Differentiating Operator Skill during Routine Fetal Ultrasound Scanning using Probe Motion Tracking

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Ultrasound Technology

Background:

• Operator skill significantly impacts diagnostic accuracy in fetal ultrasound.
• Objective assessment of ultrasound operator skill is challenging.
• Standardized methods for skill evaluation are needed.

Purpose of the Study:

• To develop and validate a deep learning framework for differentiating operator skill levels in fetal ultrasound.
• To model ultrasound probe motion for skill classification.
• To create a skill assessment tool invariant to individual scanning styles.

Main Methods:

• Acquisition of probe motion data during routine second-trimester fetal ultrasound scans.
• Utilizing a novel convolutional neural network (CNN)-based deep learning framework.
• Training the model on data from operators with known experience levels (newly-qualified and expert).

Main Results:

• The proposed deep learning model successfully identified distinct probe motion features.
• The framework achieved 95% accuracy in classifying operator skill levels.
• The model demonstrated invariance to operators' personal scanning styles.

Conclusions:

• Deep learning analysis of ultrasound probe motion is a viable method for objective skill assessment.
• This approach can differentiate between novice and expert operators in fetal ultrasound.
• The framework holds potential for improving ultrasound training and quality assurance.