Manual interactive inverse plan optimization for Intensity Modulated Radiation Therapy (IMRT) and Volumetric Modulated Arc Therapy (VMAT) is workload and time intensive, while plan quality may heavily depend on planner’s expertise, time availability for planning, and ambition of the planner and treating physician.
In the last few years, automatic planning (AP) has been introduced. Different approaches to AP can be used, the most common methods being knowledge-based planning (KBP), protocol-based automatic iterative optimization (PB-AIO), and multicriteria optimization (MCO), which can be pareto-navigation driven or automated. Some kind of solution is now available in most commercial treatment planning systems.
There are clear advantages of AP, the most significant being enormous decrease in planning workload, inter-operator variability reduction, and overall plan quality improvement, the latter depending on the applied AP system and configuration.
However, the process of introducing AP in clinical practice is generally complex and laborious, requiring extensive algorithm configuration and validation, since a suboptimal configuration translates into a plan quality systematically lower than feasible.
Moreover, a continuous effort is required to keep AP up to date, since implementation of new optimizers, cost functions or configuration options might rise opportunities for enhancing quality of automatically generated plans. AP must also be continuously adapted to new prescription strategies, fractionation schemes, volume definitions, dose constraints and delivery approaches. The need for monitoring the accuracy/consistency of AP output over time and robustness to adapted workflows is generally time consuming and not straightforward.
There are potential concerns related to the clinical use of AP. Automation of a task may result in an overrated and erroneous expectancy about easily redirecting resources/personnel towards other tasks. Safe and effective application of AP also requires continuous efforts to avoid generation of (slightly) suboptimal plans, that can remain unnoted for a long time. Another challenge is the loss of expertise in manual planning (which is also needed for AP configuration, especially for KBP). Apart from the need to maintain manual planning expertise, there is also the necessity for extensive training for optimal use of AP systems.
This presentation aims at discussing challenges of AP, and possibilities to cope with them to ensure that AP can indeed solve problems rather than increase them.