Abstract
Purpose :
The purpose of this study was to assess the accuracy of artificial intelligence (AI)-based screening for diabetic retinopathy (DR) and to explore the feasibility of applying AI-based technique to community hospital for DR screening.
Methods :
Nonmydriatic fundus photos were taken for 889 diabetic patients who were screened in community hospital clinic. According to the International Clinical Diabetic Retinopathy (ICDR) severity scale, all fundus photographs were graded independently by AI and two ophthalmologists (retina specialists, Kappa (κ)=0.899) who were masked to each other and to AI device outputs. The sensitivity and specificity of AI automatic grading were evaluated using the ophthalmologist s' grading as a reference standard.
Results :
DR was detected by ophthalmologists in 143 (16.1%) participants and by AI in 145 (16.3%) participants. Among them, there were 101 (11.4%) participants diagnosed with referable diabetic retinopathy (RDR) by ophthalmologists and 103 (11.6%) by AI. The matched diagnosis of RDR between ophthalmologists and AI-grading was observed in 91 participants. For detecting any DR, the sensitivity and specificity were 90.79% (95%, CI 86.4–94.1) and 98.5% (95%, CI 97.8–99.0), respectively. For detecting RDR, AI showed 91.18% (95%, CI 86.4–94.7) sensitivity and 98.79% (95%, CI 98.1–99.3) specificity. The area under the curve (AUC) was 0.946 (95%, CI 0.935–0.956) when testing the ability of AI to detect any DR; for the detection of RDR, the AUC was 0.950 (95%, CI 0.939–0.960).
Conclusions :
AI has high sensitivity and specificity in detecting DR and RDR, so it is feasible to carry out AI-based DR screening in outpatient clinic of community hospital.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.