The main drawback of current ECG systems is the location-specific nature of the systems due to the use of fixed/wired applications. That is why there is a critical need to improve the current ECG systems to achieve ex...The main drawback of current ECG systems is the location-specific nature of the systems due to the use of fixed/wired applications. That is why there is a critical need to improve the current ECG systems to achieve extended patient’s mobility and to cover security handling. With this in mind, Compressed Sensing (CS) procedure and the collaboration of Sensing Matrix Selection (SMS) approach are used to provide a robust ultra-low-power approach for normal and abnormal ECG signals. Our simulation results based on two proposed algorithms illustrate 25% decrease in sampling-rate and a good level of quality for the degree of incoherence between the random measurement and sparsity matrices. The simulation results also confirm that the Binary Toeplitz Matrix (BTM) provides the best compression performance with the highest energy efficiency for random sensing matrix.展开更多
心电信号容易受到采集设备和被测者状态的干扰,为此提出一种归一化最小均方差(Normalized Least Mean Square,NLMS)和自适应噪声完备集合模态分解(Complete Ensemble Empirical Mode Decomposition with Adaptive Noise,CEEMDAN)组合的...心电信号容易受到采集设备和被测者状态的干扰,为此提出一种归一化最小均方差(Normalized Least Mean Square,NLMS)和自适应噪声完备集合模态分解(Complete Ensemble Empirical Mode Decomposition with Adaptive Noise,CEEMDAN)组合的去噪方法。其中:优化的NLMS算法通过简化步长因子和输入信号的关系减少运算量,并结合迭代次数对步长因子进行优化,提高算法收敛性能;改进的CEEMDAN算法结合高斯白噪声的统计特性对所有IMF分量进行显著性检验,来识别和筛选含有噪声的成分,使干净信号与噪声信号分离。实验结果表明,在不同噪声强度下,该方法相比于CEEMDAN直接去噪效果更佳,且缓解了传统NLMS收敛速度与运算量之间的矛盾。展开更多
Remote ECG monitoring systems are becoming commonplace medical devices for remote heart monitoring. In recent years, remote ECG monitoring systems have been applied in the monitoring of various kinds of heart diseases...Remote ECG monitoring systems are becoming commonplace medical devices for remote heart monitoring. In recent years, remote ECG monitoring systems have been applied in the monitoring of various kinds of heart diseases, and the quality of the transmission and re- ception of the ECG signals during remote process kept advancing. However, there remains accompanying challenges. This report focuses on the three components of the remote ECG monitoring system： patient （the end user）, the doctor workstation, and the remote server, reviewing and evaluating the imminent challenges on the wearable systems, packet loss in remote transmission, portable ECG monitoring system, pa- tient ECG data collection system, and ECG signals transmission including real-time processing ST segment, R wave, RR interval and QRS wave, etc. This paper tries to clarify the future developmental strategies of the ECG remote monitoring, which can be helpful in guiding the research and development of remote ECG monitoring.展开更多
Recently, the development of the Internet of Things (IoT) hasenabled continuous and personal electrocardiogram (ECG) monitoring. In theECG monitoring system, classification plays an important role because it canselect...Recently, the development of the Internet of Things (IoT) hasenabled continuous and personal electrocardiogram (ECG) monitoring. In theECG monitoring system, classification plays an important role because it canselect useful data (i.e., reduce the size of the dataset) and identify abnormaldata that can be used to detect the clinical diagnosis and guide furthertreatment. Since the classification requires computing capability, the ECGdata are usually delivered to the gateway or the server where the classificationis performed based on its computing resource. However, real-time ECG datatransmission continuously consumes battery and network resources, whichare expensive and limited. To mitigate this problem, this paper proposes atiny machine learning (TinyML)-based classification (i.e., TinyCES), wherethe ECG monitoring device performs the classification by itself based onthe machine-learning model, which can reduce the memory and the networkresource usages for the classification. To demonstrate the feasibility, afterwe configure the convolutional neural networks (CNN)-based model usingECG data from the Massachusetts Institute of Technology (MIT)-Beth IsraelHospital (BIH) arrhythmia and the Physikalisch Technische Bundesanstalt(PTB) diagnostic ECG databases, TinyCES is validated using the TinyMLsupportedArduino prototype. The performance results show that TinyCEScan have an approximately 97% detection ratio, which means that it has greatpotential to be a lightweight and resource-efficient ECG monitoring system.展开更多
