188 results about "Heart rate rhythm" patented technology

The invention relates to the technical field of health status assessment, in particular to a cardiovascular disease layered early warning system, which is characterized in that blood vessel compliance and pulse wave velocity are extracted, arterial dilation capacity is measured, blood vessel elasticity is calculated, blood perfusion distribution is detected, and abnormal vasoconstriction is classified on the basis of a hemodynamic state and a heart electrical activity state; in combination with the thickness of the blood vessel wall and the blood vessel stenosis rate, the plaque deposition condition is analyzed, the arteriosclerosis score is calculated, the thrombosis tendency is detected, the local blood flow blocking degree is quantified, the isolation forest algorithm is adopted, the blood flow dynamic change is calculated, the heart rhythm abnormality is analyzed, the acute vascular occlusion probability is measured, and the sudden heart rhythm abnormality risk is evaluated. And based on a gradient boosting tree algorithm, screening high-risk cases, matching medical intervention measures, setting a risk buffer mechanism, adjusting an individual health management strategy, and optimizing a cardiovascular risk hierarchical early warning system.

Novel tools and techniques are provided for implementing intelligent assistance (“IA”) ecosystem, and, in some cases, for implementing extended reality (“XR”) for cardiac arrhythmia procedures. In various embodiments, a computing system might receive device data associated with a device(s) configured to perform a cardiac arrhythmia procedure to provide effective heart rhythm, might receive imaging data associated with an imaging device(s) configured to generate images of a portion(s) of a patient. The computing system might analyze the received device data and imaging data (collectively “received data”), might map the received data to a 3D or 4D representation of the portion(s) of the patient based on the analysis, and might generate and present (using a user experience (“UX”) device) one or more XR images or experiences based on the mapping.

The invention provides a health management method with a health data analysis function. The health management method with the health data analysis function comprises the following steps: S1, collecting health data information of a user body through existing equipment, and establishing a personal health record according to the data information, wherein the body data information comprises but is not limited to blood pressure, heart rhythm and blood sugar; s2, the collected health data need to be cleaned, sorted and preprocessed before a personal health record is established, so that subsequent data analysis is facilitated; and S3, according to the health data of the user, performing health risk assessment by using a machine learning model and a deep learning model algorithm, and predicting possible health risks, including but not limited to cardiovascular diseases and diabetes, of the user. The health management method with the health data analysis function provided by the invention has the advantages that the accuracy and individuation level of health management can be improved, and the requirements of people on health are met.

The invention relates to an intelligent heart electrophysiological signal analysis and Jiangshi connecting line optimization system and method, and belongs to the technical field of electrocardio systems, the system comprises a plurality of modules for cooperative work, a data acquisition module is used for acquiring human body 12-lead electrocardiogram and intracardiac electrogram signals and executing a pace-making stimulation function; the signal preprocessing module filters received signals, the feature extraction module extracts time domain, frequency domain and nonlinear features, the deep learning module identifies abnormal heart rhythms by using a pre-training model, the parameter tuning module dynamically adjusts model parameters and optimizes and analyzes the parameters, and the Jiangshi connecting line optimization module improves signal quality. The intelligent monitoring module automatically monitors multiple indexes, the display interface module visually displays an electrocardiogram, the system compatible module is connected with multiple devices, and the control module coordinates all the modules to ensure efficient operation of the system.

The invention discloses a rapid identification method applied to defibrillation requirements of an AED device, and belongs to the technical field of medical equipment. According to the method, multiple links including acquisition, preprocessing, feature extraction, mode classification, real-time classification and defibrillation triggering work cooperatively, and advanced signal classification and mode recognition technologies are introduced, so that efficient and accurate electrocardiosignal analysis is realized in the AED device, the response speed and accuracy in the emergency treatment process are remarkably improved, and the emergency treatment efficiency is improved. Multi-dimensional features of electrocardiosignals are extracted and input into a pattern classification model, automatic recognition of arrhythmia is achieved, whether defibrillation operation needs to be carried out or not can be accurately judged, the possibility of misjudgment is reduced, self-optimization can be carried out according to new electrocardiosignal data through a dynamic learning and retraining mechanism, and the accuracy of defibrillation is improved. The adaptive capacity to a complex heart rhythm mode is improved, the first-aid response time is effectively shortened, the safety of a patient is guaranteed, and the first-aid efficiency and the treatment effect are improved.

