AN ARTIFICIAL INTELLIGENCE SYSTEM TO SUPPORT ADAPTIVE RADIOTHERAPY
The present application describes a computing system, a computer readable medium, and/or related method for supporting decision making in adaptive therapy. An input interface of receives an input image. A machine learning module predicts, based at least in part on the input image, a dose distribution associated with a first planning technique or a first treatment modality. A comparator compares a planned dose distribution as per a current treatment plan with the predicted dose distribution, to obtain a comparison result. The comparison result enables a user to gauge whether an actual re-planning would yield a dosimetric benefit before committing time or computational resources.
1 . A computing system for replanning decision support in therapy, comprising:
an input interface for receiving an input image;
a machine learning module to predict based at least in part on the input image, a predicted dose distribution associated with a first planning technique or first treatment modality; and
a comparator to compare a planned dose distribution as per a current treatment plan with the predicted dose distribution, to obtain a comparison result.
2 . The system of claim 1 , including a graphics display generator to cause a display to display the comparison result or data derivable therefrom.
3 . The system of claim 1 , wherein the comparison result is displayed in association with the input image.
4 . The system of claim 3 , wherein the comparison result is displayed globally for the whole input image or locally per image element or locally.
5 . The system of claim 1 wherein, in response to the comparison result, or in response to a user request, a re-planning module of the system computes a new treatment plan, if there is a dosimetric benefit as per the comparison result.
6 . System of claim 1 , including a scheduler to schedule a new image session and/or a new re-planning session using the same or a planning technique, and/or new treatment session with the same or a new treatment modality.
7 . The system of claim 1 , wherein the machine learning module is one of a plurality of such modules, with different ones of the plurality of machine learning modules respectively associated with different planning techniques and/or different treatment modalities, the modules held in one or more data memories.
8 . The system of claim 7 , comprising a user interface for the user to select a different machine learning module from the plurality, and the system produces a new comparison result based at least in part on the selected machine learning module.
9 . The system of claim 1 , the machine learning module, or a further machine learning module that predicts an image representing anatomical changes due to applicable fractions.
10 . The system of claim 1 , wherein the comparison result is used by a treatment outcome predictor to estimate a treatment outcome.
11 . A computing system for training, based at least in part on training data, a machine learning module as per claim 1 .
12 . A computer-implemented method for replanning decision support in therapy, comprising:
receiving an input image;
a machine learning module, predicting, based at least in part on the input image, a predicted dose distribution associated with a first planning technique and/or first treatment modality; and
comparing a planned dose distribution as per a current treatment plan with the predicted dose distribution to obtain a comparison result.
13 . A computer-implemented method of training, based at least in part on training data, a machine learning module as per claim 1 .
14 . A non-transitory computer readable medium having stored thereon a computer program, that, when executed by at least one processor, causes the at least one process or to perform the method as per claim 11 .
15 . A non-transitory computer readable medium having stored thereon the pre-trained machine learning module of claim 1 .
16 . A non-transitory computer readable medium having stored thereon at least one of the plurality of machine learning modules of claim 7 .
17 . A non-transitory computer readable medium having stored thereon a computer program, that, when executed by at least one processor, causes the at least one processor to perform the method as per claim 12 .