Skripsi

PENINGKATAN KINERJA KLASIFIKASI MULTI KELAS INFARK MIOKARD BERBASIS DEEP LEARNING

---

Penilaian

---

Heart disease, particularly myocardial infarction (IM), is a leading cause of global death that requires early and accurate detection to prevent fatal outcomes. This study aims to develop a deep learning-based IM multi-class classification system using ECG signals from the PTB-XL dataset. The model was built with a combination architecture of Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM), and compared with CNN-BiLSTM architecture. The study included 15 classification classes based on the location of IM. The research stages include preprocessing (denoising, normalization, segmentation), model training with various layer combinations, and evaluation using accuracy, precision, recall, and f1-score metrics. The segmentation process with sliding window and denoising technique using db4 wavelet, level 8, and hard threshold proved to improve the signal quality. The best model is obtained from the combination of 25 layer CNN architecture and 1 layer LSTM. Experimental results show that the optimal combination of architecture and preprocessing can improve classification performance. The findings are expected to be a solution in assisting the automatic and efficient diagnosis of IM in clinical practice.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2507004105	T178563	T1785632025	Central Library (REFERENCE)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T1785632025

Publisher

Indralaya : Prodi Sistem Komputer, Fakultas Ilmu Komputer Universitas Sriwijaya., 2025

Collation

xxiii, 443 hlm.; ilus.; tab.; 29 cm.

Language

Indonesia

ISBN/ISSN

-

Classification

004.07

Content Type

-

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Sistem komputer

Specific Detail Info

-

Statement of Responsibility

TUTI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment