Funding pending - Randomized trial

TRIADE

AI-Assisted Triage for Informed Decision-Making

Project Description

TRIADE is a prospective cluster-randomized bicentric trial designed to evaluate the impact of AI assistance on nursing triage performance under real-world conditions. The objective is to demonstrate that AI assistance can significantly improve concordance between nursing triage and final medical evaluation.

Design

Prospective cluster-randomized bicentric trial with crossover

Population

~2000 adult patients (1000 per arm)

Centers

Lille University Hospital
Maubeuge Hospital

Funding

PHRC-I (submitted)
Doctoral grant (submitted)

Objectives

Primary: Concordance improvement from 68% to 76% (+8 points)

Secondary: Under-triage reduction, acceptability evaluation (SUS), triage time analysis, serious adverse events

Methodology

Control period: Standard nursing triage
Intervention period: Triage with AI suggestion (nurse remains decision-maker)

Project Team

Dr Edouard Lansiaux

Coordinating Investigator

Lille University Hospital

Pr Emmanuel Chazard

Methodologist

METRICS ULR 2694

Pr Daniel Balouek

Cloud Computing & Edge AI Expert

IMT Atlantique

Maubeuge Hospital Team

Partner Center

CH Sambre-Avesnois

Pr Hayfa Zgaya-Biau

Process Mining Expert

METRICS ULR 2694 & CRIStAL UMR 9189

Pr Mehdi Ammi

Man-Machine Interface Expert

LIASD