial pressure wave characteristics. The method comprises the following steps: acquiring brachial artery pressure wave data of a target object; determining key points in the brachial artery pressure wave according to the brachial artery pressure wave data, and determining feature data of the brachial artery pressure wave according to the key points; wherein the characteristic data is used to characterize a fluctuation state of the brachial artery pressure wave; and detecting the heart rhythm type according to the characteristic data to obtain the heart rhythm type of the target object. By adopting the method, the heart rhythm type can be detected based on the pressure wave data of the brachial artery.

Description

Artificial intelligence arrhythmia detection method based on arterial pressure wave characteristics

Technical Field

The application relates to the technical field of arrhythmia detection, in particular to an artificial intelligence arrhythmia detection method based on arterial pressure wave characteristics.

Background

An Electrocardiogram (ECG), which records physiological changes of electrical activity generated by each cardiac cycle of a human heart, is an important basis for diagnosing cardiovascular diseases. The clinician can determine whether the patient is arrhythmia according to the electrocardiogram so as to perform treatment.

In the related art, electrocardiography generally requires a plurality of lead lines, and is therefore mainly used in hospitals. In fact, the brachial artery is the place where the heart is closest to the heart after the aorta comes out and pressure wave data are most easily acquired from the surface, the pressure wave data of the brachial artery can be measured simultaneously with the blood pressure, the blood pumping function of the heart can be reflected, and the waveform is simpler than that of an electrocardiogram.

However, methods for arrhythmia detection based on brachial artery pressure wave data are currently lacking.

Disclosure of Invention

In view of the above, it is necessary to provide a method for detecting arrhythmia based on artificial intelligence of arterial pressure wave features, which is capable of performing heart rhythm type detection based on brachial artery pressure wave data.

A method for artificial intelligence detection of cardiac arrhythmias based on arterial pressure wave characteristics, the method comprising:

acquiring brachial artery pressure wave data of a target object;

determining key points in the brachial artery pressure wave according to the brachial artery pressure wave data, and determining characteristic data of the brachial artery pressure wave according to the key points; wherein the characteristic data is used to characterize a state of fluctuation of the brachial artery pressure wave;

and detecting the heart rhythm type according to the characteristic data to obtain the heart rhythm type of the target object.

In one embodiment, the determining key points in the brachial artery pressure wave according to the brachial artery pressure wave data and determining the feature data of the brachial artery pressure wave according to the key points includes:

converting the brachial artery pressure wave data from a time domain to a frequency domain, and performing frequency selection according to a preset frequency range to obtain frequency domain data of the brachial artery pressure wave;

determining key points in the brachial artery pressure wave according to the amplitude of the frequency domain data of the brachial artery pressure wave;

and dividing a preset frequency range into a plurality of frequency intervals, and determining characteristic data corresponding to each frequency interval according to the key points.

In one embodiment, the determining the key point in the brachial artery pressure wave according to the amplitude of the frequency domain data of the brachial artery pressure wave comprises:

determining whether each frequency domain data meets a preset condition, and determining the frequency domain data meeting the preset condition as a key point;

wherein the preset conditions include: the amplitude of the frequency domain data is greater than the amplitude of the previous frequency domain data and the amplitude of the next frequency domain data, and is greater than a preset threshold.

In one embodiment, the preset threshold is a preset percentile in the frequency domain data sequence; the frequency domain data sequence is a data sequence in which a plurality of frequency domain data are arranged from small to large in amplitude.

In one embodiment, the dividing the preset frequency range into a plurality of frequency intervals and determining the feature data corresponding to each frequency interval according to the key point includes:

dividing a preset frequency range according to the length of a preset interval to obtain a plurality of frequency intervals;

calculating characteristic data corresponding to each frequency interval according to the key points in each frequency interval; the feature data includes the number of keypoints in the frequency bin, the amplitude accumulation values of the keypoints, and the frequency accumulation values of the keypoints.

In one embodiment, the brachial artery pressure wave data includes a lifting pressure data and a constant pressure data; the buck-boost data comprises systolic pressure data and diastolic pressure data; the constant voltage data and the buck-boost data are in a linear relation.

In one embodiment, the performing the heart rhythm type detection according to the characteristic data to obtain the heart rhythm type of the target object includes:

and inputting the characteristic data into a heart rhythm type detection model trained in advance to obtain the heart rhythm type of the target object output by the heart rhythm type detection model.

In one embodiment, the process of training the heart rhythm type detection model includes:

acquiring sample characteristic data of a plurality of training objects and a gold standard corresponding to each sample characteristic data; the gold standard corresponding to the sample characteristic data is used for representing the heart rhythm type corresponding to the training object; determining the heart rhythm type corresponding to the training object according to the diagnosis result of the electrocardiogram;

and training the classification model based on the sample characteristic data of the plurality of training objects and the gold standard corresponding to each sample characteristic data to obtain the heart rhythm type detection model.

In one embodiment, the acquiring sample feature data of a plurality of training subjects includes:

acquiring characteristic data of a plurality of candidate objects;

grading the feature data of each candidate object by adopting a preset grading function to obtain a grading result corresponding to each candidate object;

and selecting a training object from the plurality of candidate objects according to the scoring result corresponding to each candidate object, and taking the feature data of the training object as sample feature data.

In one embodiment, the selecting a training object from a plurality of candidate objects according to the scoring result corresponding to each candidate object includes:

sorting the plurality of candidate objects according to the order of scores from high to low;

determining k candidate objects ranked at the top as training objects; wherein k is a positive integer.

In one embodiment, the training of the classification model based on the sample feature data of the plurality of training objects and the gold standard corresponding to each sample feature data to obtain the heart rhythm type detection model includes:

obtaining the value range and the value step length of the model parameters of the classification model;

and determining the target value of the model parameter according to the value range and the value step length of the model parameter and the gold standard corresponding to the characteristic data of each sample to obtain the heart rhythm type detection model.

