Accuracy of deep learning, a machine-learning technology, using ultra\u2013wide-field fundus ophthalmoscopy for detecting rhegmatogenous retinal detachment

Hideharu Ohsugi,Hitoshi Tabuchi,Hiroki Enno,Naofumi Ishitobi

doi:10.1038/s41598-017-09891-x

Hideharu Ohsugi, Hitoshi Tabuchi + Show 2 more

Open Access

https://doi.org/10.1038/s41598-017-09891-x

Copy DOI

Abstract

Rhegmatogenous retinal detachment (RRD) is a serious condition that can lead to blindness; however, it is highly treatable with timely and appropriate treatment. Thus, early diagnosis and treatment of RRD is crucial. In this study, we applied deep learning, a machine-learning technology, to detect RRD using ultra–wide-field fundus images and investigated its performance. In total, 411 images (329 for training and 82 for grading) from 407 RRD patients and 420 images (336 for training and 84 for grading) from 238 non-RRD patients were used in this study. The deep learning model demonstrated a high sensitivity of 97.6% [95% confidence interval (CI), 94.2–100%] and a high specificity of 96.5% (95% CI, 90.2–100%), and the area under the curve was 0.988 (95% CI, 0.981–0.995). This model can improve medical care in remote areas where eye clinics are not available by using ultra–wide-field fundus ophthalmoscopy for the accurate diagnosis of RRD. Early diagnosis of RRD can prevent blindness.

Highlights

Rhegmatogenous retinal detachment (RRD) is a highly curable condition if properly treated early[1, 2]; if it is left untreated and develops proliferative changes, it becomes an uncontrollable condition called proliferative vitreoretinopathy (PVR)
The deep learning model’s sensitivity was 97.6% [95% confidence interval (CI), 94.2–100%] and specificity was 96.5%, and the area under the curve (AUC) was 0.988
Our results showed that the deep learning technology for detecting RRD on the Optos fundus photographs had high sensitivity and high specificity

Summary

Introduction

Rhegmatogenous retinal detachment (RRD) is a highly curable condition if properly treated early[1, 2]; if it is left untreated and develops proliferative changes, it becomes an uncontrollable condition called proliferative vitreoretinopathy (PVR). PVR is a serious condition that can result in blindness regardless of repeated treatments[3,4,5]. It is important, for patients to be seen and treated at a vitreoretinal centre at the early RRD stage to preserve visual function. For patients to be seen and treated at a vitreoretinal centre at the early RRD stage to preserve visual function Establishing such vitreoretinal centres that provide advanced ophthalmological procedures is not practical because of rising social security costs, a problem that is troubling many nations around the world[6]. We assessed the ability of a deep learning technology to detect RRD using Optos images

Methods

Results

Conclusion