Automated prediction of difficult intubation from face photos and short video

PrediSuisse: Automatized Assessment of Difficult Airway Using Three Videolaryngoscopes With the Help of Facial Recognition Techniques and Neural Network

Centre Hospitalier Universitaire Vaudois · NCT06453525

This project will test an AI tool that uses face photos and short videos taken at pre‑anesthesia visits to predict whether adults having elective general anesthesia will be difficult to intubate with three common videolaryngoscopes.

Quick facts

Study typeObservational
Enrollment1800 (estimated)
Ages18 Years to 100 Years
SexAll
SponsorCentre Hospitalier Universitaire Vaudois (other)
Drugs / interventionsradiation
Locations1 site (Lausanne)
Trial IDNCT06453525 on ClinicalTrials.gov

What this trial studies

The PrediSuisse project will train and validate an AI/CNN software to predict endotracheal intubation difficulty by collecting facial photos and short video sequences during pre‑anesthesia consultations. A training set of 900 patients will be labeled for real intubation difficulty by three expert reviewers who examine recorded intubations using three commercially available videolaryngoscopes with different blade types. The model will then be applied prospectively to a separate set of 900 routine patients to validate prediction performance and produce a device‑specific difficulty score. The software aims to be ultra‑portable, radiation‑free, and to provide an objective, reproducible score to support airway management decisions.

Who should consider this trial

Good fit: Adults (≥18) presenting for elective general anesthesia requiring tracheal intubation at a participating Swiss pre‑anesthesia clinic who can give informed consent and speak the site's language (French or Italian) are ideal candidates.

Not a fit: Patients who are under 18, cannot consent, have prior airway surgery that altered anatomy, do not speak the required language, or present for emergency procedures are unlikely to benefit from participation.

Why it matters

Potential benefit: If successful, the tool could give anesthesiologists a quick, objective risk score to guide airway planning and reduce unexpected difficult intubations.

How similar studies have performed: Previous machine‑learning studies to predict difficult airways from clinical data or images have shown mixed results, so applying CNNs to photos with device‑specific labels is a relatively new and cautiously promising approach.

Eligibility criteria

Show full inclusion / exclusion criteria
Inclusion Criteria:

* Adult patients (≥ 18 years old) presenting at the pre-anesthesia consult for an elective general anesthesia necessitating a tracheal intubation
* Signed informed consent.

Exclusion Criteria:

* Patients not speaking French (in Geneva and Lausanne) or Italian (in Lugano).
* Patients previously operated on the airway with anatomical modifications (ENT Flaps, tracheotomies).
* Patients unable to follow procedures or to give consent will also be excluded.

Where this trial is running

Lausanne

Study contacts

How to participate

  1. Review the eligibility criteria above with your treating physician.
  2. Visit the official trial page on ClinicalTrials.gov for the most current contact information and recruitment status.
  3. Contact the listed study coordinator or principal investigator to request pre-screening. Pre-screening is free and never obligates you to enroll.

View on ClinicalTrials.gov →

Conditions: Anesthesia, Intubation, Difficult or Failed, Airway Complication of Anesthesia, difficult airway, videolaryngoscopy

Last reviewed 2026-05-15 by the Find a Trial editorial team. Information on this page is for educational purposes and is not medical advice. Always consult qualified healthcare professionals about clinical trial participation.