CD Skripsi

Identifikasi Penyakit Mata Berdasarkan Citra Fundus Menggunakan Convolutional Neural Network

---

ABSTRACT

Many cases of vision loss (such as those caused by infections, trauma, dangerous traditional medicines, diet-related disorders, prenatal Diseases, improper use or self-administration of topical therapy) are not always avoidable. Early Identification and proper treatment are essential for many Eye disorders, such as Diabetic retinopathy, to prevent irreparable vision loss. To detect Visual disturbances, one way is to check the fundus Image of the patient's Eye. Manual checking of fundus images requires high precision and a long time. So that various ways have begun to emerge to make it easier to identify the Image of the fundus, such as one of them, namely by Means of Deep learning. In this study, several CNN architectures were used to train fundus detection models such as Custom CNN, InceptionV3, VGG16 and VGG19. The dataset for this research model was obtained from the Kaggle site with a total of 511 images of ARMD, 1038 images of cataracts, 1098 images of Diabetic retinopathy, 1007 images of glaucoma, and 1074 images of normal. After training and testing the model, the highest accuracy was achieved by the VGG16 architecture with 91% accuracy, 91% precision, 90% recall and 90% F1-Score. Then the next test uses the fundus of patients who have been diagnosed by doctors from the Arifin Achmad Pekanbaru Hospital which totals 71 fundus images. Of the 71 images obtained from Arifin Achmad Hospital, there are 55 images of fundus that are predicted to be the same as the doctor's diagnosis.
Keywords: ARMD, fundus imagery, Convolutional Neural Network, custom, detection, glaucoma, Inception, cataracts, Diabetic retinopathy, VGG.

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

1907113293

Penerbit

Pekanbaru : Universitas Riau – F.TEKNIK – Elektro., 2025

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

1907113293

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

TEKNIK ELEKTRO

Info Detail Spesifik

-

Pernyataan Tanggungjawab

daus

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar