This patent application claims the benefit of priority of Lachaine et al.. U.S. Provisional Patent Application Serial umber 62/090,115 titled ^• MAGNETIC RESONANCE PROJECTION IMAGING. ^" filed on December 10, 2014 ( Attorney Docket No. 4186.006PRV), which is hereby incorporated by reference herein in its entirety.

BACKGROUND

Radiation therapy or "radiotherapy ^" may be used to treat cancers or other ailments in mammalian (e.g., human and animal ) tissue. One such radiotherapy technique is referred to as "gamma knife. ^" by which a patient is irradiated using a number of lower-intensity gamma rays that converge with higher intensity and high precision at a targeted region (e.g.. a tumor). In another example, radiotherapy is provided using a linear accelerator ("linac ^" ). whereby a targeted region is irradiated by high -energy particles (e.g.. electrons, protons, ions, high- energy photons, and the like). The placement and dose of the radiation beam is accurately controlled to provide a prescribed dose of radiation to the targeted region. The radiation beam is also generally controlled to reduce or minimize damage to surrounding healthy tissue, such as may be referred to as "organ(s) at risk ^" (OARs). Radiation may be referred to as "prescribed ^" because generally a physician orders a predefined dose of radiation to be delivered to a targeted region such as a tumor.

Generally, ionizing radiation in the form of a colli mated beam is directed from an external radiation source toward a patient. Modulation of a radiation beam may be provided by one or more attenuators or collimators (e.g.. a multi- leaf collimator). The intensity and shape of the radiation beam may be adj usted by collimation avoid damaging healthy tissue (e.g.. OARs) adjacent to the targeted tissue by conforming the projected beam to a profile of the targeted tissue. The treatment planning procedure may include using a three-dimensional image of the patient to identify the target region (e.g.. the tumor) and such as to identify critical organs near the tumor. Creation of a treatment plan may be a time consuming process where a planner tries to comply with various treatment objectives or constraints (e.g.. dose volume histogram (DVH) objectives or other constraints), such as taking into account importance (e.g.. weighting) of respective constraints in order to produce a treatment plan that is clinically acceptable. This task may be a time-consuming trial-and-error process that is complicated by the various organs at risk (OARs) because as the number of OARs increases (e.g.. about thirteen for a head-and-neck treatment), so does the complexity of the process. OARs distant from a tumor may be more easily spared from radiation, but OARs close to or overlapping a target tumor may be more di fficult to spare from radiation exposure during treatment.

Generally, for each patient, an initial treatment plan may be generated in an "offline ^" manner. The treatment plan may be developed well before radiation therapy is delivered, such as using one or more medical imaging techniques. Imaging information may include, for example, images from X-rays. Computed Tomography (CT). nuclear magnetic resonance (MR), positron emission tomography (PET), single-photon emission computed tomography (SPECT). or ultrasound. A health care provider, such as a physician, may use three- dimensional imaging information indicative of the patient anatomy to identify one or more target tumors along with the organs at risk near the tumor. The health care provider may delineate the target tumor that is to receive a prescribed radiation dose using a manual technique, and the health care provider may similarly delineate nearby tissue, such as organs, at risk of damage from the radiation treatment.

Alternatively or additionally, an automated tool (e.g.. ABAS provided by Elekta AB, Sweden) may be used to assist in identi fying or delineating the target tumor and organs at risk. A radiation therapy treatment plan ( ^" treatment plan ^" ) may then be created using an optimization technique based on clinical and dosimetric objectives and constraints (e.g.. the maximum, minimum, and mean doses of radiation to the tumor and critical organs). The treatment planning procedure may incl ude using a three-di mensional image of the patient to identi fy the target region (e.g.. the tumor) and to identi fy critical organs near the tumor. Creation of a treatment plan may be a time consuming process where a planner tries to comply with various treatment objectives or constraints ( e.g.. dose volume histogram ( DVH ) objectives), taking into account their individual importance ( e.g.. w eighting) in order to produce a treatment plan that is clinically acceptable. This task may be a time-consuming trial-and-error process that is complicated by the various organs at risk ( OARs ) because as the number of OARs increases ( e.g.. up to thirteen for a head-and- neck treatment), so does the complexity of the process. OARs distant from a tumor may be easily spared from radiation, while OARs close to or overlapping a target tumor may be di fficult to spare.

The treatment plan may then be later executed by positioning the patient and delivering the prescribed radiation therapy. The radiation therapy treatment plan may include dose "fractioning. ^" whereby a sequence of radiation therapy deliveries are provided over a predetermined period of time ( e.g.. 45 fractions or some other total count of fractions), such as with each therapy delivery including a speci fied fraction of a total prescribed dose. During treatment, the position of the patient or the position of the target region in relation to the treatment beam is important because such positioning in part determines whether the target region or healthy tissue is i rradiated.

OVERV I EW

In one approach, nuclear magnetic resonance (MR) imaging may be combined with a radiation therapy system such as to provide imaging information to adapt or guide radiation therapy. An example of such a combined system may be referred to generally as "MRI-linac," comprising an MR i maging system, along with linear accelerator as a source of energy for radiation therapy. In an illustrative example, i mage acquisition may be performed j ust before initiation of delivery of a speci fied radiation therapy fraction. Such imaging may provide information helpful for identi fying a position of a target region or for identi fying motion of the target region. Such contemporaneous i maging may be referred to generically as "real-time. ^" but in general a latency or time delay exists between an acquisition of an image and a delivery of radiation therapy.

The present inventors have recognized, among other things, that a problem exists in using 3 D MR i maging to plan or adapt radiation therapy. For example, i mage reconstruction of an imaged volumetric region may be adversely alTected when the target region is influenced by respiration or other motion, because the imaging duration ( ^" imaging time ^" ) is generally long enough to be affected by such motion. In addition, an acquisition latency or a long acquisition duration may be problematic because the target region may have deformed or moved signi ficantly betw een a 3D R image acquisition and a later radiation therapy delivery.

In one approach, such as w hen target region motion is periodic, a four- dimensional M R imaging technique may be used such as prior to radiation treatment. For example, i mage acquisition may be synchronized to a physiologic cycle, such as by sensing surrogate information. Examples of surrogates incl ude a signal sensed using a respiration belt or a one-dimensional ( 1 D) navigator echo indicated by M R imaging. M R imaging elements, such as acquired imaging slices, may be sorted into bins using information indicati ve of a phase or amplitude of the physiologic cycle or a surrogate correlated with such a cycle. How ever, such an approach may also have l imitations. For example, generally available slice-based 4D imaging techniques ( such as non-projection M R imaging) do not include use of an anatomical landmark such as a diaphragm location to sort or bin acquired 3D i mages with respect to a physiologic cycle. Instead, generally available 4D i maging techniques acquire images sequentially and the acquired images contain di fferent portions of the anatomy and lack common anatomical features across each image. By contrast, a projection imaging approach can include selecting or generating projection images having a common anatomical feature in each image so the common feature can be used to facilitate binning. Even i f a di fferent perspective of the feature is present in each of the proj ection images ( e.g.. di fferent view s of the feature), such feature tracking for binning can still be used in a projection imaging approach. In this manner, unl ike generally-available 4D MR i maging techniques, a surrogate (such as an external surrogate) is not required.

