Systems and methods for automatic creation of dose prediction models and therapy treatment plans as a cloud service

A method for automatically predicting a dose distribution in a radiotherapy treatment plan, the method comprising:

• selecting a dose prediction model from a plurality of dose prediction models based on a plurality of characteristics;
• predicting a dose distribution based on the dose prediction model;
• producing a plurality of optimization objectives based on the dose distribution;
• creating a patient treatment plan based on the optimization objectives, and
• verifying the patient treatment plan with the dose prediction model,
• wherein the predicting, the producing, and the creating is performed in a first computing system,
• further wherein the selecting is performed by a user through a user interface executing on a second computing system remote from the first computing system.

A system for storing a plurality of dose prediction models, the system comprising:

• a shared library of treatment data, the library of treatment data comprising a plurality of pre-generated treatment plans;
• a plurality of computing devices, a computing device of the plurality of computing devices comprising:
• a memory configured to store a patient anatomical data;
• a processor configured to execute an application, the application being configured to apply a dose prediction model to the patient anatomical data to predict a dose distribution to a patient corresponding to the patient anatomical data,
• wherein the dose prediction model is trained by analyzing a selection of treatment plans from the plurality of pre-generated treatment plans,
• further wherein the shared library of treatment data comprises a cloud storage component communicatively coupled to the plurality of computing devices.