Detection of atrial fibrillation from ECG recordings using decision tree ensemble with multi-level features

Classification Method of ECG Signals Based on RANet

BACKGROUND

Electrocardiograms (ECG) are an important source of information on human heart health and are widely used to detect different types of arrhythmias.

OBJECTIVE

With the advancement of deep learning, end-to-end ECG classification models based on neural networks have been developed. However, deeper network layers lead to gradient vanishing. Moreover, different channels and periods of an ECG signal hold varying significance for identifying different types of ECG abnormalities.

METHODS

To solve these two problems, an ECG classification method based on a residual attention neural network is proposed in this paper. The residual network (ResNet) is used to solve the gradient vanishing problem. Moreover, it has fewer model parameters, and its structure is simpler. An attention mechanism is added to focus on key information, integrate channel features, and improve voting methods to alleviate the problem of data imbalance.

RESULTS

Experiments and verifications are conducted using the PhysioNet/CinC Challenge 2017 dataset. The average F1 value is 0.817, which is 0.064 higher than that for the ResNet model. Compared with the mainstream methods, the performance is excellent.

Multifractal detrended fluctuation analysis of insole pressure sensor data to diagnose vestibular system disorders

The vestibular system (VS) is a sensory system that has a vital function in human life by serving to maintain balance. In this study, multifractal detrended fluctuation analysis (MFDFA) is applied to insole pressure sensor data collected from subjects in order to extract features to identify diseases related to VS dysfunction. We use the multifractal spectrum width as the feature to distinguish between healthy and diseased people. It is observed that multifractal behavior is more dominant and thus the spectrum is wider for healthy subjects, where we explain the reason as the long-range correlations of the small and large fluctuations of the time series for this group. We directly process the instantaneous pressure values to extract features in contrast to studies in the literature where gait analysis is based on investigation of gait dynamics (stride time, stance time, etc.) requiring long walking time. Thus, as the main innovation of this work, we detrend the data to give meaningful information even for a relatively short walk. Extracted feature set was input to fundamental classification algorithms where the Support-Vector-Machine (SVM) performed best with an average accuracy of 98.2% for the binary classification as healthy or suffering. This study is a substantial part of a big project where we finally aim to identify the specific VS disease that causes balance disorder and also determine the stage of the disease, if any. Within this scope, the achieved performance gives high motivation to work more deeply on the issue.

Assessing the Generalizability of a Deep Learning-based Automated Atrial Fibrillation Algorithm

Automated detection of atrial fibrillation (AF) from electrocardiogram (ECG) traces remains a challenging task and is crucial for telemonitoring of patients after stroke. This study aimed to quantify the generalizability of a deep learning (DL)-based automated ECG classification algorithm. We first developed a novel hybrid DL (HDL) model using the PhysioNet/CinC Challenge 2017 (CinC2017) dataset (publicly available) that can classify the ECG recordings as one of four classes: normal sinus rhythm (NSR), AF, other rhythms (OR), and too noisy (TN) recordings. The (pre)trained HDL was then used to classify 636 ECG samples collected by our research team using a handheld ECG device, CONTEC PM10 Portable ECG Monitor, from 102 (age: 68 ± 15 years, 74 male) outpatients of the Eastern Heart Clinic and inpatients in the Cardiology ward of Prince of Wales Hospital, Sydney, Australia. The proposed HDL model achieved average test F1-score of 0.892 for NSR, AF, and OR, relative to the reference values, on the CinC2017 dataset. The HDL model also achieved an average F1-score of 0.722 (AF: 0.905, NSR: 0.791, OR: 0.471 and TN: 0.342) on the dataset created by our research team. After retraining the HDL model on this dataset using a 5-fold cross validation method, the average F1-score increased to 0.961. We finally conclude that the generalizability of the HDL-based algorithm developed for AF detection from short-term single-lead ECG traces is acceptable. However, the accuracy of the pre-trained DL model was significantly improved by retraining the model parameters on the new dataset of ECG traces.

