Machine learning-based prediction of difficult laryngoscopy in infants with Pierre Robin sequence using quantitative 3D computed tomography parameters

1

Infants with Pierre Robin sequence often experience difficult laryngoscopic exposure due to anatomical challenges like mandibular hypoplasia and glossoptosis.
2

The study analyzed 214 infants with Pierre Robin sequence to identify 3D-CT parameters predictive of difficult laryngoscopic exposure.
3

Four independent predictors of difficult laryngoscopic exposure were identified: tongue length, tongue base-posterior pharyngeal wall distance, oropharyngeal area, and tongue base-epiglottic angle.
4

The Extra Trees machine learning model demonstrated superior generalizability in predicting difficult laryngoscopy with an AUC of 0.876 in an independent validation cohort.
5

Quantitative 3D-CT parameters are clinically relevant for predicting laryngoscopic exposure challenges in infants with Pierre Robin sequence.

1