RTT training in advance of online-adaptive using web-based contouring tools
Alex Beardmore,
United Kingdom
MO-0394
Abstract
RTT training in advance of online-adaptive using web-based contouring tools
Authors: Alex Beardmore1, Rachel Brooks-Pearson1, Karen Pilling1
1Newcastle upon Tyne Hospitals NHS Foundation Trust, Northern Centre for Cancer Care, Newcastle upon Tyne, United Kingdom
Show Affiliations
Hide Affiliations
Purpose or Objective
Online-adaptive radiotherapy (OART) is expected to become the treatment of choice for some complex disease sites in the near future. OART will require treatment RTTs to have advanced skills in organ at risk (OAR) contouring, critique of changes with target volumes and in plan evaluation. Current IGRT training does not prepare RTTs for all of these requirements. Providing training for these skills to RTTs should be done in a way which will limit the training burden on clinicians. This study evaluates integrating innovative web-based contouring tools (ProKnow) into an IGRT training programme to provide experience with OAR and target volume contouring. ProKnow uses peer reviewed, expert contours on anonymous CT data-sets to give users an initial and a final traffic-light result score.
Material and Methods
10 RTTs undergoing IGRT training for VMAT prostate radiotherapy were asked to complete a structured contouring exercise using ProKnow. An in-house anatomy training presentation was given to RTT trainees and was supplemented by the contouring instruction within ProKnow prior to RTT trainees completing the contouring exercises. Trainees were directed to contour 6 CT data-sets (2x prostate, 2x seminal vesicles, 1x bladder and 1x rectum). The ProKnow system compares the RTT trainee’s initial and final attempts to expert contours and reports a Dice similarity co-efficient for each contouring attempt. Paired t-tests of mean initial and final Dice scores were performed in Microsoft Excel.
Results
RTTs showed a statistically significant increase in Dice scores for the prostate and seminal vesicle target volumes (0.739 vs 0.850, p = 0.001 and 0.566 vs 0.794, p = 0.007 respectively) and for the rectum OAR (0.720 vs 0.882 p = 0.010) between the initial and final attempts. RTTs were satisfied with their initial bladder OAR contours and therefore did not reattempt this contouring exercise. The mean Dice score for the bladder OAR was 0.943.
Conclusion
Integrating contouring exercises with tools that use effective expert comparisons can help RTTs to develop some of the skills needed for a RTT-led OART service to be provided. This study shows that RTTs significantly improve in their contouring accuracy using a self-directed approach while limiting the time burden on training staff. With large investments into technology capable of performing OART, it is vital that OART is used clinically very early in the implementation to deliver a strong return on the investment. This study indicates that contouring accuracy exercises could be integrated into RTT IGRT training now to build RTT skills and experience in preparation for OART.