Artificial Intelligence Improves Patient Follow-Up in a Diabetic Retinopathy Screening Program

Abstract

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; χ²_df=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.

Keywords:

• telemedicine
• teleophthalmology
• fundus photography
• deep learning
• machine learning
• referral

Disclosure

This research was funded in part by departmental core grants from Research to Prevent Blindness and the National Institutes of Health (NIH P30 EY026877) as well as the Stanford Diabetes Research Center (SDRC). Authors with financial interests or relationships to disclose are listed after the references. Dr Diana V Do reports grants, personal fees from Regeneron, personal fees from Apellis, personal fees from Kodiak Sciences, personal fees from Iveric Bio, during the conduct of the study. Dr Vinit B Mahajan reports other from Digital Diagnostics, LLC, during the conduct of the study; Dr Theodore Leng reports grants, personal fees from Astellas, personal fees from Roche/Genentech, personal fees from Protagonist, personal fees from Nanoscope, personal fees from Alcon, personal fees from Apellis, personal fees from Regeneron, personal fees from Graybug, personal fees from Kanaph, personal fees from Boehringer Ingelheim, outside the submitted work. None of the remaining authors have any conflicts of interest to disclose.

Additional information

Funding

Theodore Leng: Institutional grants from Research to Prevent Blindness, NIH grant P30-EY026877, Astellas. Consulting fees from Roche/Genentech, Protagonist Therapeutics, Alcon, Regeneron, Graybug, Boehringer Ingelheim, Kanaph. Participation in Data Safety Monitoring or Advisory Board for Nanoscope Therapeutics, Apellis, Astellas. Diana Do: Grants from Regeneron, Kriya, Boerhinger Ingelheim. Consulting fees from Genentech, Regeneron, Kodiak Sciences, Apellis, Iveric Bio. David Myung: Institutional grants from Research to Prevent Blindness, NIH grant P30-EY026877, Stanford Diabetes Research Center (SDRC).