Loading...
Unveiling Cardiac Rhythms:
Jamjoom, Jude Qashqari, Maha Alzahrani, Mariah
Jamjoom, Jude
Qashqari, Maha
Alzahrani, Mariah
Citations
Altmetric:
Type
Supervisor
Subject
Date
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
Cardiovascular diseases (CVDs) are a leading cause of mortality worldwide, necessitating the development of advanced diagnostic tools for early detection and intervention.
Electrocardiography (ECG) serves as a critical tool for monitoring cardiac health, yet its
raw data is often noisy and complex, requiring sophisticated preprocessing and analysis.
This project presents a comprehensive framework for arrhythmia detection through ECG
signal processing and machine learning techniques.
The study involves the design and implementation of a robust system that integrates
advanced noise reduction filters, feature extraction, dimensionality reduction, and classification techniques. Utilizing the MIT-BIH Arrhythmia database, we evaluate various
filtering methods, including Savitzky-Golay, wavelet, and Wiener filters, to enhance signal quality. Features extracted from both time and frequency domains are reduced using
Principal Component Analysis (PCA) and other dimensionality reduction techniques,
optimizing computational eciency while preserving diagnostic accuracy.
The performance of machine learning classifiers such as Logistic Regression, k-Nearest
Neighbors, Decision Trees, Support Vector Machines, and Random Forests is assessed
using metrics including accuracy, precision, and recall. The results demonstrate the feasibility of real-time, resource-ecient arrhythmia detection, achieving high diagnostic
accuracy. This system has the potential to bridge gaps in traditional diagnostic methods,
providing scalable solutions for clinical and remote healthcare applications.This research
contributes to the advancement of automated ECG analysis, o↵ering a practical approach
to improving cardiac health monitoring in diverse environments.