In one embodiment, the classifying model is a multi-classifying model, and the determining the target value of the model parameter according to the value range and the value step of the model parameter and the gold standard corresponding to each sample feature data to obtain the heart rhythm type detection model includes:

initializing the model parameters according to the value range of the model parameters; the model parameters comprise input feature quantity, error tolerance and support direction measurement;

training the initialized classification model based on the sample characteristic data of a plurality of training objects to obtain the average accuracy corresponding to the classification model;

updating the value of each model parameter according to the value step of the model parameter, and repeatedly executing the step of obtaining the average accuracy corresponding to the classification model until the value of the model parameter is updated, so as to obtain a plurality of average accuracies;

and selecting candidate values of the model parameters according to the average accuracy rates, and performing performance detection on the classification model according to the candidate values of the model parameters and the gold standard corresponding to the characteristic data of each sample to obtain a heart rhythm type detection model with the model performance meeting preset detection conditions.

In one embodiment, the training the initialized classification model based on the sample feature data corresponding to the plurality of training objects to obtain the average accuracy corresponding to the classification model includes:

dividing a plurality of sample characteristic data into 10 mutually exclusive subsets;

and (3) taking 9 mutually exclusive subsets as a training set each time, taking the rest subsets as a test set to carry out 10 times of training and testing, and taking the average value of the accuracy rates of 10 times of testing as the average accuracy rate.

In one embodiment, the performing, according to the candidate values of the model parameters, performance detection on the classification model to obtain a heart rhythm type detection model whose model performance satisfies a preset detection condition includes:

calculating the model performance of the classification model according to the candidate values of the model parameters; the model performance includes accuracy, sensitivity and specificity;

and under the conditions that the accuracy is greater than the preset accuracy, the sensitivity is greater than the preset sensitivity and the specificity is greater than the preset specificity, determining that the model performance meets the preset detection condition, and determining the candidate value of the model parameter as the target value of the model parameter when the model performance meets the preset detection condition to obtain the heart rhythm type detection model.

In one embodiment, before the feature data is input into the pre-trained heart rhythm type detection model to obtain the heart rhythm type of the target object output by the heart rhythm type detection model, the method further includes:

storing the heart rhythm type detection model and the model parameters as a model file;

and running a web server, loading the model file and generating a calling interface of the heart rhythm type detection model.

An apparatus for artificial intelligence detection of cardiac arrhythmias based on arterial pressure wave characteristics, the apparatus comprising:

the pressure wave data acquisition module is used for acquiring brachial artery pressure wave data of the target object;

the characteristic data determining module is used for determining key points in the brachial artery pressure wave according to the brachial artery pressure wave data and determining characteristic data of the brachial artery pressure wave according to the key points; wherein the characteristic data is used to characterize a state of fluctuation of the brachial artery pressure wave;

and the heart rhythm type detection module is used for detecting the heart rhythm type according to the characteristic data to obtain the heart rhythm type of the target object.

In one embodiment, the characteristic data determining module includes:

the data conversion sub-module is used for converting the brachial artery pressure wave data from a time domain to a frequency domain and selecting the frequency according to a preset frequency range to obtain the frequency domain data of the brachial artery pressure wave;

the key point determining submodule is used for determining key points in the brachial artery pressure wave according to the amplitude of the frequency domain data of the brachial artery pressure wave;

and the characteristic data determining submodule is used for dividing the preset frequency range into a plurality of frequency intervals and determining the characteristic data corresponding to each frequency interval according to the key points.

In one embodiment, the keypoint determination submodule is specifically configured to determine whether each frequency domain data meets a preset condition, and determine the frequency domain data meeting the preset condition as the keypoint;

In one embodiment, the characteristic data determining submodule is specifically configured to divide a preset frequency range according to a preset interval length to obtain a plurality of frequency intervals; calculating characteristic data corresponding to each frequency interval according to the key points in each frequency interval; the feature data includes the number of keypoints in the frequency bin, the amplitude accumulation values of the keypoints, and the frequency accumulation values of the keypoints.

In one embodiment, the rhythm type detection module is specifically configured to input the feature data into a heart rhythm type detection model trained in advance, so as to obtain a rhythm type of the target object output by the heart rhythm type detection model.

In one embodiment, the apparatus further comprises:

the training sample acquisition module is used for acquiring sample characteristic data of a plurality of training objects and a gold standard corresponding to each sample characteristic data; the gold standard corresponding to the sample characteristic data is used for representing the heart rhythm type corresponding to the training object; determining the heart rhythm type corresponding to the training object according to the diagnosis result of the electrocardiogram;

and the training module is used for training the classification model based on the sample characteristic data of the plurality of training objects and the gold standard corresponding to each sample characteristic data to obtain the heart rhythm type detection model.

In one embodiment, the training sample acquiring module includes:

the characteristic data acquisition sub-module is used for acquiring the characteristic data of a plurality of candidate objects;

the scoring submodule is used for scoring the feature data of each candidate object by adopting a preset scoring function to obtain a scoring result corresponding to each candidate object;

and the training object determining submodule is used for selecting a training object from the candidate objects according to the scoring result corresponding to each candidate object, and taking the feature data of the training object as sample feature data.

In one embodiment, the training object determining sub-module is specifically configured to rank the plurality of candidate objects in order of scores from high to low; determining k candidate objects ranked at the top as training objects; wherein k is a positive integer.

In one embodiment, the training module includes:

the value obtaining submodule is used for obtaining the value range and the value step length of the model parameters of the classification model;

and the training submodule is used for determining the target value of the model parameter according to the value range and the value step length of the model parameter and the gold standard corresponding to the characteristic data of each sample to obtain the heart rhythm type detection model.

In one embodiment, the classification model is a multi-classification model, and the training submodule is specifically configured to initialize the model parameters according to a value range of the model parameters; the model parameters comprise input feature quantity, error tolerance and support direction measurement; training the initialized classification model based on the sample characteristic data of a plurality of training objects to obtain the average accuracy corresponding to the classification model; updating the value of each model parameter according to the value step of the model parameter, and repeatedly executing the step of obtaining the average accuracy corresponding to the classification model until the value of the model parameter is updated, so as to obtain a plurality of average accuracies; and selecting candidate values of the model parameters according to the average accuracy rates, and performing performance detection on the classification model according to the candidate values of the model parameters and the gold standard corresponding to the characteristic data of each sample to obtain a heart rhythm type detection model with the model performance meeting preset detection conditions.

In one embodiment, the training submodule is specifically configured to divide the plurality of sample feature data into 10 mutually exclusive subsets; and (3) taking 9 mutually exclusive subsets as a training set each time, taking the rest subsets as a test set to carry out 10 times of training and testing, and taking the average value of the accuracy rates of 10 times of testing as the average accuracy rate.

In one embodiment, the training submodule is specifically configured to calculate a model performance of the classification model according to the candidate values of the model parameters; the model performance includes accuracy, sensitivity and specificity; and under the conditions that the accuracy is greater than the preset accuracy, the sensitivity is greater than the preset sensitivity and the specificity is greater than the preset specificity, determining that the model performance meets the preset detection condition, and determining the candidate value of the model parameter as the target value of the model parameter when the model performance meets the preset detection condition to obtain the heart rhythm type detection model.

In one embodiment, the apparatus further comprises:

the model file storage module is used for storing the heart rhythm type detection model and the model parameters as a model file;

and the calling interface generation module is used for operating the web server, loading the model file and generating a calling interface of the heart rhythm type detection model.

An electronic device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:

acquiring brachial artery pressure wave data of a target object;

A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:

acquiring brachial artery pressure wave data of a target object;

In the method for detecting arrhythmia based on artificial intelligence of arterial pressure wave characteristics, a terminal obtains brachial artery pressure wave data of a target object; determining key points in the brachial artery pressure wave according to the brachial artery pressure wave data, and determining characteristic data of the brachial artery pressure wave according to the key points; and detecting the heart rhythm type according to the characteristic data to obtain the heart rhythm type of the target object. By the embodiments of the present disclosure, a way to detect heart rhythm type based on brachial artery pressure wave data is provided. Compared with the prior art that the heart rhythm type is detected according to the electrocardiogram, the data acquisition mode of the embodiment of the disclosure is simpler and easier to realize, and the terminal can be miniaturized and portable, so that the heart rhythm detection is more popular.

Drawings

FIG. 1 is a schematic flow diagram of a method for artificial intelligence based arrhythmia detection based on arterial pressure wave characteristics, according to one embodiment;

FIG. 2 is a schematic flow chart of the steps for determining key points in a brachial artery pressure wave and determining characteristic data of the brachial artery pressure wave based on the key points in one embodiment;

FIG. 3-1 is a schematic diagram of time domain data of buck-boost data in one embodiment;

3-2 are diagrams of frequency domain data of buck-boost data in one embodiment;

3-3 are diagrams of time domain data of constant pressure data in one embodiment;

FIGS. 3-4 are diagrams of frequency domain data for constant pressure data, in one embodiment;

FIG. 4 is a flowchart illustrating the steps of training a rhythm type detection model according to one embodiment;

FIG. 5 is a flowchart illustrating the training steps of the classification model based on the sample feature data of a plurality of training subjects and the gold standard corresponding to each sample feature data according to an embodiment;

FIG. 6 is a graph illustrating average accuracy obtained in one embodiment;

FIG. 7 is a diagram illustrating the update steps of model parameters in one embodiment;

FIG. 8 is a schematic flow chart illustrating a method for artificial intelligence based on arterial pressure wave characteristics for detecting cardiac arrhythmias in another embodiment;

FIG. 9 is a block diagram of an apparatus for artificial intelligence based on arterial pressure wave characteristics for detecting cardiac arrhythmia according to an embodiment;

FIG. 10 is a second block diagram illustrating an apparatus for artificial intelligence arrhythmia detection based on arterial pressure wave characteristics according to an embodiment;

FIG. 11 is a third block diagram of an apparatus for artificial intelligence arrhythmia detection based on arterial pressure wave characteristics according to an embodiment;

FIG. 12 is a diagram illustrating the internal architecture of an electronic device in one embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

In one embodiment, as shown in fig. 1, a method for detecting arrhythmia based on artificial intelligence of arterial pressure wave features is provided, and the embodiment of the disclosure is exemplified by applying the method to a terminal, and it is understood that the method can also be applied to a server, and can also be applied to a system comprising the terminal and the server, and is realized by interaction of the terminal and the server. In the disclosed embodiment, the method comprises the following steps:

step 101, obtaining brachial artery pressure wave data of a target object.

Wherein the brachial artery pressure wave data is data indicative of a pressure wave of the brachial artery. Optionally, the brachial artery pressure wave data comprises pressor and depressor data and constant pressure data; the pressure increasing and reducing data comprises systolic pressure data and diastolic pressure data, and the constant pressure data and the pressure increasing and reducing data are in a linear relation. For example, the constant pressure data is the sum of one-third of the systolic pressure data and two-thirds of the diastolic pressure data. Systolic pressure, also called high pressure, is the pressure in the artery rising when the heart of a person contracts, and the pressure in the artery is the highest in the middle stage of the systole, at which time the pressure of blood on the inner wall of the blood vessel is the systolic pressure. When the human heart relaxes, when the arterial blood vessel elastically retracts, the generated pressure is the diastolic pressure.

The terminal can acquire brachial artery pressure wave data of the target object through a sphygmomanometer; the brachial artery pressure wave data of the target object can also be acquired from a plurality of brachial artery pressure wave data stored by the server. The embodiments of the present disclosure do not limit this.

And 102, determining key points in the brachial artery pressure wave according to the brachial artery pressure wave data, and determining characteristic data of the brachial artery pressure wave according to the key points.

Wherein the characteristic data is used to characterize a state of fluctuation of the pressure wave of the brachial artery. The key points may be the peaks, valleys, etc. of the brachial artery pressure wave.

After the brachial artery pressure wave data are obtained, the wave crests and the wave troughs in the brachial artery pressure wave can be determined according to the brachial artery pressure wave data, and then the wave crest number, the wave trough number, the wave crest amplitude, the wave trough pattern, the wave crest frequency, the wave trough frequency and the like in the brachial artery pressure wave are determined, so that the characteristic data of the brachial artery pressure wave are obtained. The embodiment of the present disclosure does not limit the feature data.

