Ambulatory ECG Signals Research Articles

Optimising and validating deep learning approaches for diagnosing atrial fibrillation from few-lead ambulatory electrocardiogram signals

Abstract Funding Acknowledgements Type of funding sources: None. Background Deep learning (DL) has received much attention as a solution for automatically diagnosing atrial fibrillation (AF) from raw ECG signals. However, few studies exist to investigate how DL approaches can be optimally configured and whether their diagnostic performance is externally validated. Purpose To explore how signal-related parameter tuning affects the ability of DL approaches to diagnose AF and validate the optimal approach internally and externally. Methods We applied two dedicated DL models (InceptionTime and MINIROCKET) on a set of 7,966 AF and non-AF (normal or with other abnormalities) ambulatory ECG samples, originating from the MIT-BIH AF, MIT-BIH Normal Sinus Rhythm and Long Term AF databases. We tested the effect of different sample lengths (30sec (s), 10s, 30/10s -30s with a "sliding window" of 10s-), sampling frequencies (200, 100, 50 Hz) and lead numbers (two-, single-), and the role of denoising (Discrete Wavelet Transformation, no denoising) on the ability to diagnose AF, by measuring ROC AUC and sensitivity (SEN) after repeated model training and testing. Under the optimal configuration, we trained 10 replicas of both models on 90% of the data and tested their performance on the remaining 10% (internal validation). Finally, we applied both pre-trained models on a separate dataset (MIT-BIH Arrhythmia) to determine their external validity. Results Although the diagnostic performance did not differ between 30s and 10s signals, the 30/10s setting displayed significantly higher median AUC (0.98) and sensitivity (97.3%, p<0.05 for all comparisons). Signals sampled at 50Hz performed poorer (AUC=0.88, SEN=79.9%) than those at 100Hz (AUC=0.92, SEN=88.7%) and 200Hz (AUC=0.93, SEN=89.2%), although this difference slightly failed to reach statistical significance. Despite denoised signals showing a higher median AUC (0.95 vs. 0.92) and sensitivity (92.8% vs. 88.7%), the difference was not found significant. Similarly, two-lead signals performed better than single-lead ones (AUC=0.92 vs. 0.9 and SEN=88.7% vs. 84.1%, respectively), but without crossing the significance threshold. The internal validation with denoised, 30/10s, two-lead signals, at 100Hz, yielded similarly high performance metrics for both InceptionTime and MINIROCKET (AUC=0.98, SEN=96.9% and AUC=0.98, SEN=97.4%, respectively). In contrast, the performance on the external set dropped significantly (AUC=0.79, SEN=81.4% and AUC=0.72, SEN=83.7%, respectively, p<0.001 for all comparisons). Conclusions Both DL approaches can effectively detect AF in ambulatory ECG signals, with only 3 out of 100 cases missed, designating their promising utility as screening tools for automated AF detection. While optimising tunable parameters can enhance the internal performance of such efforts, their external validation is necessary to establish their robustness "in the wild", since their performance on "unseen" data can be, similarly to our case, notably lower.

Applications of Machine Learning in Ambulatory ECG

The ambulatory ECG (AECG) is an important diagnostic tool for many heart electrophysiology-related cases. AECG covers a wide spectrum of devices and applications. At the core of these devices and applications are the algorithms responsible for signal conditioning, ECG beat detection and classification, and event detections. Over the years, there has been huge progress for algorithm development and implementation thanks to great efforts by researchers, engineers, and physicians, alongside the rapid development of electronics and signal processing, especially machine learning (ML). The current efforts and progress in machine learning fields are unprecedented, and many of these ML algorithms have also been successfully applied to AECG applications. This review covers some key AECG applications of ML algorithms. However, instead of doing a general review of ML algorithms, we are focusing on the central tasks of AECG and discussing what ML can bring to solve the key challenges AECG is facing. The center tasks of AECG signal processing listed in the review include signal preprocessing, beat detection and classification, event detection, and event prediction. Each AECG device/system might have different portions and forms of those signal components depending on its application and the target, but these are the topics most relevant and of greatest concern to the people working in this area.

