IMRAN, AZZAM ZAHFRAN
Unknown Affiliation

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

Identifikasi Penyakit Katarak berdasarkan Citra Fundus menggunakan Siamese Convolutional Neural Network RAHMADWATI, RAHMADWATI; IMRAN, AZZAM ZAHFRAN; ASWIN, MUHAMMAD; FERDIANA, KHAIRUNISA
ELKOMIKA: Jurnal Teknik Energi Elektrik, Teknik Telekomunikasi, & Teknik Elektronika Vol 12, No 4: Published October 2024
Publisher : Institut Teknologi Nasional, Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26760/elkomika.v12i4.838

Abstract

ABSTRAKKatarak merupakan penyakit yang dipengaruhi oleh faktor-faktor tertentu seperti usia, aktivitas dan penderita penyakit genetik seperti diabetes, hipertensi, asam urat serta riwayat keluarga katarak. Diagnosis penyakit katarak ini dapat dipengaruhi oleh faktor subyektif seperti pengalaman dan keahlian dokter. Untuk mengatasi hal tersebut dan menurunkan tingkat subyektivitas diperlukan pendekatan yang akurat dan konsisten yaitu sistem identifikasi penyakit katarak terbantukan komputer. Penelitian ini bertujuan sebagai deteksi dini katarak. Metode SCNN digunakan untuk mengidentifikasi citra fundus mata katarak. Fine tuning parameter SCNN memberikan performa yang baik pada proses pelatihan dan pengujian yaitu 100 epoch, optimizer : RMS Prop dan loss function Binary Crossentropy. Performansi yang diberikan yaitu akurasi 91,25%, kepresisian 91%.Kata kunci: penyakit katarak, siamese convolutional neural network, citra fundus. ABSTRACTThe cataract is a disease that influenced by certain factors such as age, activity and people with genetic disease such as diabetes, hypertension, uric acid and family history of cataract. The diagnosis of cataracts based on opthamologist experience and expertise which signifies a level of a diagnostic subjectivities. In order to overcome that problem and reduce the level of subjectivity, the need for an accurate and consistent computer aided identification for cataract disease is inevitable. This research aims to as an early detection of cataracts. The SCNN is applied for identify the cataract disease based on eye fundus image. Fine tuning SCNN parameters which provide good performances in the training and testing process with 100 epochs, RMSProp optimizer, Binary Crossentropy Loss function.This system gives promising result with the accuracy 91,25% , precision level is 91%.Keywords: cataract disease, siamese convolutional neural network, fundus images