Klasifikasi Tingkat Keparahan Non-Proliferative Diabetic Retinopathy Berdasarkan Hard Exudate Menggunakan Extreme Learning Machine

Yani, Dinda Ulim Rizky (2017) Klasifikasi Tingkat Keparahan Non-Proliferative Diabetic Retinopathy Berdasarkan Hard Exudate Menggunakan Extreme Learning Machine. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img]
Preview
Text
TUGAS AKHIR DINDA ULIMA 1213100081.pdf - Published Version

Download (1MB) | Preview

Abstract

Diabetic Retinopathy dapat menyebabkan seseorang kehilangan kemampuan penglihatannya dan pada keadaan yang parah dapat mengakibatkan kebutaan. Tingkat keparahan Non-Proliferative Diabetic Retinopathy (NPDR) dapat diketahui dengan mendeteksi kelainan berupa hard exudate pada retina, namun diagnosanya tidak bisa dilakukan dengan cepat karena pengamatan retina harus melewati beberapa proses. Teknologi pengolahan citra digital berbasis machine learning telah banyak digunakan untuk menyelesaikan permasalahan ini. Pada Tugas Akhir ini telah dilakukan penelitian untuk mengklasifikasikan tingkat keparahan NPDR secara otomatis dengan mengekstraksi karakteristik hard exudate menggunakan Gray Level Co–occurrence Matrix (GLCM) dan Neighborhood Gray–tone Difference Matrix (NGTDM) kemudian menentukan tingkat keparahannya menggunakan Extreme Learning Machine. Hasil akurasi tertingi yang didapat sebesar 91,22% untuk ekstraksi ciri dengan menggunakan GLCM. ================================================================================================= Diabetic Retinopathy may cause vision loss ability and in severe circumstances can lead to blindness. The severity of the Non-Proliferative Diabetic Retinopathy (NPDR) can be known by detecting hard exudates on the retina. But the diagnosis can not be done quickly because the observations of this retinal must pass through several processes. Digital image processing technology and machine learning has been widely used to classify the severity of NPDR automatically by extracting the characteristic of hard exudates using Gray Level Co-occurrence Matrix (GLCM) and Neighborhood Gray-tone Difference Matrix (NGTDM) and using Extreme Learning Machine for classification. The highest accuracy results obtained is 91.22% for extract features using GLCM.

Item Type: Thesis (Undergraduate)
Additional Information: RSMa 006.425 Yan k
Uncontrolled Keywords: Diabetic Retinopathy, NPDR, hard exudate, PengolahanCitra, Extreme Learning Machine
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science. EDP
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Depositing User: Yani Dinda Ulima Rizky
Date Deposited: 12 Feb 2018 08:33
Last Modified: 05 Mar 2019 08:46
URI: http://repository.its.ac.id/id/eprint/47649

Actions (login required)

View Item View Item