# Helping specialists see urgency.

Researchers demonstrated deep-learning diagnosis and referral recommendations from retinal scans.

Canonical: https://brightaifuture.com/discoveries/retina
Format: story
Source publication: 2018-08-13
Bright publication: 2026-09-05
Substantive update: None recorded
Evidence and review: Demonstrated; confidence: unassessed; source-checked; legacy source check. Legacy source check; no named human reviewer is recorded in this projection.

## The human problem

Specialists must identify which eye conditions need urgent attention.

## The prior constraint

Clinicians interpret detailed optical coherence tomography scans.

## AI’s actual role

A two-stage model mapped scan features to diagnoses and referral suggestions.

## The documented result

The system demonstrated strong retrospective referral performance.

## Why it may matter

It suggested a possible tool for supporting specialist triage.

## Limitations

Retrospective performance does not establish routine clinical safety or benefit.

## Unresolved questions

Evaluate clinical workflows and outcomes prospectively.

## Provenance and history

{
  "dates": {
    "eventDate": null,
    "publicationDate": "2018-08-13",
    "captureDate": null,
    "lastReviewedDate": "2026-09-05"
  },
  "provenance": {
    "origin": "legacy-projection",
    "externalId": "retina"
  },
  "revisions": [],
  "corrections": []
}

## Original sources

- [Clinically applicable deep learning for diagnosis and referral in retinal disease](https://www.nature.com/articles/s41591-018-0107-6)

## Continue exploring

- [Concise evidence record](https://brightaifuture.com/discoveries/retina)