Heartbeat classification for arrhythmia detection in ambulatory monitoring: a comprehensive systematic review

This systematic literature review explores the current state of heartbeat classification for arrhythmia detection in ambulatory monitoring, with a focus on challenges and advancements in the field. Arrhythmias, being a significant contributor to cardiovascular morbidity and mortality, necessitate accurate and timely detection, especially in non-clinical settings where continuous monitoring is crucial. Despite significant progress, existing methods face substantial challenges, including issues with model accuracy, real-time data processing, and signal noise reduction. In order to accomplish this, we carried out a thorough search of academic publications from reliable sources like Scopus as well as the Web of Science (WoS), concentrating on research works released in 2024. The study's flow was organized according to the PRISMA model. The database containing the final primary data (n = 32) was examined. The results were categorized into three main themes: (1) arrhythmia detection utilizing both machine learning (ML) as well as deep learning (DL) models, (2) wearable devices and real-time monitoring systems for arrhythmia detection and (3) signal processing and noise reduction techniques for enhanced arrhythmia detection. Key findings reveal that while advancements in ML have improved detection accuracy, challenges persist in integrating these models into wearable technologies and managing data quality in real-time monitoring. Signal processing techniques have shown promise in reducing noise but require further optimization for broader application. The review concludes with recommendations for future research, emphasizing the need for enhanced model robustness, better integration of technologies, and advanced signal processing methods to improve arrhythmia detection in ambulatory settings.

Text 123004.pdf - Published Version Restricted to Repository staff only Download (821kB) Official URL or Download Paper: https://www.sciencedirect.com/science/article/pii/...

Actions (login required)

View Item