Actively Recruiting
PrediSuisse Study to Automatically Predict Difficult Airway Using Facial Recognition and Neural Networks with Three Videolaryngoscopes
Led by Centre Hospitalier Universitaire Vaudois · Updated on 2026-03-27
1800
Participants Needed
1
Research Sites
11 weeks
Total Duration
AI-Summary
What This Trial Is About
Researchers are investigating an automated software tool to predict the difficulty of tracheal intubation by analyzing facial features and movements. This study aims to create a reliable, reproducible, ultra-portable, and radiation-free system using machine learning and artificial intelligence techniques. The software will generate a difficulty score tailored to three different videolaryngoscopes and help anesthesiologists assess airway management risks objectively. The project involves collecting face and profile photos and video sequences from 900 patients during pre-anesthesia consultations. These images will be analyzed with neural networks to train the software to predict intubation difficulty, as determined by expert review of actual intubation procedures. The software will then be validated on an additional 900 patients undergoing intubation with one of three videolaryngoscopes C-Mac, MacGrath, or Airtraq. Participants will have their facial data collected during routine pre-anesthesia visits, followed by real intubations using the designated videolaryngoscope. Researchers will evaluate the softwares performance in predicting intubation difficulty and assess team communication over an 18-month period. The study does not involve any experimental treatments, and participation includes photo and video capture, data analysis, and observation of clinical intubation procedures.
CONDITIONS
Brief Title
PrediSuisse: Automatized Assessment of Difficult Airway
Research Team
P
Patrick Schoettker, PhD
Not the Right Trial for You?
Explore thousands of other clinical trials that might be a better match.
Sign up to get personalized trial recommendations delivered to your inbox.
Already have an account? Log in here