Abstract: The present disclosure is related to systems and methods for radiation. The method may include obtaining a plurality of reference images of a target of a subject and reference physiological motion information of the subject. The plurality of reference images and the reference physiological motion information may be acquired in a radiation period. The method may include establishing a correlation model based on the plurality of reference images and the reference physiological motion information. The method may include monitoring real-time motion information of the target based on the correlation model during a radiation operation performed during the radiation period.

Abstract: An installation area of a particle beam treatment system is reduced. The particle beam irradiation system includes an accelerator which is installed on a floor surface and accelerates a charged particle beam, a transport device which transports the charged particle beam emitted from the accelerator, an irradiation device which irradiates an irradiation target with the charged particle beam transported by the transport device, and a gantry which is installed on the floor surface, and to which the irradiation device is attached. The gantry includes a rotating body which rotates the irradiation device about the irradiation target, and a support device which supports the rotating body from the floor surface at a position where a projection plane of the rotating body on the floor surface and a projection plane of the accelerator or the transport device on the floor surface at least partially overlap with each other.

Abstract: A radiography and radiotherapy apparatus comprises a body unit, a gantry configured to rotate relative to the body unit, a treatment table on which an object is placed and configured to slide into holes formed in the body unit and the gantry and move in a vertical direction, a gantry driving part configured to rotate the gantry, a radiation irradiation part mounted on the gantry and configured to irradiate radiation toward the object, an image detector mounted to face the radiation irradiation part in the gantry and configured to detect the radiation irradiated from the radiation irradiation part, and a beam stopper provided at a lower end of the image detector and configured to block the radiation.

Abstract: Disclosed herein are methods and systems for predicting a virtual bolus in order to generate a radiation therapy treatment plan comprising training, by a processor, a machine-learning model using a training dataset comprising a set of medical images corresponding to a set of previously performed radiation therapy treatments, each medical image comprising at least one planning target volume and a non-anatomical region added to the medical image; and executing, by the processor, the machine-learning model using a medical image not included within the training dataset, the machine-learning model predicting an attribute of a non-anatomical region for the medical image not included in the training dataset.

Abstract: A control method for positioning is provided, which is applicable to a radiotherapy system. The method includes: acquiring a plurality of first body surface coordinates of a target site in a three-dimensional body surface image of a treatment object on a treatment couch; acquiring a plurality of second body surface coordinates under an isocentric coordinate system by transforming the first body surface coordinates using a target transformation matrix corresponding to the first placement position; determining, based on the plurality of second body surface coordinates and a contour image of the target site in the treatment plan, a placement offset of the target site, so as to control the treatment couch to move until the placement offset of the target site upon the movement meets a target offset requirement.

Abstract: The present invention relates to an ophthalmic treatment device, a control method therefor and a fundus lesion treatment method, the ophthalmic treatment device comprising: a treatment beam generation unit for generating a treatment beam; a beam delivery unit for forming a path through which the treatment beam generated from the treatment beam generating unit is irradiated into a patient's fundus; and a control unit for controlling the beam delivery unit to irradiate the treatment beam into a location adjacent to the lesion area of the fundus.

Abstract: Treatment fields can be produced as part of a treatment plan that achieves a desired balance between field delivery time and dose based on machine parameters and knowledge, such as machine-specific beam production, transport and scanning logic, and/or a maximum treatment time value. The treatment parameters can be adjusted using a graphical user interface so that treatment time or dosimetry is prioritized. As a result, the overall treatment time is reduced, and hence treatment quality and patient experience are improved.

Abstract: A method for medical imaging is provided. The method may include automatically determining that a positioning procedure of a subject has been finished based on image data of the subject captured by an image capturing device. The method may also include obtaining status information of a medical device. The method may further include causing the medical device to perform a scan on the subject based on a determination result that the positioning procedure of the subject has been finished, the status information of the medical device, and a starting signal of the scan.

Abstract: The present invention provides methods for treating, reducing the severity, or inhibiting the growth of cancer (e.g., a B-cell cancer such as Hodgkin's lymphoma or acute lymphoblastic leukemia). The methods of the present invention comprise administering to a subject in need thereof a therapeutically effective amount of an antibody or antigen-binding fragment thereof that specifically binds to programmed death 1 (PD-1) receptor in combination with a therapeutically effective amount of a bispecific antibody that specifically binds to CD20 and CD3.

Abstract: Systems and methods are disclosed for performing operations comprising: receiving a current fraction image and a reference image each depicting an anatomy of a subject, the reference image corresponding to a previous radiotherapy treatment fraction, the current fraction image corresponding to a current radiotherapy treatment fraction; obtaining a reference segmentation associated with the reference image; encoding, by a current fraction encoder, the current fraction image to generate a first set of features; encoding, by a reference encoder, the reference image and the reference segmentation to generate a second set of features; decoding, by a decoder, a combination of the first set of features and the second set of features to generate a predicted segmentation of the current fraction image; and configuring a radiotherapy treatment parameter based on the predicted segmentation of the current fraction image.

Abstract: The present invention relates to a dietary product comprising a plurality of amino acids, wherein the dietary product comprises all the essential amino acids and wherein the dietary product is substantially devoid of at least two non-essential amino acids; methods and uses thereof in the treatment of cancer; stratification methods and biomarkers for such applications.

Abstract: Disclosed herein are methods of treating venous thrombosis and vascular inflammation through administration of carbon monoxide and/or a carbon monoxide releasing molecule. Also disclosed herein are devices capable of releasing carbon monoxide for the purpose of treating microvascular, arterial and venous thromboembolism and/or inflammation.

Abstract: Information that describes a target inside a patient to be treated with radiation is accessed from computer system memory. An arrangement of spots inside the target is determined. Each the spots corresponds to a location inside the target where a respective beam of radiation is to be directed during radiation treatment of the patient. A dose rate for each of the beams is determined. The dose rate for each beam is a dose delivered in less than one second to a spot corresponding to that beam. For example, each beam can deliver at least four grays (GY) in less than one second, and may deliver as much as 20 Gy to 50 Gy or 100 Gy or more in less than one second. A radiation treatment plan, that includes the arrangement of the spots and the dose rate for each of the beams, is stored in computer system memory.

Abstract: A method is for supporting medical personnel. The method includes taking account of patient data of the patient in an electronic health record stored in a first database; determining at least one prevalence value, each relating to a diagnosis to be made for a group of people including the patient, to which the patient has been classified as belonging in dependence on at least one part of the patient data; generating, from the at least one prevalence value, at least one item of evaluation information based on at least one future examination procedure performable with respect to the diagnosis to be made, taking into account prediction information present in a second database describing at least one of sensitivity and specificity for the at least one future examination procedure with respect to the diagnosis to be made; and outputting the at least one item of evaluation information to the medical personnel.

Abstract: The invention relates to a planning apparatus for planning a radiation therapy. A medical image, in which a target to be irradiated is indicated, is reformatted based on ray geometries to be used during the radiation therapy to be planned, resulting in several reformatted medical images. Radiation therapy parameters being indicative of intensities of rays 5 to be used for irradiating a target 4 are determined based on the reformatted medical images by using a neural network unit. This allows to determine high quality radiation therapy parameters and hence allows for an improved planning of a radiation therapy. In particular, radiation and absorptions physics can be captured better, which can lead to the improved quality.

Abstract: A computer implemented method of treatment targeting includes receiving magnetic resonance (MR) images of a subject including a target region, generating at least one contour of at least one surrogate element apart from the target region in the MR images, and determining a location of the target region in each of the MR images based on a location of the at least one contour in the MR images.

Abstract: Some aspects of the disclosure provide a radiation therapy system. The system can include a radiation source configured to emit a radiation therapy beam, a collimator positioned to attenuate at least a periphery of the radiation therapy beam, a radiation fiducial marker configured to be coupled to a patient, and a first radiation detector and a second radiation detector configured to receive the radiation therapy beam after passing through a patient. The system also includes a computer configured to determine a position of the radiation fiducial marker using information from the first detector and the second detector.

Abstract: Information that describes a target inside a patient to be treated with radiation is accessed from computer system memory. An arrangement of spots inside the target is determined. Each the spots corresponds to a location inside the target where a respective beam of radiation is to be directed during radiation treatment of the patient. A dose rate for each of the beams is determined. The dose rate for each beam is a dose delivered in less than one second to a spot corresponding to that beam. For example, each beam can deliver at least four grays (GY) in less than one second, and may deliver as much as 20 Gy to 50 Gy or 100 Gy or more in less than one second. A radiation treatment plan, that includes the arrangement of the spots and the dose rate for each of the beams, is stored in computer system memory.

