In recent years, cardiovascular disease has become a common disease. With the development of machine learning and big data technologies, the processing ability of electrocardiogram (ECG) signals has been greatly enhanced through new computer technologies, enabling the auxiliary diagnosis technology for cardiovascular disease (CVD) to achieve new improvements. This article discusses the application of machine learning in ECG processing, especially in the auxiliary diagnosis of diseases. Firstly, the conventional signal preprocessing methods are introduced, and then the EEG signal processing methods based on feature extraction and fuzzy classification are explored. Secondly, the application of auxiliary diagnosis in CVD is further summarized. Finally, the advantages and disadvantages of the two methods are analyzed, and based on this, a design of an auxiliary diagnostic system compatible with the two methods is proposed, providing a new perspective for similar applied researches in the future.
Machine Learning ; Cardiovascular Diseases/diagnosis* ; Humans ; Electrocardiography ; Signal Processing, Computer-Assisted ; Diagnosis, Computer-Assisted ; Fuzzy Logic ; Electroencephalography

Hemodynamic monitoring can reflect cardiac function and blood perfusion and is an indispensable monitoring method in clinical practice. Invasive hemodynamic monitoring methods represented by the thermodilution method are limited in their clinical application scope because they require vascular cannulation. Non-invasive hemodynamic monitoring has attracted extensive attention from medical companies and clinicians at home and abroad in recent years due to its advantages such as safety, non-invasiveness, continuous monitoring, simple operation, and low cost. This paper designs a non-invasive hemodynamic monitoring system based on the impedance cardiography, including hardware, algorithm, software design, and performance parameter evaluation. Among them, the hardware part mainly includes a differential high-frequency constant current source stimulation circuit, impedance cardiogram signal acquisition, and ECG signal acquisition circuit. Signal processing includes wave filtering, impedance cardiogram signal calibration, and ECG signal and impedance cardiogram signal feature point recognition. According to the collected impedance cardiogram and ECG signals, hemodynamic parameters such as heart rate (HR), stroke volume (SV), cardiac output (CO), stroke index (SI), cardiac index (CI), and cardiac contractility index (ICON) are calculated based on the Nyboer thoracic cylinder model. After testing, the key technical indicators of the system hardware are better than that of the relevant medical device standards. The system was used to collect impedance cardiogram and ECG signal data from 40 volunteers. The calculated HR, SV, and CO, three important hemodynamic indicators, were compared with the ICONCore non-invasive cardiac output monitor of OSYPKA Medical in Germany. Their Pearson correlation coefficients were 0.992 ( P<0.001), 0.948 ( P<0.001), and 0.933 ( P<0.001), respectively, verifying that the designed system has high accuracy and reliability.
Cardiography, Impedance/methods* ; Humans ; Hemodynamic Monitoring/methods* ; Equipment Design ; Signal Processing, Computer-Assisted ; Hemodynamics ; Algorithms ; Monitoring, Physiologic/methods* ; Electrocardiography