Generally-used 4D MR imaging protocols also include relatively long acquisition times and may be time-prohibitive, such as in applications where updated imaging is to be performed prior to each radiation therapy treatment fraction. Also. 4D MR i maging techniques may not necessarily represent or predict an anatomical state of an imaging subject during a subsequent delivery of radiation therapy. For example, baseline dri fts, deformations, or changes in frequency or phase of the physiologic cycle may occur between the time at which the 4D MR i maging information is acqui red, and a later delivery of radiation therapy.

In another approach, imaging information indicati ve of intrafractional motion of the target region or other portions of the imaging subject may include imaging j ust a portion of the imaging subject, w ithout requiring full volumetric imaging, such as by acquiring tw o-dimensional ( 2D) imaging slices, such as through the target region along di fferent directions ( such as incl uding acquisition of a sequence of orthogonal slices). Such slices may be used to help localize the target region or other anatomy, generally, for delivery of radiation therapy. Such localization may be assisted in part using one or more of image segmentation or image registration techniques. However, such an approach may also have limitations. For example. M R imaging pulse sequences used to obtain 2D slices may be di fferent than those used to obtain pre-treatment interfractional volumetric 3D or 4D ' ^" reference ^"" imaging. Such di fferent pulse sequences may make registration between 2D slices and an earlier-acquired volumetric reference image challenging. Another li mitation is that out-of-slice information is lost, such as in an example w here multiple organs-at-risk ( OARs ) are present or i f retrospective dose calculations are to be made by acquiring imaging information during treatment. Yet another limitation of using 2D imaging slices is that it may be di fficult to align slices w ith target motion, particularly i f the motion varies betw een physiologic cycles such as between respiration cycles. Small targets such as tumors may be deformed or may disappear entirely from a particular acquired imaging slice. The present inventors have recognized a solution to the limitations mentioned above. Such a solution may include using an M R projection imaging approach. Such a projection imaging approach may be used intrafractionally. Altematively, or additionally. M R projection imaging may be used in a similar manner for simul ation imaging to be used for treatment planning, or p re- treatment (e.g.. "reference ^" ) imaging performed interfractionally to shi ft the patient or adapt the treatment plan prior to treatment deli very. Use of M R proj ection imaging for simulation imaging, pre-treatment reference i maging, and later intrafractional imaging may provide consistency and ease of registration or other processing. MR projection imaging may also provide imaging information in manner that more closely correlates with beam-eye-view ( BEV ) portal imaging or X-ray techniques, but without exposing the i magi ng subject to ionizing radiation during imaging. Obtaining 2D MR projection i mages may dramatically decrease imaging acquisition latency as compared to other approaches such as full 3D volumetric M R imaging, and 2D projection images may be aggregated such as to provide vol umetric imaging information using tomographic or Fourier domain (k-space) techniques, for example. Information from acquired 2D M R projection i mages or from 3 D or 4D i maging constructed from 2D M R projection images may be compared to reference imaging information, such as to localize a target region or anatomical landmarks, or to predict a later target region location. In this manner, information indicative of the target region may be used to adapt radiation therapy.

According to vari ous examples, apparatus and techniques described herein may be used to control radiation therapy delivery to a subj ect using proj ection imaging techniques. For example, reference imaging may be received, such as including imaging information obtained earlier in relation to radiation therapy treatment planning. A tw o-dimensional (2D) proj ection image may be generated using i maging information obtained via nuclear magnetic resonance (MR) imaging, the 2D projection image corresponding to a speci fied projection direction, the speci fied projection direction including a path traversing at least a portion of an imaging subject. A change between the generated 2D projection image and the reference imaging information may be determined. Delivery of the radiation therapy may be controlled at least in part (e.g., in an adapti ve manner) using the determined change between the obtained 2D projection image and the reference imaging information.

The present inventors have also recognized that reference imaging information may be obtai ned using projection imaging techniques, such as for use in spatially-registering later-obtained projection images w ith earlier-acquired imaging inf ormation. According to various examples, apparatus and techniques described herein may be used to generate four-dimensional (4D) or other imaging information, such as during one or more of obtaining reference images before radiation therapy ( e.g.. reference imaging), or later such as j ust before or during delivery of radiation therapy ( e.g.. intrafractional imaging). Generating the 4D imaging information may incl ude generating two or more two- dimensional ( 2D) images, the 2D images comprising projection images representative of di fferent proj ection angles, w here the 2D images are generated using imaging information obtained via nuclear magnetic resonance (MR) imaging. Particular 2D images may be assigned to bins at least in part using information indicative of temporal positions w ithin the physiologic cycle corresponding to the particular 2D images. Three-dimensional ( 3D ) images may be constructed using the binned 2D images. A group of 3D images may be aggregated such as to provide 4D i maging in formation.

This overview is intended to provide an overview of subject matter of the present patent application. It is not intended to provide an exclusive or exhaustive explanation of the invention. The detailed descripti on is included to provide further information about the present patent application.

BRI EF DESCRIPTION OF TH E DRAWINGS FIG. 1 A illustrates generally an example of a radiation therapy system that may include radiation therapy output configured to provide a therapy beam.

FIG. I B i llustrates generally a partially cut-aw ay view of an example of a system that including a combined radiation therapy system and a nuclear magnetic resonance (MR) imaging system. FIG. 2 illustrates generally an example of a collimator configuration, such as may be used in part to shape or colli mate a radiation therapy beam.

FIG. 3 illustrates generally an example of radiation therapy system, such as may include a radiation therapy device and an imaging acquisition device.

FIG. 4 illustrates generally an example of a system that may be used for one or more of imaging acquisition, i mage segmentation, target prediction, therapy control, or therapy adj ustment.

FIG. 5 illustrates generally an example of a system, such as may include a radiation therapy controller having an imaging input, a radiation therapy generator, and a radiation therapy output.

FIG. 6 illustrates generally a technique, such as a method, that may incl ude using M R imaging to excite a region of an i maging subject, the region defining an imaging slice, and obtaining a pixel value corresponding to a one- dimensional projection line through the slice.

FIG. 7 A illustrates generally a technique, such as a method, that may include exciting a region of an imaging subj ect using a two-dimensional ( 2D) M R imaging excitation sequence

FIG. 7B illustrates generally another technique, such as a method, that may include exciting a region of an imaging subject using a tw o-dimensional (2D) M R imaging excitation sequence.

FIG. 8A and 8B illustrate generally a technique, such as a method, that may include generating tw o-dimensional (2D) M R projection images representative of di fferent projection angles, and using such 2D projection images to construct three-di mensional (3D) images.

FIG. 9 illustrates generally a technique, such as a method, that may include generating a tw o-dimensional (2D) projection image using M R imaging and determining a change betw een the generated 2D projection image and reference imaging information.

FIG. 10A illustrates generally a spatial arrangement of a radiation therapy beam orientation with respect to one or more projection directions, such as may include two projection di rections oriented orthogonally to each other. FIG. 10B illustrates generally a spatial arrangement of MR imaging projection directions, such as corresponding to projection angles spanning an arc or circle about a specified region such as a radiation therapy treatment isocenter.

FIG. IOC illustrates generally a spatial arrangement of an R imaging projection direction, such as oriented at a specified angle with respect to a radiation therapy beam direction

FIG. 10D illustrates generally a spatial arrangement of MR imaging projection directions, such as may be specified to provide MR projection images in a manner similar to stereoscopic X-ray imaging.

In the draw ings, which are not necessarily drawn to scale, like numerals may describe similar components in different views. Like numerals having di fferent letter suffixes may represent di fferent instances of similar components. The drawings illustrate generally, by way of example, but not by way of limitation, various embodiments discussed in the present document.

