Radiotherapy planning system and method
The present invention relates to a radiotherapy planning system ( 100 ) for determining a solution ( 101 ) corresponding to a fluence profile. The invention proposes to use a Pareto frontier navigator ( 140 ) to select the best plan from a set of various auto-planned solutions. An interactive graphical user interface ( 400 ) is provided to the planner to navigate among convex combinations of auto-planned solutions. This proposed Pareto plan navigation can be considered as a further optional refinement process, which can be applied to find the best plan in those cases where auto-generated solutions are not fully satisfying the planner's requirements. The navigation tool ( 400 ) moves locally through a set of auto-generated plans and can potentially simplify the planner's decision making process and reduce the whole planning time on complex clinical cases from several hours to minutes.
1. A radiotherapy planning system for determining a solution corresponding to a fluence profile, the radiotherapy planning system comprising:
an auto-planned solution generation unit for automatically generating one or more auto-planned treatment plans based on one or more dose quality metrics;
a weight assignment unit configured to assign a predetermined plurality of weights to said one or more dose quality metrics;
a weight adjustment unit configured to adjust a selected weight from said plurality of weights; and
a Pareto frontier navigation unit configured to generate said solution corresponding to said fluence profile in response to said adjusted selected weight;
wherein said radiotherapy planning system is configured to compare said solution to a clinical goal and wherein said radiotherapy planning system is further configured to generate a comparison signal indicative of whether or not said solution satisfies said clinical goal and to provide said solution as a warm start to generate a final auto-generated plan, if said comparison signal indicates that said solution does not satisfy said clinical goal.
2. The radiotherapy planning system of claim 1 , wherein said Pareto frontier navigation unit is configured to determine a convex hull piecewise linear approximation of a Pareto front.
3. The radiotherapy planning system of claim 1 , wherein said weight adjustment unit comprises a graphical user interface, where, for each respective weight from said plurality of weights, a slider is provided to adjust said respective weight.
4. The radiotherapy planning system of claim 3 , wherein, in response to receiving a user interaction with a slider, said radiotherapy planning system is configured to optimize an inner linear programming problem based on the adjusted weight.
5. The radiotherapy planning system of claim 3 , wherein said graphical user interface is further configured to update and display respective dose maps and dose volume histograms.
6. The radiotherapy planning system of claim 1 , wherein, if said comparison signal indicates that said solution satisfies said clinical goal, said radiotherapy planning system is further configured to deliver the said solution to a radiation therapy system.
7. The radiotherapy planning system of claim 1 , wherein said radiotherapy planning system is configured to employ a Pareto-front based refinement technique.
8. The radiotherapy planning system of claim 1 , wherein said radiotherapy planning system is configured to determine a set of treatment plans which sample a Pareto frontier.
9. The radiotherapy planning system of claim 1 , wherein said radiotherapy planning system is configured to determine a set of N+1 treatment plans, wherein N corresponds to the number of dose quality metrics, wherein said radiotherapy planning system is configured to determine N anchor treatment plans by optimizing each dose quality metric individually, and wherein said radiotherapy planning system is further configured to determine one additional balance treatment plan by using the same weight for each dose quality metric.
10. The radiotherapy planning system of claim 1 , wherein said radiotherapy planning system is configured to build an approximated Pareto front by generating convex linear combinations of said one or more auto-planned treatment plans.
11. The radiotherapy planning system of claim 1 , wherein said radiotherapy planning system is configured to normalize each of said one or more dose quality metrics.
12. A radiotherapy planning method for determining a solution corresponding to a fluence profile, the radiotherapy planning method comprising the steps of
generating one or more auto-planned treatment plans based on one or more dose quality metrics;
assigning a predetermined plurality of weights to said one or more dose quality metrics;
adjusting a selected weight from said plurality of weights;
generating said solution corresponding to said fluence profile in response to said adjusted selected weight by using a Pareto frontier navigation unit;
comparing said solution to a clinical goal; and
generating a comparison signal indicative of whether or not said solution satisfies said clinical goal, wherein, if said comparison signal indicates that said solution does not satisfy said clinical goal, said solution is provided as a warm start to generate a final auto-generated plan.
13. A computer readable storage medium encoded with one or more computer executable instructions, which, when executed by a processor of a computing system causes the processor to carry out the steps of the radiotherapy planning method defined by claim 12 .