EUVIMEDEUVIMED entdecken
Crossref

INTEGRATED DEEP LEARNING AND IOT-ENABLED FRAMEWORK FOR REAL-TIME ECG ARRHYTHMIA DETECTION VIA WAVELET-BASED SIGNAL PROCESSING

International Journal of Mechatronics and Applied Mechanics · 23.06.2026 · I (24)

DOI: 10.17683/ijomam/issue24.37 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Mechatronics and Applied Mechanics, online erschienen am 26.06.2026.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

ECG analysis plays a vital role in the early detection and management of cardiovascular diseases, yet accurate interpretation is often challenged by noise, variability, and the complexity of cardiac signals.This paper presents a comprehensive methodology that integrates advanced signal processing techniques and machine learning models for robust ECG analysis.Noise reduction is achieved through digital filtering, adaptive methods, and wavelet-based denoising to suppress baseline wander, powerline interference, and electromyographic artifacts.Feature extraction employs temporal, morphological, and spectral descriptors, with wavelet transforms facilitating precise detection of fiducial points.Signal classification is performed using machine learning and deep learning algorithms, including Support Vector Machines (SVMs), Convolutional Neural Networks (CNNs), and ensemble strategies, which significantly improve accuracy in arrhythmia detection.The proposed framework is validated using benchmark ECG datasets, achieving high performance in terms of sensitivity, specificity, F1-score, and area under the ROC curve (AUC).The CNN model achieved 97.8% accuracy, 98.1% sensitivity, and 0.97 AUC on the MIT-BIH Arrhythmia Database, outperforming SVM (92.3%) and kNN baselines.Real-time deployment on Raspberry Pi demonstrated a processing latency of 4.3 ms per beat, confirming feasibility for wearable cardiac monitoring.The results highlight that deep learning models, particularly CNN-based approaches, outperform traditional techniques in handling large-scale, noisy datasets.This study underscores the potential of combining mathematical modeling, advanced signal processing, and intelligent classification to improve diagnostic accuracy and accessibility in ECG-based cardiovascular disease detection.

Zitieren

APA

(2026). INTEGRATED DEEP LEARNING AND IOT-ENABLED FRAMEWORK FOR REAL-TIME ECG ARRHYTHMIA DETECTION VIA WAVELET-BASED SIGNAL PROCESSING. International Journal of Mechatronics and Applied Mechanics, I(24). https://doi.org/10.17683/ijomam/issue24.37

Vancouver

INTEGRATED DEEP LEARNING AND IOT-ENABLED FRAMEWORK FOR REAL-TIME ECG ARRHYTHMIA DETECTION VIA WAVELET-BASED SIGNAL PROCESSING. International Journal of Mechatronics and Applied Mechanics. 2026;I(24). doi:10.17683/ijomam/issue24.37

BibTeX
@article{euvimed2026,
  title = {INTEGRATED DEEP LEARNING AND IOT-ENABLED FRAMEWORK FOR REAL-TIME ECG ARRHYTHMIA DETECTION VIA WAVELET-BASED SIGNAL PROCESSING},
  journal = {International Journal of Mechatronics and Applied Mechanics},
  year = {2026},
  volume = {I},
  number = {24},
  doi = {10.17683/ijomam/issue24.37}
}
Kostenlos starten →