DETAILED DESCRIPTION FIG. 1 A illustrates generally an example of a radiation therapy system 102 that may include radiation therapy output 104 configured to provide a therapy beam 108. The radiation therapy output 104 may include one or more attenuators or collimators, such as a multi-leaf collimator (MLC) as described in the illustrative example of FIG. 2. Referring back to FIG. I A. a patient may be positioned in a region 112, such as on a platform 116 (e.g.. a table or a couch), to recei e a prescribed radiation therapy dose according to a radiation therapy treatment plan.

The radiation therapy output 104 may be located on a gantry 106 or other mechanical support, such as to rotate the therapy output 104 around an axis ("A"). One or more of the platform 116 or the radiation therapy output 104 may be moveable to other locations, such as moveable in transverse direction ("7") or a lateral direction (" "). Other degrees of freedom are possible, such as rotation about one or more other axes, such as rotation about a transverse axis ( indicated as "R"). The coordinate system (including axes A, T, and L) shown in FIG. 1 A may have an origin located at an isocenter 110. The isocenter may be defined as a location where the radiation therapy beam 108 intersects the origin of a coordinate axis, such as to deliver a prescribed radiation dose to a location on or w ithin a patient. For example, the isocenter 1 10 may be defined as a location where the radiation therapy beam 108 intersects the patient for various rotational positions of the radiation therapy output 104 as positioned by the gantry 106 around the axis A.

In an example, a detector 1 14 may be located w ithin a field of the therapy beam 108. such as may include a fiat panel detector ( e.g.. a direct detector or a scintillation-based detector). The detector 1 14 may be mounted on the gantry 106 opposite the radiation therapy output 104. such as to maintain alignment with the therapy beam 108 as the gantry 106 rotates. In this manner, the detector 1 14 may be used to monitor the therapy beam 108 or the detector may be used 1 14 for imaging, such as portal imaging of a projection of the beam 108 through the region 1 12. The regi on 1 12 may define a pl ane and a projection of the therapy beam 108 in the region 1 12 may be referred to as a " ^" Beam Eye View" of the region 1 12.

In an illustrative example, one or more of the platform 1 16. the therapy output 104. or the gantry 106 may be automatically positioned, and the therapy output 104 may establish the therapy beam 108 according to a speci fied dose for a particular therapy delivery instance. A sequence of therapy del iveries may be speci fied according to a radiation therapy treatment plan, such as using one or more di fferent orientations or locations of the gantry 106. platform 1 16. or therapy output 104. The therapy deliveries may occur sequentially, but may intersect in a desired target region on or w ithin the patient, such as at the isocenter 1 10. A prescribed cumulative dose of radiation therapy may thereby be delivered to the target region w hile damage to tissue nearby the target region, such as one or more organs-at-risk. is reduced or avoided.

As mentioned in relation to other examples herein, the radiation therapy system 102 may incl ude or may be coupled to an imaging acquisition system, such as to provide one or more of nuclear magnetic resonance ( M R) imaging, or X-ray imaging, such as may incl ude computed tomography ( CT) imaging. In an example. MR imaging information or other imaging information may be used to generate imaging information or visualizations equivalent to CT imaging, without requiring actual CT imaging. Such imaging may be referred to as "pseudo-CT ^" i maging.

FIG. I B illustrates generally a partially cut-away view of an example of a system that including a combined radiation therapy system 102 and a nuclear magnetic resonance ( M R) imaging system 130. The M R imaging system 130 may be arranged to define a "bore ^" around an axis ("A"), and the radiation therapy system may include a radiation therapy output 104. such as to provide a radiation therapy beam 108 directed to an isocenter 1 10 within the bore along the axis. A. The radiation therapy output 104 may include a coll imator 124. such as to one or more of control or shape the radiation therapy beam 108 to direct the beam 108 to a target region w ithin a patient. The patient may be supported by a platform, such as a platform positionable along one or more of an axial direction. A, a lateral direction. L, or a transverse di rection. T. One or more portions of the radiation therapy system 102 may be mounted on a gantry 106. such as to rotate the radiation therapy output 104 about the axis A.

FIG. 1 A and FIG. 1 B ill ustrate generally examples including a configuration w here a therapy output may be rotated around a central axis ( e.g.. an axis "A"). Other radiation therapy output configurations may be used. For example, a radiation therapy output may be mounted a robotic arm or manipulator, such as having multiple degrees of freedom. In yet another example, the therapy output may be fixed, such as located in a region laterally separated from the patient, and a platform supporting the patient may be used to align a radiation therapy isocenter with a speci fied target region within the patient.

FIG. 2 illustrates generally an example of a multi-leaf collimator (MLC) configuration 132. such as may be used in part to shape or coll i mate a radiation therapy beam. In FIG. 2. leaves 132 A through 132 J may be automatically positioned to define an aperture approxi mating a tumor 140 cross section or projection. The leaves 132A through 132 J may be made of a material speci fied to attenuate or block the radiation beam in regions other than the aperture, in accordance with the radiation treatment plan. For example, the leaves 132 A through 132 J may include metallic plates, such as comprising tungsten, with a long axis of the plates oriented parallel to a beam direction, and having ends oriented orthogonall to the beam direction (as shown in the plane of the illustration of FIG. 2).

A "state" of the MIX 132 may be adjusted adaptively during a course of radiation therapy, such as to establish a therapy beam that better approximates a shape or location of the tumor 140 or other targeted region, as compared to using a static collimator configuration or as compared to using an MLC 132

configuration determined exclusively using an "offline ^" therapy planning technique. A radiation therapy technique including using the MLC 132 to produce a specified radiation dose distribution to a tumor or to speci fic areas within a tumor may be referred to as Intensity Modulated Radiation Therapy ( IMRT). As described in relation to other examples herein, imaging may be performed to locali e the target region or to determine or predict a perspective of a target region from the point-of-view of the radiation therapy beam to adaptively guide therapy.

FIG. 3 illustrates generally an example of radiation therapy system 300. such as may include a radiation therapy device 330 and an imaging acquisition device. Radiation therapy system 300 may include a training module 312. a prediction module 314. a training database 322. a testing database 324. a radiation therapy device 330. and an image acquisition device 350. Radiation therapy system 300 may also be connected to a treatment planning system (TPS) 342 and an oncology information system (OIS ) 344. which may provide patient information. In addition, radiation therapy system 300 may include a display device and a user interface.

FIG. 4 illustrates generally an example of a system 400 that may be used for one or more of imaging acquisition, image segmentation, target prediction. therapy control, or therapy adj ustment. According to some embodiments.

system 400 may be one or more high-performance computing devices capable of identifying, analyzing, maintaining, generating, or providing large amounts of data consistent with the disclosed embodiments. System 400 may be standalone, or it may be part of a subsystem, which in turn may be part of a larger system. For example, system 400 may represent distributed high-performance servers that are remotely located and communicate over a network, such as the Internet. or a dedicated netw ork, such as a local area netw ork (LAN) or a wide-area netw ork (WAN). In some embodiments, system 400 may include an embedded system, imaging scanner (e.g.. a nuclear magnetic resonance (MR) scanner or other scanner such as a computed tomography (CT) scanner), and. or touchscreen display device in communication with one or more remotely located high- performance computing devices.