Abstract: A method of radiating a plurality of masses in a patient is provided, including receiving a three-dimensional model of the patient, the model including respective locations of a plurality of OARs, receiving a set of locations in the model corresponding to the masses, respective prescribed radiation dosages for the masses, and respective radiation limits for the OARs. The method includes computing a candidate set of beams having respective beam paths that travel through at least one of the masses. The method includes scoring the candidate set of beams based on respective dosages provided to the masses, respective dosages provided to the OARs, and beams in a set of selected beams for treatment, adding a best-scoring beam among the candidate set of beams to the set of selected beams, and radiating the masses using the set of selected beams.

Abstract: A prescribing user can designate and prioritize clinical goals that can, in turn, serve as the basis for optimization objectives to guide the development of a radiation treatment plan. The prescribing user can then alter that prioritization of one or more clinical goals and view information regarding changes to fluence-based dose distributions that occur in response to those changes to relative prioritization amongst the clinical goals to thereby understand dosing tradeoffs that correspond to prioritization amongst the clinical goals.

Abstract: A virtual beam's-eye view of a planning target volume is generated based on volumetric image data acquired immediately prior to radiation therapy by a radiation therapy system. The virtual beam's-eye view can then be displayed to confirm that, with the patient disposed in the current position, the planned beam-delivered treatment extends beyond the surface of the skin. In some embodiments, the virtual beam's-eye view can be displayed in conjunction with a beam's-eye view that is generated based on volumetric image data acquired during treatment planning, to create a blended beam's-eye view. In some embodiments, a field outline of a treatment beam can be superimposed on the blended beam's-eye view, thereby illustrating whether the planned beam-delivered treatment extends beyond the surface of the skin of the patient. The blended beam's-eye view can facilitate a manual confirmation process that verifies the planned beam-delivered treatment extends beyond the surface of the skin.

Abstract: Dermatological cosmetic combination treatments with high intensity focused ultrasound, dermal fillers, fat-reducing compounds, cavitation-prone fluids, and/or septa dissection are provided. A HIFU therapy system can include an imaging system, methods adapted to alter placement and position of a line focus band therapy, multiple simultaneous cosmetic ultrasound treatment zones in tissue, and dithering ultrasound beams from a transducer to alter placement and position of multiple cosmetic treatment zones in tissue. The HIFU systems can include a hand wand, removable transducer modules, and a control module. The dermal fillers can include hydroxyapatite. The fat-reducing compounds can include adipocytolytic compounds, pentacyclic triterpenoid compounds, proapoptotic compounds, compounds impairing differentiation of pre-adipocytes, and combinations thereof. The cavitation-prone fluids can include molecules dissolved in fluids, ethanol, glycerin, and ethylene glycol.

Abstract: These teachings provide for quickly yet accurately forming an influence matrix by generating the influence matrix via integration with a Monte Carlo particle transport simulation. The resultant influence matrix can then be utilized in an ordinary manner when optimizing a radiation treatment plan. By one approach, the foregoing comprises generating the influence matrix via integration with a Monte Carlo particle transport simulation on a particle-by-particle basis. For example, for each particle, these teachings can provide for identifying a spot to which the particle belongs and then adding a dose deposited by the particle during transport to an influence matrix element that corresponds to a spot to which the particle belongs and a voxel to where the dose was deposited.

Abstract: An estimated or predicted dose for radiotherapy treatment may be generated based on a partial dose map including dose information only for one or more regions of interest within a treatment site, by use of a properly trained machine learning system such as a U-Net or a V-Net. Said partial dose map typically set to fulfil clinical goals. A method of training such a machine learning system is also disclosed.

Abstract: The present disclosure relates to a radiation system. The system may include a treatment assembly, an imaging assembly, a first gantry, and a second gantry. The treatment assembly may include a first radiation source configured to deliver a treatment beam and have a treatment region. The first gantry may be configured to support the first radiation source. The imaging assembly may include a second radiation source and a radiation detector. The second radiation source may be configured to deliver an imaging beam and the radiation detector may be configured to detect at least a portion of the imaging beam. The imaging assembly may have an imaging region. The second gantry may be configured to support the second radiation source and the radiation detector, wherein the second radiation source is located within the second gantry. The treatment region and the imaging region at least partially overlap.

