MENU

PREVIA AI: PTSD Risk Prediction via Emergency Room Narratives

PR Newswire(NYU Langone Health) USA
Overview
Researchers at NYU Langone Health developed PREVIA, an AI tool that predicts PTSD risk by analyzing patient narratives during emergency department (ED) intake. The system identified 100% of PTSD cases in a head-to-head test against a specialist, offering a scalable screening method to enable early preventive intervention.
In Depth

Key Findings

The AI tool PREVIA correctly identified more than 95.4 percent of at-risk patients in a study involving 145 emergency department (ED) patients. In a direct comparison, PREVIA flagged 100 percent of 30 cases that later developed PTSD, outperforming an ED clinician specializing in PTSD, who correctly identified 89 percent of those cases. The system utilizes spontaneous answers to the routine question, “What brought you to the ED?”, to forecast risk one month later.

Technical Details

PREVIA is built on open-source models without requiring additional training or fine-tuning. It employs a three-layer pipeline to synthesize risk factors:

  • First Layer: Scans narratives for peritraumatic symptoms (e.g., amnesia, loss of control) and early ED symptoms (e.g., feeling disconnected from the body).
  • Second Layer: Weighs risk factors, such as lower socioeconomic status and pre-existing mental health issues, against protective factors like social support and coping skills. This layer utilizes different models for specific strengths; for example, Llama emphasizes risk signs while Mistral highlights protective factors.
  • Third Layer: Synthesizes assessments into a structured summary to prevent any single model from dominating the result.

The final output combines this summary with demographic data (age, sex), trauma type, and PTSD prevalence rates to produce a risk score with a plain-language explanation. Results were validated using the Clinician-Administered PTSD Scale for DSM-5 (CAPS-5).

Background & Context

Approximately 43 million patients are admitted to US emergency departments annually for physical injuries, with over a quarter developing PTSD. While validated tools like the Posttraumatic Adjustment Scale exist, only 28 percent of EDs routinely screen for PTSD, and only two-thirds of those use validated tools. This gap is largely due to time constraints, as clinicians typically have only 15–20 minutes per patient.

Significance & Outlook

Because many patients lose contact with hospitals after the first month, the ED represents a critical window for early detection. By providing an automated screener that requires no additional intake time, PREVIA enables timely preventive interventions. The use of open-source models allows other hospitals to test the tool independently without proprietary software.

Source: https://www.prnewswire.com/news-releases/ai-predicts-ptsd-risk-based-on-stories-about-what-brought-patients-to-the-emergency-room-302926212.html

Get our weekly technology intelligence — free

Receive an infographic that lets you judge at a glance whether each field’s analysis report is worth reading.

Subscribe Free — Weekly Tech Intelligence

By subscribing, you’ll receive Troy-Technical’s weekly technology intelligence newsletter.

  • Your email and selected fields are used only to deliver the newsletter.
  • We never share your information with third parties.
  • You can unsubscribe anytime via the link in each email.

See our Privacy Policy for details.

Takes about a minute · Unsubscribe anytime

Published by Troy-Technical, an independent site run by one engineer with a career in materials development.
About the author / Contact info@troy-technical.jp
Let's share this post !

Author of this article

TOC