Artificial intelligence in cardiovascular prevention: new ways will open new doors

Prevention and effective treatment of cardiovascular disease are progressive issues that grow in tandem with the average age of the world population. Over recent decades, the potential role of artificial intelligence in cardiovascular medicine has been increasingly recognized because of the incredible amount of real-world data (RWD) regarding patient health status and healthcare delivery that can be collated from a variety of sources wherein patient information is routinely collected, including patient registries, clinical case reports, reimbursement claims and billing reports, medical devices, and electronic health records. Like any other (health) data, RWD can be analysed in accordance with high-quality research methods, and its analysis can deliver valuable patient-centric insights complementing the information obtained from conventional clinical trials. Artificial intelligence application on RWD has the potential to detect a patient's health trajectory leading to personalized medicine and tailored treatment. This article reviews the benefits of artificial intelligence in cardiovascular prevention and management, focusing on diagnostic and therapeutic improvements without neglecting the limitations of this new scientific approach.

Wearable Devices Combined with Artificial Intelligence-A Future Technology for Atrial Fibrillation Detection?

Atrial fibrillation (AF) is the most common cardiac arrhythmia in the world. The arrhythmia and methods developed to cure it have been studied for several decades. However, professionals worldwide are still working to improve treatment quality. One novel technology that can be useful is a wearable device. The two most used recordings from these devices are photoplethysmogram (PPG) and electrocardiogram (ECG) signals. As the price lowers, these devices will become significant technology to increase sensitivity, for monitoring and for treatment quality support. This is important as AF can be challenging to detect in advance, especially during home monitoring. Modern artificial intelligence (AI) has the potential to respond to this challenge. AI has already achieved state of the art results in many applications, including bioengineering. In this perspective, we discuss wearable devices combined with AI for AF detection, an approach that enables a new era of possibilities for the future.

Applying Artificial Intelligence to Wearable Sensor Data to Diagnose and Predict Cardiovascular Disease: A Review

Cardiovascular disease (CVD) is the world's leading cause of mortality. There is significant interest in using Artificial Intelligence (AI) to analyse data from novel sensors such as wearables to provide an earlier and more accurate prediction and diagnosis of heart disease. Digital health technologies that fuse AI and sensing devices may help disease prevention and reduce the substantial morbidity and mortality caused by CVD worldwide. In this review, we identify and describe recent developments in the application of digital health for CVD, focusing on AI approaches for CVD detection, diagnosis, and prediction through AI models driven by data collected from wearables. We summarise the literature on the use of wearables and AI in cardiovascular disease diagnosis, followed by a detailed description of the dominant AI approaches applied for modelling and prediction using data acquired from sensors such as wearables. We discuss the AI algorithms and models and clinical applications and find that AI and machine-learning-based approaches are superior to traditional or conventional statistical methods for predicting cardiovascular events. However, further studies evaluating the applicability of such algorithms in the real world are needed. In addition, improvements in wearable device data accuracy and better management of their application are required. Lastly, we discuss the challenges that the introduction of such technologies into routine healthcare may face.

Atrioventricular Synchronization for Detection of Atrial Fibrillation and Flutter in One to Twelve ECG Leads Using a Dense Neural Network Classifier

