Experience and Survey Results From a National Hands-On Workshop On Simulated Error Training for the Physics Plan Review
P Johnson1,2*, L Schubert3, G Kim4, J Faught5, C Buckey6, L Conroy7, S Luk8, D Schofield9, S Parker10, (1) University of Florida Health Proton Therapy Institute, Jacksonville, FL, (2) University of Florida, Gainesville, FL (3) University of Colorado Denver, Aurora, CO, (4) UC San Diego, La Jolla, CA, (5) Memphis, TN, (6) Mayo Clinic Arizona, Phoenix, AZ, (7) Princess Margaret Cancer Centre, Toronto, ON, (8) University of Vermont Medical Center, Burlington, VT, (9) AdventHealth Orlando, Clermont, FL, (10) Wake Forest Baptist Health High Point Medical Center, Winston-salem, NC
Presentations
MO-FG-201-1 (Monday, 7/11/2022) 1:45 PM - 3:45 PM [Eastern Time (GMT-4)]
Room 201
Purpose: To evaluate the use and share the experience of simulated error training for the physics plan review in a national hands-on workshop.
Methods: The workshop was held during the 2021 AAPM Annual Meeting using Zoom. After two introductory presentations, users were split into four breakout rooms to participate in two case studies. All users were registered within AAPM’s ProKnow portal to allow access/interaction with case data including images, contours, dose, registrations, and plan documentation. Users were provided 20 minutes to review each case. The first case utilized Mosaiq/RayStation and 3DCRT treatment of the mediastinum. Embedded errors included an undocumented pacemaker, pacemaker treated with high-energy fields, pacemaker receiving more than 5 Gy, incorrect wedge angle, and missing field information. The second case utilized Aria/Eclipse and a spine SBRT treatment. Embedded errors included incorrect target delineation, inconsistent prescription, poor DRR quality, wrong setup instructions, and suboptimal plan quality. Users participated in a discussion and survey following each case to share their experience.
Results: 100+ people participated in the workshop with 53 providing feedback. Experience levels were: 13% residents, 30%<10 years, 29% 10-20 years, and 9%>20 years. Error detection rates ranged from 21-92%. In general, users detected higher priority failure modes (based on TG-275 Table 1A) at a higher rate, though the correlation was weak (r=0.4). Overall, 96% of respondents agreed/strongly agreed the datasets were an effective training tool and 75% of respondents agreed/strongly agreed that they planned to utilize the datasets at their institution.
Conclusion: In this workshop, users participated in two simulated plan checks using AAPM’s ProKnow portal and mock datasets created by the WGPE. Feedback was highly positive as to the efficacy of the datasets for education and training. Work is ongoing to expand the datasets to include more content and refine the training approach to further aid implementation.
Funding Support, Disclosures, and Conflict of Interest: This work was funded by an educational grant provided by Elekta.
Keywords
Risk, Quality Assurance, Simulation
Taxonomy
Contact Email