What is it about?
Ultrasound diagnostics are non-invasive and are increasingly used each year, yet they remain very challenging to perform accurately. While existing training methods have proven useful, they lack realism as they are not conducted on human subjects. In this study, we propose a new system that enables training to 'detect lesions in an actual human body' without involving patients, by embedding virtual lesions into ultrasound images using image processing technology.
Featured Image
Photo by Bagoes Ilhamy on Unsplash
Why is it important?
Existing simulation training mimics realistic clinical experiences without using actual patients, providing learners with safe and practical training in probe manipulation. However, current content lacks opportunities for learners to practice lesion detection and clinical decision making in scenarios that most closely replicate the clinical setting, specifically involving actual human subjects where the presence of lesions is uncertain. Furthermore, most prior studies focus on improving existing training methods. These studies do not allow for training with human subjects during the practice phase. To address this issue, we introduce a novel playable design: virtual patientization. While a conventional practice allows the learner to understand scanning procedures and human anatomy, our design extends its capabilities by providing opportunities to practice lesion detection and clinical decision making, thereby enhancing diagnostic skills through real human interaction.
Perspectives
We take pride in this article as it presents a training approach from a new perspective, different from previous ultrasound training methods. In fact, some of the physicians who participated in our research have expressed interest in adopting it if realized. We hope this research will succeed and assist many physicians in their practice.
Masaki Goda
Kyoto Daigaku
Read the Original
This page is a summary of: Virtual Patientization: A Playable Design for Clinical Ultrasound Training by Embedding Virtual Lesions, October 2024, ACM (Association for Computing Machinery),
DOI: 10.1145/3665463.3678788.
You can read the full text:
Contributors
The following have contributed to this page