And 103, detecting the heart rhythm type according to the characteristic data to obtain the heart rhythm type of the target object.

Wherein the types of heart rhythm include normal heart rhythm and arrhythmia; cardiac arrhythmias including atrial and ventricular arrhythmias; atrial arrhythmias may include atrial fibrillation, premature atrial contractions, atrial flutter, and the like; ventricular arrhythmias may include ventricular premature beats and the like. The embodiments of the present disclosure do not limit the heart rate type.

Since the fluctuation state of the brachial artery pressure wave is different between the normal cardiac rhythm condition and the cardiac arrhythmia condition, and there is also a difference in the fluctuation state in each cardiac arrhythmia condition. Therefore, after the characteristic data of the brachial artery pressure wave is obtained, namely the fluctuation state of the brachial artery pressure wave is obtained, the heart rhythm type can be determined according to the fluctuation state of the brachial artery pressure wave.

In one embodiment, as shown in fig. 2, involves the steps of determining key points in a brachial artery pressure wave from the brachial artery pressure wave data, and determining feature data of the brachial artery pressure wave from the key points. On the basis of the above embodiment, the method may include:

step 201, the brachial artery pressure wave data is converted from the time domain to the frequency domain, and frequency selection is performed according to a preset frequency range to obtain the frequency domain data of the brachial artery pressure wave.

Wherein, the preset frequency range is the effective range of the frequency domain data of the brachial artery pressure wave. In practical applications, the predetermined frequency range may be 0-600 Hz.

The collected brachial artery pressure wave data is time domain data, the brachial artery pressure wave data can be converted from time domain to frequency domain by adopting discrete Fourier transform, and then frequency selection is carried out according to a preset frequency range to obtain frequency domain data of the brachial artery pressure wave. The time domain data of the voltage increasing and decreasing data is shown in fig. 3-1, the frequency domain data of the voltage increasing and decreasing data is shown in fig. 3-2, the time domain data of the constant voltage data is shown in fig. 3-3, and the frequency domain data of the constant voltage data is shown in fig. 3-4.

Step 202, determining a key point in the brachial artery pressure wave according to the amplitude of the frequency domain data of the brachial artery pressure wave.

The wave crest of the brachial artery pressure wave is used as a key point in the brachial artery pressure wave, and the wave crest of the brachial artery pressure wave is determined according to the amplitude of the frequency domain data of the brachial artery pressure wave, so that the key point in the brachial artery pressure wave can be determined.

In one embodiment, the step of determining key points in the brachial artery pressure wave may comprise: and determining whether each frequency domain data meets a preset condition, and determining the frequency domain data meeting the preset condition as a key point. Wherein the preset conditions include: the amplitude of the frequency domain data is greater than the amplitude of the previous frequency domain data and the amplitude of the next frequency domain data, and is greater than a preset threshold.

For example, the preset threshold is h, Vi is the amplitude of the ith frequency domain data, Vi-1 is the amplitude of the previous frequency domain data, and Vi +1 is the amplitude of the next frequency domain data. When Vi > Vi-1, Vi > Vi +1, and Vi > h, the data point is determined as a keypoint and labeled 1; otherwise, the data point is not a keypoint and is labeled 0.

The percentile is a statistical term, and if a group of data is sorted from small to large and the corresponding cumulative percentile is calculated, the value of the data corresponding to a certain percentile is called the percentile of the percentile. For example, if the preset percentile is the 95 th percentile, the amplitude of the frequency domain data corresponding to the 95 th percentile is the preset threshold value in the frequency domain data sequence obtained by arranging the plurality of frequency domain data from small to large according to the amplitude. The preset threshold value can also be selected according to experience, and the preset threshold value is not limited in the embodiment of the disclosure.

Understandably, the wave crest in the brachial artery pressure wave can be accurately found out according to the key point determined by the amplitude of the frequency domain data, and then the fluctuation state of the brachial artery pressure wave can be accurately determined.

Step 203, dividing the preset frequency range into a plurality of frequency intervals, and determining the characteristic data corresponding to each frequency interval according to the key points.

Dividing a preset frequency range according to the length of a preset interval to obtain a plurality of frequency intervals; and calculating the characteristic data corresponding to each frequency interval according to the key points in each frequency interval to obtain a plurality of characteristic data.

Wherein, the preset interval length may be a fixed length value, such as 50Hz, 75Hz, 100 Hz; or may be a variable length value set empirically.

The feature data includes the number of keypoints in the frequency bin, the amplitude accumulation values of the keypoints, and the frequency accumulation values of the keypoints. Namely, for each frequency interval, determining the number of key points in the frequency interval; summing the amplitudes of the plurality of key points to obtain an amplitude accumulated value of the key points; and summing the frequencies of the plurality of key points to obtain a frequency accumulated value of the key points. And taking the number of the key points, the amplitude accumulated value of the key points and the frequency accumulated value of the key points as the characteristic data of the frequency interval.

It can be understood that the more frequency intervals are divided, the more characteristic data are obtained, the more detailed the fluctuation state of the brachial artery pressure wave is described, and the more accurate the subsequent heart rhythm type detection is carried out.

In the process of determining the key point in the brachial artery pressure wave according to the brachial artery pressure wave data, and determining the characteristic data of the brachial artery pressure wave according to the key point, the terminal converts the brachial artery pressure wave data from the time domain to the frequency domain, and selects the frequency domain data of the brachial artery pressure wave in the preset frequency range; then, according to the amplitude of the frequency domain data of the brachial artery pressure wave, finding out a key point in the brachial artery pressure wave; and determining the fluctuation state of the brachial artery pressure wave in each frequency interval in each preset frequency range according to the key points to obtain a plurality of characteristic data. Through the embodiment of the disclosure, the characteristic data representing the fluctuation state of the brachial artery pressure wave is obtained, and the heart rhythm type detection can be accurately performed according to the characteristic data.

