Abstract: The dose rate of voxels within a particle beam (e.g., proton beam) treatment field delivered using pencil beam scanning (PBS) is calculated, and a representative dose rate for the particle beam treatment field is reported. The calculations account for a dose accumulation in a local region or a sub-volume (e.g., a voxel) as a function of time.

Abstract: A localized evaluation network incorporates a discriminator acting as classifier, which may be included within a generative adversarial network (GAN). GAN may include a generative network such as U-NET for creating segmentations. The localized evaluation network is trained on image pairs including medical images of organs of interest and segmentation (mask) images. The network is trained to distinguish whether an image pair does or does not represent the ground truth. GAN examines interior layers of the discriminator and evaluates how much each localized image region contributes to the final classification. The discriminator may analyze regions of the image pair that contribute to a classification by analyzing layer weights of the machine learning model. Disclosed embodiments include a visual attribute, such as a heat map, that represents contributions of localized regions of a contour to an overall confidence score. These localized regions may be highlighted and reported for quality assurance review.

Abstract: 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.

Abstract: Embodiments described herein provide for receiving a second image comprising an overlay depicting an organ-at-risk (OAR) segmentations. The overlay is generated by a first machine learning model based on a first image depicting the anatomical region of a current patient. A second machine learning model receives the second image and set of third images depicting prior patient OAR segmentations on which the second machine learning model was trained. The second machine learning model classifies the second image as one of a set of class names and characterizes the extent to which the second image is similar to, or dissimilar to, images with the same class name in the set of third images. The characterization may be based on outputs of internal layers of the second machine learning model. Dimensionality reduction may be performed on the outputs of the internal layers to present the outputs in a form comprehendible by humans.

Abstract: 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.

Abstract: Embodiments described herein provide for training an artificial intelligence model to become a preference-aware model. The artificial intelligence model preferences as the artificial intelligence model trains. Reinforcement learning is used to train experts in the artificial intelligence model such that each expert is trained to converge to a unique preference. The architecture of the artificial intelligence model is highly flexible. Upon executing a trained model, users can select automatically images according to various preferences based on medical professional preferences, geographic preferences, patient anatomy, and institutional guidelines.

Abstract: A dose rate-volume histogram can be generated for a target volume. The dose rate-volume histogram can be stored in computer system memory and used to generate a radiation treatment plan. The radiation treatment plan can be used as the basis for treating a patient using a radiation treatment system.

Abstract: A radiation therapy system includes an accelerator and beam transport system that generates a beam of particles. The accelerator and beam transport system guides the beam on a path and into a nozzle that is operable for aiming the beam toward an object. The nozzle includes a scanning magnet operable for steering the beam toward different locations within the object, and also includes a beam energy adjuster configured to adjust the beam by, for example, placing different thicknesses of material in the path of the beam to affect the energies of the particles in the beam.

Abstract: Disclosed herein are radiotherapy systems and methods that can display a workflow-oriented graphical user interface(s). In an embodiment, a system comprises a radiotherapy machine comprising: a couch; and a gantry having a camera in communication with a server, wherein the server is configured to: present for display, in real time on a screen associated with the radiotherapy machine, images received from the camera, wherein when the gantry changes its orientation, the camera revises at least one of its orientations, such that images transmitted to the server maintain an original orientation.

Abstract: A remote diagnostic monitoring of operating states for physical components of a particle accelerator system includes generating, by at least one processor, a component hierarchy corresponding to a physical arrangement of one or more physical components of a particle emitting system and including corresponding operating indicators of operating states of the physical components, identifying, by the at least one processor, a faulted physical component among the physical components, identifying, by the at least one processor, one or more fault path components among the physical components, the fault path components corresponding to a portion of the physical arrangement associated with the faulted physical component, and modifying, by the at least one processor, the operating indicators of the fault path components to fault state indicators.

Abstract: An example method for a radiation therapy system that includes a movable couch to perform radiation therapy has been disclosed. One method includes based on the X-ray images of an anatomical region of the patient that includes a target volume, reconstructing a digital volume of the anatomical region and based on a user input indicating a location of a patient origin in the digital volume, determining one or more shift values for repositioning the patient origin at an isocenter of the radiation therapy system with respect to a coordinate system. The method also includes obtaining a treatment plan that is based on the location of the patient origin and is associated with the target volume, based on the treatment plan, repositioning the movable couch so that the patient origin is disposed at the isocenter, and while the patient origin is disposed at the isocenter, directing a treatment beam to the patient origin in accordance with the treatment plan associated with the target volume.

