Evaluation of AI Models in Determining the Optimal PEEP

RecruitingOBSERVATIONAL
Enrollment

145

Participants

Timeline

Start Date

March 1, 2025

Primary Completion Date

February 1, 2026

Study Completion Date

February 2, 2026

Conditions
PEEP, Occult
Interventions
OTHER

AI-Assisted PEEP Optimization

"In this study, three artificial intelligence (AI) models (ChatGPT, DeepSeek, and Gemini) will analyze mechanical ventilation data, including compliance, elastance, auto-PEEP, time constant, and pressure-volume (P-V) curves, to generate patient-specific PEEP recommendations.~These AI-generated recommendations will be compared with manual PEEP titration performed by experienced ICU specialists. The AI models will also provide abnormal ventilation parameter detection, diagnostic suggestions, and treatment recommendations. The study aims to evaluate the reliability, accuracy, and clinical applicability of AI-generated outputs in optimizing PEEP settings for mechanically ventilated ICU patients."

Trial Locations (1)

34303

RECRUITING

Health Science University İstanbul Kanuni Sultan Süleyman Education and Training Hospital, Istanbul

All Listed Sponsors
lead

Kanuni Sultan Suleyman Training and Research Hospital

OTHER