ECG monitoring in daily life is an important means of treating heart disease. To make it easier for the medical to monitor the ECG of their patients outside the hospital, we designed and developed an ECG monitoring an...ECG monitoring in daily life is an important means of treating heart disease. To make it easier for the medical to monitor the ECG of their patients outside the hospital, we designed and developed an ECG monitoring and alarming system based on Android smart phone. In our system, an ECG device collects the ECG signal and transmits it to an Android phone. The Android phone detects alarms which come from the ECG devices. When alarms occur, Android phone will capture the ECG images and the details about the alarms, and sends them to the cloud Alarm Server (AS). Once received, AS push the messages to doctors’ phone, so the doctors could see the ECG images and alarm details on their mobile phone. In our system, high resolution ECG pictures are transmitted to doctors’ phone in a user-friendly way, which can help doctors keep track of their patient’s condition easily.展开更多
This paper describes the development of a new ECG tele-monitoring method and system based on the embedded web server. The system consists of ECG recorders with network interface and the embedded web server, internet n...This paper describes the development of a new ECG tele-monitoring method and system based on the embedded web server. The system consists of ECG recorders with network interface and the embedded web server, internet networks and computers, with the system operating on browser/server(B/S) mode. The ECG recorder was designed by ARM9 (S3C2410X) and embedded operating system (Linux). Once the ECG recorder has been connected to the internet network, medical experts can use the internet to access the server of the ECG recorder, monitor ECG signals, and diagnose patients by browsing the dynamic web pages in the embedded web server. The experimental results reveal that the designed system is stable, reliable, and suitable for the use in real-time ECG tele-monitoring for both family and community health care.展开更多
Traditional ECG acquisition system lacks for flexibility. To improve the flexibility of ECG acquisition system and the signal-to-noise ratio of ECG, a new ECG acquisition system was designed based on DAQ card and Labv...Traditional ECG acquisition system lacks for flexibility. To improve the flexibility of ECG acquisition system and the signal-to-noise ratio of ECG, a new ECG acquisition system was designed based on DAQ card and Labview and oversampling was implemented in Labview. And analog signal conditioning circuit was improved on. The result indicated that the system could detect ECG signal accurately with high signal-to-noise ratio and the signal processing methods could be adjusted easily. So the new system can satisfy many kinds of ECG acquisition. It is a flexible experiment platform for exploring new ECG acquisition methods.展开更多
2014年9月17日收到飞利浦（中国）投资有限公司报告，该公司代理的ECG管理系统（注册证号：国食药监械（进）字2013第2700072号）由于在特定情况下系统会出现错误等原因，其生产商美国Philips Medical Systems公司对该产品进行主动召回...2014年9月17日收到飞利浦（中国）投资有限公司报告，该公司代理的ECG管理系统（注册证号：国食药监械（进）字2013第2700072号）由于在特定情况下系统会出现错误等原因，其生产商美国Philips Medical Systems公司对该产品进行主动召回。该公司称此次召回产品未在中国销售。请各省、自治区、直辖市食品药品监督管理局加强对此类产品的监督管理。展开更多
Diagnosing cardiac diseases relies heavily on electrocardiogram(ECG)analysis,but detecting myocardial infarction-related arrhythmias remains challenging due to irregular heartbeats and signal variations.Despite advanc...Diagnosing cardiac diseases relies heavily on electrocardiogram(ECG)analysis,but detecting myocardial infarction-related arrhythmias remains challenging due to irregular heartbeats and signal variations.Despite advancements in machine learning,achieving both high accuracy and low computational cost for arrhythmia classification remains a critical issue.Computer-aided diagnosis systems can play a key role in early detection,reducing mortality rates associated with cardiac disorders.This study proposes a fully automated approach for ECG arrhythmia classification using deep learning and machine learning techniques to improve diagnostic accuracy while minimizing processing time.The methodology consists of three stages:1)preprocessing,where ECG