In one embodiment, system 400 may include one or more processors

414. one or more memories 410. and one or more communication interfaces

415. Processor 414 may be a processor circuit, including one or more general - purpose processing devices such as a microprocessor, central processing unit (CPU), graphics processing unit (GPU), or the like. More particularly, processor 414 may be a complex instruction set computing (CISC) microprocessor, reduced instruction set computing ( RISC) microprocessor, very long instruction Word (VLIW) microprocessor, a processor implementing other instruction sets, or processors implementing a combination of instruction sets.

Processor 414 may also be one or more special-purpose processing devices such as an application specific integrated circuit ( ASIC), a field programmable gate array ( FPGA). a digital signal processor ( DSP), a System- on-a-Chip ( SoC ). or the like. As would be appreciated by those skilled in the art. in some embodiments, processor 414 may be a special-purpose processor, rather than a general-purpose processor. Processor 414 may include one or more know n processing devices, such as a microprocessor from the Pentium™ or Xeon™ family manufactured by Intel™, the Turion™ family manufactured by AMD™, or any of various processors manufactured by other vendors such as Oracle™ (e.g.. a SPARC ^{I M} -architecture processor). Processor 414 may also include graphical processing units manufactured by Nvidia™. The disclosed embodiments are not limited to any type of processors ) otherw ise configured to meet the computing demands of identifying, analyzing, maintaining, generating. and or providing large amounts of imaging data or any other type of data consistent with the disclosed embodiments.

Memory 410 may include one or more storage devices configured to store computer-executable instructions used by processor 414 to perform functions related to the disclosed embodiments. For example, memory 410 may store computer executable softw are instructions for treatment planning softw are 411, operating system softw are 412, and training prediction softw are 413.

Processor 414 may be communicatively coupled to the memory storage device 410, and the processor 414 may be configured to execute the computer executable instructions stored thereon to perform one or more operations consistent with the disclosed embodiments. For example, processor 414 may execute training prediction softw are 413 to implement functionalities of training module 312 and prediction module 314. In addition, processor device 414 may execute treatment planning softw are 41 1 (e.g.. such as Monaco® provided by Elekta) that may interface with training prediction softw are 413.

The disclosed embodiments are not limited to separate programs or computers configured to perform dedicated tasks. For example, memory 410 may include a single program that performs the functions of the system 400 or multiple programs (e.g., treatment planning softw are 41 1 and or

training prediction softw are 413). Additionally, processor 414 may execute one or more programs located remotely from system 400. such as programs stored in database 420. such remote programs may include oncology information system softw are or treatment planning softw are. Memory 410 may also store image data or any other type of data information in any format that the system may use to perform operations consistent with the disclosed embodiments.

Communication interface 415 may be one or more devices configured to allow data to be received and or transmitted by system 400. Communication interface 415 may include one or more digital and or analog communication devices that allow system 400 to communicate with other machines and devices, such as remotely located components of system 400. database 420, or hospital database 430. For example. Processor 414 may be communicatively connected to database* s) 420 or hospital database* s) 430 through communication interface 415. For example. Communication interface 415 may be a computer network, such as the Internet, or a dedicated network, such as a LAN or a WAN.

Alternatively, the communication interface 415 may be a satellite

communications link or any form of digital or analog communications link that allows processor 414 to send receive data to/from either database! s) 420. 430.

Database* s) 420 and hospital database* s) 430 may include one or more memory devices that store information and are accessed and managed through system 400. By way of example, database* s) 420. hospital database* s) 530. or both may include relational databases such as Oracle™ databases. Sybase™ databases, or others and may include non-relational databases, such as Hadoop sequence files, HBase. Cassandra or others. The databases or other files may include, for example, one or more of raw data from MR scans or CT scans associated with an imaging subject, such as for training or providing a reference image. MR feature vectors. MR projection imaging information. CT values. reduced-dimension feature vectors. pseudo-CT prediction model (s). pseudo-CT value* s). pseudo-CT image. DICOM data. etc. Systems and methods of disclosed embodiments, how ever, are not limited to separate databases. In one aspect, system 400 may include database* s) 420 or hospital database* s) 430. Alternatively, database(s) 420 and or hospital database* s) 430 may be located remotely from the system 400. Database* s) 420 and hospital database* s) 430 may include computing components (e.g.. database management system, database server, etc. ) configured to receive and process requests for data stored in memory devices of database* s) 420 or hospital database* s) 430 and to provide data from database* s) 420 or hospital database* s) 430.

System 400 may communicate with other devices and components of system 400 over a network (not shown). The network may be any type of network ( including infrastructure) that provides communications, exchanges information, or facilitates the exchange of information and enables the sending and receiving of information between other devices and or components of system 400 over a network (not shown). In other embodiments, one or more components of system 400 may communicate directly through a dedicated communication link(s). such as a link (e.g.. hardw ired link, wireless link, or satellite link, or other communication link) betw een system 400 and database* s) 420 and hospital database* s ) 430.

The configuration and boundaries of the functional bui lding blocks of system 400 has been defined herein for the convenience of the description. Alternati ve boundaries may be defined so long as the speci fied functions and relationships thereof are appropriately performed. Alternatives ( including equivalents, extensions, variations, deviations, etc.. of those described herein ) will be apparent to persons skilled in the relevant art(s ) based on the teachings contained herein.

FIG. 5 illustrates generally an example of a system 500, such as may include a radiation therapy controller system 554 having an imaging input 560. a radiation therapy generator 556. and a radiation therapy output 504. The therapy generator 556 may include an accelerator, such as a l inear accelerator, or another source of radiation, and the therapy output 504 may be coupled to the therapy generator 556 to process a beam of energetic photons or particles provided by the therapy generator 556. For example, the therapy output 504 may incl ude or may be coupled to an output actuator 566 to one or more of rotate or translate the therapy output 504 to provide a radiation therapy beam directed to a desired target region. The therapy output 504 may include a col limator 564. such as a multi-leaf coll i mator as mentioned above in relati on to FIG. 2. Referring back to FIG. 5, the therapy controller system 554 may be configured to control one or more of the therapy generator 556. the therapy output 504. or a patient position actuator 516 (such as a movable platform including a couch or table), using an adaptive radiation treatment technique as described in other examples herein.

The therapy controller system 554 may be coupled to one or more sensors, such as using a sensor input 562. For example, a patient sensor 558 may provide physiologic information to the therapy controller system, such as information indicative of one or more of respiration ( e.g.. using a

plethysmographic sensor or respiration belt), patient cardiac mechanical or electrical activity, peripheral circulatory activity, patient position, or patient motion. S uch in formation may provide a "surrogate ^"" correlated with motion of one or more organs or other regions to be targeted by the therapy output 504. Such information may be used to control therapy such as for therapy gating or to assist in "binning ^" acquired imaging information according to one or more of a determined phase or amplitude range of a physiologic cycle as indicated by obtained information from the sensor 558.

The imaging input 560 may be coupled to an imaging system 550 (such as may include a computed tomography imaging system or a nuclear magnetic resonance (MR) imaging system, as illustrative examples). Alternatively, or in addition, the therapy controller system 554 may receive imaging information from an imaging data store 552. such as a centralized imaging database or imaging server. One or more of the therapy controller system 554 or the imaging system 550 may include elements shown and described in relation to the system 400 shown in FIG. 4.