CEEMDAN-IMFx-PCA-CICA: an improved single-channel blind source separation in multimedia environment for motion artifact reduction in ambulatory ECG

Long-term monitoring of ECG via wearable monitoring systems has already been widely adopted to detect and prevent heart diseases. However, one of the main issues faced by wearable ECG monitoring systems is that motion artifacts significantly affect the systems' stability and reliability. Therefore, motion artifact reduction is a very challenging task in filtering and processing physiological signals. Based on the existing algorithms and ECG prior knowledge, in this paper, we propose an algorithm, CEEMDAN-IMFx-PCA-CICA, for motion artifact reduction in ambulatory ECG signals using single-channel blind source separation technique. Our algorithm first utilizes CEEMDAN to decompose the mixed signals into IMFs (intrinsic mode function) containing different source signal features, thereby forming new multi-dimensional signals. Using the correlation between IMFx (IMF component with the most ECG features) and each IMF, and PCA are then applied to reduce the dimension of each IMF. Finally, the blind separation of the source ECG signals is achieved by using CICA with IMFx as the constraint reference component. The results of our experiments indicate that our algorithm outperformed CEEMDAN-CICA, CEEMDAN-PCA-CICA, and improved CEEMDAN-PCA-CICA. Besides, the number of iterations of the CICA is significantly reduced; the separated source signal is better; the obtained result is stable. Furthermore, the separated ECG signal has a higher correlation with the source ECG signal and a lower RRMSE, especially in the case of high noise-to-signal ratios.

Artefact detection and quality assessment of ambulatory ECG signals

Background and ObjectivesThe presence of noise sources could reduce the diagnostic capability of the ECG signal and result in inappropriate treatment decisions. To mitigate this problem, automated algorithms to detect artefacts and quantify the quality of the recorded signal are needed. In this study we present an automated method for the detection of artefacts and quantification of the signal quality. The suggested methodology extracts descriptive features from the autocorrelation function and feeds these to a RUSBoost classifier. The posterior probability of the clean class is used to create a continuous signal quality assessment index. Firstly, the robustness of the proposed algorithm is investigated and secondly, the novel signal quality assessment index is evaluated. MethodsData were used from three different studies: a Sleep study, the PhysioNet 2017 Challenge and a Stress study. Binary labels, clean or contaminated, were available from different annotators with experience in ECG analysis. Two types of realistic ECG noise from the MIT-BIH Noise Stress Test Database (NSTDB) were added to the Sleep study to test the quality index. Firstly, the model was trained on the Sleep dataset and subsequently tested on a subset of the other two datasets. Secondly, all recording conditions were taken into account by training the model on a subset derived from the three datasets. Lastly, the posterior probabilities of the model for the different levels of agreement between the annotators were compared. ResultsAUC values between 0.988 and 1.000 were obtained when training the model on the Sleep dataset. These results were further improved when training on the three datasets and thus taking all recording conditions into account. A Pearson correlation coefficient of 0.8131 was observed between the score of the clean class and the level of agreement. Additionally, significant quality decreases per noise level for both types of added noise were observed. ConclusionsThe main novelty of this study is the new approach to ECG signal quality assessment based on the posterior clean class probability of the classifier.

A Digital Filtering Method for Eliminating DC Offset in Ambulatory ECG Signals

The amplifier is easy to saturation because of polarization voltage in ambulatory ECG acquisition. The traditional way is using analog high-pass filter to eliminate, but the output tends to have a residue. If upgrading high-pass filter cutoff frequency, it will lead to low frequency distortion of ECG signals. In this paper, a Savitzky-Golay (SG) smoothing filter has been designed by combining the single edge point M and the polynomial order p, which can fit the polarization voltage components of ECG signals, filter useful components and get drift-free ECG signals by using the subtraction algorithm. The results of ECG filtering experiment verify the feasibility of the SG smoothing filter, and show the filtered ECG signal without any losses of useful components.

Atrial fibrillation frequency tracking in ambulatory ECG signals: The significance of signal quality assessment

An approach to atrial fibrillation (AF) frequency tracking in long-term ambulatory ECG recordings is presented, comprising f-wave extraction, dominant atrial frequency (DAF) tracking, and signal quality assessment. Since poor signal quality is commonly encountered in ambulatory monitoring, a recently proposed index is employed to assess f-wave signal quality in a database containing 38 patients with permanent AF. The index ensures that DAF outliers, typically associated with poor-quality segments, are excluded from further analysis. 40% of all 5-s signal segments were excluded from the database due to poor quality. The exclusion of DAF outliers significantly reduces the standard deviation of the frequency estimates (p≤0.01), allowing more reliable evaluation of the difference between day- and night-time DAF. The results show that signal quality assessment plays a central role in DAF tracking, and therefore should be employed in ambulatory monitoring.

Post-Ventricular Premature Contraction Phase Correction Improves the Predictive Value of Average T-Wave Alternans in Ambulatory ECG Recordings.

