AI vs. ER docs: who makes the right call?

NCT ID NCT07632859

Completed Knowledge-focused Sponsor: Marmara University Pendik Training and Research Hospital Source: ClinicalTrials.gov ↗

First seen Jun 27, 2026 · Last updated Aug 14, 2026 · Updated 1 time

Summary

This study tests whether two AI models, GPT-4o and Claude 4.6 Sonnet, can correctly diagnose patients based on emergency department notes written in Turkish. Researchers will compare the AI's diagnoses to those made by the treating doctor and a panel of three specialists. The goal is to see if AI can help reduce diagnostic errors in emergency medicine.

What this could mean

Our plain-language read of the trial. This is informational only — not medical advice or a prediction.

Active substance: Large language models (GPT-4o and Claude 4.6 Sonnet) used to analyze patient notes
What this could lead to: If successful, this could show that AI can help emergency doctors make faster or more accurate diagnoses, potentially reducing errors.
What could go wrong: This is a retrospective study using written notes, not real-time patient care. The AI may not perform as well in a busy emergency room, and results may not apply to other languages or settings.

This is an AI summary of the original study and may miss details. Read our disclaimer.

Get updates

Get notified about this study

Sign up to get updates when this study changes or when new studies for ARTIFICIAL INTELLIGENCE (AI) IN DIAGNOSIS are added.

Our safety recommendation!

By submitting, you agree to our Terms of use

Conditions

The condition(s) this trial relates to.

Emergencies

As listed by the trial registrant

The condition terms exactly as the trial's registrant entered them.

ARTIFICIAL INTELLIGENCE (AI) IN DIAGNOSIS DIAGNOSTIC ERRORS EMERGENCY MEDICINE

Contacts and locations

Locations

Marmara University Pendik Training and Research Hospital

Istanbul, Istanbul, 34899, Turkey (Türkiye)

More trials for these conditions

Other studies related to the condition(s) this trial covers.