Physics pre-treatment and weekly chart checks are used for quality assurance in radiation oncology. The process is completed manually and must be augmented by additional tools. The project aims to optimize workload during interactions of clinical personnel with physics pre-treatment and weekly chart checks by decreasing the cognitive burden of the chart check process and flagging cases that require extra cognitive scrutiny. The project focuses on predicting patient plan difficulty using a machine learning algorithm.

Learning Objective: After attending the poster session, the learner should be able to:
● Understand how machine learning can be used to augment quality assurance measures in radiation oncology.
● Understand the components of a treatment plan that contribute to plan checking difficulty.
● Understand how machine learning can help improve effectiveness of the chart checking process and reduce cognitive workload.

Authors:

Malvika Pillai (Presenter)
UNC Chapel Hill

Karthik Adapa, UNC Chapel Hill

Presentation Materials:

• Abstract/Manuscript - 1