Related Experiment Video
Updated: May 31, 2026
05:49
An Automated Squint Method for Time-syncing Behavior and Brain Dynamics in Mouse Pain Studies
Published on: November 1, 2024
Real-time face keypoint detection for pre-anesthetic assessment with optimized YOLO11 model based on DeBiFormer
1School of Automation and Intelligent Sensing, Shanghai Jiao Tong University, Shanghai, China.
Frontiers in Medical Technology
|May 29, 2026
Summary
This study introduces an AI model for automated pre-anesthetic assessment, improving airway evaluation reliability. The YOLO11-Pose framework with DeBiFormer accurately measures keypoints for better patient safety.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Anesthesiology
Background:
- Pre-anesthetic assessment is crucial for airway management but often lacks quantitative reliability.
- Current methods for evaluating airway characteristics can be subjective and prone to variability.
Purpose of the Study:
- To develop an automated, image-based framework for pre-anesthetic assessment using an optimized YOLO11-Pose model.
- To quantitatively analyze facial and hand keypoints relevant to airway assessment, including mouth opening, thyromental distance, and neck mobility.
Main Methods:
- Implementation of an optimized YOLO11-Pose framework integrated with a DeBiFormer module.
- Single-stage inference for localizing key facial and hand landmarks.
- Evaluation on a controlled dataset from Ruijin Hospital, employing Bland-Altman analysis, pixel-level error, and normalized mean error (NME).
Main Results:
- High detection performance for key facial and hand points.
- Reliable agreement with reference annotations, confirmed by Bland-Altman analysis and error metrics.
- Successful Mallampati classification with 77.34% accuracy (4-class) and 83.99% accuracy (binary), kappa=0.65.
Conclusions:
- The proposed AI method offers robust and clinically meaningful assessment of airway characteristics.
- This automated approach has the potential to enhance the reliability and efficiency of pre-anesthetic evaluations in clinical practice.