Abstract: In a radiation treatment plan that includes a plurality of treatment fields of multiple treatment modalities, such as IMRT modality and dynamic treatment path modality (e.g., VMAT and conformal arc therapy), an optimized spatial point sequence may be determined that optimizes the total treatment time, which includes both the beam-on time (i.e., during the delivery of radiation dose) and the beam-off time (i.e., during transitions between consecutive treatment fields). The result is a time-ordered field trajectory that intermixes and interleaves different treatment fields. In one embodiment, a dynamic treatment path may be cut into a plurality of sections, and one or more IMRT fields may be inserted between the plurality of sections.

Abstract: Example methods and systems for training deep learning engines for radiotherapy treatment planning are provided. One example method may comprise: obtaining a set of training data that includes unlabeled training data and labeled training data; and configuring a deep learning engine to include (a) a primary network and (b) a deep supervision network that branches off from the primary network. The method may further comprise: training the deep learning engine to perform the radiotherapy treatment planning task by processing training data instance to generate (a) primary output data and (b) deep supervision output data; and updating weight data associated with at least some of the multiple processing layers based on the primary output data and/or the deep supervision output data. The deep supervision network may be pruned prior to applying the primary network to perform the radiotherapy treatment planning task for a patient.

Abstract: Systems, devices, and methods for quality assurance for verification of radiation dose delivery in arc-based radiation therapy devices using a 3D gamma evaluation method.

Abstract: Systems, devices, and methods for dosimetric verification of radiation therapy treatments by selective evaluation of measurement points. Systems, methods, and computer program-products for providing dosimetric verification of radiation therapy treatments by evaluating measurement points using different evaluation criteria.

Abstract: Example methods and systems for deep transfer learning for radiotherapy treatment planning are provided. One example method may comprise: obtaining (310) a base deep learning engine that is pre-trained to perform a base radiotherapy treatment planning task; and based on the base deep learning engine, generating a target deep learning engine to perform a target radiotherapy treatment planning task. The target deep learning engine may be generated by configuring (330) a variable base layer among multiple base layers of the base deep learning engine, and generating (340) one of multiple target layers of the target deep learning engine by modifying the variable base layer. Alternatively or additionally, the target deep learning engine may be generated by configuring (350) an invariable base layer among the multiple base layers, and generating (360) one of multiple target layers of the target deep learning engine based on feature data generated using the invariable base layer.

Abstract: One or more medical images of a patient are processed by a first neural network model to determine a region-of-interest (ROI) or a cut-off plane. Information from the first neural network model is used to crop the medical images, which serves as input to a second neural network model. The second neural network model processes the cropped medical images to determine contours of anatomical structures in the medical images of the patient. Each of the first and second neural network models are deep neural network models. By use of cropped images in the training and inference phases of the second neural network model, contours are produced with sharp edges or flat surfaces.

Abstract: Systems, devices, and methods for dosimetric verification of radiation therapy treatments by selective evaluation of measurement points. Systems, methods, and computer program-products for providing dosimetric verification of radiation therapy treatments by evaluating measurement points using different evaluation criteria.

Abstract: A system for estimating a dose from a radiation therapy plan includes a memory that stores machine-readable instructions and a processor communicatively coupled to the memory, the processor operable to execute the instructions to subdivide a representation of a volume of interest into voxels. The processor also determines distances between a planned radiation field origin and each respective voxel. The processor further computes geometry-based expected (GED) metrics based on the distances, a plan parameter, and a field strength parameter. The processor sums the metrics to yield an estimated dose received by the volume of interest from the planned radiation field.

Abstract: Methods and systems are provided for developing radiation therapy treatment plans. A treatment template with radiation fields can be chosen for a patient based on a tumor location. Static radiation field positions can be adjusted for the patient, while arc radiation fields may remain the same. Static radiation field positions can be adjusted using dose gradient, historical patient data, and other techniques.

Abstract: Example methods and systems for tomographic data analysis are provided. One example method may comprise: obtaining first three-dimensional (3D) feature volume data and processing the first 3D feature volume data using an AI engine that includes multiple first processing layers, an interposing forward-projection module and multiple second processing layers. Example processing using the AI engine may involve: generating second 3D feature volume data by processing the first 3D feature volume data using the multiple first processing layers, transforming the second 3D volume data into 2D feature data using the forward-projection module and generating analysis output data by processing the 2D feature data using the multiple second processing layers.

Abstract: Example methods and systems for tomographic image reconstruction are provided. One example method may comprise: obtaining two-dimensional (2D) projection data and processing the 2D projection data using the AI engine that includes multiple first processing layers, an interposing back-projection module and multiple second processing layers. Example processing using the AI engine may involve: generating 2D feature data by processing the 2D projection data using the multiple first processing layers, reconstructing first three-dimensional (3D) feature volume data from the 2D feature data using the back-projection module; and generating second 3D feature volume data by processing the first 3D feature volume data using the multiple second processing layers.

Abstract: Example methods and systems for training deep learning engines for radiotherapy treatment planning are provided. One example method may comprise: obtaining a set of training data that includes unlabeled training data and labeled training data; and configuring a deep learning engine to include (a) a primary network and (b) a deep supervision network that branches off from the primary network. The method may further comprise: training the deep learning engine to perform the radiotherapy treatment planning task by processing training data instance to generate (a) primary output data and (b) deep supervision output data; and updating weight data associated with at least some of the multiple processing layers based on the primary output data and/or the deep supervision output data. The deep supervision network may be pruned prior to applying the primary network to perform the radiotherapy treatment planning task for a patient.

Abstract: Systems and methods for anatomical structure segmentation in medical images using multiple anatomical structures, instructions and segmentation models.

Abstract: A system for estimating a dose from a radiation therapy plan includes a memory that stores machine-readable instructions and a processor communicatively coupled to the memory, the processor operable to execute the instructions to subdivide a representation of a volume of interest into voxels. The processor also determines distances between a planned radiation field origin and each respective voxel. The processor further computes geometry-based expected (GED) metrics based on the distances, a plan parameter, and a field strength parameter. The processor sums the metrics to yield an estimated dose received by the volume of interest from the planned radiation field.

Abstract: A search space used for multicriteria optimization in radiation treatment planning can be expanded using extrapolation. For instance, given an initial set of base plans, one or more virtual plans can be generated by assigning a weight to each base plan such that at least one of the weights is less than zero and/or such that the sum of the weights is not normalized to 1. The dose distribution for the virtual plan is computed as the weighted sum of the dose distributions of the base plans. Virtual plans criteria can be used together with the initial set of base plans to define an expanded search space within which interpolation can be performed to identify an optimal treatment plan.

Abstract: Example methods for adaptive radiotherapy treatment planning using deep learning engines are provided. One example method may comprise obtaining treatment image data associated with a first imaging modality. The treatment image data may be acquired during a treatment phase of a patient. Also, planning image data associated with a second imaging modality may be acquired prior to the treatment phase to generate a treatment plan for the patient. The method may also comprise: in response to determination that an update of the treatment plan is required, transforming the treatment image data associated with the first imaging modality to generate transformed image data associated with the second imaging modality. The method may further comprise: processing, using the deep learning engine, the transformed image data to generate output data for updating the treatment plan.

Abstract: A system for estimating a dose from a proton therapy plan includes a memory that stores machine instructions and a processor coupled to the memory that executes the machine instructions to subdivide a representation of a volume of interest in a patient anatomy traversed by a planned proton field into a plurality of voxels. User input received by a GUI can be used to define the representation. The processor further executes the machine instructions to determine the distance from the source of the planned proton beam to one of the voxels. The processor also executes the machine instructions to compute the discrete contribution at the voxel to an estimated dose received by the volume of interest from the planned proton beam based on the distance between the source and the volume of interest.

Abstract: One or more medical images of a patient are processed by a first neural network model to determine a region-of-interest (ROI) or a cut-off plane. Information from the first neural network model is used to crop the medical images, which serves as input to a second neural network model. The second neural network model processes the cropped medical images to determine contours of anatomical structures in the medical images of the patient. Each of the first and second neural network models are deep neural network models. By use of cropped images in the training and inference phases of the second neural network model, contours are produced with sharp edges or flat surfaces.

Abstract: Streamlined and partially automated methods of setting normal tissue objectives in radiation treatment planning are provided. These methods may be applied to multiple-target cases as well as single-target cases. The methods can impose one or more target-specific dose falloff constraints around each target, taking into account geometric characteristics of each target such as target volume and shape. In some embodiments, methods can also take into account a planner's preferences for target dose homogeneity. In some embodiments, methods can generate additional dose falloff constraints in locations between two targets where dose bridging is likely to occur.

Abstract: A system for estimating a dose from a proton therapy plan includes a memory that stores machine instructions and a processor coupled to the memory that executes the machine instructions to subdivide a representation of a volume of interest in a patient anatomy traversed by a planned proton field into a plurality of voxels. The processor further executes the machine instructions to determine the distance from the source of the planned proton beam to one of the voxels. The processor also executes the machine instructions to compute the discrete contribution at the voxel to an estimated dose received by the volume of interest from the planned proton beam based on the distance between the source and the volume of interest.