RESUMO
Purpose: We examine the rate of and reasons for follow-up in an Artificial Intelligence (AI)-based workflow for diabetic retinopathy (DR) screening relative to two human-based workflows. Patients and Methods: A DR screening program initiated September 2019 between one institution and its affiliated primary care and endocrinology clinics screened 2243 adult patients with type 1 or 2 diabetes without a diagnosis of DR in the previous year in the San Francisco Bay Area. For patients who screened positive for more-than-mild-DR (MTMDR), rates of follow-up were calculated under a store-and-forward human-based DR workflow ("Human Workflow"), an AI-based workflow involving IDx-DR ("AI Workflow"), and a two-step hybrid workflow ("AI-Human Hybrid Workflow"). The AI Workflow provided results within 48 hours, whereas the other workflows took up to 7 days. Patients were surveyed by phone about follow-up decisions. Results: Under the AI Workflow, 279 patients screened positive for MTMDR. Of these, 69.2% followed up with an ophthalmologist within 90 days. Altogether 70.5% (N=48) of patients who followed up chose their location based on primary care referral. Among the subset of patients that were seen in person at the university eye institute under the Human Workflow and AI-Human Hybrid Workflow, 12.0% (N=14/117) and 11.7% (N=12/103) of patients with a referrable screening result followed up compared to 35.5% of patients under the AI Workflow (N=99/279; χ2df=2 = 36.70, p < 0.00000001). Conclusion: Ophthalmology follow-up after a positive DR screening result is approximately three-fold higher under the AI Workflow than either the Human Workflow or AI-Human Hybrid Workflow. Improved follow-up behavior may be due to the decreased time to screening result.
RESUMO
Objective: Detection of diabetic retinopathy (DR) outside of specialized eye care settings is an important means of access to vision-preserving health maintenance. Remote interpretation of fundus photographs acquired in a primary care or other nonophthalmic setting in a store-and-forward manner is a predominant paradigm of teleophthalmology screening programs. Artificial intelligence (AI)-based image interpretation offers an alternative means of DR detection. IDx-DR (Digital Diagnostics Inc) is a Food and Drug Administration-authorized autonomous testing device for DR. We evaluated the diagnostic performance of IDx-DR compared with human-based teleophthalmology over 2 and a half years. Additionally, we evaluated an AI-human hybrid workflow that combines AI-system evaluation with human expert-based assessment for referable cases. Design: Prospective cohort study and retrospective analysis. Participants: Diabetic patients ≥ 18 years old without a prior DR diagnosis or DR examination in the past year presenting for routine DR screening in a primary care clinic. Methods: Macula-centered and optic nerve-centered fundus photographs were evaluated by an AI algorithm followed by consensus-based overreading by retina specialists at the Stanford Ophthalmic Reading Center. Detection of more-than-mild diabetic retinopathy (MTMDR) was compared with in-person examination by a retina specialist. Main Outcome Measures: Sensitivity, specificity, accuracy, positive predictive value, and gradability achieved by the AI algorithm and retina specialists. Results: The AI algorithm had higher sensitivity (95.5% sensitivity; 95% confidence interval [CI], 86.7%-100%) but lower specificity (60.3% specificity; 95% CI, 47.7%-72.9%) for detection of MTMDR compared with remote image interpretation by retina specialists (69.5% sensitivity; 95% CI, 50.7%-88.3%; 96.9% specificity; 95% CI, 93.5%-100%). Gradability of encounters was also lower for the AI algorithm (62.5%) compared with retina specialists (93.1%). A 2-step AI-human hybrid workflow in which the AI algorithm initially rendered an assessment followed by overread by a retina specialist of MTMDR-positive encounters resulted in a sensitivity of 95.5% (95% CI, 86.7%-100%) and a specificity of 98.2% (95% CI, 94.6%-100%). Similarly, a 2-step overread by retina specialists of AI-ungradable encounters improved gradability from 63.5% to 95.6% of encounters. Conclusions: Implementation of an AI-human hybrid teleophthalmology workflow may both decrease reliance on human specialist effort and improve diagnostic accuracy. Financial Disclosures: Proprietary or commercial disclosure may be found after the references.