The invention relates to the field of heart rate and heart rhythm real-time monitoring methods, in particular to a dynamic arrhythmia early warning system and method based on wearable equipment, and the method comprises the steps: 1, actually measuring an electrocardiogram of a user target in real time through the wearable equipment; step 2, the wearable device receives an active positioning signal in the medical place in real time, generates a movement track of a user target, and judges whether internal interference factors in the medical place exist on the movement track or not; 3, when the internal interference factors exist, interference marking is conducted on the electrocardiogram through the wearable device, and when the internal interference factors do not exist, abnormity judgment is conducted on the electrocardiogram through the wearable device, and early warning information is sent to the background when it is judged that the electrocardiogram is abnormal; the system comprises wearable equipment for monitoring an electrocardiogram and a background for wirelessly receiving the electrocardiogram, the wearable device collects the electrocardiogram of the target user in real time, and carries out early warning according to the early warning method. According to the invention, the interference condition in the activity process can be accurately identified.

A system and method for authenticating identity during a financial transaction based, in part, on dynamic biometric factors such as body temperature, heart rhythm, blood pressure, blood sugar, vein pattern, skin pore pattern, and voice recognition. The biometric values are captured, aggregated, and patterned over time to create a unique biometric profile. When a transaction is initiated, current biometric values are captured and compared to the biometric patterns, and a biometric authentication determination is made based on the results. An overall identity authentication score is generated based on a combination of the biometric authentication determination and other identity authentication determinations such as a purchase pattern authentication determination or a location authentication determination. An identity authentication determination is made based on the overall identity authentication score and communicated to the merchant. Further, the transaction may be rejected when the overall identity authentication score violates a pre-established fraud threshold value.

Patients with either paroxysmal or persistent AFib and whose untreated AFib is greater than a threshold level of AFib are qualified as eligible for treatment with a drug, (e.g., budiodarone) capable of controlling heart rhythm in the patient. Qualified patients are treated with the drug while their heart rates are monitored with a wearable, and if a given dose of the drug is deemed ineffective, the dose is adjusted until an effective dose is determined for the patient of the patient is deemed non-responsive to the drug and removed from therapy.

The invention provides a health pre-assessment method and system based on information monitoring, and the method comprises the steps: obtaining ECG data of a user, analyzing a target ECG waveform, extracting heart electrophysiological parameters which comprise a heart rate, a heart rate variability standard deviation and a heart rate rhythm, recognizing a heart rate feature mode of the user, and carrying out the health pre-assessment of the user according to the heart rate feature mode. The periodic change of the electrocardio effect is evaluated, and heart rate characteristic mode information is obtained; based on the heart rate characteristic mode information, multiple physiological parameters of the user are collected in real time, the mutual influence relation between the multiple physiological parameters and heart mode measurement values is analyzed, the heart regulation efficiency of the user is evaluated, and a physiological correlation analysis result is obtained in combination with the correlation coefficient. Through the method and the corresponding system, important risks such as health deterioration and subclinical abnormality can be found in advance, and continuous and decision support is provided for clinical decision and individualized health management.

The invention discloses an intelligent arrhythmia recognition sphygmomanometer, and belongs to the technical field of medical equipment. Comprising a cuff type sphygmomanometer composed of a blood pressure measuring device and an electrode slice, and the cuff type sphygmomanometer is electrically connected with an intelligent recognition subsystem. The problem that existing equipment cannot give consideration to heart rhythm detection is solved, the electrode slice is additionally arranged to remind a patient whether to carry out an arrhythmia detection mode or not when the heart rate and pulse of the patient are abnormal, the electrode slice is used for obtaining the heart rhythm condition of the patient, and the heart rhythm condition is printed through the miniature printer so that the heart rhythm condition can be fed back to a doctor to assist in obtaining an accurate conclusion; analyzing the arrhythmia type of the patient by using the decision tree model so as to inform the patient of the real heart rhythm condition; when the patient is in arrhythmia, the blood pressure value of the patient is corrected by utilizing the neural network model, so that the blood pressure value of the patient is prevented from being influenced by the arrhythmia, the patient can know the real blood pressure condition of the patient, and the practicability of the sphygmomanometer is improved.

