skip to main content

Title: Automatically Detecting Arrhythmia-related Irregular Patterns using the Temporal and Spectro-Temporal Textures of ECG Signals
Arrhythmia is an abnormal heart rhythm that occurs due to the improper operation of the electrical impulses that coordinate the heartbeats. It is one of the most well-known heart conditions (including coronary artery disease, heart failure etc.) that is experienced by millions of people around the world. While there are several types of arrhythmias, not all of them are dangerous or harmful. However, there are arrhythmias that can often lead to death in minutes (e.g, ventricular fibrillation and ventricular tachycardia) even in young people. Thus, the detection of arrhythmia is critical for stopping and reversing its progression and for increasing longevity and life quality. While a doctor can perform different heart-monitoring tests specific to arrhythmias, the electrocardiogram (ECG) is one of the most common ones used either independently or in combination with other tests (to only detect, e.g. echocardiogram, or trigger arrhythmia and, then, detect, e.g. stress test). We propose a machine learning approach that augments the traditional arrhythmia detection approaches via our automatic arrhythmia classification system. It utilizes the texture of the ECG signal in both the temporal and spectro-temporal domains to detect and classify four types of heartbeats. The original ECG signal is first preprocessed, and then, the R-peaks associated with heartbeat estimation are identified. Next, 1D local binary patterns (LBP) in the temporal domain are utilized, while 2D LBPs and texture-based features extracted by a grayscale co-occurrence matrix (GLCM) are utilized in the spectro-temporal domain using the short-time Fourier transform (STFT) and Morse wavelets. Finally, different classifiers, as well as different ECG lead configurations are examined before we determine our proposed time-frequency SVM model, which obtains a maximum accuracy of 99.81%, sensitivity of 98.17%, and specificity of 99.98% when using a 10 cross-validation on the MIT-BIH database.  more » « less
Award ID(s):
Author(s) / Creator(s):
Date Published:
Journal Name:
International Conference on Pattern Recognition (ICPR)
Page Range / eLocation ID:
2301 to 2307
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Electrocardiography (ECG) is the process of recording the electrical activity of the human heart over time using electrodes that are placed over the skin. While the primary usage of electrocardiograms, the recorded signals, has been focused on the check of signs of heart-related diseases, recent studies have moved also toward their usage for human authentication. Thus, an ECG signal can be unique enough to be used independently as a biometric modality. In addition to its inherent liveness detection, it is easy to collect and can be easily captured either via sensors attached to the human body (fingertips, chest, wrist) or even passively using wireless sensors. In this paper, we propose a novel approach that exploits the spectro-temporal dynamic characteristics of the ECG signal to establish personal recognition system using both short-time Fourier transform (STFT) and generalized Morse wavelets (CWT). This process results in enriching the information extracted from the original ECG signal that is inserted in a 2D convolutional neural network (CNN) which extracts higher level and subject-specific ECG-based features for each individual. To validate our proposed CNN model, we performed nested cross-validation using eight different ECG databases. These databases are considered challenging since they include both normal and abnormal heartbeats as well as a dynamic number of subjects. Our proposed algorithms yield superior performance when compared to other state-ofart approaches discussed in the literature, i.e. the STFT-based one achieves an average identification rate, equal error rate (EER), and area under curve (AUC) of 97.86%, 0.0268, and 0.9933 respectively, whereas the CWT achieves comparable to STFT results in 97.5%, 0.0386, and 0.9882 respectively. 
    more » « less
  2. null (Ed.)
    Atrial Fibrillation (AF) is among one of the most common types of heart arrhythmia afflicting more than 3 million people in the U.S. alone. AF is estimated to be the cause of death of 1 in 4 individuals. Recent advancements in Artificial Intelligence (AI) algorithms have led to the capability of reliably detecting AF from ECG signals. While these algorithms can accurately detect AF with high precision, the discrete and deterministic classifications mean that these networks are likely to erroneously classify the given ECG signal. This paper proposes a variational autoencoder classifier network that provides an uncertainty estimation of the network's output in addition to reliable classification accuracy. This framework can increase physicians' trust in using AI-based AF detection algorithms by providing them with a confidence score which reflects how uncertain the algorithm is about a case and recommending them to put more attention to the cases with a lower confidence score. The uncertainty is estimated by conducting multiple passes of the input through the network to build a distribution; the mean of the standard deviations is reported as the network's uncertainty. Our proposed network obtains 97.64% accuracy in addition to reporting the uncertainty. 
    more » « less
  3. null (Ed.)
    Abstract Life‐threatening ventricular arrhythmias and sudden cardiac death are often preceded by cardiac alternans, a beat‐to‐beat oscillation in the T‐wave morphology or duration. However, given the spatiotemporal and structural complexity of the human heart, designing algorithms to effectively suppress alternans and prevent fatal rhythms is challenging. Recently, an antiarrhythmic constant diastolic interval pacing protocol was proposed and shown to be effective in suppressing alternans in 0‐, 1‐, and 2‐dimensional in silico studies as well as in ex vivo whole heart experiments. Herein, we provide a systematic review of the electrophysiological conditions and mechanisms that enable constant diastolic interval pacing to be an effective antiarrhythmic pacing strategy. We also demonstrate a successful translation of the constant diastolic interval pacing protocol into an ECG‐based real‐time control system capable of modulating beat‐to‐beat cardiac electrical activity and preventing alternans. Furthermore, we present evidence of the clinical utility of real‐time alternans suppression in reducing arrhythmia susceptibility in vivo. We provide a comprehensive overview of this promising pacing technique, which can potentially be translated into a clinically viable device that could radically improve the quality of life of patients experiencing abnormal cardiac rhythms. 
    more » « less
  4. Abstract