In one embodiment, performing the heart rhythm type detection according to the characteristic data to obtain the heart rhythm type of the target object may include: and inputting the characteristic data into a heart rhythm type detection model trained in advance to obtain the heart rhythm type of the target object output by the heart rhythm type detection model.

The heart rhythm type detection model may be a Support Vector Machine (SVM) multi-classification model, or may be other multi-classification models. The embodiments of the present disclosure do not limit this.

Pre-training a heart rhythm type detection model, inputting the characteristic data into the heart rhythm type detection model after obtaining the characteristic data corresponding to the target object, and carrying out classification processing and outputting a classification result by the heart rhythm type detection model according to the characteristic data. The classification result is the heart rhythm type of the target object.

In practical application, a heart rhythm type detection model can be deployed at a terminal to detect the heart rhythm type; a rhythm type detection model may also be deployed at the server side, with the terminal performing rhythm type detection by providing a call interface.

In one embodiment, the heart rhythm type detection model and model parameters are saved as a model file; and running a web server, loading the model file and generating a calling interface of the heart rhythm type detection model. The call interface may be a representational State transfer (rest), or other call interface.

Taking representational state transfer (REST) as an example, a scimit-learn framework is used for training a heart rhythm type detection model, and a joblib. The flash server is then run to invoke the rhythm type detection model and model parameters by loading the file with jobb. After that, the REST API is opened, and the heart rhythm type detection service can be provided to the terminal.

In the process of detecting the heart rhythm type according to the characteristic data to obtain the heart rhythm type of the target object, a heart rhythm type detection model is trained in advance, and after the characteristic data is obtained, the characteristic data is input into the heart rhythm type detection model to obtain the heart rhythm type of the target object output by the heart rhythm type detection model. In the embodiment of the disclosure, the heart rhythm type detection is performed through the heart rhythm type detection model trained in advance, so that important auxiliary information is provided for the diagnosis of a clinician; and is more intelligent than determining the heart rhythm type from an electrocardiogram.

In one embodiment, as shown in fig. 4, the process of training the heart rhythm type detection model may include:

step 301, sample feature data of a plurality of training objects and a gold standard corresponding to each sample feature data are obtained.

The golden standard corresponding to the sample characteristic data is used for representing the heart rhythm type corresponding to the training object; the heart rhythm type corresponding to the training object is determined according to the diagnosis result of the electrocardiogram.

Acquiring characteristic data of a plurality of candidate objects; grading the feature data of each candidate object by adopting a preset grading function to obtain a grading result corresponding to each candidate object; and selecting a training object from the plurality of candidate objects according to the scoring result corresponding to each candidate object, and taking the feature data of the training object as sample feature data.

In one embodiment, obtaining feature data of a plurality of candidate objects may include: the brachial artery pressure wave data of the candidate object is acquired, and the feature data of the candidate object is determined according to the method from the step 301 to the step 302.

Acquiring the brachial artery pressure wave data of the candidate object, acquiring electrocardiogram data of the candidate object, and determining the heart rhythm type corresponding to the candidate object according to the electrocardiogram data. After the candidate object is determined as the training object, the heart rhythm type corresponding to the candidate object is determined as the gold standard corresponding to the sample characteristic data of the training object, namely the heart rhythm type corresponding to the training object is determined according to the diagnosis result of the electrocardiogram.

In one embodiment, the preset scoring function is a Fisher scoring function, as shown in formula (1):

wherein p represents the number of categories, μ represents the mean of the characteristic data f, n_iIndicates the number of i (i: 1, 2, … … p) th class samples, μⁱAnd σⁱThe mean and variance of the feature data f in the ith type sample are shown.

It is understood that the higher the score of Fisher, which is a preset scoring function, the better the characterization feature data.

In one embodiment, selecting a training object from a plurality of candidate objects according to a scoring result corresponding to each candidate object includes: sorting the plurality of candidate objects according to the order of scores from high to low; determining k candidate objects ranked at the top as training objects; wherein k is a positive integer.

For example, k is 10, 15 candidates are ranked in order of scores from high to low, and then 10 candidates ranked at the top are taken as training objects.

Step 302, training a classification model based on the sample characteristic data of a plurality of training objects and the gold standard corresponding to each sample characteristic data to obtain a heart rhythm type detection model.

The classification model may be a Support Vector Machine (SVM) multi-classification model, or may be other classification models. The input feature quantity of the SVM multi-classification model is k, and the kernel Function adopts a Gaussian base Function (RBF).

Obtaining the value range and the value step length of the model parameters of the classification model; and determining the target value of the model parameter according to the value range and the value step length of the model parameter and the gold standard corresponding to the characteristic data of each sample to obtain the heart rhythm type detection model.

The value range of the model parameters limits the minimum value and the maximum value of the model parameters, and the value step length of the model parameters determines the length of the value of each time of updating the model parameters. The value range and the value step of the model parameter can be preset according to the empirical value.

The heart rhythm type detection method has the advantages that sample characteristic data of a plurality of training objects and the gold standard corresponding to each sample characteristic data are obtained, the classification model is trained on the basis of the sample characteristic data of the plurality of training objects and the gold standard corresponding to each sample characteristic data to obtain the heart rhythm type detection model, a more intelligent detection method is provided for heart rhythm type detection, and the heart rhythm type detection can be popularized more.

In one embodiment, as shown in FIG. 5, step 302 may comprise:

step 401, initializing the model parameters according to the value range of the model parameters.

The classification model is a multi-classification model, and the model parameters comprise input feature quantity, error tolerance and support direction measurement. The larger the value of the error tolerance, the less tolerable the error, the more overfitting and conversely the less overfitting. The support vector measurement is a parameter of a radial basis function, implicitly determines the distribution of data after being mapped to a new feature space, and the larger the support vector degree is, the fewer the support vectors are, and the smaller the support vector degree is, the more the support vectors are.

Presetting the value ranges of the model parameter input characteristic quantity k, the error tolerance c and the support direction measurement g as [ k _ begin, k _ end ], [ c _ begin, c _ end ] and [ g _ begin, g _ end ], and initializing the SVM multi-classification model by adopting the k _ begin, the c _ begin and the g _ begin.