The invention discloses a cardiovascular detection method and system, and relates to the technical field of medical detection, and the main scheme is as follows: collecting multiple items of data of a person to be detected; respectively calculating a heart beat frequency evaluation value, a heart rhythm index, a blood pressure abnormity reference value and a pulse wave conduction speed based on the multiple items of data, further obtaining a cardiovascular comprehensive index, and presetting a threshold value to judge whether the cardiovascular system is in a risk state or not; calculating a cardiovascular reserve index, further calculating a cardiovascular health index, and presetting a threshold value to judge a cardiovascular health level; according to the method, the cardiovascular state can be comprehensively and systematically evaluated, risks can be found in time, health levels can be distinguished, and a powerful basis is provided for prevention and diagnosis of cardiovascular diseases.

Method for reconstructing ECG signals through imaging for arrhythmia detection and its detection system, wherein the method for reconstructing ECG signals through imaging for arrhythmia detection involves receiving images of the human skin, which contain color changes on the skin surface caused by heartbeats and heart rhythms. The method comprises performing noise reduction on these color changes and analyzing the color variations to extract a remote photoplethysmographic (rPPG) signal that corresponds to the heart rates and heart rhythms. Frequency domain analysis is conducted to identify the frequency features within the rPPG signal, and feature extraction is performed on these frequency characteristics to obtain a photoplethysmographic (PPG) signal. The PPG signal undergoes feature extraction to derive time domain features and frequency domain features, which are then fused to output a reconstructed ECG signal. Finally, the waveform characteristics of the reconstructed ECG signal are interpreted, allowing for the identification of classifications corresponding to arrhythmias based on these waveform features.

The invention discloses a method for reconstructing electrocardiosignals through photoelectric volume pulse waves based on deep learning, and aims at solving the problems that a traditional ECG collection mode is complex, an electrode is uncomfortable to wear, and an existing signal mapping method is low in precision. PPG signals are collected through a portable device and subjected to denoising and normalization processing, a generative adversarial network is constructed, a generator maps the PPG signals into ECG signals, and a discriminator evaluates the authenticity of the generated signals. The generator loss function is combined with the mean square error and the adversarial loss, and the discriminator loss function optimizes the identification degree of the signal; a multi-scale loss function of a time domain and a frequency domain is introduced, the consistency of the generated signals in dynamic characteristics and spectral characteristics is improved, and finally, the generated ECG signals are transmitted back to the portable equipment to be displayed. Real-time signal generation can be achieved on portable equipment such as an intelligent bracelet and a watch, the generated ECG signals have high fidelity, PQRST waveforms can be accurately restored, and the method is suitable for heart rate and heart rhythm variability analysis and abnormal heart rhythm detection.

The invention belongs to the field of non-contact vital sign detection, and particularly relates to a millimeter wave radar arrhythmia detection method based on particle swarm optimization variational mode decomposition and deep learning, and the method specifically comprises the steps: S1, obtaining a human chest micro-motion signal through a millimeter wave radar, performing static clutter filtering, phase extraction, unwrapping and detrending processing on the radar echo signal to obtain a chest displacement signal containing heartbeat and breathing information; s2, aiming at the thoracic cavity displacement signal, constructing a variational mode decomposition model, and carrying out adaptive optimization on a mode number and a penalty factor through a particle swarm optimization algorithm to obtain an optimal decomposition parameter and complete signal decomposition; s3, according to the center frequency and the energy distribution characteristics of each modal component, screening the modal components in the heartbeat frequency range and reconstructing the modal components to obtain heartbeat characteristic signals representing heart mechanical activities; s4, carrying out time sequence segmentation on the heartbeat characteristic signals, extracting local heart beat morphological characteristics by utilizing a convolutional neural network, and carrying out modeling on a long-time rhythm dependency relationship of the heartbeat signals in combination with a time sequence modeling network based on an attention mechanism to obtain heart rhythm depth characteristic representation; and S5, inputting the heart rhythm depth features into a heart rhythm discrimination model, analyzing the heart rhythm state of the detected person, and outputting an arrhythmia detection result. According to the method, adaptive selection of variational mode decomposition parameters is realized by introducing a particle swarm optimization mechanism, heartbeat signals and respiration and motion interference components are effectively separated, modeling is carried out on rhythm characteristics in combination with a deep learning model, and non-contact detection of arrhythmia is realized. The method does not need to wear an electrode or contact a human body, has the advantages of strong anti-interference capability, good adaptability and high detection precision, and has a good application prospect in the fields of heart rhythm health monitoring, disease screening and the like.