Abstract: A method for adaptive treatment planning is provided. The method may include obtaining a planning image volume of a subject, a treatment image volume of the subject, and a first treatment plan related to the planning image volume of the subject, each of the planning image volume and the treatment image volume including an ROI of the subject. The method may also include registering the planning image volume and the treatment image volume, and determining a first contour of the ROI in the registered planning image volume and a second contour of the ROI in the registered treatment image volume. The method may also include evaluating whether an error exists in at least one of the registration or the contour determination based on the first contour and the second contour, and determining a second treatment plan with respect to the treatment image volume based on the evaluation result.

Abstract: The effects of acute toxicity from a radiation procedure can be reduced without altering the radiation dose. Instead, a radiation procedure can be weighted to deliver certain amounts of radiation per day through temporal feathering. A target volume for a radiation procedure can be determined based on at least one image. The radiation procedure includes a total dose of radiation to be administered in a time period. An organ outside of the target volume at risk of acute toxicity from the radiation procedure can be determined based on the at least one image. A sequence plan that ensures the total dose of radiation is administered in the time period can be calculated. The sequence plan includes one day of a high fractional dose of radiation and four days of a low fractional dose of radiation. The sequence accomplishes temporal feathering of the radiation therapy procedure.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for assessing and adjusting the composition of groups for monitoring programs. In some implementations, a computer system determines a target level of diversity for a monitoring program based on a composition of a reference group. The system determines a group level of diversity for a group based on the composition of the group, wherein the group comprises at least one of candidates or participants for the monitoring program. The system determines that the group level of diversity does not provide the target level of diversity, and in response, selects one or more actions that are configured to adjust the composition of the group to achieve the target level of diversity in the group. The system provides output data that indicates the selected one or more actions to adjust the composition of the group toward the target level of diversity for the group.

Abstract: The present invention relates to a method for irradiating a population of mammalian cells comprising at least one target mammalian cell with electron beams and/or X-rays, characterized in that: (i) a composition comprising a population of mammalian cells is irradiated in vitro with electron beams and/or X-rays, the population of mammalian cells containing at least one target mammalian cell and the dose rate being within the range from 5 Gy/sec to 107 Gy/sec, and (ii) optionally viable target mammalian cells are isolated or enriched from the population of mammalian cells, and to agents obtainable thereby and to uses thereof.

Abstract: The present disclosure, in some embodiments, relates to a non-transitory computer-readable medium storing computer-executable instructions. The computer readable medium is configured to cause a processor to access an image volume of a rectum comprising a rectal tumor. A forward mapping is generated based on non-rigidly registering a healthy rectal atlas to the image volume. The forward mapping is inverted to generate an inverse mapping from the image volume to the healthy rectal atlas. Based on the inverse mapping, a plurality of deformation vectors, associated with a deformation within a rectal wall of the rectum, are determined. Magnitude based deformation features and orientation based deformation features are computed from the plurality of deformation vectors. One or more of the magnitude based deformation features and one or more of the orientation based deformation features are utilized to determine a response of a patient to a chemoradiation treatment.

Abstract: Described herein are methods of cell therapies. Also described herein are methods of generating donor derived T cells in an organ transplant recipient, by administering bone marrow stem cells to the organ transplant recipient about 1 to about 30 days after the organ transplant recipient receives one or more organ transplants.

Abstract: CD47+ disease cells such as cancer cells are treated using a combination of CD47 blockade drug and a histone deacetylase (HDAC) inhibitor. The anti-cancer effect of one drug enhances the anti-cancer effect of the other. Specific combinations include SIRP?Fc as CD47 blockade drug, and one of depsipeptide and romidepsin as HDAC inhibitor. These combinations are useful particularly to treat blood cancers including lymphomas, leukemias and myelomas.

Abstract: An apparatus and method used for medical and/or cosmetic treatment of a tissue, such as skin tightening, skin laxity and/or wrinkle reduction and or other skin flaws, including an energy source, a vacuum chamber and rollers which may facilitate easy sliding of the apparatus from one treated area to another. The rollers may also prevent excessive friction of the skin during the treatment. The energy source can be configured to emit energy based on the position of the apparatus on the skin (i.e., the position of the rollers), as well as the pressure within the vacuum chamber. The vacuum level and energy emission can be changed or pulsed based on the treatment regime.