Step 402, training the initialized classification model based on the sample characteristic data of a plurality of training objects to obtain the average accuracy corresponding to the classification model.

And obtaining the average accuracy corresponding to the classification model by adopting a 10-fold Cross Validation (Cross Validation) mode. As shown in fig. 6, the plurality of sample feature data are divided into 10 mutually exclusive subsets; and (3) training and testing 10 times by using 9 mutually exclusive subsets as a training set and using the rest subsets as a test set each time, taking the average value of the accuracy rates of the 10 times of tests as the average accuracy rate, and storing the average accuracy rate into a data set P.

Wherein, the calculation formula of the average accuracy ac is as the formula (2):

wherein, ac_iThe accuracy of the ith test is represented, namely the ratio of the number of sample characteristic data to be paired to the total number of all sample characteristic data.

And 403, updating the value of each model parameter according to the value step of the model parameter, and repeatedly executing the step of obtaining the average accuracy corresponding to the classification model until the value of the model parameter is updated, so as to obtain a plurality of average accuracies.

And setting the value steps of the three model parameters as k _ step, c _ step and g _ step respectively. And after obtaining the average accuracy according to the initialized k, c and g, updating the value of the model parameter by adopting a Grid Search (Grid Search) mode. As shown in fig. 7, it is determined whether g is equal to g _ end, and if not, g is updated to g + g _ step according to the value step. Then, step 502 is repeatedly executed to obtain the average accuracy corresponding to the classification model, and the average accuracy is stored in the data set P.

If g equals g _ end, it is determined if c equals c _ end. If c is not equal to c _ end, c is updated to c + c _ step according to the value step, then g is determined to be g _ begin and the step 502 is repeatedly executed to obtain the average accuracy, and the average accuracy is stored in the data set P.

If c is equal to c _ end, it is determined whether k is equal to k _ end. And if k is not equal to k _ end, updating k to k + k _ step according to the value step, then repeatedly executing the steps of setting g to g _ begin, setting c to c _ begin and obtaining the average accuracy in the step 502, and storing the average accuracy in the data set P.

After the values of the model parameters are updated, a plurality of average accuracy rates can be obtained.

And step 404, selecting candidate values of the model parameters according to the average accuracy rates, and performing performance detection on the classification model according to the candidate values of the model parameters and the gold standard corresponding to each sample characteristic data to obtain a heart rhythm type detection model with the model performance meeting preset detection conditions.

And storing a plurality of average accuracy rates from the data set P, finding out the maximum value of the average accuracy rates, and determining the values of the model parameters k, c and g corresponding to the maximum value of the average accuracy rates as candidate values of the model parameters. Then, calculating the model performance of the classification model according to the candidate values of the model parameters and the gold standard corresponding to each sample characteristic data; the model performance includes accuracy, sensitivity and specificity, among others. And under the conditions that the accuracy is greater than the preset accuracy, the sensitivity is greater than the preset sensitivity and the specificity is greater than the preset specificity, determining that the model performance meets the preset detection condition, and determining the candidate value of the model parameter as the target value of the model parameter when the model performance meets the preset detection condition to obtain the heart rhythm type detection model.

Understandably, the high sensitivity indicates that the missed diagnosis is not easy, and can be used for screening, initial detection and exclusion diagnosis; the high specificity indicates that misdiagnosis is not easy, and can be used for accurate diagnosis.

In one embodiment, calculating the model performance of the classification model according to the candidate values of the model parameters and the gold standard corresponding to each sample feature data may include: adopting a binary Confusion Matrix (Confusion Matrix), assuming that the gold standard only has two types of positive examples (positive) and negative examples (negative), wherein TP (true positive) in the Confusion Matrix is a positive example and is divided into the number of samples of the positive example by a classification model; FP (false positives) is the number of samples that are counterexample but divided into positive example by the classification model; FN (false negatives) is the number of samples which are positive examples but divided into negative examples by the classification model; TN (Truenegotives) is the sample number of the counter-example and is divided into the counter-example by the classification model. Wherein, the calculation formula of the total number of samples is as formula (3):

total number of samples TP + FP + FN + TN- (3)

The calculation formula of the accuracy is formula (4):

ac＝(TP+TN)/(TP+FP+FN+TN)---------------------(4)

the calculation formula of the sensitivity is formula (5):

sn＝TP/(TP+FN)------------------------------(5)

the calculation formula of the specificity is formula (6):

sp＝TN/(FP+TN)---------------------------------(6)

in the process of training the classification model based on the sample characteristic data of the training objects and the gold standard corresponding to each sample characteristic data to obtain the heart rhythm type detection model, values of model parameters of the classification model are determined by adopting a grid searching mode and a 10-fold cross validation mode, and the performance of the model is detected by adopting a binary confusion matrix, so that the trained heart rhythm type detection model has the advantages of high accuracy, high sensitivity, high specificity and the like, and further the problems of missed diagnosis, misdiagnosis and the like of the detection result of the heart rhythm type can be avoided.

In one embodiment, as shown in fig. 8, a method for detecting arrhythmia based on artificial intelligence of arterial pressure wave features is provided, which may include, on the basis of the foregoing embodiments:

step 501, brachial artery pressure wave data of a target object is acquired.

Step 502, the brachial artery pressure wave data is converted from the time domain to the frequency domain, and frequency selection is performed according to a preset frequency range to obtain the frequency domain data of the brachial artery pressure wave.

Step 503, determining a key point in the brachial artery pressure wave according to the amplitude of the frequency domain data of the brachial artery pressure wave.

In one embodiment, whether each frequency domain data meets a preset condition is determined, and the frequency domain data meeting the preset condition is determined as a key point; wherein the preset conditions include: the amplitude of the frequency domain data is greater than the amplitude of the previous frequency domain data and the amplitude of the next frequency domain data, and is greater than a preset threshold.

Step 504, dividing the preset frequency range into a plurality of frequency intervals, and determining the characteristic data corresponding to each frequency interval according to the key points.