A cardiac rhythm condition recognition method based on a seismocardiogram signal comprises the steps that 1, the seismocardiogram signal and a gold standard electrocardiogram signal are collected, and filtering and standardization preprocessing are carried out on the seismocardiogram signal and the gold standard electrocardiogram signal; step 2, automatically labeling the heart rhythm condition based on the gold standard electrocardiogram signal: extracting at least one heart rhythm feature through an electrocardiogram analysis algorithm, and labeling the heart rhythm condition of the electrocardiogram signal based on a multi-evidence fusion labeling rule; step 3, mapping the labeling result of the electrocardiogram signal to the synchronous seismocardiogram signal: mapping the electrocardiogram labeling result to a corresponding time window of the synchronous seismocardiogram signal through a time alignment and priority mapping strategy, and endowing each seismocardiogram signal time window with a heart rhythm condition label; 4, constructing a deep learning model fusing a residual convolutional neural network and a long-short-term memory network, and training by using the marked seismocardiogram signal data; and 5, training model deployment and heart rhythm condition real-time identification.

Disclosed are various examples and embodiments of systems, devices, components and methods configured to estimate the action potential wave propagation in a patient's heart, and subsequently to detect at least one location or type of at least one source of, or rotational phenomenon associated with, at least one cardiac rhythm disorder using intracardiac electrodes and a modified multi-frame Horn-Schunck algorithm to generate a map corresponding to a spatial map, the map being configured to reveal on a monitor or display to a user the at least one location of the at least one source of the at least one cardiac rhythm disorder.

The invention relates to a multi-sensor fused vital sign and motion posture monitoring system and method. The system comprises a photoelectric volume pulse wave acquisition module, an electrocardio acquisition module, an inertia measurement module, a synchronization module, an artifact suppression module, a contact state recognition module, a data quality evaluation module, a vital sign extraction module and an anomaly detection module. Synchronously acquiring a photoelectric volume pulse wave signal, an electrocardiosignal and motion information; suppressing motion artifacts in the photoplethysmography signal and / or the electrocardiosignal based on the motion information; identifying the contact state and generating a data quality indicator related to the measurement reliability in combination with the de-artifact physiological signal; when the data quality meets a preset condition, heart rate, oxyhemoglobin saturation and / or breathing information are extracted, and vital sign parameters are formed; and carrying out conjoint analysis on the artifact-removed physiological signal and the motion information, and marking heart rate rhythm abnormity and / or posture abnormity. According to the invention, the reliability and availability of vital sign monitoring can be improved.

The present invention relates to the technical field of dynamic electrocardiogram analysis, specifically a dynamic electrocardiogram data processing method and system, comprising the following steps: segmenting the acquired dynamic electrocardiogram data, adjusting the signal strength of each segment, calculating the energy distribution of each segment, and judging its stability, and generating stability-adjusted electrocardiogram data segments. The present invention, by extracting bioinformatics features based on deep learning for electrocardiogram signals, can identify complex patterns and features in electrocardiogram signals in the early detection of abnormal heart rhythms, thereby improving the accuracy and sensitivity of abnormal heart rhythm identification. It provides strong support for the accurate classification of abnormal heart rhythms. The fractal dimension theory is used to conduct an in-depth analysis of dynamic electrocardiogram data, revealing the nonlinear characteristics and self-similarity of electrocardiogram signals. The theoretical depth of electrocardiogram analysis is enhanced.

A Wearable Medical System (WMS) includes one or more pacing capabilities. The WMS may detect when the patient's heart rhythm starts to deteriorate, but not necessarily in a way that requires defibrillation. In particular, the WMS may detect bradycardia of one or more types, and then confirm the detection before pacing to treat the detected bradycardia.

The invention discloses a smart watch arrhythmia early warning method and system based on deep learning. The method comprises the steps that firstly, PPG signals are collected through a photoelectric volume pulse wave sensor, and denoising and standardization processing are conducted through an adaptive filtering algorithm; then, the heart beat position is recognized through a wavelet transform peak value detection algorithm, and a heart rate variability feature sequence is obtained; converting the one-dimensional heart rate variability feature sequence into a two-dimensional heart rhythm image by using an improved Gramian angular field transformation algorithm; and training an arrhythmia classification model by adopting a ResNet-50 convolutional neural network to realize automatic identification of different types of arrhythmia. And finally, continuous monitoring and timely alarming are realized through a sliding window technology and an early warning mechanism. The key technical problems of low detection precision, poor real-time performance, high equipment cost, poor user experience and the like in the existing arrhythmia detection technology are effectively solved, and an innovative solution is provided for heart rhythm health monitoring of the intelligent wearable equipment.

