Text

KLASIFIKASI ARITMIA PADA SINYAL FETAL ECG BERBASIS DEEP LEARNING

---

Fetal Electrocardiogram (ECG) is a type of signal that reflects the electrophysiological activity of the fetal heart. The non-invasive abdominal electrode method is employed to obtain fetal EKG signals. Classification was conducted between normal and arrhythmia categories using two main approaches, which are autoencoder-Deep Neural Network (DNN) and Convolutional Neural Network (CNN). The autoencoder was used for feature extraction before classification with DNN. Evaluation results indicated that the DNN with autoencoder achieved a best accuracy of 76%, precision of 83%, recall of 83%, and F1 score of 79%. In contrast, the CNN approach demonstrated higher performance with an accuracy of 97%, precision of 100%, recall of 99%, and F1 score of 97%. Although the CNN approach yielded better metric results, the training loss and accuracy graphs were more optimal in the autoencoder-DNN procedure. This study indicates that the CNN method is superior for classifying fetal ECG signals, yet the autoencoder-DNN approach also shows potential with better optimization. The procedures used in this study were conducted to fulfill deep learning methods in artificial intelligence for the Non-Invasive Fetal ECG Arrhythmia Database (NIFEA DB).

---

Availability

Detail Information

Series Title

-

Call Number

T1565642024

Publisher

Indralaya : Prodi Sistem Komputer, Fakultas Ilmu Komputer Universitas Sriwijaya., 2024

Collation

xii, 87 hlm.; ilus.; tab, 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.307

Content Type

Text

Media Type

unmediated

Carrier Type

other (computer)

Edition

-

Subject(s)

Kecerdasan Buatan
Prodi Sistem Komputer

Specific Detail Info

-

Statement of Responsibility

SEW

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment