Related Experiment Video
Updated: Jul 28, 2025
10:43
Eye-tracking Technology and Data-mining Techniques used for a Behavioral Analysis of Adults engaged in Learning Processes
Published on: June 10, 2021
5.4K
Skill, or Style? Classification of Fetal Sonography Eye-Tracking Data
Clare Teng1, Lior Drukker2,3, Aris T Papageorghiou2
1Institute of Biomedical Engineering, University of Oxford, United Kingdom.
Proceedings of Machine Learning Research
|May 30, 2023
Summary
This study introduces a novel method to classify fetal ultrasound skill using eye-tracking data, achieving high accuracy for expert and trainee sonographers. The findings suggest that years of experience correlate with sonographer expertise.
Area of Science:
- Medical Imaging
- Human-Computer Interaction
- Ophthalmology
Background:
- Clinical skill in fetal ultrasound scanning is often assessed by years of experience, categorizing sonographers as experts (10+ years) or beginners (0-5 years).
- Previous research on eye-tracking data for skill assessment required separating data into specific eye movements like fixations and saccades.
Purpose of the Study:
- To develop and validate a new method for classifying human skill in fetal ultrasound scanning.
- To assess skill classification using eye-tracking and pupillary data without prior assumptions on experience-skill relationships or data segmentation.
Main Methods:
- Utilized eye-tracking and pupillary data from sonographers performing fetal ultrasound scans.
- Developed a classification model that does not require pre-segmentation of eye-tracking data or assumptions about the correlation between experience and skill.
Main Results:
- The best-performing skill classification model achieved an F1 score of 98% for the expert class and 70% for the trainee class.
- Demonstrated a significant correlation between years of professional experience and sonographer expertise.
Conclusions:
- The proposed method effectively classifies sonographer skill in fetal ultrasound scanning using eye-tracking and pupillary data.
- The findings support the use of objective measures derived from eye-tracking for skill assessment in medical sonography, potentially refining traditional experience-based evaluations.