Systems, devices, and methods provide up-to-date defibrillation shock recommendations. In an example method, multiple segments of an electrocardiogram (ECG) of an individual are detected from an individual receiving chest compressions. The multiple segments are evaluated to determine whether the individual is exhibiting a shockable heart rhythm. A medical device outputs a recommendation indicating whether a defibrillation shock is advised based on the most recent determination of the individual's heart rhythm. For example, the medical device outputs an up-to-date recommendation on-demand in response to an input signal from a user. In some examples, the medical device updates the recommendation based on ongoing analysis of the ECG.

The invention discloses an electrocardiosignal anomaly classification method and device. The method comprises the following steps: receiving a multi-lead electrocardiosignal data set acquired by multi-lead acquisition equipment; aiming at each lead electrocardiosignal sequence, constructing a lead internal graph structure for representing the internal rhythm change of the electrocardiosignal waveform; based on waveform synchronism and correlation between different leads, an external graph structure between the leads is constructed; further, based on the lead internal graph structure and the lead external graph structure, determining an external distribution mode of the electrocardiosignals under different physiological states or individual differences; fusing the graph structure and the external distribution mode to generate global graph representation of the electrocardiosignal; and inputting the graph representation into a classifier to obtain an abnormal heart rhythm classification result. According to the method, the local waveform morphological characteristics, the cross-lead correlation characteristics and the distribution change caused by cross-crowd difference of the electrocardiosignals can be modeled at the same time, so that the accuracy, the robustness and the cross-subject generalization ability of an electrocardiograph anomaly classification task are effectively improved.

To provide suitable devices, systems, and methods for percutaneous pneumatic cardiac support. [Solution] The cardiac support system includes a pneumatic effector implanted beneath the pericardium and across the myocardial surface covering the patient's left ventricle. A port receives an implanted, percutaneously introduced cannula. The port is connected to supply the drive gas received from the cannula to the pneumatic effector. An external drive unit includes a pump assembly and a control circuit that operates the pump to activate the pneumatic effector in response to the patient's sensed cardiac rhythm. A connecting tube has a pump end connected to the pump and a percutaneous port connection end attached to the implantable port.

A card capable of measuring ECG signals includes a card body, a control unit, an electronic control unit, a conduction unit, an ECG signal processing unit, a storage unit, a link unit and a light source processing unit. The card allows a person to be tested to touch the conduction unit simultaneously with the fingers of both hands, so that a closed circuit is formed between the heart of the person to be tested, the arms of the person to be tested, the conduction unit and the ECG signal processing unit, so as to carry out the measurement of the heart rhythm, the ECG data generated after measurement is stored in the storage unit, the reading device reads the ECG data through the link unit, relevant signals are generated to the light source processing unit for a light source demonstration during the measurement process and after measurement.

The invention relates to an electrocardiosignal processing method, electronic equipment, a medium and a product, and the electrocardiosignal processing method comprises the steps that an electrocardiosignal is obtained, the heart rate type to which the electrocardiosignal belongs is recognized through a preset model, and probability information of the heart rate type to which the electrocardiosignal belongs is determined to obtain a processed electrocardiosignal, the probability information comprises a probability value of the heart rhythm type to which the electrocardiosignal belongs, and obtaining the heart rhythm type to which the processed electrocardiosignal belongs and the probability information of the heart rhythm type to which the processed electrocardiosignal belongs. According to the invention, the wearable device prompts the user of the possibility of the heart rhythm type to which the obtained electrocardiosignal belongs in a probabilistic manner, so that the misjudgment and false alarm of the heart rhythm type to which the obtained electrocardiosignal belongs due to the limitation of the hardware condition or the use condition of the electronic device are effectively avoided; panic and pressure of customers caused by false alarms are avoided, and user experience is improved.

