Embodiments described herein provide for revising radiation therapy treatment plans, and in particular, revising beam angles used during radiation therapy treatment. A computer may receive a radiation therapy treatment plan based on a particular patient's diagnosis. The computer may use a machine learning model to revise radiation therapy treatment parameters such as a beam angle indicating a direction of radiation into the patient. The machine learning model may use reinforcement learning to optimize an initial beam angle from the radiation therapy treatment plan, revising the beam angle. The performance of the machine learning model is measured against metrics including fulfilling dosimetric clinical goals. The machine learning model may present the revised beam angle for display to a medical professional, or transmit the revised beam angle to downstream applications to further revise the radiation therapy treatment plan.

Embodiments described herein provide for recommending radiotherapy treatment attributes. A machine learning model predicts the preference of a medical professional and provides relevant suggestions (or recommendations) of radiotherapy treatment attributes for various categories of radiotherapy treatment. Specifically, the machine learning model predicts field geometry attributes from various field geometry attribute options for various field geometry attribute categories. The machine learning model is conditioned on patient data such as medical images and patient information. The machine learning model is trained in response to cumulative reward information associated with a medical professional accepting the provided/displayed recommendations.

The present disclosure provides a system and method for generating radiation treatment plan. The method may include obtaining a plurality of beam angles of an arc for radiation treatment and preliminary segment parameters of control points associated with the plurality of beam angles. The method may also include grouping the plurality of beam angles into at least two sets so that each pair of two consecutive beam angles of the plurality of beam angles belong to different sets of the at least two sets, and determining the target segment parameters of the control points in each of the at least two sets by optimizing, based on a leaf motion constraint, the preliminary segment parameters of the control points associating with the plurality of beam angles. The method may further include generating a treatment plan based on the target segment parameters.

The disclosure provides systems and methods for adjusting a multi-leaf collimator (MLC). The MLC includes a plurality of cross-layer leaf pairs, each cross-layer leaf pair of the plurality of cross-layer leaf pairs includes a first leaf located in a first layer of leaves and a second leaf opposingly located in a second layer of leaves. For at least one cross-layer leaf pair, an effective cross-layer leaf gap to be formed between the first leaf and the second leaf may be determined; at least one of the first leaf or the second leaf may be caused to move to form the effective cross-layer leaf gap; and an in-layer leaf gap may be caused, based on the effective cross-layer leaf gap, to be formed between the first leaf and an opposing first leaf in the first layer. A size of the in-layer leaf gap may be no less than a threshold.

A radiation therapy system comprising a therapeutic radiation system (e.g., an MV X-ray source, and/or a linac) and a co-planar imaging system (e.g., a kV X-ray system) on a fast rotating ring gantry frame. The therapeutic radiation system and the imaging system are separated by a gantry angle, and the gantry frame may rotate in a direction such that the imaging system leads the MV system. The radiation sources of both the therapeutic and imaging radiation systems are each collimated by a dynamic multi-leaf collimator (DMLC) disposed in the beam path of the MV X-ray source and the kV X-ray source, respectively. In one variation, the imaging system identifies patient tumor(s) positions in real-time. The DMLC for the imaging radiation source limits the kV X-ray beam spread to the tumor(s) and/or immediate tumor regions, and helps to reduce irradiation of healthy tissue (e.g., reduce the dose-area product).

This invention provides a method applied for the new dynamic arc radiotherapy treatment planning to calculate an optimal arc angle. With this invention, an operator without rich experience is able to reach the expected low dose in lungs easily and quickly. This invention can not only estimate the distribution of low radiation dose in lungs but also reduce the shortcomings like consumption of time and inaccuracy caused by manual trial and error.

Example methods and systems for radiotherapy treatment planning based on treatment delivery efficiency are described. One example method may comprise a computer system configuring dosimetric planning objective(s) and non-dosimetric planning objective(s) associated with efficiency of treatment delivery. A set of multiple treatment plan variants may be generated based on the dosimetric planning objective(s) and non-dosimetric planning objective(s). A first treatment plan associated with a first tradeoff and a second treatment plan associated with a second tradeoff may then be identified from the set of multiple treatment plan variants. The second treatment plan may be associated with improved efficiency of treatment delivery compared to the first treatment plan.

Systems and methods for radiation treatment planning can include a computing system determining an estimate of radiation dose distribution within an anatomical region of a patient, and determining a cost matrix representing an objective function, using the estimate of radiation dose distribution. The computing system can project the cost matrix on each of a plurality of fluence planes. Each of the plurality of fluence planes can be associated with a corresponding gantry-couch orientation of a plurality of gantry-couch orientations of a medical linear accelerator. The computing system can determine, using projections of the cost matrix on each of the plurality of fluence planes, a sequence of gantry-couch orientations among the plurality of gantry-couch orientations representing a treatment path.

The present application provides an initial, or first, packed arc setup to be compared with predefined arc setup constraints. These predefined arc setup constraints constrain at least one or more of the number of patient table angles per target volume, the number of times the gantry moves along one arc per table angle, the sum of gantry span per metastasis over all arcs, and the minimum table span. Based on the result of the comparison between the first packed arc setup with the predefined arc setup constraints, a second arc setup is automatically suggested. The automatically suggested second arc setup may then be compared with the first arc setup by calculating a score for both setups. Several iterations of such a method can be carried out based on the comparison between an arc setup and the following, subsequent arc setup in the iteration.

Disclosed is a computer-implemented method of determining a treatment plan, encompassing acquiring patient image data, acquiring target data describing targets, acquiring position data describing control points which define one or more arcs, and determining target projection data which describes outlines of the target in a beam's-eye view. Margin data is acquired. For the outlines, margins are applied to determine auxiliary outlines. Beam shaping device data is determined describing configurations of the collimator leaves so that irradiation of the auxiliary outlines is enabled. Based on these configurations, the irradiation amount is simulated for voxels of the patient image data. Constraints to be fulfilled by the treatment plan may be set. Configurations of blockings, arc-weights and margins are proposed. Only different combinations of these parameters are proposed while additional possible parameters are neglected. An optimization algorithm is used to minimize an objective function. The best configuration is selected as the treatment plan.