TUHH Open Research
Help
  • Log In
    New user? Click here to register.Have you forgotten your password?
  • English
  • Deutsch
  • Communities & Collections
  • Publications
  • Research Data
  • People
  • Institutions
  • Projects
  • Statistics
  1. Home
  2. TUHH
  3. Publications
  4. Pre-operative three-dimensional face scans for predicting difficult facemask ventilation: a prospective development study
 
Options

Pre-operative three-dimensional face scans for predicting difficult facemask ventilation: a prospective development study

Citation Link: https://doi.org/10.15480/882.17318
Publikationstyp
Journal Article
Date Issued
2026-06-05
Sprache
English
Author(s)
Wünsch, Viktor A.  
Bommes, Hannes
Germer, Sofia  
Piltan, Milan
Breitfeld, Philipp  
Latus, Sarah  orcid-logo
Medizintechnische und Intelligente Systeme E-1  
Krause, Linda
Zöllner, Christian  
Gundler, Christopher  
Petzoldt, Martin  
TORE-DOI
10.15480/882.17318
TORE-URI
https://hdl.handle.net/11420/63495
Journal
Anaesthesia  
Citation
Anaesthesia (in Press): (2026)
Publisher DOI
10.1111/anae.70261
Scopus ID
2-s2.0-105041067986
Publisher
Wiley-Blackwell
Introduction: Facemask ventilation is a key airway management skill but predicting difficulty can be challenging. Pre-operative three-dimensional face scanning may have diagnostic value. We aimed to identify interpretable facial shape features and to quantify their value for predicting difficult facemask ventilation. Methods: In this prospective observational single-centre study, pre-operative three-dimensional face scans were obtained, and a structured airway assessment was performed on patients undergoing ear, nose and throat or maxillofacial surgery. The primary outcome was difficult facemask ventilation documented as an alert in the patient health record. After postprocessing, three-dimensional face scans were fitted to an established, non-clinical facial model to identify interpretable shape coefficients. The area under the receiver operating characteristic (AUROC) curve for the DIFFMASK score was calculated before and after enrichment with three facial shape features and the added diagnostic value was assessed using likelihood ratios. Results: Data from 398 patients were analysed. The optimism-corrected AUROC was 0.73 (95%CI 0.65–0.80) for the DIFFMASK score and 0.74 (95%CI 0.66–0.82) for selected facial shape features. Enrichment of the DIFFMASK score with three facial shape features improved goodness of model fit (p = 0.002) and achieved an optimism-corrected AUROC of 0.76 (95%CI 0.68–0.82). Generated face meshes with superimposed colour mapping revealed that morphological features of the nose, lower mandible, neck region and facial convexity were most predictive of difficult facemask ventilation. Discussion: Pre-operative three-dimensional face scans predicted difficult facemask ventilation at least as well as the DIFFMASK score. Integrating the features of three selected facial shapes enriched the DIFFMASK score and improved its diagnostic value. Digital phenotyping can complement traditional clinical assessment.
Subjects
airway management
facial recognition
respiratory system
ventilation
DDC Class
616: Diseases
617: Surgery, Regional Medicine, Dentistry, Ophthalmology, Otology, Audiology
Lizenz
https://creativecommons.org/licenses/by/4.0/
Publication version
publishedVersion
Loading...
Thumbnail Image
Name

Anaesthesia - 2026 - Wünsch - Pre‐operative three‐dimensional face scans for predicting difficult facemask ventilation a.pdf

Type

Main Article

Size

410.26 KB

Format

Adobe PDF

TUHH
Weiterführende Links
  • Contact
  • Send Feedback
  • Cookie settings
  • Privacy policy
  • Impress
DSpace Software

Built with DSpace-CRIS software - Extension maintained and optimized by 4Science
Design by effective webwork GmbH

  • Deutsche NationalbibliothekDeutsche Nationalbibliothek
  • ORCiD Member OrganizationORCiD Member Organization
  • DataCiteDataCite
  • Re3DataRe3Data
  • OpenDOAROpenDOAR
  • OpenAireOpenAire
  • BASE Bielefeld Academic Search EngineBASE Bielefeld Academic Search Engine
Feedback