OBJECTIVE: To investigate cisplatin-induced cardiovascular toxicity and explore the protective effects and potential mechanism of curcumin co-treatment. METHODS: Forty adult male Sprague-Dawley rats were numbered and randomly divided into control group, cisplatin group (7.5 mg/kg, once a week, for 2 weeks), curcumin group (200 mg/kg per day, for 2 weeks) and cisplatin+curcumin group (cisplatin 7.5 mg/kg, once a week, and curcumin 200 mg/kg per day for 2 weeks) by a random number table method, with 10 rats in each group. Cardiac and vascular morphology and functions were assessed using hematoxylin-eosin and Masson's trichrome staining, serum indexes detection, echocardiography, electrocardiogram (ECG), blood pressure monitoring, vascular ring isometric tension measurement, and left ventricular pressure evaluation. The expressions of extracellular signal-regulated kinases (ERK) and phosphorylated-ERK (p-ERK) were analyzed by immunohistochemical staining. RESULTS: Cisplatin treatment induced notable cardiac alteration, as evidenced by changes in cardiac morphology, elevated serum enzymes (P<0.05), ECG abnormalities, and increased left ventricular end-diastolic pressure (P<0.05). Meanwhile, cisplatin significantly increased arterial pulse pressure (P<0.01), primarily due to a decrease in diastolic blood pressure. Severe fibrosis was also observed in the thoracic aorta wall. In vascular ring experiments, cisplatin treatment led to a significant reduction in phenylephrine-induced contraction (P<0.05) and acetylcholine-induced relaxation (P<0.01). Notably, Curcumin co-administration significantly alleviated cisplatin-induced cardiovascular damages, as demonstrated by improvement in these parameters. Furthermore, ERK expression in the myocardium and p-ERK expression in vascular smooth muscle cells were significantly upregulated following curcumin co-treatment. CONCLUSIONS Curcumin protects the heart and vasculature from cisplatin-induced damages, likely by upregulating ERK/p-ERK expression. These findings suggest that curcumin may serve as a promising therapeutic strategy for mitigating cisplatin-associated cardiovascular toxicity during tumor chemotherapy. In vitro cell culture experiments are needed to clarify the underlying mechanism.
Animals ; Curcumin/therapeutic use* ; Cisplatin/adverse effects* ; Rats, Sprague-Dawley ; Male ; Up-Regulation/drug effects* ; Extracellular Signal-Regulated MAP Kinases/metabolism* ; Phosphorylation/drug effects* ; Electrocardiography ; Blood Pressure/drug effects* ; Rats ; MAP Kinase Signaling System/drug effects*

OBJECTIVES: To develop an early atrial fibrillation (AF) risk prediction model based on large-scale electrocardiogram (ECG) data from the Chinese population. METHODS: The data of multiple ECG records of 30 383 patients admitted in the Chinese PLA General Hospital between 2009 and 2023 were randomly divided into the training set and the internal testing set in a 7:3 ratio. The predictive factors were selected based on the training set using univariate analysis, LASSO regression, and the Boruta algorithm. Cox proportional hazards regression was used to establish the ECG model and the composite model incorporating age, gender, and ECG model score. The discrimination power, calibration, and clinical net benefits of the models were evaluated using the area under the receiver operating characteristic curve (AUROC), calibration curves, and decision curves. RESULTS: The cohort included 51.1% male patients with a median age of the patients of 51 (36, 62) years and an AF incidence of 4.5% (1370/30 383). In the ECG model, the parameters related to the P wave and QRS complex were identified as significant predictors. In the testing set, the AUROC of the ECG model for predicting 5-year AF risk was 0.77 (95% CI: 0.74-0.80), which was increased to 0.81 (95% CI: 0.78-0.83) after incorporating age and gender, with a net reclassification improvement of 0.123 and an integrated discrimination improvement of 0.04 (P<0.05). The calibration curve of the model was close to the diagonal line. Decision curve analysis showed that the clinical net benefit of the composite model was higher than that of the ECG model across the majority of threshold probability. CONCLUSIONS The composite model incorporating quantitative ECG features during sinus rhythm, along with age and gender, can effectively predict AF risk in the Chinese population, thus providing a low-cost screening tool for early AF risk assessment and management.
Humans ; Atrial Fibrillation/epidemiology* ; Electrocardiography ; Middle Aged ; Male ; Female ; China/epidemiology* ; Proportional Hazards Models ; Adult ; Risk Factors ; Risk Assessment ; East Asian People

OBJECTIVES: To design a deep learning model that balances model complexity and performance to enable its integration into wearable ECG monitoring devices for automated diagnosis of atrial fibrillation. METHODS: This study was performed based on data from 84 patients with atrial fibrillation, 25 patients with atrial fibrillation, and 18 subjects without obvious arrhythmia collected from the publicly available datasets LTAFDB, AFDB, and NSRDB, respectively. A lightweight attention network based on depthwise separable convolution and fusion of channel-spatial information, namely DSC-AttNet, was proposed. Depthwise separable convolution was introduced to replace standard convolution and reduce model parameters and computational complexity to realize high efficiency and light weight of the model. The multilayer hybrid attention mechanism was embedded to compute the attentional weights of the channels and spatial information at different scales to improve the feature expression ability of the model. Ten-fold cross-validation was performed on LTAFDB, and external independent testing was conducted on AFDB and NSRDB datasets. RESULTS: DSC-AttNet achieved a ten-fold average accuracy of 97.33% and a precision of 97.30% on the test set, both of which outperformed the other 4 comparison models as well as the 3 classical models. The accuracy of the model on the external test set reached 92.78%, better than those of the 3 classical models. The number of parameters of DSC-AttNet was 1.01M, and the computational volume was 27.19G, both smaller than the 3 classical models. CONCLUSIONS This proposed method has a smaller complexity, achieves better classification performance, and has a better generalization ability for atrial fibrillation classification.
Atrial Fibrillation/diagnosis* ; Humans ; Electrocardiography ; Deep Learning ; Wearable Electronic Devices ; Neural Networks, Computer

