Preoperative Airway Images for Difficult Airway Prediction (AI-AIRWAY)

Multimodal Artificial Intelligence for Image-Based Prediction of Difficult Airway: A Prospective Observational Study

This prospective observational study will evaluate whether commonly available multimodal artificial intelligence models can predict difficult laryngoscopy and difficult intubation using standardized preoperative airway photographs. Adult patients scheduled for elective surgery requiring endotracheal intubation will undergo an eight-view preoperative airway photography protocol. The anonymized image sets will be assessed by ChatGPT, Gemini, and Grok using the same structured prompt. Their predictions will be compared with expert anesthesiologist image-based assessments, conventional airway evaluation findings, and prospectively recorded intraoperative airway outcomes. The primary aim is to determine the diagnostic performance of AI models for predicting difficult intubation. A key secondary aim is to evaluate their performance for predicting difficult laryngoscopy. The study is intended to explore whether image-based AI assessment may support preoperative airway risk stratification as a clinician-supervised screening tool.

Study Overview

• Status: Active, not recruiting
• Conditions: Difficult Intubation; Difficult Laryngoscopy

Detailed Description

Preoperative airway assessment is important for identifying patients at risk for difficult laryngoscopy or difficult intubation. However, conventional bedside airway predictors have limited accuracy when used alone. Multimodal artificial intelligence models may provide additional image-based information by evaluating visible anatomical features from standardized preoperative airway photographs.

In this prospective observational study, adult patients undergoing elective surgery requiring endotracheal intubation will be enrolled between June and September 2026. Each participant will undergo standardized eight-view airway photography during the pre-anesthetic evaluation. The image set will include frontal facial, lateral profile, maximal mouth opening, modified Mallampati, neck extension, and anterior neck views. Images will be anonymized before assessment.

The same image sets will be independently evaluated by multimodal AI models, including ChatGPT, Gemini, and Grok, using an identical structured prompt. The AI models will provide categorical and binary predictions for difficult laryngoscopy and difficult intubation based only on visible image-based anatomical features. No intraoperative outcome data, expert predictions, or conventional airway assessment results will be provided to the AI models.

AI-generated predictions will be compared with expert anesthesiologist image-based assessments, conventional airway evaluation parameters, and prospectively recorded intraoperative reference outcomes. Difficult laryngoscopy will be defined as Cormack-Lehane grade III or IV. Difficult intubation will be defined using objective intraoperative criteria, including more than one intubation attempt, need for bougie or stylet assistance, rescue use of video laryngoscopy or supraglottic airway device, intubation time exceeding 60 seconds, or Intubation Difficulty Scale score greater than 5.

The study will assess the sensitivity, specificity, positive predictive value, negative predictive value, accuracy, receiver operating characteristic performance, and agreement between AI models and expert anesthesiologist assessments. The findings may help clarify whether multimodal AI can serve as a clinician-supervised adjunct for preoperative difficult airway risk stratification.

• Study Type: Observational
• Enrollment (Estimated): 319

Contacts and Locations

Study Locations

• Turkey (Türkiye)
  • Kadıköy
    • Istanbul, Kadıköy, Turkey (Türkiye), 34734
      • Dr. Siyami Ersek Thoracic and Cardiovascular Surgery Training and Research Hospital

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

Adult patients scheduled for elective surgical procedures requiring endotracheal intubation at Dr. Siyami Ersek Thoracic and Cardiovascular Surgery Training and Research Hospital.

