A reinforcement learning agent facilitates optimization of a radiation-delivery treatment plan. The reinforcement learning agent is configured to generate a radiation-delivery treatment plan that can exceed the quality of a plan or plans employed to train the reinforcement learning agent. The reinforcement learning agent is trained to evaluate a radiation-delivery treatment plan that is output by an optimization software application, modify one or more dose-volume objective parameters of the evaluated radiation-delivery treatment plan, and then input the modified radiation-delivery treatment plan to the optimization software application for further optimization. The reinforcement learning agent adaptively adjusts the one or more dose-volume objective parameters based on an action policy learned during a reinforcement learning training process.

Systems and methods can include a method for training a deep convolutional neural network to provide a patient radiation treatment plan, the method comprising collecting patient data from a group of patients, the patient data including at least one image of patient anatomy and a prior treatment plan, wherein the treatment plan includes predetermined machine parameters, and training a deep convolution neural network for regression by using the prior treatment plans and the corresponding collected patient data to determine a new treatment plan. Systems and methods can also include a method of using a deep convolutional neural network to provide a radiation treatment plan, the method comprising retrieving a trained deep convolution neural network previously trained on patient data from a group of patients, collecting new patient data, wherein the new patient data includes at least one image of patient anatomy, and determining a new treatment plan for the new patient using the trained deep convolutional neural network for regression, wherein the new treatment plan has a new set of machine parameters.

A correlation between a CT value and a water equivalent thickness ratio distribution for each patient can be corrected without increasing a treatment time, and more accurate treatment can be realized. A treatment planning system 112 which generates a treatment plan for irradiating an irradiation target with a particle beam calculates a correction amount of a water equivalent thickness ratio of a first treatment plan created in advance, calculates a water equivalent thickness ratio distribution based on the correction amount and the first treatment plan, and creates a second treatment plan from the water equivalent thickness distribution.

An example method of radiation therapy in a radiation therapy system that includes a gantry with a treatment-delivering X-ray source and an imaging X-ray source mounted thereon is described. The method includes rotating the gantry in a first direction at a first rotational velocity about an open bore and concurrently rotating an annular support structure at a second rotational velocity about the open bore, wherein the second rotational velocity is less than the first rotational velocity. While continuing to rotate the gantry in the first direction about the open bore from a first position to a treatment position, the method also includes generating multiple images of a target volume disposed in the bore using the imaging X-ray source. Upon rotating the gantry to the treatment position, the method includes initiating delivery of a treatment beam to the target volume with the treatment-delivering X-ray source.

Systems and methods for implementing an adaptive therapy workflow that minimizes time needed to create a session patient model, select an appropriate plan for the treatment session, and treat the patient.

An improved x-ray cone-beam CT image reconstruction by end-to-end training of a multi-layered neural network is proposed, which employs cone-beam CT images of many patients as input training data, and precalculated scattering projection images of the same patients as output training data. After the training is completed, scattering projection images for a new patient are estimated by inputting a cone-beam CT image of the new patient into the trained multi-layered neural network. Subsequently, scatter-free projection images for the new patient are obtained by subtracting the estimated scattering projection images from measured projection images, beam angle by beam angle. A scatter-free cone-beam CT image is reconstructed from the scatter-free projection images.

In one aspect, the present disclosure relates to a method of adaptive treatment of a subject with a tumor. The method may include administering a first pulse dose of radiation to a tumor within a subject; administering a second pulse dose of radiation to the tumor, wherein the second pulse dose is administered after an observation period, the observation period having a duration of at least 7 days; and concurrently treating the subject with an immunotherapy.

A method, system and computer readable medium of: providing feature data of at least one organ at risk or target volume of said patient from a database of non-transitory data stored on a data storage device of prior patients data; generating, using a data processor, a distribution of dose points of the at least one organ at risk or target volume of said patient based on said feature data; calculating, using the data processor, at least one of (i) a probability of toxicity for the at least one organ at risk or (ii) a probability of treatment failure for the at least one target volume, based on said distribution of dose points; assessing, using the data processor, a dosimetric-outcome relationship based on the calculated probability; and automatically formulating, using the data processor, a treatment plan using the dosimetric-outcome relationship to minimize the at least one treatment-related risk.

It is provided a method for checking quality of a treatment plan, wherein a treatment plan specifies a distribution of radiation to thereby provide radiation to a planning target volume. The method is performed by a quality assurance device and comprises the steps of: obtaining a treatment plan and a corresponding first dose, the treatment plan having been calculated in a treatment planning system, the first dose being a predicted dose to be deposited in the patient using the treatment plan; initiating a calculation of a secondary dose, being a dose deposited by the treatment plan, using a secondary dose calculation algorithm; repeatedly calculating a confidence interval of a comparative statistical measurement by comparing the first dose and the secondary dose over a defined geometric volume; and interrupting the calculation of the secondary dose when the confidence interval is better than at least one predefined criterion.

Systems and methods are disclosed for monitoring anatomic position of a human subject and modifying a radiotherapy treatment based on anatomic position changes, as determined with a regression model trained to estimate movement of a region of interest. Example operations for movement monitoring and therapy control include: obtaining 3D image data for a subject, which provides a reference volume and at least one defined region of interest; obtaining real-time 2D image data corresponding to the subject, captured during the radiotherapy treatment session; extracting features from the 2D image data; producing a relative motion estimation of a region of interest with a machine learning regression model, the model trained to estimate a spatial transformation from the 2D image data based on training from the reference volume; and controlling a radiotherapy beam of a radiotherapy machine used in the radiotherapy session, based on the relative motion estimation.