Errors in the planning and delivery of radiation therapy can result in inferior outcomes for patients. The magnitude of this problem is now well-appreciated from cooperative group trials and other sources. At present, however, there are few concrete tools to reduce the chance of plan errors passing undetected. Current practice uses a variety of error mitigation means such as human inspection of the radiation treatment plan or rule-based checking of plan variables. However, research has shown that many errors are still missed. As radiation therapy becomes more complex and varied, and as the demands on staff increases, new methods are needed. Our development will result in an independent means of verifying that the type of radiotherapy treatment and the way it is prescribed are in keeping with the standards already in place within a specific clinic. The software system will be capable of comparing every treatment variable with what would be expected for a given type of disease and location. With the large number of variables now created for complex radiotherapy treatments ranging into the hundreds, human checking for consistency of treatment technique after treatment plan generation is impractical. The tools created in this proposed project will provide the ability to cross check an entire area of treatment planning that is largely overlooked in today?s radiotherapy clinics.
This proposed project will have a significant impact on public health. It will catch human errors in the development of complex radiotherapy treatment preparation and delivery that heretofore go unnoticed, where the errors range from minor to life threatening. Conservative estimates range from 1% to 2% of patients undergoing radiotherapy treatment may benefit from the reduction of errors that could be achieved with the implementation of the proposed independent probabilistic monitoring system.