1067 / 1161
Different plan optimization approaches
•
Fluence map optimization (FMO)
➢
Linear relationship between fluence and dose
➢
Less computational effort
➢
Still leaf-sequencing needs to be done either by optimizing MU or aperture
•
Direct machine parameter optimization (DMPO)
➢
leaf positions and segment weights as variables during optimization
➢
More difficult to solve this problem as more physical constraints exist
➢
Uses simulated annealing, column generation, gradient-based methods or
genetic algorithms or heuristic methods
•
For IMPT FMO is used
•
Problem is to translate objectives and constraints with non-clinical
meaningful weights into objective function
→
plan quality still
depends on time commitment and experience of planer
→
multi
criteria optimization (weight factors avoided)