signals undergo noise reduction and feature extraction;2)feature Identification,where deep convolutional neural network(CNN)blocks,combined with data augmentation and transfer learning,extract key parameters;3)classification,where a hybrid CNN-SVM model is employed for arrhythmia recognition.CNN-extracted features were fed into a binary support vector machine(SVM)classifier,and model performance was assessed using five-fold cross-validation.Experimental findings demonstrated that the CNN2 model achieved 85.52%accuracy,while the hybrid CNN2-SVM approach significantly improved accuracy to 97.33%,outperforming conventional methods.This model enhances classification efficiency while reducing computational complexity.The proposed approach bridges the gap between accuracy and processing speed in ECG arrhythmia classification,offering a promising solution for real-time clinical applications.Its superior performance compared to nonlinear classifiers highlights its potential for improving automated cardiac diagnosis.展开更多
Background:Irregular heartbeats can have serious health implications if left undetected and untreated for an extended period of time.Methods:This study leverages machine learning(ML)techniques to classify electrocardi...Background:Irregular heartbeats can have serious health implications if left undetected and untreated for an extended period of time.Methods:This study leverages machine learning(ML)techniques to classify electrocardiogram(ECG)heartbeats,comparing traditional feature-based ML methods with innovative image-based approaches.The dataset underwent rigorous preprocessing,including down-sampling,frequency filtering,beat segmentation,and normalization.Two methodologies were explored:(1)handcrafted feature extraction,utilizing metrics like heart rate variability and RR distances with LightGBM classifiers,and(2)image transformation of ECG signals using Gramian Angular Field(GAF),Markov Transition Field(MTF),and Recurrence Plot(RP),enabling multimodal input for convolutional neural networks(CNNs).The Synthetic Minority Oversampling Technique(SMOTE)addressed data imbalance,significantly improving minority-class metrics.Results:The handcrafted feature approach achieved notable performance,with LightGBM excelling in precision and recall.Image-based classification further enhanced outcomes,with a custom Inception-based CNN,attaining an 85%F1 score and 97%accuracy using combined GAF,MTF,and RP transformations.Statistical analyses confirmed the significance of these improvements.Conclusion:This work highlights the potential of ML for cardiac irregularities detection,demonstrating that combining advanced preprocessing,feature engineering,and state-of-the-art neural networks can improve classification accuracy.These findings contribute to advancing AI-driven diagnostic tools,offering promising implications for cardiovascular healthcare.展开更多
Atrial Fibrillation(AF)is a cardiac disorder characterized by irregular heart rhythms,typically diagnosed using Electrocardiogram(ECG)signals.In remote regions with limited healthcare personnel,automated AF detection ...Atrial Fibrillation(AF)is a cardiac disorder characterized by irregular heart rhythms,typically diagnosed using Electrocardiogram(ECG)signals.In remote regions with limited healthcare personnel,automated AF detection is extremely important.Although recent studies have explored various machine learning and deep learning approaches,challenges such as signal noise and subtle variations between AF and other cardiac rhythms continue to hinder accurate classification.In this study,we propose a novel framework that integrates robust preprocessing,comprehensive feature extraction,and an ensemble classification strategy.In the first step,ECG signals are divided into equal-sized segments using a 5-s sliding window with 50%overlap,followed by bandpass filtering between 0.5 and 45 Hz for noise removal.After preprocessing,both time and frequency-domain features are extracted,and a custom one-dimensional Convolutional Neural Network—Bidirectional Long Short-Term Memory(1D CNN-BiLSTM)architecture is introduced.Handcrafted and automated features are concatenated into a unified feature vector and classified using Support Vector Machine(SVM),Random Forest(RF),and Long Short-Term Memory(LSTM)models.A Quantum Genetic Algorithm(QGA)optimizes weighted averages of the classifier outputs for multi-class classification,distinguishing among AF,noisy,normal,and other rhythms.Evaluated on the PhysioNet 2017 Cardiology Challenge dataset,the proposed method achieved an accuracy of 94.40%and an F1-score of 92.30%,outperforming several state-of-the-art techniques.展开更多