Generally-available radiation therapy equipment can be used to acquire projection images using X-ray imaging techniques. For example, linear accelerator (linac) systems can acquire X-ray projection images using one or more of the megavoltage (MV) treatment beam itself combined with a portal imaging device (such as shown illustratively in FIG. I A) or using one or more separate kilovolt (kV) X-ray sources. In an example, a kV X-ray source can be mounted on a gantry such as oriented at a ( ⁽ -degree angle with respect to the treatment beam orientation. In another example, two independent X-ray source imager pairs can be located to provide stereoscopic X-ray imaging.

Projection images acquired using X-ray imaging represent a divergent X-ray path from the imaging source, which can be referred to as a "point source ^" or focal point.

Prior to delivery of radiation therapy, such as prior to a particular radiation therapy treatment fraction. X-ray computed tomography (CT) images may be acquired. For example, a cone-beam CT (CBCT) imaging technique can be used to obtain projection images at various projection angles during a rotation of a gantry-mounted X-ray source around an imaging subject. A three- dimensional (3D) image can be reconstructed from such cone beam projections. For imaging subjects that exhibit significant motion, such as respiratory motion. 3D CBCT images may be blurred because each projection may capture a snapshot of the patient at a di fferent point in the respiration cycle or other physiologic cycle. To reduce motion blurring, four-dimensional (4D) CBCT imaging may be used, such as by binning proj ections according to a phase or amplitude of the physiologic cycle corresponding to the time at which the projection image was acquired.

The present inventors have recognized, among other things, that nuclear magnetic resonance (MR) imaging projections can be similarly acquired and processed, such as reducing exposure of the imaging subject to ionizing radiation and providing enhanced soft tissue contrast as compared to X-ray-based i maging approaches. The present inventors have also recognized, among other things, that MR proj ection images are not degraded by scattered radiation, and the projection imaging di rection is not l imited by physical constraints such as having to be oriented at 90 degrees relative to the treatment beam. M R projection imaging can be used to acquire a single imaging perspective ( e.g.. a 2D projection image) of all of the information contained within a depth extent of an excited imaging region, as opposed to using relatively thin 2D M R i maging slices which capture only a portion of the information in a depth direction. MR proj ection imaging does have l imitations, such as that information in an acquired 2D projection image is not localized in the depth direction orthogonal to the proj ection imaging plane, and structures surroundi ng a targeted region can appear to partially mask it.

M R Projection Imaging such as using 1 D Projection Lines FIG. 6 illustrates generally a technique 600. such as a method, that may include at 602 using M R imaging to excite a region of an imaging subject. For example, a two-dimensional (2 D) excitation sequence may be used. At 604, a readout gradient may be appl ied to the i maging subject, and a one-dimensional ( 1 D) projection line (e.g.. a "ray") through the 2D excited region may be acquired. At 606. another projection line axis may be selected, and a 2D region is again excited at 602, and a readout gradient is applied at 604 corresponding to the updated projection line axis. Referring to the inset diagrams at 606. the proj ection lines may be established in a divergent manner or a parallel manner. For example, i f di vergent projection line orientations are used, a resultin projection image defined in the plane 670 A may provide a projection representation that is similar to projection images produced by divergent X-ray imaging techniques, or similar to a projection image produced by a radiation therapy output beam during portal i maging. In this manner. M R projection imaging can be used to simulate X-ray imaging, but with enhanced contrast and without undesired scattering, for example.

In the example of divergent MR projection i maging using 1 D projection lines, the I D projection lines can be speci fied to converge at a location 650, such as corresponding to a location of a radiation therapy beam source or

corresponding to a location where an X-ray imaging source would generally reside. A scale and spatial resol ution of information defined in a resulting 2D projection image established in the plane 670 A may be determined by the distance betw een the source location 650 and a selected imaging plane 670 A location. For example, a first projection line orientation 660 A can be orthogonal to the projection i maging plane 670A, and a corresponding acquired I D projection line can be used to establish a pixel value at a location 680 A. All information acquired along the first proj ection line is generally incorporated and compressed into the pixel value, thus depth selecti vity is lost in the direction along the projection line.

The line orientation 660 A orthogonal to the projection imaging plane 670A may generally be referred to as the projection "direction ^" or "angle ^" even though in divergent examples, other projection l ine orientations are not parallel. A second projection line orientation 662A can si milarly establish a second pixel value at a location 682A, and a third projection line orientation 664A can similarly establish a third pixel value at a location 684 A. In a reconstructed image, the pixel locations 680 A. 682B. and 684B are determined at least in part by a speci fied separation between the plane 670 A location and the source location 650. To achieve higher spatial resolution in a lateral direction in plane 670A, a greater number of separate 1 D proj ection line directions can be acqui red at the cost of total 2D projection image acquisition duration, because particular acquired I D projections are aggregated to create a full 2D projection in the plane 670A.

In a parallel I D projection line example, such as at 606. a first projection line orientation 660 B can be established to provide a first pixel value at a location 680B in a resulting projection image defined in the plane 670B. Other parallel lines 662B and 664 B can be used to provide information for

corresponding location 682B and 684B in the projection image. As a practical consideration, if exclusively parallel 1 D projection lines are being used to construct a particular 2D projection image, the techniques of FIG. 7 A or FIG. 7B may provide enhanced efficiency as compared to the example of FIG. 6 because a 2D projection image can be reconstructed directly by suppressing a slice selection gradient or by using a large slice thickness (relative to a depth extent of interest) without requiring excitation and readout of particular I D projection lines.

Parallel projection can provide one or more of a simplified geometry compared to a divergent approach, one-to-one correspondence between projection image pixels and a plane defining a "beam eye view," or easier tomographic reconstruction from multiple projections. By comparison. X-ray - based CBCT tomographic reconstruction is generally only approximate due to divergence of the acquired cone-beam projection images. In either the divergent or parallel 1 D projection line examples, the spacing or orientation of the projection lines need not be uniform and may be specified depending on a variety of factors. For example, a spatial resolution in a direction of predicted motion of a target w ithin the field of view of the projection image and parallel to the projection image plane may be enhanced by increasing a spatial frequency of projection lines in the direction of predicted motion. Similarly, shorter total acquisitions can be provided by using a more sparse set of divergent projection lines.

MR Projection Imaging such as using 2D Excitation without requiring Slice

Selection Gradient or using Large Slice Depth Encompassin a Depth of Interest FIG. 7 A illustrates generally a technique 700 A. such as a method, that may include exciting a region of an imaging subject using a tw o-dimensional (2D) MR imaging excitation sequence at 702. F IG. 7B illustrates generally another technique 700 B, such as a method, that may also include exciting a region of an i maging subject usi ng a two-dimensional ( 2D) MR imagi ng excitation sequence at 702.

In the example of FIG. 7 A. a 2D R projection image can be obtained at 704A by using a 2D imaging sequence w ithout requiring use of a slice selection gradient ( e.g.. the slice selection gradient pulse sequence can be suppressed or omitted) so that information in a depth direction ( e.g.. in the projection imaging direction and perpendicular to a plane of a resulting projection image) is acquired at all depths within the excited region. S uch an approach does not require excitation and gradient readout of 1 D proj ection lines and can therefore reduce image acquisition duration as compared to a 1 D projection approach in the case where parallel I D proj ection lines are desired.

