Search citation statements
Paper Sections
Citation Types
Year Published
Publication Types
Relationship
Authors
Journals
Electrocardiogram (ECG) data's high dimensionality challenges real-time arrhythmia classification. Our approach employs functional approximation to condense ECG recordings into a compact feature set for simpler classification using Chebyshev polynomials. These polynomials, with 200 time points and 80 coefficients, accurately represent arrhythmias in an 81 × 1 feature vector. We prove Chebyshev polynomials act as implicit low-pass filters on input signals. Using MIT-BIH Arrhythmia and MIT-BIH Supraventricular Arrhythmia datasets, we introduce classifiers that achieve significant accuracy. A threelayered Artificial Neural Network yields high F1-scores (0.99, 0.90, 0.93, and 0.76 for classes N, S, V, and F) with minimal parameters (20,964), surpassing existing models. Furthermore, our proposed ECG classification system exhibits minimal computational demands, requiring only 0.1 MIPS per beat. We also propose efficient signal reconstruction methods, with the iterative approach showcasing accurate reconstruction with negligible error. This approach accommodates various data sampling types and determines optimal Chebyshev coefficients for capturing signal bandwidth.INDEX TERMS level-crossing ADC, electrocardiograms, functional approximation, Chebyshev polynomials, artificial neural networks, arrhythmia, support vector machines, bandwidth analysis I. INTRODUCTION A RRHYTHMIA, an abnormal heart rhythm, is a significant medical condition. Atrial fibrillation, the most common form of arrhythmia, is projected to affect millions of people in the United States and Europe in the coming decades [1]. Automated arrhythmia classification is a widely researched area that can help with early diagnosis and improve patient care by providing long-term remote cardiac monitoring. Several hardware-friendly designs have been proposed for real-time arrhythmia classification. Notably, the literature includes real-time patient-specific ECG classifiers as demonstrated in prior works such as [2]-[4]. Abubakar and colleagues introduced a wearable long-term ECG processor for arrhythmia classification, employing a reduced feature set [5]. Additionally, a low-complexity antidictionary-based ECG classifier was proposed by Duforest et al. [6]. Tang et al. presented a patient-specific arrhythmia classifier with low complexity, utilizing support vector machines [7]. Another promising approach for real-time wearable arrhythmia classification involves utilising a novel sampling technique at the analogue front end, employing level-crossing ADCs.Recent research has demonstrated a growing interest in level-crossing ADCs (LC-ADCs) due to their potential to reduce data streams and battery consumption. Li et al. [8] introduced an ECG front-end featuring LC-ADC, showcasing its potential for low-power, high-performance applications. In a different approach, Marisa et al. presented a pseudo-asynchronous LC-ADC with dynamic comparators for implantable biomedical sensing. This design offers energy efficiency, a smaller chip area, and robust performance in noisy co...
Electrocardiogram (ECG) data's high dimensionality challenges real-time arrhythmia classification. Our approach employs functional approximation to condense ECG recordings into a compact feature set for simpler classification using Chebyshev polynomials. These polynomials, with 200 time points and 80 coefficients, accurately represent arrhythmias in an 81 × 1 feature vector. We prove Chebyshev polynomials act as implicit low-pass filters on input signals. Using MIT-BIH Arrhythmia and MIT-BIH Supraventricular Arrhythmia datasets, we introduce classifiers that achieve significant accuracy. A threelayered Artificial Neural Network yields high F1-scores (0.99, 0.90, 0.93, and 0.76 for classes N, S, V, and F) with minimal parameters (20,964), surpassing existing models. Furthermore, our proposed ECG classification system exhibits minimal computational demands, requiring only 0.1 MIPS per beat. We also propose efficient signal reconstruction methods, with the iterative approach showcasing accurate reconstruction with negligible error. This approach accommodates various data sampling types and determines optimal Chebyshev coefficients for capturing signal bandwidth.INDEX TERMS level-crossing ADC, electrocardiograms, functional approximation, Chebyshev polynomials, artificial neural networks, arrhythmia, support vector machines, bandwidth analysis I. INTRODUCTION A RRHYTHMIA, an abnormal heart rhythm, is a significant medical condition. Atrial fibrillation, the most common form of arrhythmia, is projected to affect millions of people in the United States and Europe in the coming decades [1]. Automated arrhythmia classification is a widely researched area that can help with early diagnosis and improve patient care by providing long-term remote cardiac monitoring. Several hardware-friendly designs have been proposed for real-time arrhythmia classification. Notably, the literature includes real-time patient-specific ECG classifiers as demonstrated in prior works such as [2]-[4]. Abubakar and colleagues introduced a wearable long-term ECG processor for arrhythmia classification, employing a reduced feature set [5]. Additionally, a low-complexity antidictionary-based ECG classifier was proposed by Duforest et al. [6]. Tang et al. presented a patient-specific arrhythmia classifier with low complexity, utilizing support vector machines [7]. Another promising approach for real-time wearable arrhythmia classification involves utilising a novel sampling technique at the analogue front end, employing level-crossing ADCs.Recent research has demonstrated a growing interest in level-crossing ADCs (LC-ADCs) due to their potential to reduce data streams and battery consumption. Li et al. [8] introduced an ECG front-end featuring LC-ADC, showcasing its potential for low-power, high-performance applications. In a different approach, Marisa et al. presented a pseudo-asynchronous LC-ADC with dynamic comparators for implantable biomedical sensing. This design offers energy efficiency, a smaller chip area, and robust performance in noisy co...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.