Article Per Year (5 Year)
p-Index From 2019 - 2024
0.23
P-Index
This Author published in this journals
All Journal IJRETINA - International Journal of Retina
Yeni Dwi Lestari
Ophthalmology Department, Faculty of Medicine Universitas Indonesia Cipto Mangunkusumo General Hospital
Health Professions Medicine & Pharmacology Public Health

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

Found 1 Documents
Search

 1 
The Validity of Artificial Intelligence (AI) for Diabetic Retinopathy Screening in Asia: A Systematic Review Yeni Dwi Lestari; Andi Putra Jayanegara
International Journal of Retina Vol 6 No 2 (2023): International Journal of Retina (IJRetina) - INAVRS
Publisher : Indonesian Vitreoretinal Society

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35479/ijretina.2023.vol006.iss002.241

Abstract

Introduction: One of the major diabetes mellitus (DM) complications is diabetic retinopathy (DR) which may lead to visual impairment. Unfortunately, estimated that more than one-third of diabetic patients in Asia have never undergone an ophthalmological examination. Screening for DR is an essential aspect of DM management. With advanced technological development, the use of artificial intelligence (AI) for DR screening have been increasing in popularity, especially since Deep Learning (DL) technologies in AI can process a large amount of data automatically without human intervention. Therefore, this review aims to evaluate the validity of AI as DR screening tool in Asian countries compared to ophthalmologist DR grading as a reference standard. Methods: A comprehensive search was conducted using relevant search terms from the electronic database (Pubmed, Cochrane, and EMBASE). Included studies were selected based on predefined inclusion criteria. Result: Twelve studies reported a good sensitivity of AI for detecting Referable DR (RDR). The lowest sensitivity was 79.2 %, and the highest was 100 %. For specificity, eleven studies reported good specificity, with only one study reporting a low value, with only 68.8% for detecting RDR. Conclusion: The AI can detect DR by screening large amounts of retina images quickly without the assistance of a trained retina specialist. In this review of local Asia population settings, AI has a good result for detecting RDR in almost all studies.
Co-Authors Andi Putra Jayanegara