In the example of FIG. 7B. a 2D M R proj ection i mage can be obtained at 704 B by using a 2D imaging sequence using a slice selection gradient defining a slice sufficiently large in depth to encompass a region of depth of interest, such as corresponding to a portion or an entirety of a radiation therapy target extent in a dimension parallel to the projection angle. As the slice thickness is increased, the depth dimension of the slice includes more and more anatomical contribution of information that was previously out-of-field in depth. Such depth information is compressed into a single point or pixel location in the resulting 2D proj ection image. The technique 700 B of FIG. 7B similarly offers a reduced image acquisition duration as compared to the 1 D proj ection approach and can be referred to as a "very thick slice ^" projection approach.

MR imaging data can also be obtained in "k-space. ^" representing a coordinate space corresponding to the spatial Fourier transform of the imaging information. For example. MR imaging data can be naturally collected in k- space by varying image gradients; a particular combination of x, y and z gradients generally corresponds to a single point in k-space. By sequentially filling the points in k-space. an inverse Fourier Transform can then be appl ied to the k-space representation to generate an image. A 2D plane in k-space corresponds to a 2D projection in image space. Accordingly, a 2D projection can also be obtained by acquiring k-space points that lie in a plane in k-space, and generating a 2D inverse Fourier Transform on the plane in k-space (a k- space slice) to obtain a 2D projection in image space.

Three-dimensional (3D) and Four-dimensional (4D) Imaging using MR Projection Imaging such as Correlated with a Physiologic C'vcle FIG. 8 A illustrates generally a technique 800 A. such as a method, that may include generating tw o-dimensional (2D) MR projection images such as representative of different projection angles, and using such 2D projection images to construct three-dimensional (3D) images. A corresponding technique 800 B is show n schematically in FIG. SB.

At 802A a series of 2D R projection images can be generated, such as using one or more techniques mentioned elsewhere herein as show n, for example, in FIG. 6 (by aggregating I D projection lines), or as show n in FIG. 7 A or FIG. 7B. Referring to FIG. SB at 802 B. 2D projection images Pi. P ₂ , P .. ... ,PN can be acquired at different projection angles. For example, projection angles can be specified to capture projection directions around the imaging subject. Tomographic reconstruction may then be performed to obtain a 3D image. As the projection direction rotates around the patient, tomographic reconstruction techniques, such as similar to X-ray techniques including CT or CBCT reconstruction, can used to either create a new 3D image, or update a previous 3D image with new information.

How ever, motion may induce blurring in reconstructed 3D images. Accordingly, in FIG. SA at 804 A, particular acquired 2D projection images can be assigned to bins using information indicative of a temporal position within a physiologic cycle, such as respiration. Such binning can be accomplished using information obtained from one or more of a surrogate, an external marker, or an internal marker or feature. For example, to obtain information indicative of a respiration cycle, a respiration belt can be used to provide a surrogate signal or diaphragm motion can be tracked in acquired imaging information.

Referring to FIG. SB. at 804B. f(t) can represent a plot of a signal representative of a portion of a physiologic cycle such as respiration. Various bins such as phase bins φι, φ2, (j>3, ..., φ _η can be established, such as

corresponding to portions (e.g.. a range At) along f(t). Acquired 2D projection images can be assigned to bins φι, φ2, φ3, ..., φ _η such as by determining a portion of f(t) on which a particular acquired image falls. The use of phase-based bins is merely illustrative and amplitude bins could similarly be used, such as corresponding to amplitude ranges (e.g.. a range Af) along f(t).

Referring to FIG. 8A, at 806 A a 3D image can be constructed using a binned series of 2D projection images corresponding to different projection angles. In the context of FIG. 8B, at 806B. the 3D images . h. h 1„ can correspond to each of the bins φι, φ2, φ3, ..., φ _α Referring to FIG. 8A, at 808 A. 4D imaging information can be constructed by aggregating the 3D images constructed at 806 A. In the context of FIG. 8B, the series of 3D images can provide a 4D representation of imaged region of the subject throughout the physiologic cycle. Motion induced by a physiologic cycle such as respiration may generally be highly periodic and reproducible. R Projection Imaging for Radiation Therapy Control

FIG. 9 illustrates generally a technique, such as a method, that may include generating a tw o-dimensional (2D) projection image using MR imaging and determining a change betw een the generated 2D projection image and reference imaging information. At 902. reference imaging information can be received. For example, the techniques 800A or 800 B of FIG. 8 A or FIG. 8B can be used to obtain reference imaging information, such as prior to treatment. In another example, a particular 3D reference image can also be generated, without requiring generation of other 3D images or aggregation of the acquired 3D images into 4D imaging information. For example, if respiration-gated therapy is to be delivered during a particular phase or amplitude of a respiration cycle. one or more 3D images can be constructed corresponding to a portion of the respiration cycle of interest, either during pre-treatment plannin or

intrafractionally.

At 904. a 2D projection image can be generated using techniques show n and described elsewhere herein (e.g.. using a 2D MR imaging sequence with a large slice selection gradient or no slice selection gradient, or by aggregating information acquired corresponding to multiple I D projection lines). At 906, a change betw een a generated 2D projection image and reference imaging information can be determined. At 908. delivery of the radiation therapy may be controlled at least in part using information indicative of the determined change.

The determined change can provide information indicative of one or more of an updated location of a target region, an anatomical feature or landmark, or a motion of the target region, anatomical feature, or landmark, as illustrative examples. In an example, the 2D MR projection image generated at 904 can include or can be related to target motion from the perspective of a radiation therapy "beam eye view" ( BEV) plane. There are various ways that the target motion in the BEV plane can be extracted from a 2D MR projection image.

In one approach, a 2D 3D registration can be performed between the 2D R projection image and 3D MR imaging information, such as in a manner similar to techniques used for registration betw een an X-ray projection image and a reference CT or CBCT image. Such an approach can be used, for example, to identify one or more translations that provide a match betw een the 2D projection image and shifted 3D MR imaging information, and the identi fied translation can be used as a the "change ^" in the context of FIG. 9 at 906 and 908 to control delivery , such as by repositioning one or more of the therapy beam output or the patient, or by modifying the therapy beam aperture. A quality of the match can be defined such as using one or more metrics, such as may include determining normalized cross-correlation or mutual information. Rotations and deformations may be included in the registration technique at the cost of simplicity and computational efficiency.

In another approach, a dimensionality reduction can be performed, such as to transform a 2D 3D registration problem into a 2D 2D registration problem. In one approach, reference projections can be extracted from 3D reference MR imaging information, in a manner similar to digitally reconstructed radiographs (DRRs) in X-ray based radiotherapy imaging. Segmentation can be used, such as to identify a target or surrounding structures such as OARs. though one more of the radiation therapy target or OARs may be masked by structures that lie in the path of the projection direction. Once the target or other structure has been segmented, a motion of the target or other structure can be identified. Such motion an also be used to predict a future location of the target.

A challenge can exist in attempting to register or otherw ise compare later-acquired MR projection images w ith reference 3D or 4D MR imaging information. Later- acquired MR projection images may have different image quality than the reference imaging information, particularly when the reference imaging information was acquired without use of projection imaging.

Registration techniques are generally more effective in comparing images having similar image quality or characteristics. The present inventors have recogni/ed. among other things, that the reference imaging information (such as received at 902 in FIG. 9) can be acquired using MR projection imaging, such as using a rotating set of projections.

As mentioned in relation to FIG. 8 A and FIG. 8B, such MR projections can be used to reconstruct a 3D MR image, such as using a tomographic reconstruction technique. In this manner, the reference 3D MR image will have similar image quality as later-acquired MR projection images. In an example, a later-acquired MR projection image can be compared directly to an acquired reference MR projection image without requiring use of 3D or 4D reference imaging information.