This study investigates the use of atrioventricular (AV) synchronization as an important diagnostic criterion for atrial fibrillation and flutter (AF) using one to twelve ECG leads. Heart rate, lead-specific AV conduction time, and P-/f-wave amplitude were evaluated by three representative ECG metrics (mean value, standard deviation), namely RR-interval (RRi-mean, RRi-std), PQ-interval (PQi-mean, PQI-std), and PQ-amplitude (PQa-mean, PQa-std), in 71,545 standard 12-lead ECG records from the six largest PhysioNet CinC Challenge 2021 databases. Two rhythm classes were considered (AF, non-AF), randomly assigning records into training (70%), validation (20%), and test (10%) datasets. In a grid search of 19, 55, and 83 dense neural network (DenseNet) architectures and five independent training runs, we optimized models for one-lead, six-lead (chest or limb), and twelve-lead input features. Lead-set performance and SHapley Additive exPlanations (SHAP) input feature importance were evaluated on the test set. Optimal DenseNet architectures with the number of neurons in sequential [1st, 2nd, 3rd] hidden layers were assessed for sensitivity and specificity: DenseNet [16,16,0] with primary leads (I or II) had 87.9-88.3 and 90.5-91.5%; DenseNet [32,32,32] with six limb leads had 90.7 and 94.2%; DenseNet [32,32,4] with six chest leads had 92.1 and 93.2%; and DenseNet [128,8,8] with all 12 leads had 91.8 and 95.8%, indicating sensitivity and specificity values, respectively. Mean SHAP values on the entire test set highlighted the importance of RRi-mean (100%), RR-std (84%), and atrial synchronization (40-60%) for the PQa-mean (aVR, I), PQi-std (V2, aVF, II), and PQi-mean (aVL, aVR). Our focus on finding the strongest AV synchronization predictors of AF in 12-lead ECGs would lead to a comprehensive understanding of the decision-making process in advanced neural network classifiers. DenseNet self-learned to rely on a few ECG behavioral characteristics: first, characteristics usually associated with AF conduction such as rapid heart rate, enhanced heart rate variability, and large PQ-interval deviation in V2 and inferior leads (aVF, II); second, characteristics related to a typical P-wave pattern in sinus rhythm, which is best distinguished from AF by the earliest negative P-peak deflection of the right atrium in the lead (aVR) and late positive left atrial deflection in lateral leads (I, aVL). Our results on lead-selection and feature-selection practices for AF detection should be considered for one- to twelve-lead ECG signal processing settings, particularly those measuring heart rate, AV conduction times, and P-/f-wave amplitudes. Performances are limited to the AF diagnostic potential of these three metrics. SHAP value importance can be used in combination with a human expert's ECG interpretation to change the focus from a broad observation of 12-lead ECG morphology to focusing on the few AV synchronization findings strongly predictive of AF or non-AF arrhythmias. Our results are representative of AV synchronization findings across a broad taxonomy of cardiac arrhythmias in large 12-lead ECG databases.

Cardiovascular Diseases in the Digital Health Era: A Translational Approach from the Lab to the Clinic

Translational science has been introduced as the nexus among the scientific and the clinical field, which allows researchers to provide and demonstrate that the evidence-based research can connect the gaps present between basic and clinical levels. This type of research has played a major role in the field of cardiovascular diseases, where the main objective has been to identify and transfer potential treatments identified at preclinical stages into clinical practice. This transfer has been enhanced by the intromission of digital health solutions into both basic research and clinical scenarios. This review aimed to identify and summarize the most important translational advances in the last years in the cardiovascular field together with the potential challenges that still remain in basic research, clinical scenarios, and regulatory agencies.

Artificial Intelligence-Driven Algorithm for Drug Effect Prediction on Atrial Fibrillation: An in silico Population of Models Approach

Background: Antiarrhythmic drugs are the first-line treatment for atrial fibrillation (AF), but their effect is highly dependent on the characteristics of the patient. Moreover, anatomical variability, and specifically atrial size, have also a strong influence on AF recurrence. Objective: We performed a proof-of-concept study using artificial intelligence (AI) that enabled us to identify proarrhythmic profiles based on pattern identification from in silico simulations. Methods: A population of models consisting of 127 electrophysiological profiles with a variation of nine electrophysiological variables (G _Na , I _NaK , G _K1, G _CaL , G _Kur , I _KCa , [Na] _ext , and [K] _ext and diffusion) was simulated using the Koivumaki atrial model on square planes corresponding to a normal (16 cm²) and dilated (22.5 cm²) atrium. The simple pore channel equation was used for drug implementation including three drugs (isoproterenol, flecainide, and verapamil). We analyzed the effect of every ionic channel combination to evaluate arrhythmia induction. A Random Forest algorithm was trained using the population of models and AF inducibility as input and output, respectively. The algorithm was trained with 80% of the data (N = 832) and 20% of the data was used for testing with a k-fold cross-validation (k = 5). Results: We found two electrophysiological patterns derived from the AI algorithm that was associated with proarrhythmic behavior in most of the profiles, where G _K1 was identified as the most important current for classifying the proarrhythmicity of a given profile. Additionally, we found different effects of the drugs depending on the electrophysiological profile and a higher tendency of the dilated tissue to fibrillate (Small tissue: 80 profiles vs Dilated tissue: 87 profiles). Conclusion: Artificial intelligence algorithms appear as a novel tool for electrophysiological pattern identification and analysis of the effect of antiarrhythmic drugs on a heterogeneous population of patients with AF.