Wearable signal analysis is an important technology for monitoring physiological signals without interfering with an individual’s daily behavior.As detecting cardiovascular diseases can dramatically reduce mortality,...Wearable signal analysis is an important technology for monitoring physiological signals without interfering with an individual’s daily behavior.As detecting cardiovascular diseases can dramatically reduce mortality,arrhythmia recognition using ECG signals has attracted much attention.In this paper,we propose a singlechannel convolutional neural network to detect Atrial Fibrillation(AF)based on ECG signals collected by wearable devices.It contains 3 primary modules.All recordings are firstly uniformly sized,normalized,and Butterworth low-pass filtered for noise removal.Then the preprocessed ECG signals are fed into convolutional layers for feature extraction.In the classification module,the preprocessed signals are fed into convolutional layers containing large kernels for feature extraction,and the fully connected layer provides probabilities.During the training process,the output of the previous pooling layer is concatenated with the vectors of the convolutional layer as a new feature map to reduce feature loss.Numerous comparison and ablation experiments are performed on the 2017 PhysioNet/CinC Challenge dataset,demonstrating the superiority of the proposed method.展开更多
Background:The accurate identification of cardiac abnormalities is essential for proper diagnosis and effective treatment of cardiovascular diseases.Method:This work introduces an advanced methodology for detecting ca...Background:The accurate identification of cardiac abnormalities is essential for proper diagnosis and effective treatment of cardiovascular diseases.Method:This work introduces an advanced methodology for detecting cardiac abnormalities and estimating electrocardiographic age(ECG Age)using sophisticated signal processing and deep learning techniques.This study looks at six main heart conditions found in 12-lead electrocardiogram(ECG)data.It addresses important issues like class imbalances,missing lead scenarios,and model generalizations.A modified residual neural network(ResNet)architecture was developed to enhance the detection of cardiac abnormalities.Results:The proposed ResNet demonst rated superior performance when compared with two linear models and an alternative ResNet architectures,achieving an overall classification accuracy of 91.25%and an F1 score of 93.9%,surpassing baseline models.A comprehensive lead loss analysis was conducted,evaluating model performance across 4096 combinations of missing leads.The results revealed that pulse rate-based factors remained robust with up to 75%lead loss,while block-based factors experienced significant performance declines beyond the loss of four leads.Conclusion:This analysis highlighted the importance of addressing lead loss impacts to maintain a robust model.To optimize performance,targeted training approaches were developed for different conditions.Based on these insights,a grouping strategy was implemented to train specialized models for pulse rate-based and block-based conditions.This approach resulted in notable improvements,achieving an overall classification accuracy of 95.12%and an F1 score of 95.79%.展开更多
文摘The main drawback of current ECG systems is the location-specific nature of the systems due to the use of fixed/wired applications. That is why there is a critical need to improve the current ECG systems to achieve extended patient’s mobility and to cover security handling. With this in mind, Compressed Sensing (CS) procedure and the collaboration of Sensing Matrix Selection (SMS) approach are used to provide a robust ultra-low-power approach for normal and abnormal ECG signals. Our simulation results based on two proposed algorithms illustrate 25% decrease in sampling-rate and a good level of quality for the degree of incoherence between the random measurement and sparsity matrices. The simulation results also confirm that the Binary Toeplitz Matrix (BTM) provides the best compression performance with the highest energy efficiency for random sensing matrix.
