A SYSTEM FOR GENERATING OBJECTIVE FUNCTIONS FOR TREATMENT PLANNING
System (OGS) and related method for generating an objective function for use in radiation treatment, RT, planning. The system may include a machine learning model (M) and a training system (TS) for training the model (M) based on training data. The training data may include previous (at least partial) RT plans, and a user awarded ranking thereof. The model, once trained, may be used as the objective function. The system allows a user to turn, in a defined manner, clinical objectives or goals into a computable objective function which can be used for RT planning.
1 . System configured for generating an objective function for radiation treatment planning, wherein the system comprises a user input interface configured to allow a user to provide training data, the training data including:
i) a plurality of previous at least partial radiation treatment plans, and
ii) a user defined ranking of the plurality of previous at least partial radiation treatment plans,
and wherein the system is based on a machine learning model, and the system comprises a training system configured to train the machine learning model based on training data.
2 . System as claimed in claim 1 , wherein the objective function is, or comprises, the trained machine learning model.
3 . System as claimed in claim 1 , wherein the user interface includes a graphical user interface.
4 . System as claimed in claim 3 , wherein the, or a, user interface is configured to allow the user to store the representation of the generated objective function in a data storage, preferably in association with a user selectable identifier that identifies an objective criterion.
5 . System as claimed in claim 3 , wherein the, or a, user interface is configured to allow the user to select the objective criterion, wherein, upon such selection, the objective function is included as an objective into a radiation treatment planner system, the radiation treatment planner system to run an optimizing procedure driven at least by the objective function to compute a new treatment plan.
6 . System as claimed in claim 5 , wherein the optimizing procedure is of the inverse planning type.
7 . System as claimed in claim 1 , wherein the machine learning model is a regression type model or is of the generative type.
8 . System for computer-assisted radiation treatment planning, the system comprising:
a system as claimed in claim 1 ; and
a radiation treatment planning module configured to run an optimizing procedure driven at least by the selected objective function to compute a new treatment plan.
9 . A radiation treatment arrangement, comprising the system as claimed in claim 8 , and a radiation delivery apparatus controllable by the new treatment plan.
10 . A computer implemented method for supporting radiation treatment planning, the method comprising generating an objective function for radiation treatment planning, wherein the method allows a user to provide training data, the training data including:
i) a plurality of previous at least partial radiation treatment plans, and
ii) a user defined ranking of the plurality of previous at least partial radiation treatment plans,
and wherein the method further comprises a machine learning model based on training data.
11 . Method as claimed in claim 10 , further comprising providing the objective function to a radiation treatment planning module for computing a radiation treatment plan based on the objective function.
12 . A computer program element, which, when being executed by a processing unit, is adapted to cause the at least one processing unit to perform the method as claimed in claim 10 .
13 . A computer readable medium having stored thereon the program element as claimed in claim 12 .
14 . A method of training a machine learning model Mj based on training data, wherein the training data comprises:
i) a plurality of previous at least partial radiation treatment plans, and
ii) a user defined ranking of the plurality of previous at least partial radiation treatment plans.
15 . A method of training as claimed in claim 14 , wherein user weighting inside and outside a delineated region of at least partial radiation treatment plans is applied to the training of the machine learning model Mj.