Considerations on Performance Evaluation of Atrial Fibrillation Detectors

OBJECTIVE

A large number of atrial fibrillation (AF) detectors have been published in recent years, signifying that the comparison of detector performance plays a central role, though not always consistent. The aim of this study is to shed needed light on aspects crucial to the evaluation of detection performance.

METHODS

Three types of AF detector, using either information on rhythm, rhythm and morphology, or segments of ECG samples, are implemented and studied on both real and simulated ECG signals. The properties of different performance measures are investigated, for example, in relation to dataset imbalance.

RESULTS

The results show that performance can differ considerably depending on the way detector output is compared to database annotations, i.e., beat-to-beat, segment-to-segment, or episode-to-episode comparison. Moreover, depending on the type of detector, the results substantiate that physiological and technical factors, e.g., changes in ECG morphology, rate of atrial premature beats, and noise level, can have a considerable influence on performance.

CONCLUSION

The present study demonstrates overall strengths and weaknesses of different types of detector, highlights challenges in AF detection, and proposes five recommendations on how to handle data and characterize performance.

Application of Dense Neural Networks for Detection of Atrial Fibrillation and Ranking of Augmented ECG Feature Set

Considering the significant burden to patients and healthcare systems globally related to atrial fibrillation (AF) complications, the early AF diagnosis is of crucial importance. In the view of prominent perspectives for fast and accurate point-of-care arrhythmia detection, our study optimizes an artificial neural network (NN) classifier and ranks the importance of enhanced 137 diagnostic ECG features computed from time and frequency ECG signal representations of short single-lead strips available in 2017 Physionet/CinC Challenge database. Based on hyperparameters' grid search of densely connected NN layers, we derive the optimal topology with three layers and 128, 32, 4 neurons per layer (DenseNet-3@128-32-4), which presents maximal F1-scores for classification of Normal rhythms (0.883, 5076 strips), AF (0.825, 758 strips), Other rhythms (0.705, 2415 strips), Noise (0.618, 279 strips) and total F1 relevant to the CinC Challenge of 0.804, derived by five-fold cross-validation. DenseNet-3@128-32-4 performs equally well with 137 to 32 features and presents tolerable reduction by about 0.03 to 0.06 points for limited input sets, including 8 and 16 features, respectively. The feature reduction is linked to effective application of a comprehensive method for computation of the feature map importance based on the weights of the activated neurons through the total path from input to specific output in DenseNet. The detailed analysis of 20 top-ranked ECG features with greatest importance to the detection of each rhythm and overall of all rhythms reveals DenseNet decision-making process, noticeably corresponding to the cardiologists' diagnostic point of view.

