CARDIAC ARRHYTHMIA DETECTION SYSTEM

OKEKE, DAVID ODIRACHUKWMMA (2025) CARDIAC ARRHYTHMIA DETECTION SYSTEM. Other thesis, GODFREY OKOYE UNIVERSITY, ENUGU.

Text REPO152.docx Download (4MB)

Abstract

Cardiac arrhythmia is a serious health situation, which takes place when the regular rhythm of the heartbeat becomes in interruption, and, therefore, can lead to such life threatening conditions as stroke, cardiac arrest, or heart failure in the case the irregular heartbeat rhythms remain unidentified and unsolved respectively. Traditional manual methods of arrhythmia diagnosis through electrocardiogram (ECG) interpretation are time-consuming and prone to human error. The following paper demonstrates how to design and implement an intelligent Cardiac Arrhythmia Detection System on the machine learning datasets and the ECG signals. The system utilizes 1D Convolutional Neural Networks (1D CNNs), Long Short-Term Memory (LSTM) networks, and a hybrid 1D CNN-LSTM architecture to classify ECG signals into normal and abnormal rhythms. We used the data preprocessing stage, which includes filtering, normalization, segmentation, and augmentation, as the techniques that enhance the quality of signals and performance of the model. Its user-friendly interface is elegant web-based, and the user performs simple operations of uploading the ECG data in it and receives a graphical waveform of ECG in real time, as well as the results of the predictions with the interpretability of the saliency map and the scores of the confidence. It is being operated on a cloud platform where there is security in the API integration, thereby, making system scalable and remote. It demonstrates the value of the achieved performances of modeling the 1D CNN-LSTM on the benchmark datasets, such as the MIT-BIH and Kaggle Arrhythmia datasets, testing the performance of the 94.7 with an F1-score of 92.3, in addition to the AUC-ROC of 95.1, which suggests the reliability of the system as well as the likelihood of its eventual clinical use in real-time through screening of arrhythmia.

Item Type:	Thesis (Other)
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions:	Faculty of Natural Sciences and Environmental Studies
Depositing User:	HILARY OBIEKWE
Date Deposited:	13 Aug 2026 11:21
Last Modified:	13 Aug 2026 11:21
URI:	http://eprints.gouni.edu.ng/id/eprint/6253

Actions (login required)

View Item