As mentioned above, if projection directions are rotated around the imaging subject. 4D MR reference imaging information can be compiled in manner similar to 4D-CBCT. because particular projections will generally contain different view s of the anatomy of the imaging subject. Such anatomy may include landmarks such as show ing a diaphragm location, or a region to be targeted by radiation. Common anatomical landmarks can then be used to bin the projections to form the 4D M RI sequence, rather than using an independent surrogate.

An acquisition duration to obtain 3D or 4D imaging information can be controlled using M R projection imaging techniques. For example, an acquisition duration can be shortened signi ficantly such as by acqui ring a more l i mi ted number of tomographic projections and using sparse tomographic reconstruction techniques such as compressed sensing or prior image compressed sensing (PICCS). An acquisition duration can also be improved such as by using parallel imaging strategies including one or more of multiple transmit or receive coils with di fferent sensitivity profiles.

In the examples described herein. M R projections need not include projection profiles that encompass an entirety of the pati ent in the depth dimension along the proj ection di rection. For example, a particular MR projection image may use a finite sl ice thickness encompassing an entirety of the region of interest in the depth dimension. Reducing an extent of the depth dimension can hel p to reduce obscuration of the region of interest ( e.g..

shadow ing) by overlying or underlying anatomy in the depth dimension, but at the expense of reducing or eliminating an ability to provide full tomographic reconstruction.

M R Projection Imaging Spatial Arrangements such as relative

to Radiation Therap ^y Beam Orientation FIG. 10A ill ustrates generally a spatial arrangement of a radiation therapy beam orientation 1090 with respect to one or more projection directions. such as may include two projection directions oriented orthogonally to each other. In the simplest approach, an MR proj ection image can be acquired using a proj ection line orientation 1060 A that coincides with the therapy beam orientation 090. at fi rst angular position ΘΑ. AS mentioned in relation to other examples, a projection imaging plane 070 A can include information acquired using parallel projection ( e.g.. such as corresponding to lines 1064A and 062 A ), or using a divergent proj ection line orientation, converging at a location 1050. The MR projection image can obtain information along the projection lines to encompass a region of interest 1012, such as incl udin a treatment isocenler 1010. In this manner, the projection imaging plane 1070 A can provide an imaging representation similar to a beam eye view ( BEV ) or portal image.

The configuration shown in FIG. I OA can be static, or the beam orientation and projection line orientation can be rotating together around the patient (as in the example of a gantry-mounted treatment beam output providing portal imaging). An orientation for MR projection imaging aligned with the BEV is generally a useful di rection because the aperture of the therapy beam is generally shaped to provide a speci fied profile in the plane parallel to the proj ection imaging plane 1070A. Without imaging from other directions.

motion of a target or imaging features may not be explicitly determined in the depth direction ( e.g.. Y di rection), but approaches exist to estimate such motion i f in-plane information indicative of target motion is available. Otherw ise, additional projections can be acquired having other projection directions.

In an example, one or more MR projections perpendicular to a BEV plane may be acquired, such as at various di fferent ti mes. Such orthogonal images can help to obtain information missing in the depth direction ( e.g.. Y direction ) along the fi rst projection line orientation 1060 A. For example, as shown ill ustratively in FIG. 10A, a second projection line orientation 1060B can be used at an orthogonal angular position ΘΒ, defining a projection imaging plane 1070B orthogonal to the first projection imaging plane 1070A. Again, parallel or di vergent projection lines can be established, such as the parallel l ines 1064B and 1062B shown in FIG. 10A.

Alternating or otherw ise sequencing betw een proj ections parallel and perpendicular to the BEV plane can provide full depth information, at the expense of a reduced acquisition frequency of projections parallel to the BE V. The orthogonal configuration show n in FIG. lOA can simulate gantry-mounted x-ray based stereoscopic imaging. As an illustrative example, the perpendicular projections need not be acquired alternately for every BEV proj ection acquisition. For example, the orthogonal projection orientation may be used for acquisition only occasionally to establish or update a correlation betw een target motion in the BEV and motion in the depth direction. FIG. 10B illustrates generally a spatial arrangement of MR imaging projection directions 1060 A. 1060B. and 1060C, such as corresponding to projection angle positions θ \, ΘΒ, ΘΓ spanning an arc or circle about a specified region such as a radiation therapy treatment isocenter 1010. The respective projection directions 1060 A. 1060B. and 1060C can provide particular projection imaging plane orientations 1070A. 1070B, and 1070C. For rotational radiation therapy treatment deliveries, the BEV naturally rotates around the patient, such as when the radiation therapy source is mounted on a gantry.

Acquired projection images can have more than one purpose. For example, as mentioned above, a particular BEV projection image can provide information indicative of a radiation therapy target position or shape from the perspective of the radiation therapy beam source. Also, if a series of projection images are acquired, a 3D tomographic R image can be reconstructed. The MR projection orientations shown in FIG. I OB are not limited to examples where the radiation therapy beam source is rotated. For example, for radiation therapy involving one or more static therapy fields, rotating MR projections can be acquired separately from the BEV projections, such as in an alternating fashion or according to another specified imaging sequence.

FIG. IOC illustrates generally a spatial arrangement of an MR imaging projection direction 1060D. such as oriented at a specified angle, a. with respect to a radiation therapy beam direction. The radiation therapy beam 1090 can diverge from a source location 1050, and a plane 1092 can define the BEV. By contrast with other examples, the MR imaging projection direction 1060D can be specified to capture an imaging perspective slightly different from the BEV projection, such as to obtain imaging information corresponding to a temporally- advanced BEV offset from a current BEV. Such temporally-advanced MR projection imaging can include an angle a specified to account for time lag associated with one or more of MR projection imaging acquisition, correction of the radiation therapy delivery protocol, or updating of the radiation therapy delivery protocol in response to acquired MR projection imaging. As in other examples, parallel or divergent MR projection imaging schemes can be used, and also as in other examples, the proj ection line orientation 1060D can be rotated relative to the radiation therapy beam orientation as the radiation therapy beam is rotated around the patient.

The advance angle a can be determined using intbrmation about one or more of a known lag duration or the angular speed of a beam-positioning gantry. as an illustrative example. A prediction technique can be applied to information acquired from the "advance BEV plane ^" 1070D such as to predict the most likely target position that will occur by the time the therapy output beam position catches up with the alignment of the advance BEV plane 1070D. Examples of prediction techniques can include one or more of kernel density estimation. wavelet-based techniques, or relevance vector machine (RVM) techniques. A dimensionality of the prediction problem can be reduced from three dimensions to two dimensions, because projected motion may be confined to the advance BEV plane 1070D perspective rather than having to predict target motion in a three-di mensional coordinate space.

