Feasibility of a semi-automated approach to grading point of care ultrasound image generation skills

Zizui Chen, Mohamed S. Shehata, Minglun Gong, Heather Carnahan, Adam Dubrowski, Andrew J. Smith

Abstract: Development and evaluation of Point of Care Ultrasound (PoCUS) skill is a resource intensive undertaking. Current practice involves expert supervision of trainees performing numerous practice scans in the clinical setting. Computer augmented training represents the potential for significant reduction in resources utilization. Multiple ultrasound training simulators exist however of unclear value in teaching image generation skills. This paper describes the concept of using a standard ultrasound machine and human subjects, combined with image processing and depth sensing technologies, to develop a realistic PoCUS training tool. In addition to the concept, we describe the initial data collection experiment and preliminary work integrating ultrasound imagery and probe movement. Operator assessment metrics explored in this paper include image quality, stability, and acquisition time, demonstrate good potential to differentiate between novice and experienced sonographers.