Ophthalmology · Diabetic retinopathy screening

LumineticsCore

Digital Diagnostics (formerly IDx Technologies)

CEFDAProspective

The first FDA-cleared AI to operate without physician interpretation (De Novo DEN180001, April 2018), LumineticsCore is the regulatory and clinical reference point for autonomous AI in medicine — no expert in the loop, validated across 10 primary-care sites in the Abramoff pivotal trial (n=900), and followed by the ACCESS RCT demonstrating 100% screening completion in youth. Eight years of post-market deployment across major US health systems, a Class IIa CE mark, and an expanding Labcorp laboratory-channel footprint confirm durable real-world clinical utility.

Performance Metrics

1stAUTONOMOUS AI IN ANY MEDICAL FIELDFDA De Novo DEN180001, April 2018 — no human in the loop
87.2%SENSITIVITY (PIVOTAL TRIAL)Abràmoff et al., npj Digital Medicine 2018 — 10 sites, n=900
100%SCREENING COMPLETION (ACCESS RCT)vs 22% in control arm — Nature Communications, 2024
162%INCREASE IN DIABETIC EYE EXAMSOhioHealth real-world deployment, 40 sites

Clinical Evidence

The pivotal study underpinning FDA De Novo authorisation was published by Abràmoff et al. in *npj Digital Medicine* (August 2018; NCT-registered prospective observational; n=900 adults with diabetes at 10 primary care sites across the US). Participants had no prior diagnosis of diabetic retinopathy. Retinal images were compared against a Wisconsin Fundus Photograph Reading Center (FPRC) reference standard using widefield stereoscopic photography and macular OCT graded to ETDRS severity scale. IDx-DR exceeded all three pre-specified superiority endpoints: sensitivity 87.2% (95% CI 81.8–91.2%; pre-specified threshold >85%), specificity 90.7% (95% CI 88.3–92.7%; threshold >82.5%), and imageability rate 96.1% (95% CI 94.6–97.3%). *This was the first prospective pivotal trial to demonstrate that an autonomous AI — with no clinician interpretation — could meet or exceed specialist-level diagnostic thresholds in a real primary-care workflow.* The ACCESS RCT (Merin et al., *Nature Communications*, January 2024; NCT05131451; n=163 youth aged 8–21 with type 1 or type 2 diabetes at an academic paediatric diabetes centre) randomised participants to autonomous AI exam at the point of care versus standard scripted eye-care referral. Diabetic eye exam completion at 6 months was 100% (95% CI 95.5%–100%) in the AI intervention arm versus 22% in the control arm — a statistically significant difference — with follow-through with an eye-care provider significantly higher in the AI arm. This constitutes the only published RCT of an autonomous AI device in any field of medicine. OSF HealthCare (multi-site US health system) deployed LumineticsCore across its primary care network; 25% of nearly 1,500 patients with diabetes screened positive for diabetic retinopathy. Following the initial rollout, OSF expanded to 24 additional sites. OhioHealth scaled to approximately 40 camera placements across its footprint, achieving a 162% increase in diabetic eye exams within 12 months of adoption. A 2025 *Ophthalmology Science* study on health-system implementation identified LumineticsCore as one of three FDA-cleared AI systems actively evaluated across academic health systems (alongside EyeArt and AEYE-DS), with LumineticsCore cited as the reference benchmark for autonomous deployment models. In 2023, Digital Diagnostics expanded into Labcorp patient service centres (beginning with nine sites in Alabama), enabling diabetic retinopathy screening within laboratory visits — a channel extension made possible precisely by the autonomous operating model: no eye specialist needs to be present to interpret results.

StudyDesignnSensitivitySpecificityAUCPublished
Abràmoff MD, Lavin PT, Birch M, et al. (IDx Technologies / University of Iowa)
ProspectiveProspective
90087.2%90.7%npj Digital Medicine, Aug 2018
Merin SR, Patel A, Serrano CI, et al. (ACCESS RCT, NCT05131451)
RCTRCT
163Nature Communications, Jan 2024

Clinical Pulse

Verified clinician sentiment

Verified clinician reviews launching · Apply to contribute →

Reviews are signed by clinicians verified against their regulator's public register.

Inside the Auris+ Listing

Six more sections complete this device’s Auris+ Listing.

  • Inside the algorithm

    Pro

    Pro unlocks the five-stage editorial walkthrough of how this device's AI reaches a verdict — model architecture, training data, inference latency, and where the clinician stays in the loop.

  • Decision Ledger

    Pro

    Pro unlocks a private, cross-vendor log of every case you read on this device — what the AI called, what you concluded, and a one-line reason if you overrode it. Ready for the EU AI Act's deployer logging obligations when they land in 2028.

  • Clinical Evidence Deep Dive

    Pro

    Pro unlocks the structured clinical-evidence summary — study count, target patient population, and a tabular accuracy-metrics view drawn from peer-reviewed sources.

  • Peer-Reviewed Publications

    Pro

    Pro unlocks the curated peer-reviewed publication list with PubMed cross-links — the citation backbone of every editorial verdict.

  • Post-Market & Regulatory Conditions

    Pro

    Pro unlocks the post-market surveillance summary, recall record, and the conditions of approval that bound real-world use.

  • AI Algorithm Version History

    Pro

    Pro unlocks the chronological record of algorithm version changes — what changed when, drawn from manufacturer changelogs and regulatory filings.

Upgrade to Pro →

Regulatory Approvals

CE
LumineticsCore (formerly IDx-DR) — Class IIa CE mark; available via Topcon Harmony imaging platform in Europe

Class IIa

FDA
IDx-DR — first autonomous AI diagnostic system authorised in any field of medicine (Class II, Product Code PIB)

DEN180001

Class II

Safety Record

No safety alerts or recalls on record.

No Class I or Class II FDA safety recalls specific to LumineticsCore or IDx-DR have been identified in publicly available FDA enforcement databases as of the review date. The MAUDE database does not contain a significant adverse-event cluster attributable to the device; the binary refer/rescreen output is designed to over-refer rather than under-refer (sensitivity-weighted design), which limits the systematic harm surface for false-negative screening outcomes. The pivotal trial itself enrolled only patients with no prior DR diagnosis, and the intended use is explicitly bounded to that population — the device is not indicated for patients with known retinopathy or for diagnostic grading beyond the binary threshold. Procurement officers should note that LumineticsCore is paired to a specific camera (Topcon TRC-NW400); image quality is not validated on other devices, and imageability rates may vary in populations with media opacity or small pupils. Digital Diagnostics remains a private company (Series B, $75M, led by KKR, August 2022; 97 employees as of February 2026) with no announced acquisition; operational continuity risk is standard for a mid-stage private medtech firm.

Intended Use & Indications

LumineticsCore (originally marketed as IDx-DR) is a software-only medical device intended for use by healthcare providers in primary care and laboratory settings to screen adults with diabetes for diabetic retinopathy. The provider acquires retinal images with the paired Topcon TRC-NW400 camera; the algorithm then analyses the images and issues an autonomous binary result — "refer to specialist" or "rescreen in 12 months" — without requiring the acquiring clinician or any other physician to interpret the fundus photographs. The device is indicated for patients with no prior diagnosis of diabetic retinopathy; positive screens are referred for comprehensive ophthalmological evaluation. The autonomous operating model means LumineticsCore can be deployed in settings where no eye-care specialist is present — primary care clinics, endocrinology practices, and laboratory patient-service centres.