Encrypted login | home

Program Information

Implementation of An Efficient Workflow for the Analysis of Alerts Observed During Large Scale EPID-Based 3D in Vivo Dosimetry

B Mijnheer

B Mijnheer*, A van Mourik , I Olaciregui Ruiz , A Mans , The Netherlands Cancer Institute, Amsterdam, The Netherlands


SU-K-205-10 (Sunday, July 30, 2017) 4:00 PM - 6:00 PM Room: 205

Purpose: In our institution, more than 5000 RT treatments per year are verified using EPID-based 3D in vivo dosimetry. With our current set of tolerance levels, deviations are detected in about 30% of the treatments. The purpose of this study is to investigate the usefulness of a newly developed workflow to analyze the alerted treatments.

Methods: In the new workflow, in vivo reports are automatically created almost immediately after treatment delivery. If no deviation is detected, the treatment is automatically approved. In case of an alert, the new framework links extra sources of information (e.g., cone-beam CT and trends per patient) and imposes treatment-site specific checks to help medical physicists to explain the reason of the deviation. In cases where the clinical relevance of a deviation is still in doubt, the workflow prompts the scheduling of extra actions such as EPID or ion chamber based 3D phantom dosimetry. Radiation oncologists and therapists can be consulted at various levels during this workflow.

Results: In 2016 about 20 alerts had to be analyzed daily by a medical physicist requiring one to two hours. From these alerted deviations, 83% could immediately be explained following the protocol of the workspace, while 13% and 4% of the alerted cases were approved after an EPID or ion chamber based phantom check, respectively. The extra time for these phantom measurements amounts to about 30 min per day. The plan was adapted in 0.8% of the alerts.

Conclusion: The automated workflow ensures the timely inspection of each palliative and curative RT treatment at a minimal amount of inspection work. Furthermore, the structured alert handling approach facilitates consistent decision making and is an excellent tool to check the overall clinical process.

Contact Email: