MRI-guided radiotherapy

Abstract

Disclosed are systems, methods, and computer software related to gate control using non-parallel imaging planes, determining the accumulated dose to tissue during radiation therapy using actual beam delivery information, stopping / modifying / re-optimizing treatment based on the accumulated dose, and generating and using radiation therapy plans with prognostic motion models and prognostic motion adaptation.

Description

Technical Field

[0001] Related Applications This application claims the benefit of priority of U.S. Provisional Application No. 63 / 270,862, filed on October 22, 2021, entitled "MRI-Guided Radiation Therapy", the contents of which are incorporated herein by reference.

[0002] Description of Related Art Magnetic resonance imaging (MRI) or nuclear magnetic resonance imaging is a non-invasive imaging technique that uses radiofrequency pulses, a strong magnetic field (modified by weaker gradient magnetic fields applied across it for encoding), and interactions between body tissues to acquire projection images, spectral signals, and images of planes or volumes from within a patient's body. MRI is particularly useful for imaging soft tissues and can be used for the diagnosis of diseases. Real-time MRI or cine MRI can be used for the diagnosis of medical conditions that require imaging of moving structures within a patient. Real-time MRI can also be used in conjunction with interventional procedures such as radiation therapy or image-guided surgery to help guide such procedures.

Summary of the Invention

[0003] Gate control using non-parallel imaging planes and determining the accumulated dose to tissue during radiation therapy using actual beam delivery information, and a system, method, and computer software for generating and using a prognostic motion model and a prognostic motion-adaptive radiation therapy plan are disclosed. In one aspect, the system and software may be configured to perform operations including acquiring real-time images of a patient from a magnetic resonance imaging system in at least two non-parallel planes and depicting the contours of the patient's anatomical structures in at least two non-parallel planes. Spatial limits may be set for movement of anatomical structures in non-parallel planes. The radiation therapy device may be controlled to deliver a radiation therapy beam to the patient, and the radiation therapy beam may be gated off when the anatomical structure exceeds a spatial limit in any of the non-parallel planes. In some variants, there may be three orthogonal planes, and the depiction of the contours regarding the patient's anatomical structures may be performed via automatic contouring of the machine.

[0004] In other variants, the radiation therapy device may further be controlled to deliver stereotactic radiosurgery (SRS) to the patient. The magnetic resonance imaging system may operate at a magnetic field strength of less than 1.0 Tesla, and the spatial limits may be set within 0.5 mm of the boundaries of anatomical structures in at least two non-parallel planes.

[0005] In related aspects, the system and software may be configured to perform operations including delivering radiation therapy to a patient from a radiation therapy device, the patient's image may be acquired from a magnetic resonance imaging system during radiation therapy, and the actual beam delivery information may be acquired during radiation therapy. The actual beam delivery information may include actual beam measurements including one or more of the measured monitor unit, the measured MLC leaf position, the measured gantry position, the measured couch position, and the measured fluence profile. The dose to the tissue may be calculated during radiation therapy based on the acquired image and the acquired actual beam delivery information. The dose may also be accumulated to the tissue during radiation therapy. In some variants, the accumulated dose may be displayed in three orthogonal planes.

[0006] In other variants, the anatomical structure of the patient can be delineated. The accumulated dose to the delineated anatomical structure can be determined, and a notification or alarm can be provided if the accumulated dose to the delineated anatomical structure exceeds a specified limit. Delivery of radiotherapy can be stopped if the accumulated dose to the delineated anatomical structure exceeds a specified limit, and a re-optimized treatment plan can be determined.

[0007] In another related aspect, pre-treatment images can be acquired using a magnetic resonance imaging system, and the pre-treatment images capture patient movement. A prognostic motion model can then be generated based at least on the pre-treatment images, and a radiotherapy plan with prognostic motion adaptation can be generated based at least on the prognostic motion model.

[0008] In some variants, the prognostic motion model can include a model of the expected patient movement during treatment. The prognostic motion model can also be generated to include multiple types of motion observed in pre-treatment images, such as regular motion due to breathing, motion due to deep breathing, motion due to GI system gas movement, motion due to bladder filling, motion due to patient movement, motion due to swallowing, chest wall breathing, diaphragmatic breathing, conversation, eye movement, cardiac motion, or voluntary muscle motion.

[0009] In other variants, the radiotherapy plan with prognostic motion adaptation can take into account the expected patient movement of the prognostic motion model during treatment, including deformation of the target or organ of interest.

[0010] The radiotherapy plan with prognostic motion adaptation can be configured to adjust or stop delivery when irregular patient movement of the prognostic motion model is observed during treatment. In other variants, the radiotherapy plan with prognostic motion adaptation can direct the aiming of the radiation beam to where the target is expected to be based on the prognostic motion model and system latency, rather than at a specific point in space.

[0011] During radiotherapy, a treatment image of a patient can be acquired from a magnetic resonance imaging system, and the treatment image captures the movement of the patient. The radiotherapy can then be delivered from a radiotherapy device to the patient according to a radiotherapy plan with a prognostic motion adaptation based at least on a prognostic motion model. In some variants, the delivery of radiotherapy can be interrupted when the movement of the patient does not match the expected movement of the patient. In other variants, a second prognostic motion model can be generated using the acquired treatment image. A radiotherapy plan with a second prognostic motion adaptation can be generated based at least on the second prognostic motion model, and the delivery of radiotherapy can be resumed using the radiotherapy plan with the second prognostic motion adaptation.

[0012] In yet other variants, by combining the techniques described herein, actual beam delivery information can be acquired during radiotherapy, and the dose to tissue during radiotherapy can be calculated based on the acquired image and the acquired actual beam delivery information. The dose to tissue can be accumulated during radiotherapy, and a radiotherapy plan with a second prognostic motion adaptation can be generated if the accumulated dose to an anatomical structure exceeds a specified limit. The delivery of radiotherapy can be continued using the radiotherapy plan with the second prognostic motion adaptation. The radiotherapy plan with the second prognostic motion adaptation can also be generated taking into account the accumulated dose and any underdosed or overdosed regions.

[0013] Implementations of the subject matter may include, but are not limited to, tangible, machine-readable media that are operable to cause one or more machines (e.g., computers, etc.) to behave in a manner consistent with the descriptions provided herein and to cause one or more operations to implement one or more of the features described. Similarly, a computer system may be envisioned that includes one or more processors and one or more memories connected to the one or more processors. The memory, which may include a computer-readable storage medium, may include, encode, store, or otherwise perform one or more programs that cause one or more processors to perform one or more of the operations described herein. A computer-implemented method consistent with one or more implementations of the subject matter may be implemented by one or more data processors present in a single computing system or across multiple computing systems. Such multiple computing systems may be connected and may exchange data and / or commands or other instructions or the like via one or more connections, including but not limited to connections in a network (e.g., the Internet, a wireless wide area network, a local area network, a wide area network, a wired network, or the like), or direct connections between one or more of the multiple computing systems.

[0014] Details regarding one or more variants of the subject matter described herein are set forth in the accompanying drawings and the description below. Other features and advantages of the subject matter described herein will be apparent from the specification, the drawings, and the claims. It should be immediately understood that specific features described in connection with particular implementations are not intended to be limiting. The claims that follow this disclosure are intended to define the scope of the protected subject matter.

Brief Description of the Drawings

[0015] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate specific embodiments of the subject matter disclosed herein and, together with the specification, serve to explain some of the principles associated with the implementations of the present disclosure. In the drawings,

[0016]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

[0017] The present disclosure provides improvements to the delivery of therapeutic radiation through improved tissue tracking and treatment gating control. In particular, improvements to the calculation and accumulation of radiation dose are also disclosed by determining the actual dose of radiation delivered to a patient during the course of radiation therapy. Further disclosed are techniques for creating a radiation treatment plan with prognostic motion adaptation and utilizing a prognostic patient motion model that enables further improved radiation delivery using such a plan.

[0018] FIG. 1 shows one implementation of a magnetic resonance imaging-guided radiation therapy system 100 (MRgRT system) that combines a magnetic resonance imaging system (MRI) 101 and a radiation therapy source 150, in accordance with certain aspects of the present disclosure. In FIG. 1, the MRI 101 includes a main electromagnet 102, a gradient coil assembly 104, and an RF coil system 106. A patient couch 108 on which a patient 110 can lie is within the MRI 101.

[0019] An exemplary main electromagnet 102 of the MRI 101 can be a solenoid electromagnet with a gap, separated by a yoke 114 having a gap 116, as shown in FIG. 1. "Gap", when the term is used herein, refers to the type of gap 116 of the solenoid magnet depicted in FIG. 1. As also depicted in FIG. 1, the current in the main electromagnet 102 can be in either the first direction 118 or the second direction 120 so as to generate a main magnetic field B0 shown along axis 122, where the direction of the magnetic field depends on the current direction of the main electromagnet.

[0020] FIG. 1 also depicts a simplified exemplary radiation therapy device 150 for the delivery of radiation therapy. Examples of radiation therapy devices can include, for example, a linear accelerator (linac) for the delivery of high-energy photons (such as x-rays, gamma rays, etc.), a particle beam source (e.g., protons, heavy ions, neutrons, electrons, etc.). The radiation therapy device 150 can be configured to move to various locations around a patient and deliver radiation at various angles. For example, the radiation therapy device can be mounted on a rotatable gantry disposed between the halves of an MRI magnet, such that the gantry can rotate around the patient and enable imaging using MRI while delivering radiation at varying gantry angles. A simplified depiction of the rotatable gantry is shown in FIG. 5. In other embodiments, the radiation therapy device 150 can be mounted on a robotic arm or can be in a fixed position.

[0021] As used herein, the phrase "MRgRT system" refers to the hardware and / or software associated with the operation of a magnetic resonance imaging system and a radiation therapy device. In contrast, the more general phrase "system" used throughout this disclosure encompasses any hardware and / or software necessary to implement the concepts of this disclosure that refer to that system. For example, while an MRgRT system may be capable of delivering radiation therapy and performing imaging, the MRgRT system may not necessarily be capable of providing the analysis or display of data as described in certain embodiments herein. Thus, the use of the term "a / the system" encompasses the concepts of this disclosure, such as a processor and / or computer program (and, optionally, an MRgRT system) for enabling radiation therapy gating, dose calculation, generation of a radiation therapy plan, etc. FIG. 1 depicts an exemplary system, but the improvements to MRI-guided therapy disclosed herein may be implemented using other MRgRT designs.

[0022] FIG. 2 shows an example of imaging a patient by an MRgRT system that utilizes non-parallel imaging planes, which is consistent with a particular aspect of the present disclosure. The imaging can be performed in varying planes through the patient to provide various views of the patient's anatomical structures that can be utilized for diagnosis, radiation delivery, etc. FIG. 2 depicts three exemplary non-parallel planes (e.g., sagittal plane 200, coronal plane 202, and transverse plane 204) through patient 110. In some embodiments, such planes may correspond to a natural coordinate system for a particular system (e.g., having dimensions along or transverse to the main magnetic field axis), but this is not essential and any orientation with respect to the planes of the present disclosure is contemplated. Generally, an inset showing an exemplary imaged anatomical structure (e.g., a target tumor) 210 centered at the intersection of the exemplary non-parallel planes is depicted below the patient.

[0023] As depicted in further inserted figures, the 3D anatomical structure 210 has 2D projection images (220, 222, 224) that are imaged in corresponding non-parallel planes (200, 202, 204). Such projections can be utilized to facilitate radiation delivery or gating with respect to specific spatial limits (e.g., 230, 232, 234) depicted in FIG. 2 by the dashed boundaries. Also, in certain embodiments, the system can cause a graphical representation regarding any combination of non-parallel planes, projection images of anatomical structures, spatial limits, etc. to be displayed on a display device 250. Examples of display devices can include computer monitors, touch screen monitors, smartphone screens, etc.

[0024] As shown in the example of FIG. 2, some embodiments can include three orthogonal planes. More generally, advantages can be obtained when the system is configured to image in at least two non-parallel planes. In some embodiments, as shown, the non-parallel planes can be orthogonal. In other embodiments, the non-parallel planes can be oblique (i.e., not orthogonal). It is contemplated that the above combinations can be utilized. For example, in an embodiment having three planes, in addition to a third plane (similar to a cross-section), there can be a coronal plane and a sagittal plane that are orthogonal, and the third plane can be oblique with respect to one or both of the other two planes. Generally, the terms ("orthogonal" and "oblique") are used to describe the relationship of the planes with respect to each other, rather than with respect to any particular coordinate system. Further details of the embodiments and concepts disclosed in FIG. 2 are described below with reference to the contour depiction / gating control embodiments described with respect to FIG. 4.

[0025] FIG. 3 shows an embodiment of a process for imaging non-parallel planes according to certain aspects of the present disclosure. The use of multiple non-parallel planes can be made possible, for example, by energizing and controlling various subsystems of an MRI gradient coil system. As depicted by the exemplary process in FIG. 3, at 310, a portion of the Y-Z plane can be imaged by establishing a gradient magnetic field in the X direction by an X gradient coil. Next, at 320, a portion of the X-Z plane can be imaged by establishing a gradient magnetic field in the Y direction by a Y gradient coil. At 330, a portion of the X-Y plane can be imaged by establishing a gradient magnetic field in the Z direction by a Z gradient coil.

[0026] In some embodiments, for volumetric imaging applications and the like, multiple images can be acquired at offset planes (e.g., by repeatedly imaging the planes with different gradient coil configurations to shift the location of slice selection). FIG. 3 depicts a process of imaging the i-th plane by process loops (312, 322, 332) associated with each respective gradient coil (where there are a total of N planes with the plane being imaged at the instant designated by subscript (i)). The present disclosure contemplates that this can be done in any combination, e.g., the orientation of the planes can be sequentially changed (e.g., Y-Z, then X-Z, then X-Y), and then an offset is applied to acquire a new set of non-parallel images (e.g., Y-Z+ΔX, then X-Z+ΔY, then X-Y+ΔZ).

[0027] Accordingly, the present disclosure contemplates that any combination of X, Y, and Z gradient coils can be energized (including multiple coils simultaneously, such as X and Y). Such use of the gradient coil system can enable a series of rapid imaging to be acquired by switching the imaging plane. Accordingly, this rapid switching can enable the acquisition of real-time images from a magnetic resonance imaging system at at least two non-parallel planes.

[0028] In some implementations, orthogonal acquisitions can be performed as separate groups of imaging planes (e.g., sagittal, coronal, or transverse), each containing one or more imaging planes passing through the patient. For example, in an embodiment, there may be a single imaging plane (e.g., a total of three planes) for each group. In other embodiments, there may be multiple planes in each group (e.g., a total of 30 planes, 10 parallel planes for each non-parallel group) that can facilitate volumetric imaging. The system can independently set the orientation (sagittal, coronal, or transverse) of the imaging plane and the imaging plane offset (e.g., X, Y, Z coordinates) (optionally based on user input). In some embodiments, the system can set the phase encoding direction based on the slice orientation for best image quality and speed. Also, in some embodiments, other parameters can be common between imaging plane groups and can allow for rapid switching of orientations. In some embodiments, the imaging speed can be improved by using a combination of accelerations in the phase encoding direction (e.g., generalized autocalibrating partially parallel acquisition (GRAPPA) and partial Fourier). Also, the image quality can be improved by incorporating phase oversampling when it is necessary to prevent aliasing. Thus, the system can perform image analysis to detect the presence of aliasing and implement phase oversampling to reduce or prevent the aliasing.

[0029] The system can perform various types of MRI to enable more accurate identification of the patient's anatomical structures. In some embodiments, for example, the acquisition of real-time images from a magnetic resonance imaging system in at least two non-parallel planes can include T1 and T2 weighted volumetric scans at the SRS isocenter. Other types of planar scans or volumetric scans can also be implemented by the MRI of the present disclosure.

[0030] Figure 4 shows an embodiment of a process for gate controlling a radiotherapy beam that is consistent with certain aspects of the present disclosure. At 410, the system may acquire real-time images of a patient from a magnetic resonance imaging system in at least two non-parallel planes. At 420, the system may delineate the contours of the patient's anatomical structures in the non-parallel planes. At 430, the system may set spatial limits for movement of the anatomical structures in the non-parallel planes. At 440, the system may control a radiotherapy device to deliver a radiotherapy beam to the patient. At 450, the system may turn off the gate of the radiotherapy beam when the anatomical structures exceed the spatial limits in any of the non-parallel planes.

[0031] As used herein, "real-time" means minimal delay between image data capture and image display. In particular, the acquisition, reconstruction, and processing times may be less than the frame rate such that the imaging can follow the procedure. Thus, the use of such real-time images can inform a clinician or system about the patient's nearly current state. In fact, implementations of real-time imaging may include a delay of less than one second, sometimes much less, e.g., 0.25 seconds, to reconstruct and display volumetric images. The required frame rate may depend on the speed of the motion being tracked in the patient, and 0.25 seconds is sufficient to resolve typical cardiac and respiratory motions.

[0032] Referring to operation 420, in some embodiments, the delineation of the contours of the patient's anatomical structure can be performed via automated contouring by the machine. For example, from the MRI image, the system can perform image analysis (e.g., using gradient analysis or other edge detection algorithms) to detect the edges of the anatomical structure and generate a contour around the detected edges to isolate and identify the anatomical structure. In other embodiments, the delineation of the contours of the patient's anatomical structure can be performed via a machine that receives manual user input. For example, a clinician can receive a patient image on a computing device and generate or edit contours around various anatomical structures via various input methods, such as a mouse, a touch screen, a stylus, etc. In yet other embodiments, the two methods can be combined to allow the clinician to add, remove, or edit machine-generated contours. Along with both the automated contouring of the machine and the manual user input concept, a non-rigid image registration concept can also be utilized to facilitate the delineation of contours in multiple images. For example, the contours in a previous image can be registered deformably in the current image. The present disclosure often refers to a tumor or object that is intended to receive radiation as the "target", which can also be considered an "anatomical structure" whose contours can be delineated and tracked by the MRgRT system.

[0033] Referring to operation 450, gating control (e.g., stopping the delivery of radiation) can be utilized by the system, for example, when it is required to protect the patient from receiving an inappropriate radiation dose. Such gating control can occur, for example, when the system determines that the radiation is not being delivered appropriately, based on an analysis of the movement of the patient or the target, or the radiation dose being delivered.

[0034] To facilitate the gating function, the spatial limits around the anatomical structure can be established by the system or the clinician, where the system can perform gating if the anatomical structure exceeds one or more spatial limits. Simplified depictions (230, 232, 234) of the spatial limits are shown in FIG. 2 by dashed lines at the individual planes (200, 202, 204), where the planar projections (220, 222, 224) of the anatomical structure can be tracked by the system. In the example shown, the spatial limits are depicted as an expansion region around the anatomical structure and represent the allowable region for delivering radiation. In the example, the anatomical structure is depicted as being shifted such that the projections of the anatomical structure in two of the non-parallel planes (200, 204) are outside their respective spatial limits (230, 234). Thus, the system can identify this as meeting the gating conditions and abort the delivery of radiation. In some embodiments, the difference defining the spatial limits can be set to zero or nearly zero to substantially match the contour of the anatomical structure. In other embodiments, the spatial limits can be set in the system as a percentage increase (e.g., 1%, 2%, 5%, etc.) around the volume of the anatomical structure or a specific distance (e.g., a difference of 0.1 mm, 0.2 mm, etc.).

[0035] In some embodiments, the system can be configured to allow the spatial limits to be breached for a limited period without performing gating, for example, to allow short-duration operations that do not have an excessive impact on the radiation dose. For example, in various embodiments, the spatial limits can be exceeded for up to 0.1 seconds, 0.25 seconds, 0.5 seconds, or 1.0 seconds. In other embodiments, a particular anatomical structure (e.g., the prostate) can be allowed to exceed the spatial limits for tens of seconds, e.g., 10 seconds, 20 seconds, or 30 seconds. In further embodiments, the determination of the period during which the spatial gate can be breached can be determined by the system based on the allowable dose to nearby anatomical structures that can drift into the radiation beam when the target moves.

[0036] The present disclosure contemplates that imaging using multiple surfaces can be performed anywhere on a patient, but there are certain applications that benefit from the accuracy provided by the multiple-surface imaging / gating control of the present disclosure. An example of such an application is stereotactic radiosurgery (SRS), which is typically utilized to treat a patient's cranial region (e.g., a brain tumor). Conventional SRS often involves physically fixing the patient by mechanically clamping the patient's head in a fixed position. Despite such techniques, the patient's internal anatomical structures can be connected to (or be) soft tissue. Thus, for example, when the patient yawns, coughs, or changes position, the radiation target can still move relative to the mechanically clamped patient's skull. Since conventional SRS assumes that the patient's anatomical structures are stationary (or simply assumes that the anatomical structures move rigidly with the skull and tracks the displacement of the skull), the actual movement of the internal soft tissue during treatment is unknown, which can thereby result in an inappropriate delivery of radiation dose.

[0037] The techniques and radiation treatment systems disclosed herein may be particularly well-suited for performing SRS, and thus, the system may be controlled to deliver stereotactic irradiation to a patient. Since the brain tumor is very small and the margin of error is also small, the treatment of the tumor in SRS may involve delivering a very small beam of radiation (e.g., having a very small cross-section). In some cases, such delivery may be facilitated by the use of a high-resolution multi-leaf collimator to provide highly precise stereotactic radiotherapy to a very small target. Also, due to the need to gate-control the radiation quickly while considering strict tolerances for radiation delivery, non-parallel tracking / gate control in multiple planes may be combined with the delivery of small beam radiation to deliver highly precise stereotactic radiotherapy during an SRS procedure. The highly precise delivery of radiation may further be facilitated by a system that utilizes a magnetic resonance imaging system at a magnetic field strength of less than 2.0 Tesla, 1.0 Tesla, 0.75 Tesla, 0.5 Tesla, or 0.35 Tesla. Operating the MRI at such a relatively low magnetic field (compared to high-field MRI, which can be 3.5T or higher) may enable imaging with reduced artifact generation and thus may enable a more spatially accurate representation of anatomical structures in the MRI image.

[0038] In other embodiments that may be utilized in SRS, the system may be configured to set spatial limits within 0.5 mm of the boundaries of anatomical structures in at least two non-parallel planes. Other examples of similar stereotactic spatial limits may include 0.1 mm, 0.25 mm, 0.75 mm, or 1.0 mm, etc.

[0039] FIG. 5 shows an MRgRT system that includes a multi-leaf collimator (MLC) 520 that collimates a beam within a specific shape during treatment. The exemplary system may also include a mechanism for acquiring actual beam delivery information. For example, some embodiments may include a beam output sensor 510 (e.g., a monitor / ion chamber). Additionally, the shape of the beam (or the position of the MLC leaves) may be measured by a beam shape sensor or fluence sensor 522 (e.g., a scintillator). The MRgRT system may include a gantry 530, and the gantry 530 may include a gantry sensor 532 for measuring the gantry angle / position. An exit radiation detector 542 may also be present, and the exit radiation detector 542 may be attached to rotate with the gantry to measure radiation that is not absorbed or scattered by the patient (e.g., an electronic portal imaging device or EPID). Also, the patient couch 108 may include a patient couch sensor 552 for measuring the location / orientation of the patient couch during treatment. The present disclosure assumes that the sensors described are exemplary and that the sensors may be incorporated in any combination in the MRgRT system. Also, the examples given do not exclude other sensors that may be incorporated within the system to provide actual beam delivery information. Exemplary types and uses of such sensors for determining actual beam delivery information are further described with reference to FIG. 6.

[0040] FIG. 6 shows an embodiment of a system and a process for using actual beam delivery information to calculate an accumulated radiation dose in software that is consistent with a particular aspect of the present disclosure. As further described herein, another improvement over current radiation therapy techniques is that rather than assuming that a planned radiation dose has been delivered, the systems and processes of the present disclosure utilize information regarding what the radiation therapy system has actually delivered. By calculating the actual radiation dose delivered (e.g., at every instant), the system can further calculate the actual dose that has actually been accumulated in the tissue during radiation therapy. As further described below, this facilitates other technological developments including re-optimization of the improved radiation therapy plan.

[0041] The process flowchart of FIG. 6 depicts one preferred embodiment, where, at 610, the system may deliver radiation therapy to a patient from a radiation therapy device. At 620, the system may acquire an image of the patient from a magnetic resonance imaging system during radiation therapy. At 630, the system may acquire actual beam delivery information during radiation therapy, where the actual beam delivery information includes actual beam measurements including one or more of a measured monitor unit, a measured MLC leaf position, a measured gantry position, a measured couch position, and a measured fluence profile. At 640, the system may calculate the dose to tissue during radiation therapy based on the acquired image and the acquired actual beam delivery information. At 650, the system may accumulate the dose to tissue during radiation therapy.

[0042] Referring to operation 620, examples of the acquired image may include a single plane, multiple planes (e.g., multiple parallel or non-parallel planes as described herein), volumetric (e.g., 3D patient volume over time, e.g., imaging sufficient to generate 4 volumes per second or 8 volumes per second, etc.). As used herein, the term "acquired image" excludes modalities that may rely on features, e.g., fiducials, tracking of skin surfaces, etc., where the treatment or planning relies on interference with the internal anatomical structures of the patient.

[0043] Referring to operation 630, rather than relying on the characteristics of the planned radiation delivery (e.g., planned monitor units, planned MLC leaf positions, etc.), the system can access or determine the actual beam delivery information. For example, the measured monitor units can be obtained from the beam output sensor 510, which can include a radiation detector, such as a monitor / ion chamber, a diode, a pickup coil, etc., that measures the actual output of the radiation therapy device. Similarly, rather than relying on the planned MLC leaf positions, the determination of the actual MLC leaf positions can be obtained, for example, by a beam shape sensor 522 that can analyze light from intervening scintillators and a camera system, or by obtaining MLC leaf position encoder data. The measured gantry position can be determined using a gantry sensor 532, which can include gantry angle position encoder data and, for example, an exit radiation analysis (using an exit radiation detector that does not rotate with the gantry and can thus identify changes in the angle of incidence of the radiation). The measured fluence profile can be obtained using an exit radiation detector 542, such as an EPID or a scintillator. The measured couch position can be obtained from a couch sensor 552, which can provide couch position encoder data.

[0044] Referring to operation 640, the system can obtain any of the actual beam delivery information and provide it to a dose calculator for calculating the actual dose delivered during radiation therapy. As used herein, the term "dose calculator" refers to a software program or module programmed to calculate the radiation dose to a patient from information regarding beam delivery and the patient's anatomical structure. Examples of the radiation dose output by the dose calculator can include 2D and / or 3D mapping of the dose to the patient, and optionally, can be obtained over time to represent dose accumulation.

[0045] FIG. 7 provides an exemplary diagram relating to the determination and display of the cumulative radiation dose to a patient tissue that is consistent with certain aspects of the present disclosure. Since radiation may not always be delivered precisely to the target tissue, determining the actual accumulated dose can improve the ability of the system to meet treatment specifications that can be defined in terms of the desired dose to the target and the tolerance dose to other tissues / organs. Prior to describing how embodiments of the present disclosure perform re-optimization of a radiation treatment plan based on the actual dose delivered to a patient, FIG. 7 depicts a system for determining and displaying the actual cumulative radiation dose.

[0046] FIG. 7 depicts the radiation dose accumulation in three planes (200, 202, 204) showing the projected images of targets (220, 222, 224) similar to those described with reference to FIG. 2. The area around the target is the corresponding area of radiation dose deposition that can be created to physically cover the target and the slight differences in the area around the target. FIG. 7 depicts this accumulation at three different times (Time A, Time B, and Time C). The amount of accumulated radiation is represented by the degree of the shaded area. At Time A, in the upper panel, the radiation is delivered in the areas (720a, 722a, 724a) bounded by the dashed lines. At Time B, in the middle panel, the patient shifts upward. The radiation is delivered again to the same location under the assumption that the target was stationary. Since the radiation is delivered to different locations in the patient tissue (i.e., below where the target (220, 222, 224) is currently located), there may be an area of underdose 730 to the target and an area of overdose 740 to the nearby tissue. At Time C, in the lower panel, when the patient shifts to the right, this process continues and the radiation is still delivered to different locations. The accumulated radiation depicted in FIG. 7 is not ideal, but with real-time imaging, the utilization of the present disclosure regarding the actual beam delivery information for calculating the accumulated dose enables the system to measure it. Using this information, embodiments of the present disclosure may enable the re-optimization or replanning of treatment to take into account the known actual accumulated radiation dose. Thus, such improved measurement and / or re-optimization procedures may reduce or eliminate deviations from radiation regulations.

[0047] The system of the present disclosure can display the accumulated dose to tissue during radiation therapy (e.g., on a display device such as a computer monitor). The display can occur during and / or after radiation therapy. In some embodiments, the display of the accumulated dose can be overlaid on an image acquired during radiation therapy. Thus, the patient's anatomical structures appearing in the MRI image can be associated with the delivered dose. To facilitate the determination and display of the accumulated dose during radiation therapy, some embodiments can include displaying the accumulated dose in the most recent set of MRI images. Thus, a dose computer and / or clinician can be provided with an up-to-date dose map of the patient's tissue. The dose can be displayed in a number of formats by displaying the accumulated dose on multiple imaging planes, e.g., three orthogonal planes, as depicted, for example, in FIG. 7.

[0048] In some implementations, determination of the accumulated dose can be incorporated with the gating operation (e.g., to prevent an undesirable excessive delivery of radiation outside of an acceptable dose limit). In still other implementations, additionally or alternatively, the system can re-optimize the radiation treatment plan to take into account the actual accumulated dose at a particular stage of the radiation treatment. One exemplary process that can incorporate both gating and re-optimization can include a system that depicts the contours of a patient's anatomical structures, such as depicted in FIG. 7 and described elsewhere herein. The system can then determine the accumulated dose to the anatomically structured depicted contours, for example, via a dosimetry computer. In some embodiments, the system can provide a notification or an alarm if the accumulated dose to the anatomically structured depicted contours exceeds a specified limit. Such a specified limit can be provided by radiation regulations that can be set prior to the delivery of the radiation. Similarly, the system can be configured to stop the delivery of the radiation treatment if the accumulated dose to the anatomically structured depicted contours exceeds a specified limit / when the accumulated dose to the anatomically structured depicted contours exceeds a specified limit. For example, if a particular organ at risk (e.g., the rectum) during prostate treatment accumulates more than the exemplary 10.8 Gy limit, rather than continuing the treatment as done in conventional systems that do not have the capabilities described herein, the radiation treatment can be ended.

[0049] Furthermore, the system of the present disclosure can determine a re-optimized treatment plan if the accumulated dose to the anatomically structured depicted contours exceeds a specified limit. In some advantageous embodiments, determination of the re-optimized treatment plan takes into account the accumulated dose and any underdosed or overdosed regions. For example, referring to FIG. 7, when an underdosed region 730 or an overdosed region 740 is identified (relative to what should be in the current treatment plan), the treatment plan can adjust subsequent radiation delivery parameters (e.g., MLC position, gantry angle, etc.) such that the re-optimized treatment plan meets the specifications. The re-optimized plan can be implemented immediately, although in some implementations, the gating (e.g., interruption of the radiation treatment) can continue until the system receives clinician approval regarding the re-optimized treatment plan.

[0050] In embodiments with sufficient processing power, the interruption of radiation delivery can be very short, such that the system can continuously re-optimize the radiation treatment plan as rapidly as when patient imaging (or other limiting processes) can occur.

[0051] FIG. 8 shows an exemplary system that utilizes non-rigid image registration to track accumulated dose during radiation therapy, in accordance with certain aspects of the present disclosure. In some embodiments, techniques including non-rigid image registration can further improve the utilization of accumulated dose by accurately tracking the deposited dose even when tissue is deforming during radiation therapy. As depicted in FIG. 8, a system that calculates and / or accumulates dose to tissue can utilize non-rigid image registration and a most recent set of MRI images. When the present disclosure contemplates numerous embodiments related to real-time imaging and dose calculation, the term "most recent" can refer to MRI images that are generated at a time very close to the current time during radiation therapy, such that the MRI images substantially represent the patient's current state. Thus, "most recent" can literally include the most recently acquired MRI images, although it is contemplated that the MRI images can also include MRI images that were acquired up to 0.25 seconds prior.

[0052] In FIG. 8, the upper panel 810 shows an exemplary target 812 and a dose distribution 814 generally corresponding to the target. Some exemplary pixels (or voxels) representing locations 816 that have received a given dose are depicted. The middle panel 820 shows a deformed target 822 that represents a stretching of the target 812. This panel may also be considered an example of a latest MRI image. By utilizing non-rigid image registration, the system can generate a mapping between the MRI image showing the deformed target and the locations in the previous MRI image. The lower panel 830 depicts the system applying the mapping to the dose distribution 814 and then deformably registering the dose distribution onto the subsequent (or latest) MRI image to form an updated dose distribution 834. The updated locations 836 of the locations 816 are also depicted. By continuously performing this process during radiation delivery, the system can accurately accumulate the dose in the latest set of MRI images.

[0053] In some embodiments, the system can also perform such dose calculation / accumulation based on accounting for the assigned relative electron density. As part of a radiation treatment plan, different substances within a patient (e.g., water, tissue, bone, etc.) can have different electron densities that affect their ability to receive a radiation dose. The system can optionally assign such relative electron densities to the identified anatomical structures using input manually provided by a clinician. Similarly, a mapping can also be applied to the relative electron density to facilitate accurate dose calculation for the deformed structure to the extent that such a structure deforms as described above.

[0054] Similar to other embodiments disclosed herein, the system can provide a display of the dose accumulated in the images acquired during radiation treatment by using non-rigid image registration between the latest set of MRI images and the images acquired during radiation treatment.

[0055] In some embodiments, the calculation of the dose to the tissue further utilizes independent measurements of the magnetic resonance imaging system (e.g., measuring the control system / gradient pulses / etc., rather than using the MRI timestamp) and independent measurements of the radiation therapy device (e.g., ion chamber readings) to synchronize the acquired image and the acquired actual beam information.

[0056] The concepts described above with respect to dose calculation and accumulation can be integrated into embodiments of a radiotherapy plan with a prognostic motion model / prognostic motion adaptation as described below. An example of such integration is provided in the following description with respect to FIG. 13.

[0057] One limitation in previous MRI-guided radiotherapy delivery was that, even when the patient was imaged in real time, the treatment device had a limited ability to deliver treatment that adapted to and was consistent with such motion. However, the present disclosure addresses these issues, at least in part, through the use of a prognostic motion model. Using the prognostic motion model, the expected patient movement is known to the MRgRT system, and the radiotherapy plan and delivery can take such motion into account, improve the dose distribution, accelerate the treatment, and account for the latency in the system between patient imaging and radiation delivery. A radiotherapy plan that utilizes such a prognostic motion model is referred to herein as a prognostic motion adaptation radiotherapy plan.

[0058] FIG. 9 shows the generation of a prognostic motion model and a prognostic motion adaptation radiotherapy plan that is consistent with certain aspects of the present disclosure. The prognostic motion model may include a model of the expected patient movement during treatment. In some embodiments, the MRgRT system may be configured to perform operations including, at step 910, acquiring a pre-treatment image using a magnetic resonance imaging system, the pre-treatment image capturing patient movement. Then (at step 920), the system may generate a prognostic motion model based at least on the pre-treatment image. At step 930, the system may generate a prognostic motion adaptation radiotherapy plan based at least on the prognostic motion model.

[0059] Referring to operation 910, pre-treatment images can include cine MRI (e.g., 2D or 3D MRI images of a patient over a certain time span), phase-binned MRI (e.g., MRI images grouped to effectively generate a representation of a patient's motion over a certain time span, not necessarily acquired sequentially), etc. Ideally, the pre-treatment images enable reconstruction regarding the patient's anatomical structure and movement with high temporal resolution. However, the present disclosure assumes that gaps may exist in the acquired pre-treatment images. Thus, it is not essential for the pre-treatment images to be a continuous (or nearly continuous) representation of the patient's movement. Rather, there may be a series of images sufficient to enable the development and implementation of motion-adaptive radiotherapy planning as described herein. In certain embodiments, the pre-treatment images can be acquired at exemplary rates of, for example, 2 frames per second, 4 frames per second, 8 frames per second, 16 frames per second, or 32 frames per second for 2D images, and 1 volume per second, 2 volumes per second, or 4 volumes per second for the reconstruction of a typical 3D patient volume. In some embodiments, the time-dependent 3D volume reconstruction can occur at the same rate as the acquisition of its composite 2D images. Operations 920 and 930 of FIG. 9 are described below with reference to FIG. 10.

[0060] In some embodiments, generating the prognostic motion model further includes excluding pre-treatment images that do not represent the expected patient motion during treatment. For example, during acquisition of the pre-treatment images, if the patient performs an action such as coughing, changing position, or otherwise that is irregular and not expected to be reproduced during an actual treatment session, such images can be excluded from the model. In some embodiments, such images can be deleted by a clinician who examines the pre-treatment images on a display device. In other embodiments, such images can be deleted by the system by detecting excessive differences between images or groups of images. Such detection can occur, for example, by performing image analysis to identify anatomical structures, reference points, etc., and determining deviations from expected locations. For example, if the location of an anatomical structure changes over 2 cm during a typical respiratory cycle, but a portion of a pre-treatment image depicts the anatomical structure 5 cm from its expected (or average) location, such pre-treatment images can be excluded from the prognostic motion model.

[0061] In certain embodiments, the prognostic motion model can include multiple types of motion observed in the pre-treatment images. For example, the multiple types of motion can include expected patient motion and irregular patient motion. These motions can be considered in planning and delivery as further described below. Some examples of multiple types of motion that can be observed and included in the model are regular motion due to breathing, motion due to deep breathing, motion due to GI system gas movement, motion due to bladder filling, chest wall breathing, diaphragmatic breathing, motion due to swallowing, cardiac motion, patient motion (e.g., position change, rotation, etc.), motion due to conversation, eye movement, voluntary muscle motion (e.g., contraction or tension / relaxation), etc.

[0062] Conventional radiation treatment planning assumes a stationary patient's anatomical structure, but embodiments of the systems of the present disclosure can generate a prognostic motion adaptive radiation treatment plan that takes into account the expected patient motion of the prognostic motion model during treatment.

[0063] The expected patient motion during treatment can include the motion of the target being treated. When a radiotherapy plan with motion adaptation in the future takes into account the expected motion of the target when determining the location and time at which treatment is delivered, it can provide a more accurate dose delivery than a plan that is not thus adapted. For example, a radiotherapy plan with motion adaptation in the future can take into account the motion of the target that is small enough to avoid beam gating control in a conventional system but large enough that the dose will be delivered outside the target. For example, a conventional MRgRT system may have the ability to turn off the radiation gate when the beam significantly deviates from the target, but nevertheless, it can allow the beam slightly away from the target to continue radiation delivery for a significant period. Thus, a conventional system can deliver an excessive dose outside the target. In contrast, a radiotherapy plan with motion adaptation in the future can avoid such doses outside the target by more fully considering the location and shape of the target according to its motion model in the future.

[0064] Not only do the radiotherapy target and nearby organs at risk move during treatment, but they also deform, so a radiotherapy plan with motion adaptation in the future can also take into account the expected patient motion during treatment that is the deformation of the target or an organ of interest.

[0065] FIG. 10 shows, in a simplified manner, one way in which such a radiotherapy plan with motion adaptation in the future can more fully consider tissue deformation than the treatment of conventional radiotherapy. At the top of FIG. 10, the target 1010 is surrounded by a spatial gating control limit 1020, and the organ 1030 of interest (e.g., an organ that should not receive radiation, such as a blood vessel) is on the right side.

[0066] Conventional radiation treatment plans based on static imaging assume that the target 1010 and the organ of interest 1030 maintain the same spatial relationship to each other (e.g., 5 mm apart). Such conventional systems can adequately handle the movement described in the second row of FIG. 10, where both the target and the organ of interest move to the left and the deviation of the target from the spatial gate control limit 1020 turns off the beam gate. However, conventional systems do not adequately handle the movement shown in the third row of FIG. 10, where both movement and deformation occur and the target does not exceed the spatial gate control limit 1020, and as a result, the beam is not turned off, and then the organ of interest 1030 is exposed to radiation and damage. Instead, the beneficial technique of the present disclosure that utilizes a prospective motion-adaptive radiation treatment plan recognizes that the organ of interest deforms into the beam path during a particular patient's motion phase, and thus the plan stops delivery during that motion phase and the organ of interest is spared from damaging radiation.

[0067] Similarly, some embodiments regarding prospective motion-adaptive radiation treatment plans can account for expected patient motion of the prospective motion model during treatment, as well as irregular patient motion. As an example of how a prospective motion-adaptive radiation treatment plan can account for irregular patient motion, the prospective motion-adaptive radiation treatment plan can adjust or stop delivery when irregular patient motion in the prospective motion model is observed by the imaging system during treatment. As an example of adjusting radiation delivery, the patient motion model can include patient motion identified as irregular patient motion such as gas movement through the patient's gastrointestinal tract that results in movement of the target prostate. The prospective motion-adaptive radiation treatment plan can then stop the delivery of radiation when such irregular patient motion is detected. In another embodiment, the prospective motion-adaptive radiation treatment plan can deliver radiation to the MRgRT system along the irregular patient motion identified by the prospective motion model. Such an implementation has the advantage of avoiding interruptions in radiation delivery that can lengthen the patient's treatment time.

[0068] This disclosure assumes that a radiotherapy plan with prospective motion adaptation can take into account the latency between the movement of a patient observed using a magnetic resonance imaging system and the delivery of a radiation beam to the patient. There is a delay or latency during the time the patient moves, the time MRI image reconstruction is performed, and the time during which radiation delivery can be adjusted (e.g., beam on / off, adjustment of MLC leaf positions, gantry rotation, etc.). None of these operations occur instantaneously, and herein, latency is understood to include any combination related to hardware and / or software delays. By taking these delays into account, a radiotherapy plan with prospective motion adaptation can aim the radiation beam towards where the target is expected to be based on the prospective motion model and latency, rather than at a specific point in the space where one might think the target is located.

[0069] FIG. 11 shows one embodiment of a process for delivering radiotherapy according to a radiotherapy plan with prospective motion adaptation that is consistent with a particular aspect of this disclosure. At 1110, the system may obtain a treatment image of the patient from a magnetic resonance imaging system, the treatment image capturing the movement of the patient. At 1120, the system may deliver radiotherapy from a radiotherapy device to the patient according to a radiotherapy plan with prospective motion adaptation based at least on a prospective motion model.

[0070] In some implementations, the system can be configured to identify the movement of the current patient from the treatment images. Various embodiments of the system of the present disclosure can identify the movement of the current patient (e.g., through any of the processes described below with respect to the description of FIG. 12). The system can then be configured to initiate the delivery of radiation therapy when the movement of the current patient matches the expected patient movement included in the prognostic motion model. As used herein, the term "matches" does not necessarily require an exact match between the movement of the current patient and the expected patient movement (however, in some embodiments, it can be utilized). Instead, in various embodiments, there can be an acceptable deviation between the movement determined from the treatment images and the expected patient movement. The system can be configured to accept movement variations such as 1%, 2%, 5%, or 10% in any particular direction. As another example, the system can be configured to initiate the delivery of radiation therapy when the temporal overlap between the location of the actual target volume and the location of the expected target volume is as small as an overlap of 1%, 2%, 5%, or 10%. Further, the initiation of radiation therapy can further take into account the latency between the movement of the current patient observed using a magnetic resonance imaging system and the delivery of the radiation beam to the patient.

[0071] FIG. 12 shows one embodiment of a process for delivering radiation therapy when the patient's movement follows a prognostic motion model. In an exemplary process, at 1210, the system can identify the movement of the patient from the acquired treatment images. The present disclosure contemplates a number of algorithms for identifying whether the movement of the patient in the acquired treatment images is expected based on the prognostic motion model.

[0072] In some embodiments, the system can directly compare the patient's image to that of the prognostic motion model and determine whether a given image matches the expected patient motion based on a similarity calculation between the images. Such similarity calculations can include pixel comparison, determination of correlation coefficients, and the like.

[0073] In other embodiments, the system may determine whether a patient's movement matches a radiation treatment plan adapted to prognostic behavior by utilizing an organizational probability distribution based on a prognostic behavior model. An exemplary process may include utilizing a prognostic behavior model to determine an organizational probability distribution, which quantifies the possible locations of a patient's tissue during an expected patient movement. The system may then compare the patient's tissue distribution from the treatment image with the organizational probability distribution. Radiation treatment may be delivered when the comparison indicates that the tissue distribution is likely to represent the expected patient movement.

[0074] In some embodiments, the system may determine whether the trajectory of tissue in a patient matches a radiation treatment plan adapted to prognostic behavior by utilizing an expected patient tissue trajectory based on a prognostic behavior model. An exemplary process may include utilizing a prognostic behavior model to determine an expected patient tissue trajectory, which quantifies the expected trajectory of a patient's tissue during an expected patient movement. The system may then compare the patient's tissue trajectory from the treatment image with the expected patient tissue trajectory. Radiation treatment may be delivered when the comparison indicates that the patient tissue trajectory is sufficiently similar to the expected patient tissue trajectory.

[0075] In yet other embodiments, the system may determine whether the differential movement of tissue in a patient matches a radiation treatment plan adapted to prognostic behavior by utilizing an expected differential movement of a patient's tissue based on a prognostic behavior model. An exemplary process may include utilizing a prognostic behavior model to determine an expected differential movement of a patient's tissue, which quantifies the expected differential movement of a patient's tissue during an expected patient movement. The system may then compare the differential movement of the patient's tissue from the treatment image with the expected differential movement of the patient's tissue. Radiation treatment may be delivered when the comparison indicates that the differential movement of the patient's tissue is sufficiently similar to the expected differential movement of the patient's tissue.

[0076] In step 1220 of FIG. 12, the system can determine whether the patient's movement is the expected movement. For example, the identified movement can be compared to the type of movement (i.e., the expected movement) associated with the radiotherapy plan for prognostic motion adaptation. If the identified movement matches the expected movement, the treatment can then be delivered according to the portion of the plan created for the expected movement. If the identified movement does not match the expected movement, certain embodiments can interrupt the treatment in step 1230 (e.g., turn off the beam gate). In another embodiment, if the identified movement is not the expected movement, the system can check in step 1240 to determine whether the identified movement is a known irregular movement. In some embodiments, the identification of a known irregular movement can cause treatment interruption / gate control in step 1250. In other embodiments, when a known irregular movement is identified, the radiotherapy plan for prognostic motion adaptation can deliver treatment along that known irregular movement in action 1260.

[0077] In response to treatment interruption or gate control, the system can continue to monitor and identify the patient's movement and, at the stage when the patient's movement returns to the expected patient movement, resume delivery of radiotherapy according to the radiotherapy plan for prognostic motion adaptation.

[0078] As used herein, the term "in a phase" means when the observed patient movement / location of the anatomical structure returns to a state similar to that before the interruption of treatment. For example, if the prognostic motion model includes normal patient breathing and the patient coughs and gates the system, the resumption of radiation therapy may not begin until further imaging and analysis determines that the patient has resumed normal breathing. In embodiments, this may further include waiting until an action in a particular state is identified. For example, if radiation therapy is gated at full exhalation, the system may then resume delivery of radiation therapy only when another full exhalation is the identified current patient movement. In some embodiments, this may include requiring that the patient's movement match the expected movement for a particular period of time, such as 1 second, 30 seconds, 1 minute, etc., to ensure that unacceptable movements are aborted.

[0079] If the movement of the patient during treatment does not sufficiently match the prognostic motion model, the system may be configured to generate a second prognostic motion model using the acquired treatment images. The system may then generate a radiation therapy plan with a second prognostic motion adaptation based at least on the second prognostic motion model. In some embodiments, the second prognostic motion model may be an entirely new model, while in other embodiments, it may be a modified model. The system may generate such a modified second prognostic motion model based on inclusion / editing / replacement of images constituting the prognostic motion, such as expanding the image set to include images of deeper breaths.

[0080] After creating a radiation therapy plan with a second prognostic motion adaptation, the system may resume delivery of radiation therapy using the radiation therapy plan with the second prognostic motion adaptation. In one implementation, the radiation therapy plan with the second prognostic motion adaptation may be presented to the clinician for approval (e.g., via a display device) before resumption of treatment. In another embodiment, the system may continuously update the prognostic motion model and the radiation therapy plan with the prognostic motion adaptation to best match the current movement of the patient on the treatment table.

[0081] Many of the technical improvements provided by the present disclosure can be combined to provide even better accuracy and efficiency when delivering radiation therapy. For example, the present disclosure describes improved techniques for dose calculation and accumulation based on the acquisition of actual beam delivery information during radiation therapy (see, e.g., FIG. 6). Such features can be incorporated into the generation and use of a radiation therapy plan with prognostic motion adaptation as described herein. FIG. 13 depicts a preferred embodiment of a process that combines such features.

[0082] At 1310, the system can be configured to acquire actual beam delivery information during radiation therapy, where the actual beam delivery information comprises actual beam measurements including one or more of a measured monitor unit, a measured MLC leaf position, a measured gantry position, a measured couch position, and a measured fluence profile.

[0083] At 1320, the system can calculate the dose to tissue during radiation therapy based on the acquired image and the acquired actual beam delivery information.

[0084] At 1330, the system can accumulate the dose to tissue during radiation therapy. At 1340, the system can generate a second radiation therapy plan with prognostic motion adaptation if the accumulated dose to the anatomical structure exceeds a specified limit. Similar to the operations described above, the second radiation therapy plan with prognostic motion adaptation can be a modified / edited plan or a completely new plan using a new prognostic model. Additionally, the system can generate a second radiation therapy plan with prognostic motion adaptation taking into account the accumulated dose and any underdosed or overdosed regions. For example, the number, orientation, and duration of the radiation beams delivered can be adjusted to best meet the specifications. In addition, the system can generate a second radiation therapy plan with prognostic motion adaptation taking into account a second prognostic motion model using the acquired treatment images, similar to the techniques described above.

[0085] At 1350, the system may continue to deliver radiation therapy using a second prognostic motion adaptive radiation therapy plan. In one embodiment, the new radiation therapy plan may be displayed to the clinician for review, editing, and approval. However, the system may also be configured to reference a set of tolerances (e.g., dose volume histogram or total dose constraints), and if the system can generate a re-optimized treatment plan that remains within the tolerances and dose constraints, the system may then implement the new re-optimized treatment plan and continue to deliver radiation therapy with little or no interruption due to the re-optimization. Thus, the radiation therapy plan may be re-optimized for optimal target coverage, organ preservation, and maximum efficiency.

[0086] In the following, further features, characteristics, and exemplary technical solutions of the present disclosure are described from the perspective of items that may optionally be claimed in any combination.

[0087] Item 1: A non-transitory machine-readable medium storing instructions that, when executed by at least one programmable processor, cause the at least one programmable processor to perform operations, the operations including obtaining real-time images of a patient from a magnetic resonance imaging system in at least two non-parallel planes, depicting the contours of the anatomical structures of the patient in the at least two non-parallel planes, setting spatial limits for movement of the anatomical structures in the at least two non-parallel planes, controlling a radiation therapy device to deliver a radiation therapy beam to the patient, and turning off the gate of the radiation therapy beam when the anatomical structure exceeds a spatial limit in any of the at least two non-parallel planes.

[0088] Item 2: The machine-readable medium according to item 1, wherein the at least two non-parallel planes are three orthogonal planes.

[0089] Item 3: The machine-readable medium according to any one of items 1 to 2, wherein the at least two non-parallel planes are orthogonal.

[0090] Item 4: The at least two non-parallel surfaces are inclined, the machine-readable medium according to any one of Items 1 to 3.

[0091] Item 5: The acquisition of the real-time image from the magnetic resonance imaging system on the at least two non-parallel surfaces is performed by energizing and controlling one or more subsystems of the gradient coil system of the magnetic resonance imaging system, the machine-readable medium according to any one of Items 1 to 4.

[0092] Item 6: The acquisition of the real-time image from the magnetic resonance imaging system on the at least two non-parallel surfaces includes T1 and T2 weighted volumetric scans at the SRS isocenter, the machine-readable medium according to any one of Items 1 to 5.

[0093] Item 7: The depiction of the contour regarding the anatomical structure of the patient is performed via automatic contour depiction of the machine, the machine-readable medium according to any one of Items 1 to 6.

[0094] Item 8: The depiction of the contour regarding the anatomical structure of the patient is performed via a machine that receives manual user input, the machine-readable medium according to any one of Items 1 to 7.

[0095] Item 9: The radiation therapy device is further controlled to deliver stereotactic radiosurgery (SRS) to the patient, the machine-readable medium according to any one of Items 1 to 8.

[0096] Item 10: The magnetic resonance imaging system operates at a magnetic field strength of less than 1.0 tesla, the machine-readable medium according to any one of Items 1 to 9.

[0097] Item 11: The spatial limit is set within 0.5 mm of the boundary of the anatomical structure on the at least two non-parallel surfaces, the machine-readable medium according to any one of Items 1 to 10.

[0098] Item 12: A non-transitory machine-readable medium storing instructions that, when executed by at least one programmable processor, cause the at least one programmable processor to perform operations, the operations including delivering radiation therapy from a radiation therapy device to a patient, acquiring an image of the patient from a magnetic resonance imaging system during the radiation therapy, acquiring actual beam delivery information during the radiation therapy, the actual beam delivery information comprising actual beam measurements including one or more of a measured monitor unit, a measured MLC leaf position, a measured gantry position, a measured couch position, and a measured fluence profile, calculating a dose to tissue during the radiation therapy based on the acquired image and the acquired actual beam delivery information, and accumulating a dose to tissue during the radiation therapy, optionally including any one of Items 1 to 11.

[0099] Item 13: The machine-readable medium according to any one of Items 1 to 12, wherein the operations further include displaying an accumulated dose to tissue during the radiation therapy.

[0100] Item 14: The machine-readable medium according to any one of Items 1 to 13, wherein the operations further include displaying an accumulated dose in the image acquired during the radiation therapy.

[0101] Item 15: The machine-readable medium according to any one of Items 1 to 14, wherein the operations further include displaying an accumulated dose in the most recent set of MRI images.

[0102] Item 16: The machine-readable medium according to any one of Items 1 to 15, wherein the operations further include displaying an accumulated dose at a plurality of imaging planes.

[0103] Item 17: The machine-readable medium according to any one of Items 1 to 16, wherein the operation further includes displaying the dose accumulated on three orthogonal planes.

[0104] Item 18: The machine-readable medium according to any one of Items 1 to 17, wherein the operation further includes depicting the contour of the patient's anatomical structure, determining the accumulated dose for the anatomically structured contour, providing a notification or alarm when the accumulated dose for the anatomically structured contour exceeds a specified limit, stopping the delivery of radiation therapy when the accumulated dose for the anatomically structured contour exceeds the specified limit, and determining a re-optimized treatment plan when the accumulated dose for the anatomically structured contour exceeds the specified limit.

[0105] Item 19: The machine-readable medium according to any one of Items 1 to 18, wherein the determination of the re-optimized treatment plan takes into account the accumulated dose and any underdose or overdose.

[0106] Item 20: The machine-readable medium according to any one of Items 1 to 19, wherein the calculation of the dose to the tissue further utilizes non-rigid image registration and a current set of MRI images including assigned relative electron densities.

[0107] Item 21: The machine-readable medium according to any one of Items 1 to 20, wherein the accumulation of the dose to the tissue further utilizes non-rigid image registration and a current set of MRI images including assigned relative electron densities.

[0108] Item 22: The machine-readable medium according to any one of Items 1 to 21, wherein the dose is accumulated in the current set of MRI images.

[0109] Item 23: The machine-readable medium according to any one of items 1 to 22, wherein the operation further includes using non-rigid image registration between the latest set of MRI images and the images acquired during radiotherapy to display the dose accumulated in the images acquired during radiotherapy.

[0110] Item 24: The machine-readable medium according to any one of items 1 to 23, wherein the calculation of the dose to the tissue further utilizes independent measurements regarding the magnetic resonance imaging system and the radiotherapy device to synchronize the acquired images and the acquired actual beam information.

[0111] Item 25: A non-transitory machine-readable medium storing instructions that, when executed by at least one programmable processor, cause the at least one programmable processor to perform operations, the operations including acquiring pre-treatment images using a magnetic resonance imaging system, wherein the pre-treatment images capture patient movement, generating a prognostic motion model based at least on the pre-treatment images, and generating a radiotherapy plan with prognostic motion adaptation based at least on the prognostic motion model, and optionally including any one of items 1 to 24.

[0112] Item 26: The machine-readable medium according to any one of items 1 to 25, wherein the pre-treatment images include cine MRI.

[0113] Item 27: The machine-readable medium according to any one of items 1 to 26, wherein the prognostic motion model includes a model of the expected patient movement during treatment.

[0114] Item 28: The machine-readable medium according to any one of items 1 to 27, wherein the generation of the prognostic motion model further includes excluding pre-treatment images that do not represent the expected patient movement during treatment.

[0115] Item 29: The machine-readable medium according to any one of Items 1 to 28, wherein generating a prognostic motion model further includes including a plurality of types of motions observed in a pre-treatment image.

[0116] Item 30: The machine-readable medium according to any one of Items 1 to 29, wherein the plurality of types of motions include expected patient movement and irregular patient movement.

[0117] Item 31: The machine-readable medium according to any one of Items 1 to 30, wherein the plurality of types of motions include one or more of regular motion due to breathing, motion due to deep breathing, motion due to GI system gas movement, motion due to bladder instillation, motion due to patient movement, motion due to swallowing, chest wall breathing, diaphragmatic breathing, conversation, eye movement, cardiac motion, or voluntary muscle motion.

[0118] Item 32: The machine-readable medium according to any one of Items 1 to 31, wherein a radiation treatment plan for prognostic motion adaptation takes into account the expected patient movement of the prognostic motion model during treatment.

[0119] Item 33: The machine-readable medium according to any one of Items 1 to 32, wherein the expected patient movement during treatment includes deformation of a target or an organ of interest.

[0120] Item 34: The machine-readable medium according to any one of Items 1 to 33, wherein the expected patient movement during treatment includes movement of a target.

[0121] Item 35: The machine-readable medium according to any one of Items 1 to 34, wherein a radiation treatment plan for prognostic motion adaptation takes into account the movement of the target that is small enough to avoid beam gating but large enough for the dose to be delivered outside the target.

[0122] Item 36: The machine-readable medium according to any one of Items 1 to 35, wherein a radiation treatment plan for prognostic motion adaptation takes into account the expected patient movement and irregular patient movement of the prognostic motion model during treatment.

[0123] Item 37: A radiation treatment plan for adaptive post-treatment motion is a machine-readable medium according to any one of Items 1 to 36 that adjusts or stops delivery when irregular patient motion of a post-treatment motion model is observed during treatment.

[0124] Item 38: A radiation treatment plan for adaptive post-treatment motion is a machine-readable medium according to any one of Items 1 to 37 that takes into account the latency between patient motion observed using the magnetic resonance imaging system and the delivery of a radiation beam to the patient.

[0125] Item 39: A radiation treatment plan for adaptive post-treatment motion is a machine-readable medium according to any one of Items 1 to 38 that determines the aiming of the radiation beam not at a specific point in space but at the location where the target is expected to be based on the post-treatment motion model and latency.

[0126] Item 40: A non-transitory machine-readable medium storing instructions that, when executed by at least one programmable processor, cause the at least one programmable processor to perform operations including obtaining a treatment image of a patient from a magnetic resonance imaging system, the treatment image capturing patient motion, and delivering radiation treatment from a radiation treatment device to the patient according to a radiation treatment plan for adaptive post-treatment motion based at least on a post-treatment motion model, optionally including any one of Items 1 to 39.

[0127] Item 41: The operations further include identifying current patient motion from a treatment image and starting delivery of radiation treatment when the current patient motion matches the expected patient motion included in the post-treatment motion model, the machine-readable medium according to any one of Items 1 to 40.

[0128] Item 42: The start of radiotherapy further takes into account the latency between the current movement of the patient observed using the magnetic resonance imaging system and the delivery of the radiation beam to the patient, the machine-readable medium according to any one of Items 1 to 41.

[0129] Item 43: The radiotherapy to be delivered is targeted not at a specific point in space, but at a location where the target is expected to be based on a prognostic motion model, the latency between the movement of the patient observed using the magnetic resonance imaging system, and the delivery of the radiation beam to the patient, the machine-readable medium according to any one of Items 1 to 42.

[0130] Item 44: The operation further includes identifying the movement of the patient from the acquired treatment images, the machine-readable medium according to any one of Items 1 to 43.

[0131] Item 45: The operation further includes delivering radiotherapy to the patient when the movement of the patient matches the expected movement of the patient, the machine-readable medium according to any one of Items 1 to 44.

[0132] Item 46: The operation further includes interrupting the delivery of radiotherapy when the movement of the patient does not match the expected movement of the patient, the machine-readable medium according to any one of Items 1 to 45.

[0133] Item 47: The operation further includes interrupting the delivery of radiotherapy when the movement of the patient matches the irregular movement of the patient, the machine-readable medium according to any one of Items 1 to 46.

[0134] Item 48: The operation further includes resuming the delivery of radiotherapy according to a radiotherapy plan with prognostic motion adaptation at a stage when the movement of the patient returns to the expected movement of the patient, the machine-readable medium according to any one of Items 1 to 47.

[0135] Item 49: The operation further includes generating a second prognostic operation model using the acquired treatment image, generating a radiation treatment plan for a second prognostic operation adaptation based at least on the second prognostic operation model, and resuming delivery of radiation treatment using the radiation treatment plan for the second prognostic operation adaptation. The machine-readable medium according to any one of Items 1 to 48.

[0136] Item 50: The operation is obtaining actual beam delivery information during radiation treatment, where the actual beam delivery information includes actual beam measurement values including one or more of a measured monitor unit, a measured MLC leaf position, a measured gantry position, a measured couch position, and a measured fluence profile; calculating a dose to tissue during the radiation treatment based on the acquired image and the acquired actual beam delivery information; accumulating the dose to tissue during the radiation treatment; generating a radiation treatment plan for a second prognostic operation adaptation when the accumulated dose to an anatomical structure exceeds a specified limit; and continuing delivery of the radiation treatment using the radiation treatment plan for the second prognostic operation adaptation. The machine-readable medium according to any one of Items 1 to 49.

[0137] Item 51: Generation of a radiation treatment plan for a second prognostic operation adaptation takes into account the accumulated dose and any underdose or overdose. The machine-readable medium according to any one of Items 1 to 50.

[0138] Item 52: Generation of a radiation treatment plan for a second prognostic operation adaptation takes into account a second prognostic operation model using the acquired treatment image. The machine-readable medium according to any one of Items 1 to 51.

[0139] Item 53: A method including the operation according to any one of Items 1 to 52.

[0140] Item 54: A system comprising at least one programmable processor and a non-transitory machine-readable medium storing instructions that, when executed by the at least one programmable processor, cause the at least one programmable processor to perform operations, wherein the operations comprise those described in any one of Items 1 to 52.

[0141] This disclosure assumes that the calculations disclosed in the embodiments herein can be performed in a number of ways by applying the same concepts taught herein, and that such calculations are equivalent to the disclosed embodiments.

[0142] One or more aspects or features of the subject matter described herein can be implemented in digital electronic circuitry, integrated circuitry, specially designed application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), computer hardware, firmware, software, and / or combinations thereof. These various aspects or features can include implementations in one or more computer programs executable and / or interpretable in a programmable system including at least one programmable processor, where the at least one programmable processor can be dedicated or general purpose and is connected to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device. A programmable system or computing system can include a client and a server. Clients and servers are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on respective computers and having a client-server relationship to each other.

[0143] The computer program, which may also be referred to as a program, software, software application, application, component, or code, includes machine instructions for a programmable processor and may be implemented in a high-level procedural language, an object-oriented programming language, a functional programming language, a logical programming language, and / or an assembly / machine language. As used herein, the term "machine-readable medium" (or "computer-readable medium") refers to any computer program product, apparatus, and / or device, such as, for example, magnetic disks, optical disks, memory, and programmable logic devices (PLDs), used to provide machine instructions and / or data to a programmable processor, and includes a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" (or "computer-readable signal") refers to any signal used to provide machine instructions and / or data to a programmable processor. A machine-readable medium may store such machine instructions non-transitorily, such as, for example, in a non-transitory solid-state memory or a magnetic hard drive or any equivalent storage medium. Alternatively or additionally, a machine-readable medium may store such machine instructions in a non-transitory manner, such as, for example, in a processor cache or other random access memory associated with one or more physical processor cores.

[0144] To provide interaction with a user, one or more aspects or features related to the subject matter described herein can be implemented on a computer having a display device, such as a cathode ray tube (CRT), a liquid crystal display (LCD), or a light emitting diode (LED) monitor, for displaying information to the user, as well as a keyboard and a pointing device, such as a mouse or a trackball, by which the user can provide input to the computer. Similarly, other types of devices can be used to provide interaction with a user. For example, the feedback provided to the user can be any form of sensor feedback, such as visual feedback, audio feedback, or tactile feedback, and the input received from the user can be in any form including, but not limited to, acoustic, speech, or tactile input. Other possible input devices include, but are not limited to, touch screens or other touch-sensitive devices, such as single or multi-point resistive or capacitive trackpads, speech recognition hardware and software, optical scanners, optical pointers, digital image capture devices and associated interpretation software, and the like.

[0145] In the above specification and claims, phrases such as "at least one of" or "one or more of" may be followed by an conjunctive list of elements or features. The term "and / or" may also be present in a list of two or more elements or features. In the context in which it is used, such phrases are intended to mean either any one of the recited elements or features individually, or any one of the recited elements or features in combination with any one of the other recited elements or features, unless specifically implicitly or explicitly negated. For example, the phrases "at least one of A and B", "one or more of A and B", and "A and / or B" are each intended to mean "A alone, B alone, or A and B together". A similar interpretation is intended for lists containing three or more items. For example, the phrases "at least one of A, B, and C", "one or more of A, B, and C", and "A, B, and / or C" are each intended to mean "A alone, B alone, C alone, A and B together, A and C together, B and C together, or A, B, and C together". The use of the term "based on" in the above and the claims is intended to mean "at least partially based on" so that features or elements not described are also permitted.

[0146] The subject matter described in this specification can be embodied in a system, an apparatus, a method, a computer program, and / or an article, depending on the desired configuration. Any method or logic flow depicted in the accompanying figures and / or described in this specification need not necessarily be in the particular order or sequential order shown in order to achieve the desired result. The implementations described in the foregoing specification do not represent all implementations consistent with the subject matter described in this specification. Instead, the implementations are merely some examples consistent with aspects related to the described subject matter. Although some variations have been described in detail above, other modifications or additions are possible. In particular, further features and / or variations can be provided in addition to those described in this specification. The above implementations can be directed to various combinations and sub - combinations of the features of this disclosure, and / or combinations and sub - combinations of the further features described above. Further, the above advantages are not intended to limit the application of any of the claims issued from this disclosure to processes and structures that achieve any or all of the advantages.

[0147] Furthermore, section headings shall neither limit nor characterize the inventions described in any claims that may issue from this disclosure. Further, the description of the technology in the "Background Art" should not be construed as an admission that the technology is prior art to any invention in this disclosure. The "Summary of the Invention" should also not be regarded as a characterization of the features of the inventions described in the issued claims. Further, any reference to the present disclosure in general, or the use of the singular form of the word "invention" is not intended to imply any limitation to the scope of the claims described below. Multiple inventions can be described in accordance with the limitations of multiple claims issued from this disclosure, and accordingly, such claims define the inventions protected thereby and their equivalents.

Claims

1. A non - transitory machine - readable medium storing instructions that, when executed by at least one programmable processor, cause the at least one programmable processor to perform operations, the operations including: Acquiring real - time images of a patient from a magnetic resonance imaging system in at least two non - parallel planes; Depicting the contours of the anatomical structures of the patient in the at least two non - parallel planes; Setting spatial limits for the movement of the anatomical structures in the at least two non - parallel planes; Controlling a radiation therapy device to deliver a radiation therapy beam to the patient; Turning off the gate of the radiation therapy beam when the anatomical structure exceeds the spatial limit in any of the at least two non - parallel planes; A machine - readable medium comprising the above.

2. The machine - readable medium according to claim 1, wherein the at least two non - parallel planes are three orthogonal planes.

3. The machine - readable medium according to claim 1, wherein the at least two non - parallel planes are orthogonal.

4. The machine - readable medium according to claim 1, wherein the at least two non - parallel planes are oblique.

5. The acquisition of the real - time images from the magnetic resonance imaging system in at least two non - parallel planes is performed by energizing and controlling one or more subsystems of the gradient coil system of the magnetic resonance imaging system, according to claim 1.

6. The acquisition of the real - time images from the magnetic resonance imaging system in at least two non - parallel planes includes T1 and T2 - weighted volumetric scans at the SRS isocenter, according to claim 1.

7. The depiction of the contours regarding the anatomical structures of the patient is performed via automatic contouring of the machine, according to claim 1.

8. The depiction of the contours regarding the anatomical structures of the patient is performed via a machine that receives manual user input, according to claim 1.

9. The radiation therapy device is further controlled to deliver stereotactic radiosurgery (SRS) to the patient, according to claim 1.

10. The magnetic resonance imaging system operates at a magnetic field strength of less than 1.0 tesla, according to claim 1.

11. The machine-readable medium according to claim 1, wherein the spatial limit is set within 0.5 mm of the boundary of the anatomical structure in the at least two non-parallel planes.

12. A non-transitory machine-readable medium storing instructions that, when executed by at least one programmable processor, cause the at least one programmable processor to perform operations, the operations including: delivering radiation therapy to a patient from a radiation therapy device; acquiring an image of the patient from a magnetic resonance imaging system during the radiation therapy; acquiring actual beam delivery information during the radiation therapy, the actual beam delivery information comprising actual beam measurements including one or more of a measured monitor unit, a measured MLC leaf position, a measured gantry position, a measured couch position, and a measured fluence profile; calculating a dose to tissue during the radiation therapy based on the acquired image and the acquired actual beam delivery information; accumulating the dose to tissue during the radiation therapy; and including a machine-readable medium.

13. The machine-readable medium according to claim 12, wherein the operations further include displaying an accumulated dose to tissue during the radiation therapy.

14. The machine-readable medium according to claim 13, wherein the operations further include displaying the accumulated dose in the image acquired during the radiation therapy.

15. The machine-readable medium according to claim 13, wherein the operations further include displaying the accumulated dose in a most recent set of MRI images.

16. The machine-readable medium according to claim 13, wherein the operations further include displaying the accumulated dose on a plurality of imaging planes.

17. The machine-readable medium according to claim 16, wherein the operations further include displaying the accumulated dose on three orthogonal planes.

18. The operations include: depicting a contour of the anatomical structure of the patient; determining an accumulated dose to the anatomically structured contour; providing a notification or alarm when the accumulated dose to the anatomically structured contour exceeds a specified limit; and stopping delivery of the radiation therapy when the accumulated dose to the anatomically structured contour exceeds the specified limit. Determining a re-optimized treatment plan when the accumulated dose to the anatomical structure depicted by the contour exceeds the specified limit; The machine-readable medium according to claim 12, further comprising.

19. The determination of the re-optimized treatment plan takes into account the accumulated dose and any underdose or overdose, the machine-readable medium according to claim 18.

20. The calculation of the dose to the tissue further utilizes a non-rigid image registration and a latest set of MRI images including the assigned relative electron density, the machine-readable medium according to claim 12.

21. The accumulation of the dose to the tissue further utilizes a non-rigid image registration and a latest set of MRI images including the assigned relative electron density, the machine-readable medium according to claim 12.

22. The dose is accumulated in the latest set of MRI images, the machine-readable medium according to claim 21.

23. The operation further includes using non-rigid image registration between the latest set of MRI images and the images acquired during radiotherapy to display the dose accumulated in the images acquired during radiotherapy, the machine-readable medium according to claim 22.

24. The calculation of the dose to the tissue further utilizes independent measurements regarding the magnetic resonance imaging system and the radiotherapy device to synchronize the acquired images and the acquired actual beam information, the machine-readable medium according to claim 23.

25. A non-transitory machine-readable medium storing instructions that, when executed by at least one programmable processor, cause the at least one programmable processor to perform operations, the operations including: Acquiring pre-treatment images using a magnetic resonance imaging system, the pre-treatment images capturing patient movement; Generating a prognostic motion model based on at least the pre-treatment images; Generating a radiotherapy plan with prognostic motion adaptation based on at least the prognostic motion model; The machine-readable medium comprising.

26. The pre-treatment images comprise cine MRI, the machine-readable medium according to claim 25.

27. The prognostic motion model comprises a model of the expected patient movement during treatment, the machine-readable medium according to claim 25.

28. The machine-readable medium according to claim 25, wherein the generation of the prognostic motion model further includes excluding pre-treatment images that do not represent the expected patient movement during treatment.

29. The machine-readable medium according to claim 25, wherein the generation of the prognostic motion model further includes including a plurality of types of motions observed in the pre-treatment images.

30. The machine-readable medium according to claim 29, wherein the plurality of types of motions include the expected patient movement and the irregular patient movement.

31. The machine-readable medium according to claim 30, wherein the plurality of types of motions include one or more of regular motion due to respiration, motion due to deep breathing, motion due to GI system gas movement, motion due to bladder infusion, motion due to patient movement, motion due to swallowing, chest wall respiration, diaphragmatic respiration, conversation, eye movement, cardiac motion, or voluntary muscle motion.

32. The machine-readable medium according to claim 30, wherein the radiation treatment plan with prognostic motion adaptation takes into account the expected patient movement of the prognostic motion model during treatment.

33. The machine-readable medium according to claim 32, wherein the expected patient movement during treatment includes deformation of the target or the organ of interest.

34. The machine-readable medium according to claim 32, wherein the expected patient movement during treatment includes movement of the target.

35. The machine-readable medium according to claim 34, wherein the radiation treatment plan with prognostic motion adaptation takes into account the movement of the target that is small enough to avoid beam gating but large enough for the dose to be delivered outside the target.

36. The machine-readable medium according to claim 34, wherein the radiation treatment plan with prognostic motion adaptation takes into account the expected patient movement and the irregular patient movement of the prognostic motion model during treatment.

37. The machine-readable medium according to claim 36, wherein the radiation treatment plan with prognostic motion adaptation adjusts or stops delivery when the irregular patient movement of the prognostic motion model is observed during treatment.

38. The machine-readable medium according to claim 34, wherein the radiation treatment plan with prognostic motion adaptation takes into account the latency between the patient movement observed using the magnetic resonance imaging system and the delivery of the radiation beam to the patient.

39. The radiation therapy plan for the prognostic motion adaptation determines the aiming of the radiation beam not at a specific point in space, but towards the location where the target is expected to be based on the prognostic motion model and the latency, the machine-readable medium according to claim 38.

40. A non-transitory machine-readable medium storing instructions that, when executed by at least one programmable processor, cause the at least one programmable processor to perform operations, the operations being obtaining a treatment image of a patient from a magnetic resonance imaging system, the treatment image capturing the movement of the patient, delivering radiation therapy from a radiation therapy device to the patient according to a radiation therapy plan for prognostic motion adaptation based at least on a prognostic motion model, A machine-readable medium comprising.

41. The operations are identifying the current movement of the patient from the treatment image, starting the delivery of radiation therapy when the current movement of the patient matches the expected movement of the patient included in the prognostic motion model, The machine-readable medium according to claim 40, further comprising.

42. The start of the radiation therapy further takes into account the latency between the current movement of the patient observed using the magnetic resonance imaging system and the delivery of the radiation beam to the patient, the machine-readable medium according to claim 41.

43. The delivered radiation therapy is aimed not at a specific point in space, but towards the location where the target is expected to be based on the prognostic motion model, the movement of the patient observed using the magnetic resonance imaging system, and the latency between the delivery of the radiation beam to the patient, the machine-readable medium according to claim 40.

44. The operations further include identifying the movement of the patient from the obtained treatment image, the machine-readable medium according to claim 40.

45. The operations further include delivering the radiation therapy to the patient when the movement of the patient matches the expected movement of the patient, the machine-readable medium according to claim 44.

46. The operations further include interrupting the delivery of the radiation therapy when the movement of the patient does not match the expected movement of the patient, the machine-readable medium according to claim 45.

47. The machine-readable medium of claim 45, wherein the operation further includes interrupting delivery of the radiation therapy when the movement of the patient coincides with an irregular patient movement.

48. The machine-readable medium of claim 47, wherein the operation further includes resuming delivery of the radiation therapy according to a radiation therapy plan adapted to the prognostic operation at a stage when the movement of the patient returns to the expected patient movement.

49. The operation is generating a second prognostic operation model using the acquired treatment image; generating a radiation therapy plan adapted to the second prognostic operation based at least on the second prognostic operation model; resuming delivery of the radiation therapy using the radiation therapy plan adapted to the second prognostic operation; The machine-readable medium of claim 46, further comprising:

50. The operation is acquiring actual beam delivery information during the radiation therapy, the actual beam delivery information comprising actual beam measurement values including one or more of a measured monitor unit, a measured MLC leaf position, a measured gantry position, a measured couch position, and a measured fluence profile; calculating a dose to tissue during the radiation therapy based on the acquired image and the acquired actual beam delivery information; accumulating the dose to tissue during the radiation therapy; generating a radiation therapy plan adapted to the second prognostic operation when the accumulated dose to the anatomical structure exceeds a specified limit; continuing delivery of the radiation therapy using the radiation therapy plan adapted to the second prognostic operation; The machine-readable medium of claim 40, further comprising:

51. The generation of the radiation therapy plan adapted to the second prognostic operation takes into account the accumulated dose and any underdose or overdose. The machine-readable medium of claim 50.

52. The generation of the radiation therapy plan adapted to the second prognostic operation takes into account a second prognostic operation model using the acquired treatment image. The machine-readable medium of claim 50.

Images