文摘心电信号容易受到采集设备和被测者状态的干扰,为此提出一种归一化最小均方差(Normalized Least Mean Square,NLMS)和自适应噪声完备集合模态分解(Complete Ensemble Empirical Mode Decomposition with Adaptive Noise,CEEMDAN)组合的去噪方法。其中:优化的NLMS算法通过简化步长因子和输入信号的关系减少运算量,并结合迭代次数对步长因子进行优化,提高算法收敛性能;改进的CEEMDAN算法结合高斯白噪声的统计特性对所有IMF分量进行显著性检验,来识别和筛选含有噪声的成分,使干净信号与噪声信号分离。实验结果表明,在不同噪声强度下,该方法相比于CEEMDAN直接去噪效果更佳,且缓解了传统NLMS收敛速度与运算量之间的矛盾。
文摘Remote ECG monitoring systems are becoming commonplace medical devices for remote heart monitoring. In recent years, remote ECG monitoring systems have been applied in the monitoring of various kinds of heart diseases, and the quality of the transmission and re- ception of the ECG signals during remote process kept advancing. However, there remains accompanying challenges. This report focuses on the three components of the remote ECG monitoring system： patient （the end user）, the doctor workstation, and the remote server, reviewing and evaluating the imminent challenges on the wearable systems, packet loss in remote transmission, portable ECG monitoring system, pa- tient ECG data collection system, and ECG signals transmission including real-time processing ST segment, R wave, RR interval and QRS wave, etc. This paper tries to clarify the future developmental strategies of the ECG remote monitoring, which can be helpful in guiding the research and development of remote ECG monitoring.
基金supported by National Research Foundation (NRF)of Korea Grant funded by the Korean Government (MSIP) (No.2022R1F1A1063183).
文摘Recently, the development of the Internet of Things (IoT) hasenabled continuous and personal electrocardiogram (ECG) monitoring. In theECG monitoring system, classification plays an important role because it canselect useful data (i.e., reduce the size of the dataset) and identify abnormaldata that can be used to detect the clinical diagnosis and guide furthertreatment. Since the classification requires computing capability, the ECGdata are usually delivered to the gateway or the server where the classificationis performed based on its computing resource. However, real-time ECG datatransmission continuously consumes battery and network resources, whichare expensive and limited. To mitigate this problem, this paper proposes atiny machine learning (TinyML)-based classification (i.e., TinyCES), wherethe ECG monitoring device performs the classification by itself based onthe machine-learning model, which can reduce the memory and the networkresource usages for the classification. To demonstrate the feasibility, afterwe configure the convolutional neural networks (CNN)-based model usingECG data from the Massachusetts Institute of Technology (MIT)-Beth IsraelHospital (BIH) arrhythmia and the Physikalisch Technische Bundesanstalt(PTB) diagnostic ECG databases, TinyCES is validated using the TinyMLsupportedArduino prototype. The performance results show that TinyCEScan have an approximately 97% detection ratio, which means that it has greatpotential to be a lightweight and resource-efficient ECG monitoring system.
文摘ECG monitoring in daily life is an important means of treating heart disease. To make it easier for the medical to monitor the ECG of their patients outside the hospital, we designed and developed an ECG monitoring and alarming system based on Android smart phone. In our system, an ECG device collects the ECG signal and transmits it to an Android phone. The Android phone detects alarms which come from the ECG devices. When alarms occur, Android phone will capture the ECG images and the details about the alarms, and sends them to the cloud Alarm Server (AS). Once received, AS push the messages to doctors’ phone, so the doctors could see the ECG images and alarm details on their mobile phone. In our system, high resolution ECG pictures are transmitted to doctors’ phone in a user-friendly way, which can help doctors keep track of their patient’s condition easily.
基金Education Committee Foundation of Beijing grant number: KM200610005022+1 种基金Young Backbone Teacher Foundation of Beijing grant number: 102KB00845
文摘This paper describes the development of a new ECG tele-monitoring method and system based on the embedded web server. The system consists of ECG recorders with network interface and the embedded web server, internet networks and computers, with the system operating on browser/server(B/S) mode. The ECG recorder was designed by ARM9 (S3C2410X) and embedded operating system (Linux). Once the ECG recorder has been connected to the internet network, medical experts can use the internet to access the server of the ECG recorder, monitor ECG signals, and diagnose patients by browsing the dynamic web pages in the embedded web server. The experimental results reveal that the designed system is stable, reliable, and suitable for the use in real-time ECG tele-monitoring for both family and community health care.