Advances of ECG Sensors from Hardware, Software and Format Interoperability Perspectives

It is well-known that cardiovascular disease is one of the major causes of death worldwide nowadays. Electrocardiogram (ECG) sensor is one of the tools commonly used by cardiologists to diagnose and detect signs of heart disease with their patients. Since fast, prompt and accurate interpretation and decision is important in saving the life of patients from sudden heart attack or cardiac arrest, many innovations have been made to ECG sensors. However, the use of traditional ECG sensors is still prevalent in the clinical settings of many medical institutions. This article provides a comprehensive survey on ECG sensors from hardware, software and data format interoperability perspectives. The hardware perspective outlines a general hardware architecture of an ECG sensor along with the description of its hardware components. The software perspective describes various techniques (denoising, machine learning, deep learning, and privacy preservation) and other computer paradigms used in the software development and deployment for ECG sensors. Finally, the format interoperability perspective offers a detailed taxonomy of current ECG formats and the relationship among these formats. The intention is to help researchers towards the development of modern ECG sensors that are suitable and approved for adoption in real clinical settings.

Automatic screening method for atrial fibrillation based on lossy compression of the electrocardiogram signal

OBJECTIVE

Compressed sensing (CS) is a low-complexity compression technology that has recently been proposed. It can be applied to long-term electrocardiogram (ECG) monitoring using wearable devices. In this study, an automatic screening method for atrial fibrillation (AF) based on lossy compression of the electrocardiogram signal is proposed.

APPROACH

The proposed method combines the CS with the convolutional neural network (CNN). The sparse binary sensing matrix is first used to project the raw ECG signal randomly, transforming the raw ECG data from high-dimensional space to low-dimensional space to complete compression, and then using CNN to classify the compressed ECG signal involving AF. Our proposed model is validated on the MIT-BIH atrial fibrillation database.

MAIN RESULTS

The experimental results show that the model only needs about 1 s to complete the 24 h ECG recording of AF, which is 3.41%, 69.84% and 67.56% less than the time required by AlexNet, VGGNet and GoogLeNet, respectively. Under different compression ratios of 10% to 90%, the maximum and minimum F1 scores reach 96.25% and 88.17%, respectively.

SIGNIFICANCE

The CS-CNN (compressed sensing convolutional neural network) model has high computational efficiency while ensuring prediction accuracy, and is a promising method for AF screening in wearable application scenarios.

Towards the Prediction of Rearrest during Out-of-Hospital Cardiac Arrest

A secondary arrest is frequent in patients that recover spontaneous circulation after an out-of-hospital cardiac arrest (OHCA). Rearrest events are associated to worse patient outcomes, but little is known on the heart dynamics that lead to rearrest. The prediction of rearrest could help improve OHCA patient outcomes. The aim of this study was to develop a machine learning model to predict rearrest. A random forest classifier based on 21 heart rate variability (HRV) and electrocardiogram (ECG) features was designed. An analysis interval of 2 min after recovery of spontaneous circulation was used to compute the features. The model was trained and tested using a repeated cross-validation procedure, on a cohort of 162 OHCA patients (55 with rearrest). The median (interquartile range) sensitivity (rearrest) and specificity (no-rearrest) of the model were 67.3% (9.1%) and 67.3% (10.3%), respectively, with median areas under the receiver operating characteristics and the precision–recall curves of 0.69 and 0.53, respectively. This is the first machine learning model to predict rearrest, and would provide clinically valuable information to the clinician in an automated way.

A Wearable Electrocardiogram Telemonitoring System for Atrial Fibrillation Detection

In this paper we proposed a wearable electrocardiogram (ECG) telemonitoring system for atrial fibrillation (AF) detection based on a smartphone and cloud computing. A wearable ECG patch was designed to collect ECG signals and send the signals to an Android smartphone via Bluetooth. An Android APP was developed to display the ECG waveforms in real time and transmit every 30 s ECG data to a remote cloud server. A machine learning (CatBoost)-based ECG classification method was proposed to detect AF in the cloud server. In case of detected AF, the cloud server pushed the ECG data and classification results to the web browser of a doctor. Finally, the Android APP displayed the doctor's diagnosis for the ECG signals. Experimental results showed the proposed CatBoost classifier trained with 17 selected features achieved an overall F1 score of 0.92 on the test set (n = 7,270). The proposed wearable ECG monitoring system may potentially be useful for long-term ECG telemonitoring for AF detection.