In one embodiment, a preset frequency range is divided according to a preset interval length to obtain a plurality of frequency intervals; calculating characteristic data corresponding to each frequency interval according to the key points in each frequency interval; the feature data includes the number of keypoints in the frequency bin, the amplitude accumulation values of the keypoints, and the frequency accumulation values of the keypoints.

And 505, inputting the characteristic data into a heart rhythm type detection model trained in advance to obtain the heart rhythm type of the target object output by the heart rhythm type detection model.

In the method for artificially and intelligently detecting arrhythmia based on arterial pressure wave characteristics, a terminal acquires brachial artery pressure wave data of a target object, performs data conversion on the brachial artery pressure wave data, and selects effective frequency domain data of the brachial artery pressure wave; then determining key points in the brachial artery pressure wave, and determining characteristic data corresponding to each frequency interval in a preset frequency range according to the key points; and finally, inputting the characteristic data into a heart rhythm type detection model trained in advance to obtain the heart rhythm type of the target object. The embodiment of the disclosure provides a method for detecting the type of the heart rhythm based on brachial artery pressure wave data, and the method can be easily used for easily acquiring the brachial artery pressure wave data, so that the method for artificially and intelligently detecting the arrhythmia based on the artery pressure wave characteristics is suitable for various scenes and is easier to popularize.

It should be understood that although the various steps in the flowcharts of fig. 1-8 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 1-8 may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed in turn or alternately with other steps or at least some of the other steps or stages.

In one embodiment, as shown in fig. 9, there is provided an apparatus for artificial intelligence detection of cardiac arrhythmia based on arterial pressure wave characteristics, comprising:

a pressure wave data acquisition module 601 for acquiring brachial artery pressure wave data of the target object;

the characteristic data determining module 602 is configured to determine a key point in a brachial artery pressure wave according to the brachial artery pressure wave data, and determine characteristic data of the brachial artery pressure wave according to the key point; wherein the characteristic data is used to characterize a state of fluctuation of the brachial artery pressure wave;

and a heart rhythm type detection module 603, configured to perform heart rhythm type detection according to the feature data to obtain a heart rhythm type of the target object.

In one embodiment, the characteristic data determining module 602 includes:

In one embodiment, the rhythm type detection module 603 is specifically configured to input the feature data into a previously trained rhythm type detection model, so as to obtain the rhythm type of the target object output by the rhythm type detection model.

In one embodiment, as shown in fig. 10, the apparatus further comprises:

a training sample obtaining module 604, configured to obtain sample feature data of multiple training objects and a gold standard corresponding to each sample feature data; the gold standard corresponding to the sample characteristic data is used for representing the heart rhythm type corresponding to the training object;

the training module 605 is configured to perform training of the classification model based on the sample feature data of the multiple training objects and the gold standard corresponding to each sample feature data, so as to obtain a heart rhythm type detection model.

In one embodiment, the training sample acquiring module 704 includes:

In one embodiment, the training module 605 includes:

In one embodiment, the classification model is a Support Vector Machine (SVM) multi-classification model, and the training submodule is specifically configured to initialize the model parameters according to the value ranges of the model parameters; the model parameters comprise input feature quantity, error tolerance and support direction measurement; training the initialized classification model based on the sample characteristic data of a plurality of training objects to obtain the average accuracy corresponding to the classification model; updating the value of each model parameter according to the value step of the model parameter, and repeatedly executing the step of obtaining the average accuracy corresponding to the classification model until the value of the model parameter is updated, so as to obtain a plurality of average accuracies; and selecting candidate values of the model parameters according to the average accuracy rates, and performing performance detection on the classification model according to the candidate values of the model parameters and the gold standard corresponding to the characteristic data of each sample to obtain a heart rhythm type detection model with the model performance meeting preset detection conditions.

In one embodiment, as shown in fig. 11, the apparatus further comprises:

a model file saving module 606, configured to save the heart rhythm type detection model and the model parameters as a model file;

and the calling interface generation module 607 is used for running the web server, loading the model file and generating a calling interface of the heart rhythm type detection model.

For specific limitations of the apparatus for detecting arrhythmia based on artificial intelligence of arterial pressure wave characteristics, reference may be made to the above limitations of the method for detecting arrhythmia based on artificial intelligence of arterial pressure wave characteristics, which are not repeated herein. The modules in the device for detecting arrhythmia based on artificial intelligence of arterial pressure wave characteristics can be wholly or partially realized by software, hardware and combination thereof. The modules can be embedded in a hardware form or independent of a processor in the electronic device, or can be stored in a memory in the electronic device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, an electronic device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 12. The electronic device comprises a processor, a memory, a communication interface, a display screen and an input device which are connected through a system bus. Wherein the processor of the electronic device is configured to provide computing and control capabilities. The memory of the electronic equipment comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the electronic device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a method of artificial intelligence detection of cardiac arrhythmias based on arterial pressure wave characteristics. The display screen of the electronic equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the electronic equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the electronic equipment, an external keyboard, a touch pad or a mouse and the like.

Those skilled in the art will appreciate that the structure shown in fig. 12 is a block diagram of only a portion of the structure relevant to the present disclosure, and does not constitute a limitation on the electronic device to which the present disclosure may be applied, and that a particular electronic device may include more or less components than those shown, or combine certain components, or have a different arrangement of components.

