DETEKSI ELEVASI DAN DEPRESI SEGMEN ST SINYAL ELEKTROKARDIOGRAM BERBASIS MEDICAL RULES DENGAN PENDEKATAN DELINEASI MENGGUNAKAN MODEL DEEP LEARNING

ISDWANTA, RENDY and Darmawahyuni, Annisa (2024) DETEKSI ELEVASI DAN DEPRESI SEGMEN ST SINYAL ELEKTROKARDIOGRAM BERBASIS MEDICAL RULES DENGAN PENDEKATAN DELINEASI MENGGUNAKAN MODEL DEEP LEARNING. Undergraduate thesis, Sriwijaya University.

	Image RAMA_55201_09021282126073_cover.jpg - Cover Image Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (257kB) | Request a copy
	Text RAMA_55201_09021282126073.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (6MB) | Request a copy
	Text RAMA_55201_09021282126073_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5MB) | Request a copy
	Text RAMA_55201_09021282126073_8968340022_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB)
	Text RAMA_55201_09021282126073_8968340022_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (651kB) | Request a copy
	Text RAMA_55201_09021282126073_8968340022_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55201_09021282126073_8968340022_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (860kB) | Request a copy
	Text RAMA_55201_09021282126073_8968340022_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55201_09021282126073_8968340022_06.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (179kB) | Request a copy
	Text RAMA_55201_09021282126073_8968340022_07_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (187kB) | Request a copy
	Text RAMA_55201_09021282126073_8968340022_08_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (474kB) | Request a copy

Abstract

An accurate early detection of ST-segment elevation and depression in electrocardiogram (ECG) signals is a crucial step in diagnosing cardiovascular disease, such as myocardial infarction and myocardial ischemia. In clinical practice, cardiac conditions diagnosis is typically performed by cardiologists, or trained medical professionals. However, this way of diagnosing manually has some apparent limitations, such as shortage of trained medical professionals and variability amongst observers. Therefore, an accurate and consistent method for detecting ST-segment elevation and depression is necessary. This study proposes a method to detect ST-segment elevation and depression based on medical rules through delineation approach using ConvBiLSTM model. Twelve delineation models were developed for 12-lead ECG signals, segmenting the ECG waves into seven classes, namely Pon – Poff, Poff – QRSon, QRSon – Rpeak, Rpeak – QRSoff, QRSoff – Ton, Ton – Toff, and Toff – Pon. The delineation results are used to determine J-point and baseline amplitude, which are used to interpret ST-segment elevation and depression according to the predefined medical rules. The lead V6 delineation model demonstrated the best performance, achieving an accuracy of 99.54%, an error rate of 0.46%, a precision of 95.45%, and an F1-score of 95.43%. The detection process using this approach effectively interprets ST-segment elevation and depression. Future research may address the identified limitations by improving the delineation model’s performance or exploring alternative detection methods beyond delineation.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Elevasi segmen ST, Depresi segmen ST, Deteksi berbasis medical rules, Sinyal ECG, Delineasi sinyal ECG, ConvBiLSTM
Subjects:	T Technology > T Technology (General) > T1-995 Technology (General)
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Rendy Isdwanta
Date Deposited:	13 Jan 2025 07:54
Last Modified:	13 Jan 2025 07:54
URI:	http://repository.unsri.ac.id/id/eprint/164381

Actions (login required)

View Item