We proposed and evaluated a method for correcting possible phase shifts provoked by the presence of ventricular premature contractions (VPCs) for a better assessment of T-wave alternans (TWA). Methods: First, we synthesized ECG signals with artificial TWA in the presence of different noise sources. Then, we assessed the prognostic value for sudden cardiac death (SCD) of the long-term average of TWA amplitude (the index of average alternans, ) in ambulatory ECG signals from congestive heart failure (CHF) and evaluated whether it is sensitive to the presence of VPCs. The inclusion of the phase correction after VPC in the processing always improved estimation accuracy of the under different noisy conditions and regardless of the number of the VPCs included in the sequence. It also presented a positive impact on the prognostic value of with increased hazard ratios (from 17% to 29%, depending of the scenario) in comparison to the noninclusion of this step. The proposed methodology for estimation, which corrects for the possible phase reversal on TWA after the presence of VPCs, represents a robust TWA estimation approach with a significant impact on the prognostic value of for SCD stratification in CHF patients. An accurate TWA estimation has a potential direct clinical impact on noninvasive SCD stratification, allowing better identification of patients at higher risk and helping clinicians in adopting the most appropriate therapeutic strategy.

Physical activities recognition from ambulatory ECG signals using neuro-fuzzy classifiers and support vector machines

The use of wearable recorders for long-term monitoring of physiological parameters has increased in the last few years. The ambulatory electrocardiogram (A-ECG) signals of five healthy subjects with four body movements or physical activities (PA)—left arm up down, right arm up down, waist twisting and walking—have been recorded using a wearable ECG recorder. The classification of these four PAs has been performed using neuro-fuzzy classifier (NFC) and support vector machines (SVM). The PA classification is based on the distinct, time-frequency features of the extracted motion artifacts contained in recorded A-ECG signals. The motion artifacts in A-ECG signals have been separated first by the discrete wavelet transform (DWT) and the time–frequency features of these motion artifacts have then been extracted using the Gabor transform. The Gabor energy feature vectors have been fed to the NFC and SVM classifiers. Both the classifiers have achieved a PA classification accuracy of over 95% for all subjects.

STUDY OF SPECTRAL CHARACTERISTICS OF MOTION ARTIFACTS IN AMBULATORY ECG BY ANALYZING PCA RESIDUAL ERRORS

In this paper, the spectral characteristics of motion artifacts occurring in an ambulatory ECG signal have been studied using principal component analysis (PCA). The PCA residual errors characterize the spectral behavior of the motion artifacts occurring in ambulatory ECG signals. The ECG signals have been acquired from Biopac MP-36 system and a self-developed wearable ECG recorder. The performance is evaluated by power spectral density (PSD) plots of PCA residual errors as well as statistical parameters like mean, median and variance of PCA errors. The PSD plots clearly indicate that the peak frequency of the motion artifacts occurring due to various body movements (like left and right arms up–down, left and right legs up–down, waist twist, walking and sitting up–down) is located around 20–25 Hz against the ECG peak frequency around 5–10 Hz.

Classification of Body Movements in Ambulatory ECG Using Wavelet Transform, Adaptive Filter and Artificial Neural Networks

Ambulatory ECG (A-ECG) monitoring provides electrical activity of the heart when a person is involved in doing normal routine activities. Thus, the recorded ECG signal consists of cardiac signal along with motion artifacts introduced due to person’s body movements during routine activities. Detection of motion artifacts due to different physical activities might help in further cardiac diagnosis. Ambulatory ECG signal analysis for detection of various body movements using Discrete Wavelet Transform (DWT) and adaptive filtering approaches has been addressed in this paper. The ECG signals of five healthy subjects (aged between 22 to 30 years) were recorded while the person performs various body movements like up and down movement of left hand, up and down movement of right hand, waist twisting movement while standing and change from sitting down on chair to standing up movement in lead I configuration using BIOPAC MP 36 data acquisition system. The features of motion artifact signal, extracted using Gabor transform, have been fed to the train the artificial neural network (ANN) for classifying body movements.

Artificial neural network-based classification of body movements in ambulatory ECG signal

Ambulatory ECG monitoring provides electrical activity of the heart when a person is involved in doing normal routine activities. Thus, the recorded ECG signal consists of cardiac signal along with motion artifacts introduced due to a person’s body movements during routine activities. Detection of motion artifacts due to different physical activities might help in further cardiac diagnosis. Ambulatory ECG signal analysis for detection of various motion artifacts using adaptive filtering approach is addressed in this paper. We have used BIOPAC MP 36 system for acquiring ECG signal. The ECG signals of five healthy subjects (aged between 22–30 years) were recorded while the person performed various body movements like up and down movement of the left hand, up and down movement of the right hand, waist twisting movement while standing and change from sitting down on a chair to standing up movement in lead I configuration. An adaptive filter-based approach has been used to extract the motion artifact component from the ambulatory ECG signal. The features of motion artifact signal, extracted using Gabor transform, have been used to train the artificial neural network (ANN) for classifying body movements.