Description

Inclusion Criteria:

• Age 18 years or older
• Scheduled for elective surgery requiring endotracheal intubation
• Able to cooperate with the standardized preoperative airway photography protocol
• Able to provide written informed consent

Exclusion Criteria:

• Age younger than 18 years
• Emergency surgery
• Refusal or inability to provide informed consent
• Inability to cooperate with the standardized photographic protocol
• Known craniofacial or cervical deformity
• History of major head and neck surgery or radiotherapy
• Obstruction of key anatomical landmarks by facial hair, dressings, cervical collars, or other external devices
• Incomplete or poor-quality image sets despite repeated acquisition
• Missing clinical airway assessment data
• No endotracheal intubation performed
• Airway difficulty could not be reliably evaluated
• Planned awake fiberoptic intubation or other preplanned advanced airway technique because of known difficult airway

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort
Elective Surgery Patients Requiring Endotracheal Intubation; Adult patients scheduled for elective surgery requiring endotracheal intubation who will undergo standardized preoperative airway photography and prospective intraoperative airway outcome recording.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Diagnostic Performance of Multimodal AI Models for Predicting Difficult Intubation	The primary outcome is the diagnostic performance of multimodal artificial intelligence models for predicting true difficult intubation based on standardized preoperative airway photographs. Difficult intubation will be determined using prospectively recorded intraoperative reference criteria, including more than one intubation attempt, need for bougie or stylet assistance, rescue use of video laryngoscopy or supraglottic airway device, intubation time exceeding 60 seconds, or Intubation Difficulty Scale score greater than 5. Diagnostic performance will be assessed using sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and receiver operating characteristic analysis.	From preoperative airway photography to completion of intraoperative endotracheal intubation, up to 1 day

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Diagnostic Performance of Multimodal AI Models for Predicting Difficult Laryngoscopy	The key secondary outcome is the diagnostic performance of multimodal artificial intelligence models for predicting true difficult laryngoscopy based on standardized preoperative airway photographs. Difficult laryngoscopy will be defined as Cormack-Lehane grade III or IV recorded during intraoperative airway management. Diagnostic performance will be assessed using sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and receiver operating characteristic analysis.	From preoperative airway photography to completion of intraoperative laryngoscopy, up to 1 day

Collaborators and Investigators

• Sponsor: Memorial Atasehir Hospital

Publications and helpful links

General Publications

• Tavolara TE, Gurcan MN, Segal S, Niazi MKK. Identification of difficult to intubate patients from frontal face images using an ensemble of deep learning models. Comput Biol Med. 2021 Sep;136:104737. doi: 10.1016/j.compbiomed.2021.104737. Epub 2021 Aug 4.
• Wang Z, Jin Y, Zheng Y, Chen H, Feng J, Sun J. Evaluation of preoperative difficult airway prediction methods for adult patients without obvious airway abnormalities: a systematic review and meta-analysis. BMC Anesthesiol. 2024 Jul 17;24(1):242. doi: 10.1186/s12871-024-02627-1.
• Garcia-Garcia F, Lee DJ, Mendoza-Garces FJ, Garcia-Gutierrez S. Reliable prediction of difficult airway for tracheal intubation from patient preoperative photographs by machine learning methods. Comput Methods Programs Biomed. 2024 May;248:108118. doi: 10.1016/j.cmpb.2024.108118. Epub 2024 Mar 12.

Study record dates

Study Major Dates

• Study Start (Actual): June 25, 2026
• Primary Completion (Estimated): September 1, 2026
• Study Completion (Estimated): September 1, 2026

Study Registration Dates

• First Submitted: July 7, 2026
• First Submitted That Met QC Criteria: July 7, 2026
• First Posted (Actual): July 14, 2026

Study Record Updates

• Last Update Posted (Actual): July 14, 2026
• Last Update Submitted That Met QC Criteria: July 7, 2026
• Last Verified: July 1, 2026

More Information

Terms related to this study

• Keywords: Artificial intelligence; Difficult intubation; Difficult airway; Difficult laryngoscopy; Multimodal artificial intelligence; Preoperative airway assessment; Airway photographs; Image-based prediction
• Other Study ID Numbers: E-28001928-604.01-318945497

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO
• IPD Plan Description: Individual participant data will not be shared due to privacy and confidentiality considerations, particularly because the study involves preoperative airway images. De-identified aggregate data will be presented in the final analysis and publication.

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No