Varian’s RapidPlan™ knowledge-based planning brings the power of machine learning to the challenge of treatment planning and has been helping clinicians at centers around the world since 2014 to achieve new levels of consistency, efficiency, and quality. RapidPlan was designed to streamline the treatment planning process by enabling clinicians to use quality models as a guideline and starting point in the treatment planning process, reducing or even eliminating the need for multiple, time-consuming iterations seen with traditional planning approaches.
“RapidPlan leverages machine learning algorithms to convert previously-created high-quality plans into models that expedite the creation of new plans,” said Mu Young Lee, vice president, Radiation Oncology Software Solutions at Varian. “This enables a self-perpetuating cycle of continuous improvement in plan quality, as new and better inputs can be added to the models.”
Implementing RapidPlan using the models provided by Varian or models shared across the radiation oncology community creates minimal impact on the physics team and is the fastest route to implementation of RapidPlan. However, many centers want to develop their own RapidPlan models. Northwell Health, New York’s largest network of community-based healthcare facilities with seven that offer radiation oncology services, was one such organization. Recently, Varian interviewed the physics team at Northwell Health about their experience utilizing the services of a third-party physics consulting company to aggregate data from their best head & neck treatment plans and train their RapidPlan model for use going forward. Read the case study.