文摘Traditional ECG acquisition system lacks for flexibility. To improve the flexibility of ECG acquisition system and the signal-to-noise ratio of ECG, a new ECG acquisition system was designed based on DAQ card and Labview and oversampling was implemented in Labview. And analog signal conditioning circuit was improved on. The result indicated that the system could detect ECG signal accurately with high signal-to-noise ratio and the signal processing methods could be adjusted easily. So the new system can satisfy many kinds of ECG acquisition. It is a flexible experiment platform for exploring new ECG acquisition methods.
文摘2014年9月17日收到飞利浦（中国）投资有限公司报告，该公司代理的ECG管理系统（注册证号：国食药监械（进）字2013第2700072号）由于在特定情况下系统会出现错误等原因，其生产商美国Philips Medical Systems公司对该产品进行主动召回。该公司称此次召回产品未在中国销售。请各省、自治区、直辖市食品药品监督管理局加强对此类产品的监督管理。
文摘Diagnosing cardiac diseases relies heavily on electrocardiogram(ECG)analysis,but detecting myocardial infarction-related arrhythmias remains challenging due to irregular heartbeats and signal variations.Despite advancements in machine learning,achieving both high accuracy and low computational cost for arrhythmia classification remains a critical issue.Computer-aided diagnosis systems can play a key role in early detection,reducing mortality rates associated with cardiac disorders.This study proposes a fully automated approach for ECG arrhythmia classification using deep learning and machine learning techniques to improve diagnostic accuracy while minimizing processing time.The methodology consists of three stages:1)preprocessing,where ECG signals undergo noise reduction and feature extraction;2)feature Identification,where deep convolutional neural network(CNN)blocks,combined with data augmentation and transfer learning,extract key parameters;3)classification,where a hybrid CNN-SVM model is employed for arrhythmia recognition.CNN-extracted features were fed into a binary support vector machine(SVM)classifier,and model performance was assessed using five-fold cross-validation.Experimental findings demonstrated that the CNN2 model achieved 85.52%accuracy,while the hybrid CNN2-SVM approach significantly improved accuracy to 97.33%,outperforming conventional methods.This model enhances classification efficiency while reducing computational complexity.The proposed approach bridges the gap between accuracy and processing speed in ECG arrhythmia classification,offering a promising solution for real-time clinical applications.Its superior performance compared to nonlinear classifiers highlights its potential for improving automated cardiac diagnosis.
文摘Background:Irregular heartbeats can have serious health implications if left undetected and untreated for an extended period of time.Methods:This study leverages machine learning(ML)techniques to classify electrocardiogram(ECG)heartbeats,comparing traditional feature-based ML methods with innovative image-based approaches.The dataset underwent rigorous preprocessing,including down-sampling,frequency filtering,beat segmentation,and normalization.Two methodologies were explored:(1)handcrafted feature extraction,utilizing metrics like heart rate variability and RR distances with LightGBM classifiers,and(2)image transformation of ECG signals using Gramian Angular Field(GAF),Markov Transition Field(MTF),and Recurrence Plot(RP),enabling multimodal input for convolutional neural networks(CNNs).The Synthetic Minority Oversampling Technique(SMOTE)addressed data imbalance,significantly improving minority-class metrics.Results:The handcrafted feature approach achieved notable performance,with LightGBM excelling in precision and recall.Image-based classification further enhanced outcomes,with a custom Inception-based CNN,attaining an 85%F1 score and 97%accuracy using combined GAF,MTF,and RP transformations.Statistical analyses confirmed the significance of these improvements.Conclusion:This work highlights the potential of ML for cardiac irregularities detection,demonstrating that combining advanced preprocessing,feature engineering,and state-of-the-art neural networks can improve classification accuracy.These findings contribute to advancing AI-driven diagnostic tools,offering promising implications for cardiovascular healthcare.
