KLASIFIKASI MULTICLASS SINYAL ELEKTROKARDIOGRAM BERBASIS DEEP LEARNING MENGGUNAKAN DENOISING AUTOENCODER DAN CONVOLUTIONAL NEURAL NETWORK

AL HADI, MUHAMMAD ISRA and Nurmaini, Siti (2023) KLASIFIKASI MULTICLASS SINYAL ELEKTROKARDIOGRAM BERBASIS DEEP LEARNING MENGGUNAKAN DENOISING AUTOENCODER DAN CONVOLUTIONAL NEURAL NETWORK. Undergraduate thesis, Sriwijaya University.

	Text RAMA_56201_09011181924006.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy
	Text RAMA_56201_09011181924006_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (8MB) | Request a copy
	Text RAMA_56201_09011181924006_0002085908_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (754kB)
	Text RAMA_56201_09011181924006_0002085908_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (349kB) | Request a copy
	Text RAMA_56201_09011181924006_0002085908_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (425kB) | Request a copy
	Text RAMA_56201_09011181924006_0002085908_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (942kB) | Request a copy
	Text RAMA_56201_09011181924006_0002085908_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (45kB) | Request a copy
	Text RAMA_56201_09011181924006_0002085908_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (113kB) | Request a copy
	Text RAMA_56201_09011181924006_0002085908_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (735kB) | Request a copy

Abstract

The use of Deep Learning (DL) models in signal processing aims to improve efficiency and effectiveness in extracting information from available data. This research aims to classify heart disease classes in a dataset of medical signals derived from Electrocardiograms. In this study, classification is performed on three heart disease classes: Normal, AF, and AFL. The method used in this research is Denoising Autoencoder, which reduces noise in the dataset, and Convolutional Neural Network (CNN), used for the classification process. The constructed model will be trained and validated with 9,222 signal recordings and tested with 1,025 signal recordings. The best results obtained for the 2-class classification model have average values of Accuracy 97.09%, Sensitivity 95.54%, Specificity 95.54%, Precision 94.40%, F1 Score 94.92%, and Error 2.90%. While for the 3-class classification model, the average values are Accuracy 97.22%, Sensitivity 93.56%, Specificity 96.93%, Precision 92.98%, F1 Score 93.24%, and Error 2.78%. Meanwhile, the best result when tested using unseen data, the model achieves accuracy values of 66.34% for the Normal class, 68.68% for the AF class, and 93.17% for the AFL class.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Klasifikasi, Deep Learning, Denoising Autoencoder, Convolutional Neural Network
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics > Q325.5 Machine learning Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions:	09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User:	Muhammad Isra Al Hadi
Date Deposited:	12 Jul 2023 01:59
Last Modified:	12 Jul 2023 01:59
URI:	http://repository.unsri.ac.id/id/eprint/115844

Actions (login required)

View Item