Abstract: A statistical learning technique that does not rely upon paired imaging information is described herein. The technique may be computer-implemented and may be used in order to train a statistical learning model to perform image synthesis, such as in support of radiation therapy treatment planning. In an example, a trained statistical learning model may include a convolutional neural network established as a generator convolutional network, and the generator may be trained at least in part using a separate convolutional neural network established as a discriminator convolutional network. The generator convolutional network and the discriminator convolutional network may form an adversarial network architecture for use during training. After training, the generator convolutional network may be provided for use in synthesis of images, such as to receive imaging data corresponding to a first imaging modality type, and to synthesize imaging data corresponding to a different, second imaging modality type.

Abstract: Methods of treatment trajectory optimization for radiotherapy treatment of multiple targets include determining beam's eye view (BEV) regions and a BEV region connectivity manifold for each target group of a plurality of target groups separately. The information contained in the BEV regions and the BEV region connectivity manifolds for all target groups is used to guide an optimizer to find optimal treatment trajectories. To improve the visibility of insufficiently exposed voxels of planning target volumes (PTVs), a post-processing step may be performed to enlarge certain BEV regions, which are considered for exposing during treatment trajectory optimization.

Abstract: Methods of beam angle optimization for intensity modulated radiotherapy (IMRT) treatment include determining beam's eye view (BEV) regions and a BEV region connectivity manifold by evaluating dose response of each region of interest for each vertex in a delivery coordinate space (DCS). The information contained in the BEV regions and the BEV region connectivity manifold is used to guide an optimizer to find optimal field geometries in the IMRT treatment. To improve the visibility of insufficiently exposed voxels of planning target volumes (PTVs), a post-processing step may be performed to enlarge certain BEV regions, which are considered for exposing during treatment trajectory optimization.

Abstract: A control circuit accesses information regarding a plurality of pre-existing vetted radiation treatment plans for a variety of patients and uses that information to train at least one model (such as a dose volume histogram estimation model). The control circuit then uses that model to develop estimates for a radiation treatment plan for a particular patient. The control circuit can then use those estimates to develop a candidate radiation treatment plan.

Abstract: A photon source emits a flattening filter-free photon beam. A control circuit operably couples to a multi-layer multi-leaf collimator that is disposed between the photon source and a treatment area of a patient. The control circuit automatically arranges operation of some, but not all, of the layers of the multi-layer multi-leaf collimator to serve as a virtual flattening filter with respect to the flattening filter-free photon beam emitted by the photon source. By one approach, another of the layers of the multi-layer multi-leaf collimator serves to form a treatment aperture corresponding to a shape of the treatment area of the patient. By one approach the control circuit comprises an integral part of a treatment platform (as versus a dedicated treatment planning platform) and can carry out most or even essentially all of the planning steps that lead to administration of the treatment to a patient.

Abstract: A phantom, calibration system and calibration method are described. The phantom having a phantom body and an X-ray luminescent material, wherein at least a portion of the X-ray luminescent material is on a surface of the phantom.

Abstract: A computer based method of obtaining a 3D image of a part of a patient's body is disclosed, based on a fraction image having a limited field-of-view and extending the field of view with information from an image of the patient's outline, obtained from a surface scan of the patient. Anatomical data from the planning image are preferably used to fill in the outline image, by means of a contour-guided deformable registration between the planning image and contour.

Abstract: A method and apparatus for radiation therapy using functional measurements of branching structures. The method includes determining a location of each voxel of a plurality of voxels in a reference frame of a radiation device. The method further includes obtaining measurements that indicate a tissue type at each voxel. The method further includes determining a subset of the voxels based on an anatomical parameter of a respective branching structure of a set of branching structures indicated by the measurements. The method further includes determining a subset of the voxels that enclose an organ-at-risk (OAR) volume. The method further includes determining a value of a utility measure at each voxel. The method further includes determining a series of beam shapes and intensities which minimize a value of an objective function based on a computed dose delivered to each voxel and the utility measure for that voxel summed over all voxels.

Abstract: Systems and methods for delivering a radiation beam using a linear accelerator (LINAC). Optimal beam alignment parameters may be determined and stored for each of N gantry angles. The beam alignment parameters may adjust a current supplied to one or more bending magnets of the LINAC and, thus, change an angle and direction of the radiation beam. An optimum beam alignment parameter for a gantry angle may be determined by adjusting the beam alignment parameter until a center of a radiation field of the radiation beam in a radiation transmission image is at a center of shadow of a radiation opaque marker, which may be placed at a radiation isocenter. The beam alignment parameters stored for the N gantry angles may be used to adjust the beam steering current as the gantry is rotated through any arbitrary gantry angle.

