Automatic Screening for Diabetic Retinopathy Severity
Deep learning models for grading diabetic retinopathy severity from retinal fundus images, built on EfficientNet.
Diabetic retinopathy is a leading cause of preventable blindness, and screening at scale is limited by the number of clinicians available to read retinal images. Automatic severity grading offers a way to triage.
At the Biomedical Research Group at ITB, I built and trained models in TensorFlow for automatic grading of diabetic retinopathy severity from fundus photographs, using EfficientNet backbones. The work was published at IEEE TENCON 2020.
Publications from this project
IEEE Region 10 Conference (TENCON), 2020
Automatic Diabetic Retinopathy Classification with EfficientNet
pp. 756–760.
@inproceedings{lazuardi2020automatic,
title = {Automatic diabetic retinopathy classification with EfficientNet},
author = {Lazuardi, Rachmadio Noval and Abiwinanda, Nyoman and Suryawan, Tafwida Hesaputra and Hanif, Muhammad and Handayani, Astri},
booktitle = {2020 IEEE Region 10 Conference (TENCON)},
pages = {756--760},
year = {2020},
publisher = {IEEE}
}