OBJECTIVES: We propose a myocardial infarction (MI) detection and localization model for improving the diagnostic accuracy for MI to provide assistance to clinical decision-making. METHODS: The proposed model was constructed based on multi-scale field residual blocks fusion modified channel attention (MSF-RB-MCA). The model utilizes lead II electrocardiogram (ECG) signals to detect and localize MI, and extracts different levels of feature information through the multi-scale field residual block. A modified channel attention for automatic adjustment of the feature weights was introduced to enhance the model's ability to focus on the MI region, thereby improving the accuracy of MI detection and localization. RESULTS: A 5-fold cross-validation test of the model was performed using the publicly available Physikalisch-Technische Bundesanstalt (PTB) dataset. For MI detection, the model achieved an accuracy of 99.96% on the test set with a specificity of 99.84% and a sensitivity of 99.99%. For MI localization, the accuracy, specificity and sensitivity were 99.81%, 99.98% and 99.65%, respectively. The performances of the model for MI detection and localization were superior to those of other comparison models. CONCLUSIONS The proposed MSF-RB-MCA model shows excellent performance in AI detection and localization based on lead II ECG signals, demonstrating its great potential for application in wearable devices.
Myocardial Infarction/diagnosis* ; Humans ; Electrocardiography/methods* ; Signal Processing, Computer-Assisted ; Algorithms ; Sensitivity and Specificity

INTRODUCTION: Elevated troponin, while essential for diagnosing myocardial infarction, can also be present in non-myocardial infarction conditions. The myocardial-ischaemic-injury-index (MI3) algorithm is a machine learning algorithm that considers age, sex and cardiac troponin I (TnI) results to risk-stratify patients for type 1 myocardial infarction. METHOD: Patients aged ≥25 years who presented to the emergency department (ED) of Singapore General Hospital with symptoms suggestive of acute coronary syndrome with no diagnostic 12-lead electrocardiogram (ECG) changes were included. Participants had serial ECGs and high-sensitivity troponin assays performed at 0, 2 and 7 hours. The primary outcome was the adjudicated diagnosis of type 1 myocardial infarction at 30 days. We compared the performance of MI3 in predicting the primary outcome with the European Society of Cardiology (ESC) 0/2-hour algorithm as well as the 99th percentile upper reference limit (URL) for TnI. RESULTS: There were 1351 patients included (66.7% male, mean age 56 years), 902 (66.8%) of whom had only 0-hour troponin results and 449 (33.2%) with serial (both 0 and 2-hour) troponin results available. MI3 ruled out type 1 myocardial infarction with a higher sensitivity (98.9, 95% confidence interval [CI] 93.4-99.9%) and similar negative predictive value (NPV) 99.8% (95% CI 98.6-100%) as compared to the ESC strategy. The 99th percentile cut-off strategy had the lowest sensitivity, specificity, positive predictive value and NPV. CONCLUSION The MI3 algorithm was accurate in risk stratifying ED patients for myocardial infarction. The 99th percentile URL cut-off was the least accurate in ruling in and out myocardial infarction compared to the other strategies.
Humans ; Male ; Female ; Emergency Service, Hospital ; Middle Aged ; Electrocardiography ; Machine Learning ; Singapore ; Chest Pain/blood* ; Troponin I/blood* ; Myocardial Infarction/blood* ; Risk Assessment/methods* ; Aged ; Algorithms ; Acute Coronary Syndrome/blood* ; Adult ; Sensitivity and Specificity