Abstract: A deformable phantom, according to the present invention, has a housing made of a MRI invisible material enclosing a sealed reservoir filled with a MRI signal producing material, a piston slidably mounted within a sleeve and extending into the sealed reservoir, wherein the sleeve is slidably mounted to the housing and extends into the sealed reservoir, and a deformable structure within the sealed reservoir. The piston and sleeve move opposite to one another to conserve a constant fluid volume within the sealed reservoir as the piston moves in and out of the sealed reservoir to cause motion and/or deformation of the deformable structure.

Abstract: A method for EPID-based verification, correction and minimization of the isocenter of a radiotherapy device includes the following: Positioning a measurement body; applying an irradiation field; capturing a common dose image of the measurement body; creating a dose profile on the basis of the captured dose image; determining an inflection point in a plot of the dose profile; linking positions of the inflection points to bodily limits of the measurement body; determining position of a center point of the measurement body relative to an EPID-center; determining a differential vector from a deviation in position of the center point of the measurement body from the EPID-center and from a deviation in position of the field center point of the irradiation field from the EPID-center; and correcting the current radiological isocenter.

Abstract: Systems and methods are provided for registering images. The systems and methods perform operations comprising: receiving, at a first time point in a given radiation session, a first imaging slice corresponding to a first plane; encoding the first imaging slice to a lower dimensional representation; applying a trained machine learning model to the encoded first imaging slice to estimate an encoded version of a second imaging slice corresponding to a second plane at the first time point to provide a pair of imaging slices for the first time point; simultaneously spatially registering the pair of imaging slices to a volumetric image, received prior to the given radiation session, comprising a time-varying object to calculate displacement of the object; and generating an updated therapy protocol to control delivery of a therapy beam based on the calculated displacement of the object.

Abstract: Information associated with a radiation treatment plan includes, for example, values of dose per voxel in a target volume, values of dose rate per voxel in the target volume, and values of parameters used when generating the values of dose per voxel and the values of dose rate per voxel. Renderings that include, for example, a rendering of an image of or including the target volume, and a rendering of selected values of the radiation treatment plan, are displayed. When a selection of a region of one of the renderings is received, a displayed characteristic of another one of the renderings is changed based on the selection.

Abstract: Systems and methods include a method for providing, for presentation to a user, a graphical user interface (GUI) for defining and generating machine learning-based proxy models as surrogates for process-based reactive transport modeling (RTM). User selections of training parameters for generating training sample data are received. The training sample data is generated in response to receiving a parameter files generation indication. A training cases generation indication is received. Training sample cases are executed using the training sample data. User selections of proxy models training parameters are received. A set of parameter-specific proxy models represented by a neural network are trained. Each parameter-specific proxy model corresponds to a specific RTM parameter from a set of RTM parameters. Blind tests are performed using the set of parameter-specific proxy models. Each blind test tests a specific one of the parameter-specific proxy models.

Abstract: Nominal values of parameters, and perturbations of the nominal values, that are associated with previously defined radiation treatment plans are accessed. For each treatment field of the treatment plans, a field-specific planning target volume (fsPTV) is determined based on those perturbations. At least one clinical target volume (CTV) and at least one organ-at-risk (OAR) volume are also delineated. Each OAR includes at least one sub-volume that is delineated based on spatial relationships between each OAR and the CTV and the fsPTV for each treatment field. Dose distributions for the sub-volumes are determined based on the nominal values and the perturbations. One or more dose prediction models are generated for each sub-volume. The dose prediction model(s) are trained using the dose distributions.

Abstract: 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.

Abstract: A medical instrument includes an instrument shaft with exit holes near a distal end of the shaft, a tool coupled to the distal end of the shaft, and a backend. The backend may include a mechanism that manipulates a drive element that extends through the shaft and couples to the tool, a fluid inlet, and a fluid channel assembly providing fluid communication between the fluid inlet and the proximal end of the shaft. Cleaning fluid is directed into the fluid inlet, through the fluid channel assembly, and into the shaft. A chassis or other structural piece of the backend may form part of the fluid channel assembly.