UNAIR NEWS – Science and technology are growing rapidly. The development has penetrated various aspects, including in the field of health. Application of technology can help medical personnel to diagnose a disease and abnormalities in human organs.

Diabetes Mellitus (DM) as a metabolic disease is also growing. DM has the potential to cause complications and the emergence of advanced disease. Diabetic retinopathy  is one of the complications of DM that attacks the retina of the eye and is one of the leading causes of blindness.

Examination of diabetic retinopathy is currently done by fundus photography technique using a tool called fundus cameras. The image generated from this technique is studied by the ophthalmologist manually. The current fundus camera device is not currently equipped with software, specifically which can detect diabetic retinopathy automatically.

It takes an effective and efficient method to help the ophthalmologist in diagnosing diabetic retinopathy. Systems of diagnosis of a disease based on medical image processing with the help of computers are being currently developed. The latest image processing techniques based on feature extraction analysis can ot only diagnose but also detect the severity of diabetic retinopathy (Walvekar and Salunke , 2015).

Departing from that important requirement, Universitas Airlangga’s Technology Student Creativity Program (PKM-T) team consisting of Bestia Kumala (leader), Debrina Rizka, Nurrahmah Wida, Nalindra Berliani and Nitasya Ayu Alamanda Putri, created an automatic diabetic retinopathy detection system named Sadar Diri ( Smart Automated Detection Software for Diabetic Retinopathy ).

This detection system is based on artificial neural network that is considered effective in recognizing complex and specific patterns in classifying data through the learning process. After passing a rigorous selection by Directorate General of Higher Education, PKMT innovation proposal entitled “Sadar Diri ( Smart Automated Detection Software For Diabetic Retinopathy ) Based on Artificial Neural Network at UPT East Java Eye Hospital” was entitled to research grant from Ministry of Technology Research and Higher Education (Kemenristekdikti) in PKM program 2017-2018.

The Mechanism

“The working mechanism of this device is very simple, so it can save time, ie retinal images captured by fundus camera will be connected to the Personal Computer (PC) pre-installed with the program,” said Bestia Kumala.

The software will perform some image processing and classify it as normal image or diabetic retinopathy image. In addition, this program can also classify the severity of Nonproliferative Diabetic Retinopathy (NPDR) and Proliferative Diabetic Retinopathy (PDR).

The software has a user friendly interface design that is easy to operate. This software technology can be used as a quick screening and early detection because the diagnostic results can be directly seen on the PC / computer screen.

Bestia also said that blindness due to diabetic retinopathy will reduce the quality of life and productivity of patients, which eventually cause social burden society.

“With the application of this innovation, we hope it could be used as a second opinion in the diagnosis of diabetic retinopathy disease, thus improving the quality of health services. And good health services will improve the quality of life of Indonesian people, “said Bestia Kumala. (*)

Editor: Bambang Bes