Muna, Niyalatul (2013) Identifikasi diabetic retinopathy menggunakan support vector machine. Undergraduate thesis, Universitas Islam Negeri Maulana Malik Ibrahim.
|
Text (Fulltext)
09650092.pdf - Accepted Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (4MB) | Preview |
Abstract
INDONESIA:
Salah satu dampak dari penyakit Diabetic Mellitus (DM) adalah penyumbatan pembuluh darah mata, dikenal sebagai penyakit Diabetic Retinopathy (DR. Gejala yang dapat ditemui oleh orang yang terkena penyakit ini adalah kesulitan dalam membaca, penglihatan kabur, penglihatan tiba-tiba menurun pada satu mata, melihat lingkaran-lingkaran cahaya, melihat bintik gelap, dan cahaya berkedip. Hal ini terjadi karena ada rembesan darah yang mengenai lensa mata.
Penelitian dilakukan untuk membuat suatu aplikasi yang dapat menampilkan hasil identifikasi Diabetic Retinopathy sesuai dengan tingkat stadiumnya yaitu normal, Non-Proliferate Diabetic Retinopathy (NPDR), dan Proliferate Diabetic Retinopathy (PDR. Proses identifikasi dari DR, meliputi tiga tahap utama yaitu preprocessing, ekstraksi fitur, dan klasifikasi. Metode preprocessing yang digunakan diantaranya citra kanal hijau (green channel), operasi morfologi, pengambilan nilai ekstraksi fitur dengan menghitung area dari masing-masing objek, dan proses klasifikasi menggunakan metode support vector machine sehingga DR dapat dideteksi.
Aplikasi dapat mendeteksi DR berdasarkan stadiumnya yaitu normal, NPDR, dan PDR dalam citra fundus yang diujikan dengan persentase keberhasilan sebesar 90 % dari seluruh citra.
ENGLISH:
One of the effects of the Diabetic Mellitus (DM) disease is a blockage of eye blood vessels, known as the Diabetic retinopathy (DR) disease. Symptoms that can be encountered by people affected by this disease are difficulty in reading, blurred vision, sudden decreased vision in one eye, seeing halos, seeing dark spots, and a flashing light. This happens because there is seepage of blood on the lens of the eye.
The research was conducted to create an application that can display the results of the identification diabetic retinopathy according to the stage that is normal, Non-Proliferate Diabetic Retinopathy (NPDR), and Proliferate Diabetic Retinopathy (PDR). The process of identification of the DR, includes three main phases, namely preprocessing, feature extraction and classification. Preprocessing methods used include green channel image, morphological operations, taking the value of feature extraction by calculating the area of each object, and classification using the support vector machine method so that the DR can be detected.
Applications can detect DR based on the stage that is normal, NPDR, and PDR in fundus images were tested with success percentage of 90% of the entire image.
Item Type: | Thesis (Undergraduate) | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
Supervisor: | Faisal, Muhammad and Kusumawati, Ririen | |||||||||
Contributors: |
|
|||||||||
Keywords: | Diabetic Retinopathy; Support Vector Machine | |||||||||
Departement: | Fakultas Sains dan Teknologi > Jurusan Teknik Informatika | |||||||||
Depositing User: | Desi Latifah Hamzah | |||||||||
Date Deposited: | 14 Aug 2017 15:43 | |||||||||
Last Modified: | 14 Aug 2017 15:43 | |||||||||
URI: | http://etheses.uin-malang.ac.id/id/eprint/7719 |
Downloads
Downloads per month over past year
Actions (login required)
View Item |