The invention relates to the technical field of auxiliary detection devices, in particular to an auxiliary device for electrocardiogram detection. Electrocardio data of a patient can be monitored through the dynamic monitor assembly, when an electrocardio numerical value is abnormal, a miniature vibration motor can vibrate and sense, meanwhile, defibrillation current generated by the multifunctional dynamic monitor conducts electric shock defibrillation on the heart through chest lead pieces on the two sides, and then the heart rhythm of the patient is recovered to be normal; the guide block is supported through the elastic interlayer and the monitoring vest, the chest lead sheet is prevented from displacement caused by friction, and the chest lead sheet is isolated from the external environment through the anti-radiation coating, so that the influence of the external environment on the monitoring result is reduced; the position of the lower limb lead clamp is adjusted through the electric control air cylinder so that the lower limb of the human body can be conveniently fixed, the upper limb lead clamp drives the sliding block to transversely move on the inner wall of the sliding rail, the upper limb of the patient can be conveniently clamped into the upper limb lead clamp to be fixed, and the lead clamp fixing of the four limbs of the patient is simpler and more convenient.

The invention relates to a sudden cardiac arrest treatment device based on an ultrasonic Doppler feedback technology, and belongs to the field of medical instruments. The cardiac arrest treatment device based on the ultrasonic Doppler feedback technology comprises a first monitoring module used for monitoring the blood flow condition of the neck blood vessel of a target organism; and / or the second monitoring module is used for monitoring the blood oxygen condition of the neck blood vessel of the target organism; the acquisition module is used for acquiring blood supply and / or oxygen supply of the target organism according to the blood flow condition and the blood oxygen condition; the system comprises a blood supply module, a cardio-pulmonary resuscitation pressing module, and / or a ventilation module, and / or a defibrillation module, and the cardio-pulmonary resuscitation pressing module is used for adjusting the pressing depth or pressing frequency according to blood supply and / or oxygen supply; the ventilation module is used for adjusting the ventilation frequency in the cardio-pulmonary resuscitation pressing process according to blood supply and / or oxygen supply; the defibrillation module is used for executing defibrillation operation under the condition that the defibrillable heart rhythm exists. According to the technical scheme, the treatment rate can be increased, and the cardiac arrest treatment quality can be improved.

The wearable device obtains heart rate data of a user; and determines a first feature and a second feature of the heart rate data based on the heart rate data and a Poincare plot. The first feature is a standard deviation of a distance from an origin of the Poincare plot to a first straight line, and the second feature is a standard deviation of a distance from the origin of the Poincare plot to a second straight line. The wearable device inputs the target feature to an identification model, to obtain an identification result. The identification model is configured to identify, based on the input, whether a heart rhythm is abnormal.

The invention provides an abnormal electrocardio position positioning and disease prediction classification method based on deep learning, belongs to the technical field of artificial intelligence and medical health crossing, and solves the technical problems that an electrocardiogram signal is high in dependence on manual labeling of professional doctors, high in labeling cost and low in efficiency. Comprising the following steps: S1, preprocessing a data set; s2, dividing a training set and a test set; s3, a CNN-LSTM-Attention deep learning model is constructed according to the MIT-BIH data; s4, constructing a multi-layer perceptron (MLPS) model for the PTB-XL data; s5, adopting an early stop strategy to avoid overfitting of the model; and S6, inputting test set data into the trained deep learning model, and outputting an abnormal heart rhythm position and a disease prediction result. The method has the beneficial effect that the problem that the electrocardiogram depends on manual labeling is relieved through the deep learning structure.

The invention provides a millimeter-wave radar heart rhythm detection method based on comparative learning. The millimeter-wave radar heart rhythm detection method comprises the steps that S1, echo signals generated after millimeter-wave radar signals act on a to-be-detected object are acquired; s2, based on the obtained echo signals, reflected signals from the chest area in the three-dimensional space are extracted; s3, weakening the influence of respiratory movement based on the phase change of the reflected signal to obtain a heart mechanical activity signal; s4, creating a model, pre-training the heart mechanical activity signal based on comparative learning, and training the ability of a feature extractor in the model to extract features related to heart mechanical activity; s5, on the basis of the feature extractor obtained through training, heart rhythm feature tags are added for fine adjustment, and model training is completed through a loss function; s6, heart mechanical activity signals of the to-be-tested object are input into the trained model, and a heart rhythm classification result of the to-be-tested object is obtained.