Fathonah, Diah (2024) KLASIFIKASI SUARA JANTUNG MENGGUNAKAN METODE CONVOLUTIONAL NEURAL NETWORK (CNN) BERDASARKAN EKSTRAKSI CIRI MEL FREQUENCY CEPSTRUM COEFFICIENTS (MFCC). S1 thesis, Fakultas Teknik Universitas Sultan Ageng Tirtayasa.

	Text (SKRIPSI) Diah Fathonah_3332170044_Fulltext.pdf Restricted to Registered users only Download (3MB) | Request a copy
	Text (SKRIPSI) Diah Fathonah_3332170044_01.pdf Restricted to Registered users only Download (560kB) | Request a copy
	Text (SKRIPSI) Diah Fathonah_3332170044_02.pdf Restricted to Registered users only Download (477kB) | Request a copy
	Text (SKRIPSI) Diah Fathonah_3332170044_03.pdf Restricted to Registered users only Download (460kB) | Request a copy
	Text (SKRIPSI) Diah Fathonah_3332170044_04.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text (SKRIPSI) Diah Fathonah_3332170044_05.pdf Restricted to Registered users only Download (35kB) | Request a copy
	Text (SKRIPSI) Diah Fathonah_3332170044_Ref.pdf Restricted to Registered users only Download (136kB) | Request a copy
	Text (SKRIPSI) Diah Fathonah_3332170044_Lamp.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text (SKRIPSI) Diah Fathonah_3332170044_CP.pdf Restricted to Registered users only Download (13MB) | Request a copy

Abstract

According to the World Health Organization (WHO), heart disease is the deadliest disease in the world and one of the three leading causes of mortality annually. The traditional auscultation method of diagnosing heart sounds is subjective, as it is dependent on the hearing sensitivity of experienced experts and the stethoscope generates a weak sound, so a system is required to assist in the early detection of heart conditions through heart sounds. Heart sound signals can be detected through a phonocardiogram. The aim of this study is to classify cardiac sounds into 23 types using the Convolutional Neural Network (CNN) method, which is based on the extraction of Mel Frequency Cepstrum Coefficients (MFCC) features. the classification system's research phases include preprocessing, segmentation, feature extraction, and classification. In total, 230 cardiac sounds were utilized, including 184 training data and 46 test data. The results indicate that the classification system implementing the Convolutional Neural Network (CNN) method achieves an accuracy level of 100% during training and 97.8% during testing.

Item Type:	Thesis (S1)
Contributors:	Contribution Contributors NIP/NIM Thesis advisor Alfanz, Rocky 198103282010121001
Additional Information:	Penyakit jantung menurut data WHO (World Health Organization) adalah salah satu dari tiga penyebab kematian di dunia setiap tahun dan merupakan penyakit pembunuh nomor satu di dunia. Diagnosis suara jantung dengan metode auskultasi tradisional bersifat subjektif berdasarkan sensitivitas pendengaran ahli berpengalaman, dan stetoskop menghasilkan suara yang lemah, sehingga dibutuhkan suatu sistem yang dapat membantu mendeteksi dini kondisi jantung melalui suara jantung. Deteksi dapat dilakukan melalui sinyal suara jantung phonocardiogram. Penelitian ini bertujuan untuk mengklasifikasikan suara jantung dengan 23 kelas suara jantung menggunakan metode Convolutional Neural Network (CNN) berdasarkan ekstraksi ciri Mel Frequency Cepstrum Coefficients (MFCC). Tahapan penelitian yang dilakukan dalam sistem klasifikasi adalah preprocessing, segmentasi, ekstraksi ciri dan klasifikasi. Data yang digunakan sebanyak 230 suara jantung yang terdiri dari 184 data latih dan 46 data uji. Hasil menunjukkan bahwa sistem klasifikasi menggunakan metode Convolutional Neural Network (CNN) memiliki tingkat akurasi pada pelatihan yaitu 100% dan hasil pada pengujian yaitu 97,8%.
Subjects:	Q Science > QP Physiology T Technology > T Technology (General)
Divisions:	03-Fakultas Teknik 03-Fakultas Teknik > 20201-Jurusan Teknik Elektro
Depositing User:	Mrs. Diah Fathonah
Date Deposited:	14 Aug 2024 10:58
Last Modified:	14 Aug 2024 10:58
URI:	http://eprints.untirta.ac.id/id/eprint/40667

Actions (login required)

View Item