Abstract: A remote diagnostic control of physical components of a particle accelerator system includes presenting, by at least one processor at a first physical location, a fault control interface including at least one control affordance corresponding to a physical component associated with a particle emitting system and a particle delivery system each located at a second physical location remote from the first physical location, and at least one arrangement presentation corresponding to the physical component and at least one physical device including the physical component, the arrangement presentation including a first operating state indicator associated with the physical component and a second operating state indicator associated with the physical device, and in response to activating the control affordance, generating a device command for transmission to the physical component to modify an operating state of the physical component at one or more of the particle emitting system and the particle delivery system, m

Abstract: A remote diagnostic monitoring and control of physical components of a particle accelerator system has a particle emitting system located at a first physical site and includes one or more particle emitting system components to operate the particle emitting system, a particle delivery system located at the first physical site and including one or more particle delivery system components to operate the particle delivery system, a particle system gateway located at the first physical site and operatively coupled to the particle emitting system components and the particle delivery system components by a first network interface, and a diagnostic monitoring system located at a second physical site remote from the first physical site, operatively coupled to the particle system gateway by a second network interface, and operable to monitor one or more first operating states corresponding to one or more of the particle emitting system components and one or more second operating states corresponding to one or more of t

Abstract: A multileaf collimator includes a plurality of movable leaves for shaping a radiotherapy beam, wherein each leaf is independently movable in a same linear travel direction. Each leaf includes a linear array of magnets disposed on a measurement surface of the leaf and an array of magnetoresistive sensors that is disposed proximate the measurement surfaces of the leaves.

Abstract: Example methods and apparatuses of controlling a user interface with a plurality of input mechanisms are disclosed. One example method includes causing a first set of input mechanisms in the plurality of input mechanisms to be visually emphasized via a first visual technique while a second set of input mechanisms in the plurality of input mechanisms is not visually emphasized via the first visual technique, receiving a user input via an input mechanism that is included in the first set, based on the user input, determining a third set of input mechanisms in the plurality of input mechanisms and a fourth set of input mechanisms in the plurality of input mechanisms, and causing the third set of input mechanisms to be visually emphasized via the first visual technique while the fourth set of available input mechanisms is not visually emphasized via the first visual technique.

Abstract: A method for implementing a radiation therapy knowledge exchange starts with searching a database of cases studies and selecting a case study. The selected case study is downloaded. The downloaded case study is applied to a medical case, wherein the downloaded case is applied using deformable image registration to deform reference images of the downloaded case to medical images of the medical case. After application of the downloaded case study, the medical case is uploaded to the network, wherein uploading the medical case allows at least the submitting clinician to download, review, and edit at least a portion of the medical case to create a reviewed medical case. Finally, the reviewed medical case is downloaded and applied to the medical case to create a final medical case.

Abstract: Presented systems and methods facilitate efficient and effective monitoring of particle beams. In some embodiments, a radiation gun system comprises: a particle beam gun that generates a particle beam, and a gun control component that controls the gun particle beam generation characteristics, including particle beam fidelity characteristics. The particle beam characteristics can be compatible with FLASH radiation therapy. Resolution control of the particle beam generation can enable dose delivery at an intra-pulse level and micro-bunch level. The micro-bunch can include individual bunches per each 3 GHz RF cycle within the 5 to 15 ?sec pulse-width. The FLASH radiation therapy dose delivery can have a bunch level resolution of approximately 4.4×10{circumflex over (?)}?6cGy/bunch.

Abstract: An example computer-implemented method for tuning a beam spot in a radiation therapy system based on radiation field measurements has been disclosed. The example method includes configuring an electron beam to generate a first beam spot on an electron-beam target of the radiation therapy system, determining a value for one or more radiation field quality metrics for a first radiation beam that originates from the first beam spot, and based on the value, determining whether the first radiation beam is outside a specified quality range.

Abstract: An example computer-implemented method for tuning a beam spot in a radiation therapy system has been disclosed. The example method includes configuring an electron beam to generate a first beam spot on an electron-beam target of the radiation therapy system, generating, using an imager of the radiation therapy system, a first plurality of projection images of the first beam spot, wherein each of the projection images of the first beam spot is generated with a line of sight blocked between the imager and a different respective portion of the beam spot, based on the first plurality of projection images, determining a value for one or more beam spot quality metrics associated with the first beam spot, and based on the value, determining whether the first beam spot is outside a specified quality range.