Cardiovascular diseases and psychological disorders represent two major threats to human physical and mental health. Research on electrocardiogram (ECG) signals offers valuable opportunities to address these issues. However, existing methods are constrained by limitations in understanding ECG features and transferring knowledge across tasks. To address these challenges, this study developed a multi-resolution feature encoding network based on residual networks, which effectively extracted local morphological features and global rhythm features of ECG signals, thereby enhancing feature representation. Furthermore, a model compression-based continual learning method was proposed, enabling the structured transfer of knowledge from simpler tasks to more complex ones, resulting in improved performance in downstream tasks. The multi-resolution learning model demonstrated superior or comparable performance to state-of-the-art algorithms across five datasets, including tasks such as ECG QRS complex detection, arrhythmia classification, and emotion classification. The continual learning method achieved significant improvements over conventional training approaches in cross-domain, cross-task, and incremental data scenarios. These results highlight the potential of the proposed method for effective cross-task knowledge transfer in ECG analysis and offer a new perspective for multi-task learning using ECG signals.
Humans ; Electrocardiography/methods* ; Mental Health ; Algorithms ; Signal Processing, Computer-Assisted ; Machine Learning ; Arrhythmias, Cardiac/diagnosis* ; Cardiovascular Diseases ; Neural Networks, Computer ; Mental Disorders

Deep learning method can be used to automatically analyze electrocardiogram (ECG) data and rapidly implement arrhythmia classification, which provides significant clinical value for the early screening of arrhythmias. How to select arrhythmia features effectively under limited abnormal sample supervision is an urgent issue to address. This paper proposed an arrhythmia classification algorithm based on an adaptive multi-feature fusion network. The algorithm extracted RR interval features from ECG signals, employed one-dimensional convolutional neural network (1D-CNN) to extract time-domain deep features, employed Mel frequency cepstral coefficients (MFCC) and two-dimensional convolutional neural network (2D-CNN) to extract frequency-domain deep features. The features were fused using adaptive weighting strategy for arrhythmia classification. The paper used the arrhythmia database jointly developed by the Massachusetts Institute of Technology and Beth Israel Hospital (MIT-BIH) and evaluated the algorithm under the inter-patient paradigm. Experimental results demonstrated that the proposed algorithm achieved an average precision of 75.2%, an average recall of 70.1% and an average F ₁-score of 71.3%, demonstrating high classification accuracy and being able to provide algorithmic support for arrhythmia classification in wearable devices.
Humans ; Arrhythmias, Cardiac/diagnosis* ; Algorithms ; Electrocardiography/methods* ; Neural Networks, Computer ; Signal Processing, Computer-Assisted ; Deep Learning ; Classification Algorithms

Heart rate variability time and frequency indices are widely used in functional assessment for autonomic nervous system (ANS). However, this method merely analyzes the effect of cardiac dynamics, overlooking the effect of cardio-pulmonary interplays. Given this, the present study proposes a novel cardiopulmonary coupling (CPC) algorithm based on cross-wavelet transform to quantify cardio-pulmonary interactions, and establish an assessment system for ANS aging effects using wearable electrocardiogram (ECG) and respiratory monitoring devices. To validate the superiority of the proposed method under nonstationary and low signal-to-noise ratio conditions, simulations were first conducted to demonstrate the performance strength of the proposed method to the traditional one. Next, the proposed CPC algorithm was applied to analyze cardiac and respiratory data from both elderly and young populations, revealing that young populations exhibited significantly stronger couplings in the high-frequency band compared with their elderly counterparts. Finally, a CPC assessment system was constructed by integrating wearable devices, and additional recordings from both elderly and young populations were collected by using the system, completing the validation and application of the aging effect assessment algorithm and the wearable system. In conclusion, this study may offers methodological and system support for assessing the aging effects on the ANS.
Humans ; Autonomic Nervous System/physiology* ; Algorithms ; Aging/physiology* ; Electrocardiography/methods* ; Heart Rate/physiology* ; Wavelet Analysis ; Aged ; Signal Processing, Computer-Assisted ; Wearable Electronic Devices