Computer-assisted selection of importance factors in inverse planning

The clinical objectives are usually multifaceted and incommensurable. A set of importance factors (IFs) is often incorporated in the objective function in inverse planning to parameterize tradeoff strategies and to prioritize the dose conformality in different structures. Whereas the general formalism remains the same, different sets of IFs characterize plans of obviously different flavor and critically influence the final plan. Up to now, the determination of these parameters has been a “guessing” game based on empirical knowledge because the influence of these parameters on the plan is not known until the optimization is completed. To compromise properly the conflicting requirements of the target and sensitive structures, the parameters are usually adjusted through trial-and-error. Here, the authors report a computational algorithm to automate the selection of the parameters. The plan selection is done in two steps. First, a set of IFs are chosen and the corresponding beam parameters are optimized under the guidance of an objective function using an iterative algorithm. The “optimal” plan is then evaluated by an additional scoring function. The IFs in the objective function are adjusted accordingly to improve the ranking of the plan. For every change in the IFs, the beam parameters need to be re-optimized. This process continues in an iterative fashion until the scoring function is saturated. The algorithm was applied to two clinical cases and the results demonstrated that it has the potential to improve significantly the existing method of inverse planning