基金supported and funded by the Deanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University(IMSIU)(grant number IMSIU-DDRSP2501)。
文摘Atrial Fibrillation(AF)is a cardiac disorder characterized by irregular heart rhythms,typically diagnosed using Electrocardiogram(ECG)signals.In remote regions with limited healthcare personnel,automated AF detection is extremely important.Although recent studies have explored various machine learning and deep learning approaches,challenges such as signal noise and subtle variations between AF and other cardiac rhythms continue to hinder accurate classification.In this study,we propose a novel framework that integrates robust preprocessing,comprehensive feature extraction,and an ensemble classification strategy.In the first step,ECG signals are divided into equal-sized segments using a 5-s sliding window with 50%overlap,followed by bandpass filtering between 0.5 and 45 Hz for noise removal.After preprocessing,both time and frequency-domain features are extracted,and a custom one-dimensional Convolutional Neural Network—Bidirectional Long Short-Term Memory(1D CNN-BiLSTM)architecture is introduced.Handcrafted and automated features are concatenated into a unified feature vector and classified using Support Vector Machine(SVM),Random Forest(RF),and Long Short-Term Memory(LSTM)models.A Quantum Genetic Algorithm(QGA)optimizes weighted averages of the classifier outputs for multi-class classification,distinguishing among AF,noisy,normal,and other rhythms.Evaluated on the PhysioNet 2017 Cardiology Challenge dataset,the proposed method achieved an accuracy of 94.40%and an F1-score of 92.30%,outperforming several state-of-the-art techniques.
基金funded by the National Natural Science Foundation of China(No.62171114)the Fundamental Research Funds for the Central Universities(No.DUT22RC(3)099)Xiaomi Young Talents Program.
文摘Wearable signal analysis is an important technology for monitoring physiological signals without interfering with an individual’s daily behavior.As detecting cardiovascular diseases can dramatically reduce mortality,arrhythmia recognition using ECG signals has attracted much attention.In this paper,we propose a singlechannel convolutional neural network to detect Atrial Fibrillation(AF)based on ECG signals collected by wearable devices.It contains 3 primary modules.All recordings are firstly uniformly sized,normalized,and Butterworth low-pass filtered for noise removal.Then the preprocessed ECG signals are fed into convolutional layers for feature extraction.In the classification module,the preprocessed signals are fed into convolutional layers containing large kernels for feature extraction,and the fully connected layer provides probabilities.During the training process,the output of the previous pooling layer is concatenated with the vectors of the convolutional layer as a new feature map to reduce feature loss.Numerous comparison and ablation experiments are performed on the 2017 PhysioNet/CinC Challenge dataset,demonstrating the superiority of the proposed method.
文摘Background:The accurate identification of cardiac abnormalities is essential for proper diagnosis and effective treatment of cardiovascular diseases.Method:This work introduces an advanced methodology for detecting cardiac abnormalities and estimating electrocardiographic age(ECG Age)using sophisticated signal processing and deep learning techniques.This study looks at six main heart conditions found in 12-lead electrocardiogram(ECG)data.It addresses important issues like class imbalances,missing lead scenarios,and model generalizations.A modified residual neural network(ResNet)architecture was developed to enhance the detection of cardiac abnormalities.Results:The proposed ResNet demonst rated superior performance when compared with two linear models and an alternative ResNet architectures,achieving an overall classification accuracy of 91.25%and an F1 score of 93.9%,surpassing baseline models.A comprehensive lead loss analysis was conducted,evaluating model performance across 4096 combinations of missing leads.The results revealed that pulse rate-based factors remained robust with up to 75%lead loss,while block-based factors experienced significant performance declines beyond the loss of four leads.Conclusion:This analysis highlighted the importance of addressing lead loss impacts to maintain a robust model.To optimize performance,targeted training approaches were developed for different conditions.Based on these insights,a grouping strategy was implemented to train specialized models for pulse rate-based and block-based conditions.This approach resulted in notable improvements,achieving an overall classification accuracy of 95.12%and an F1 score of 95.79%.