In one embodiment, an electronic device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:

acquiring brachial artery pressure wave data of a target object;

In one embodiment, the processor, when executing the computer program, further performs the steps of:

In one embodiment, the brachial artery pressure wave data includes a buck-boost pressure data and a constant pressure data; the buck-boost data comprises systolic pressure data and diastolic pressure data; the constant voltage data and the buck-boost data are in a linear relation.

acquiring characteristic data of a plurality of candidate objects;

In one embodiment, the classification model is a multi-classification model, and the processor executes the computer program to further implement the following steps:

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:

acquiring brachial artery pressure wave data of a target object;

In one embodiment, the computer program when executed by the processor further performs the steps of:

acquiring characteristic data of a plurality of candidate objects;

In one embodiment, the classification model is a multi-classification model, and the computer program when executed by the processor further performs the steps of:

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A method for artificial intelligence detection of cardiac arrhythmias based on arterial pressure wave characteristics, the method comprising:

acquiring brachial artery pressure wave data of a target object;

determining key points in the brachial artery pressure wave according to the brachial artery pressure wave data, and determining feature data of the brachial artery pressure wave according to the key points; wherein the characteristic data is used to characterize a fluctuation state of the brachial artery pressure wave;

2. The method of claim 1, wherein determining key points in a brachial artery pressure wave from the brachial artery pressure wave data and determining feature data of the brachial artery pressure wave from the key points comprises:

determining a key point in the brachial artery pressure wave according to the amplitude of the frequency domain data of the brachial artery pressure wave;

and dividing the preset frequency range into a plurality of frequency intervals, and determining the characteristic data corresponding to each frequency interval according to the key points.

3. The method of claim 2, wherein determining keypoints in the brachial artery pressure wave from amplitudes of frequency domain data of the brachial artery pressure wave comprises:

determining whether each frequency domain data meets a preset condition, and determining the frequency domain data meeting the preset condition as the key point;

wherein the preset conditions include: the amplitude of the frequency domain data is greater than the amplitude of the previous frequency domain data and the amplitude of the next frequency domain data and is greater than a preset threshold value.

4. The method of claim 3, wherein the predetermined threshold is a predetermined percentile of the frequency domain data sequences; the frequency domain data sequence is a data sequence obtained by arranging a plurality of frequency domain data from small amplitude to large amplitude.

5. The method according to claim 2, wherein the dividing the preset frequency range into a plurality of frequency intervals and determining feature data corresponding to each of the frequency intervals according to the key point comprises:

dividing the preset frequency range according to the length of a preset interval to obtain a plurality of frequency intervals;

calculating characteristic data corresponding to each frequency interval according to the key points in each frequency interval; the feature data includes the number of keypoints in the frequency interval, the amplitude accumulated values of the keypoints, and the frequency accumulated values of the keypoints.

6. The method of claim 1, wherein the brachial artery pressure wave data comprises supra-systolic pressure data and supra-systolic pressure data; the buck-boost data comprises systolic pressure data and diastolic pressure data; the constant voltage data and the voltage increasing and decreasing data are in a linear relation.

7. The method according to any one of claims 1-6, wherein said performing a heart rhythm type detection based on said characteristic data to obtain a heart rhythm type of said target subject comprises:

8. The method of claim 7, wherein training the rhythm type detection model comprises:

acquiring sample characteristic data of a plurality of training objects and a gold standard corresponding to each sample characteristic data; the gold standard corresponding to the sample characteristic data is used for representing the heart rhythm type corresponding to the training object; the heart rhythm type corresponding to the training object is determined according to the diagnosis result of the electrocardiogram;

and training a classification model based on the sample characteristic data of the plurality of training objects and the gold standard corresponding to each sample characteristic data to obtain the heart rhythm type detection model.

9. The method of claim 7, wherein the obtaining sample feature data for a plurality of training subjects comprises:

acquiring characteristic data of a plurality of candidate objects;

and selecting the training object from the plurality of candidate objects according to the scoring result corresponding to each candidate object, and taking the feature data of the training object as the sample feature data.

10. The method of claim 9, wherein selecting the training object from the plurality of candidate objects according to the scoring result corresponding to each candidate object comprises:

determining k candidate objects ranked at the top as the training objects; wherein k is a positive integer.

11. The method according to claim 8, wherein the training of the classification model based on the sample feature data of the training subjects and the gold standard corresponding to each sample feature data to obtain the heart rhythm type detection model comprises:

and determining the target value of the model parameter according to the value range and the value step length of the model parameter and the gold standard corresponding to each sample characteristic data to obtain the heart rhythm type detection model.

12. The method of claim 11, wherein the classification model is a multi-classification model, and determining the target values of the model parameters according to the value ranges and the value step lengths of the model parameters and the gold standards corresponding to the sample feature data to obtain the heart rhythm type detection model comprises:

training the initialized classification model based on the sample characteristic data of the training objects to obtain the average accuracy corresponding to the classification model;

and selecting candidate values of the model parameters according to the average accuracy rates, and performing performance detection on the classification model according to the candidate values of the model parameters and gold standards corresponding to the sample characteristic data to obtain the heart rhythm type detection model with model performance meeting preset detection conditions.

13. The method of claim 12, wherein training the initialized classification model based on the sample feature data corresponding to the plurality of training objects to obtain an average accuracy corresponding to the classification model comprises:

dividing a plurality of the sample feature data into 10 mutually exclusive subsets;

14. The method according to claim 12, wherein the performing performance detection on the classification model according to the candidate values of the model parameters to obtain the heart rhythm type detection model with model performance satisfying a preset detection condition comprises:

and under the condition that the accuracy is determined to be greater than the preset accuracy, the sensitivity is determined to be greater than the preset sensitivity, and the specificity is determined to be greater than the preset specificity, determining that the model performance meets the preset detection condition, and determining the candidate value of the model parameter when the model performance meets the preset detection condition as the target value of the model parameter to obtain the heart rhythm type detection model.

15. The method of claim 7, wherein before the inputting the feature data into a pre-trained heart rhythm type detection model to obtain the heart rhythm type of the target object output by the heart rhythm type detection model, the method further comprises:

and operating a web server, loading the model file, and generating a calling interface of the heart rhythm type detection model.

16. An apparatus for artificial intelligence detection of cardiac arrhythmia based on arterial pressure wave characteristics, the apparatus comprising:

the characteristic data determining module is used for determining key points in the brachial artery pressure wave according to the brachial artery pressure wave data and determining characteristic data of the brachial artery pressure wave according to the key points; wherein the characteristic data is used to characterize a fluctuation state of the brachial artery pressure wave;

and the heart rhythm type detection module is used for carrying out heart rhythm type detection according to the characteristic data to obtain the heart rhythm type of the target object.

17. An electronic device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor realizes the steps of the method of any of claims 1 to 15 when executing the computer program.

18. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 15.