    Atrial fibrillation (AF) is the most common arrhythmia and is associated with inflammation. AF patients have elevated levels of inflammatory cytokines known to promote vascular leak, such as vascular endothelial growth factor A (VEGF). However, the contribution of vascular leak and consequent cardiac edema to the genesis of atrial arrhythmias remains unknown. Previous work suggests that interstitial edema in the heart can acutely promote ventricular arrhythmias by disrupting ventricular myocyte intercalated disk (ID) nanodomains rich in cardiac sodium channels (NaV1.5) and slowing cardiac conduction. Interestingly, similar disruption of ID nanodomains has been identified in atrial samples from AF patients. Therefore, we tested the hypothesis that VEGF-induced vascular leak can acutely increase atrial arrhythmia susceptibility by disrupting ID nanodomains and slowing atrial conduction. Treatment of murine hearts with VEGF (30–60 min, at clinically relevant levels) prolonged the electrocardiographic P wave and increased susceptibility to burst pacing-induced atrial arrhythmias. Optical voltage mapping revealed slower atrial conduction following VEGF treatment (10 ± 0.4 cm/s vs. 21 ± 1 cm/s at baseline,p < 0.05). Transmission electron microscopy revealed increased intermembrane spacing at ID sites adjacent to gap junctions (GJs; 64 ± 9 nm versus 17 ± 1 nm in controls,p < 0.05), as well as sites next to mechanical junctions (MJs; 63 ± 4 nm versus 27 ± 2 nm in controls,p < 0.05) in VEGF–treated hearts relative to controls. Importantly, super-resolution microscopy and quantitative image analysis revealed reorganization of NaV1.5 away from dense clusters localized near GJs and MJs to a more diffuse distribution throughout the ID. Taken together, these data suggest that VEGF can acutely predispose otherwise normal hearts to atrial arrhythmias by dynamically disrupting NaV1.5-rich ID nanodomains and slowing atrial conduction. These data highlight inflammation-induced vascular leak as a potential factor in the development and progression of AF.

    more » « less
  5. Abstract Background

    Chagas disease, caused by the protozoan parasiteTrypanosoma cruzi, causes sudden death and chronic heart disease with no currently approved treatment.


    To report epidemiologic and select cardiac characteristics associated withT. cruziinfection in dogs presenting to a teaching hospital in Texas.


    Three hundred seventy‐five client‐owned dogs.


    A retrospective search of medical records identified dogs tested forT. cruziantibodies or with histologicT. cruziparasites. Data retrieved included signalment, location of residence, reported reason for testing, cardiac troponin I (cTnI) concentration, and ECG abnormalities.


    Trypanosoma cruzi‐infected dogs (N = 63, 16.8%) were significantly younger than negative dogs (N = 312) (mean, 5.9 ± 3.8 versus 7.4 ± 4.0 years;P = .007) with no difference by sex or breed. Ninety‐one breeds were tested; the highest percent infected were non‐sporting (10/35; 29%) and toy breed (10/42; 24%) groups. The odds of infection were 13 times greater among dogs with an infected housemate or littermate (95% confidence interval [CI], 3.94‐50.45;P< .001). Infected dogs were more likely to have ventricular arrhythmias (odds ratio [OR], 2.19; 95% CI, 1.15‐4.33,P = .02), combinations of ECG abnormalities (OR, 2.91; 95% CI, 1.37‐5.99;P = .004), and cTnI >0.129 ng/mL (ADVIA; OR, 10.71; 95% CI, 1.60‐212.21;P = .035).

    Conclusions and Clinical Importance

    Dogs infected withT. cruziwere identified in Texas in many breed groups including breeds affected by well‐described heart diseases that mimic Chagas disease suggesting a need for increased awareness, including knowledge of when to consider testing.

    more » « less