FIG. 10D ill ustrates generally a spatial arrangement of M R imaging projection di rections 1060E and 1060F. such as may be speci fied to provide M R proj ection images in projection planes 1070E and 1070F in a manner similar to stereoscopic X-ray imaging. In the example of FIG. 10D, a projection image need not be acquired in the BEV di rection, but may still be acquired using fixed orientations such as simulating room-mounted stereoscopic X-ray imaging techniques. As an illustrative example, alternating M R projections in the anterioposterior and lateral di rections can be acquired such as to help locate a radiation therapy target or other anatomical features. In an example, a combination MR projection directions such as fixed orientations and rotating orientations corresponding to gantry position can be used. As an illustrative example, three or more projections can be acqui ed, such as in an alternating fashion, including a projection oriented to coincide with the BEV; an anterioposterior proj ection; and a lateral projection. Each of the projections can be selected to include a path traversing a speci fied region of the imaging subj ect. such as the treatment isocenter 1010. Such projections do not need each need to be acquired at the same imaging rate. Various Notes & Examples

Example 1 can include or use subj ect matter (such as an apparatus, a method, a means for performing acts, or a device readable medium including instructions that, when performed by the device, can cause the device to perform acts), such as can include a method for generating a four-dimensional (4D) imaging information representative of a physiologic cycle of a subject, the method comprising: generating two or more tw o-dimensional ( 2D) images, the 2D images comprising projection images representative of di fferent projection angles, and the 2D images generated using imaging information obtained via nuclear magnetic resonance (MR) imaging; assigning the particular 2D i mages to bins at least in part using in formation indicative of temporal positions w ithin the physiologic cycle corresponding to the particular 2D images; constructing three-dimensional ( 3D) i mages using the binned 2D images: and constructing the 4D imaging information, comprising aggregating the 3D images.

In Example 2. the subj ect matter of Example 1 optionally incl udes: a physiologic cycle comprising a respi ration cycle; and obtaining the two or more 2D images comprising obtaining 2D i mages representative of di fferent projection angles over a duration spanning multi ple respi ration cycles.

In Example 3. the subject matter of any one or more of Examples 1 -2 optionally includes generating two or more 2D projection images including aggregating acquired one-dimensional ( 1 D) projection lines into a particular 2D image, the 1 D projection lines oriented spatially parallel to one another.

In Example 4. the subject matter of any one or more of Examples 1 -3 optionally includes generating two or more 2D projection images including aggregating acquired one-dimensional ( I D) projection lines into a particular 2D image, the 1 D projection lines oriented to spatially diverge from one another.

In Example 5. the subject matter of any one or more of Examples 1 -4 optionally includes generating two or more 2D projection images including acquiring a 2D MR imaging slice perpendicular to a projection angle without requi ring a sl ice selection gradient. In Example 6, the subject matter of any one or more of Examples 1-5 optionally includes generating two or more 2D projection images including acqui ring a 2D MR imaging slice perpendicular to a projection angle using a slice selection gradient defining a sl ice sufficiently large in depth to encompass an entiretN of a radiation therapy target extent in a dimension parallel to the projection angle.

In Example 7. the subject matter of any one or more of Examples 1-6 optionally includes projection angles spanning an arc rotating about a speci fied central axis.

In Example 8, the subject matter of any one or more of Examples 1 -7 optionally includes determining a phase of a portion of the physiologic cycle corresponding to particular 2D images; and assigning the particul ar 2D images to bins using information indicative of the determined phase.

In Example 9, the subject matter of any one or more of Examples 1-8 optionally includes determining an amplitude of a portion of the physiologic cycle corresponding to particular 2D images; and assigning the particular 2D images to bins using information indicati ve of the determined amplitude.

In Example 10, the subject matter of any one or more of Examples 1-9 optionally incl udes one or more of the phase or ampl itude of the portion of the physiologic cycle corresponding to particular 2D images determined using a feature extracted from the particular 2D i mages.

In Example 1 1 , the subject matter of Example 10 optionally includes an extracted feature corresponding to a diaphragm of an i maging subject.

In Example 12. the subject matter of any one or more of Examples 1 - 1 I optionally includes, assigning the particular 2D images to bins using a dimensionality reduction of acquired imaging information.

In Example 13. the subject matter of any one or more of Examples 1 - 1 2 optionally includes assigning the particular 2D images to bins using a Fourier Transform of the particular 2 D images. In Example 14. the subject matter of an one or more of Examples 1 - 1 3 optionally incl udes constructing a 3 D image from acquired 2D proj ection images using a tomographic image reconstruction technique.

In Example 15, the subject matter of any one or more of Examples 1 - 14 optionally includes constructing the 3D i mage from the acqui red 2D projection images including performing the 3D i mage construction using transformed imaging information represented in a Fourier space.

In Example 16, the subject matter of any one or more of Examples 1-15 optionally includes constructing the 3D i mage from the acquired 2D projection images including performing the 3D image construction using a filtered back- projection technique.

In Example 1 7. the subject matter of any one or more of Examples 1-16 optionally includes constructing the 3 D image from the acquired 2D projection images including performing the 3D image construction using a compressed sensing technique.

In Example 18, the subject matter of any one or more of Examples 1 - 1 7 optionally including constructing the 3D image from the acquired 2D projection images including performing the 3 D image construction using Feldman-Davis- Kress construction.

In Example 19, the subject matter of any one or more of Examples 1-18 optionally includes constructing the 3D i mage from the acquired 2D projection images including performing the 3D image construction using an iterative approach.

In Example 20, the subject matter of any one or more of Examples 1-19 optionally includes providing the 4D imaging information for use in generating or adapting a radiation therapy treatment plan.

In Example 2 1 . the subject matter of any one or more of Examples 1-20 optionally includes using the 4D i maging information to assign or determine a position of the patient prior to delivery of a radiation therapy treatment fraction.

Example 22 can include, or can optionally be combined with the subj ect matter of one or any combination of Examples I -2 1 to incl ude, subj ect matter (such as an apparatus, a method, a means for performing acts, or a machine readable medi um including instructions that, when performed by the machine, that can cause the machine to perform acts), such as can include a method to control radiation therapy delivery to a subject using projection imaging, the method comprising: receivin reference i maging information; generating a tw o- di mensional (2D) projection image using i maging information obtained via nuclear magnetic resonance (MR) imaging, the 2D projection image corresponding to a speci fied projection direction, the speci fied projection direction including a path traversing at least a portion of an imaging subject; determining a change between the generated 2D projection image and the reference i maging information; controlling del ivery of the radiation therapy at least in part using the determined change betw een the obtained 2D projection image and the reference i maging information.

In Example 23. the subject matter of Example 22 optionally includes, generating the 2D proj ection image comprising aggregating acquired one- dimensional ( I D) projection lines.

In Example 24. the subject matter of Example 23 optionally includes that the speci fied projection direction is speci fied at least in part to provide 1 D projection l ines defined by respective paths traversing a radiation therapy treatment isocenter.

In Example 25, the subject matter of any one or more of Examples 23-24 optionally includes that the I D projection lines are oriented to spatially di verge from one another.

In Example 26, the subject matter of any one or more of Examples 23-25 optionally includes that the di rections corresponding to particular I D proj ection lines are speci fied to converge in a location corresponding to an available position of a radiation therapy beam output.

In Example 27. the subject matter of any one or more of Examples 22-26 optionally incl udes that generating the 2D projection image comprises acquiring a 2D MR imaging slice perpendicular to a proj ection angle w ithout requi ring a slice selection gradient.

In Example 28, the subject matter of any one or more of Examples 22-27 optionally incl udes that generating the 2D projection image comprises acquiring a 2D MR imaging slice perpendicular to a projection angle using a slice selection gradient defining a slice sufficiently large in depth to encompass an entirety of a radiation therapy target extent in a dimension parallel to the projection angle.

In Example 29, the subject matter of any one or more of Examples 22-28 optionally includes that the speci fied proj ection direction corresponds to a present or a future radiation therapy beam direction.

In Example 30, the subject matter of any one or more of Examples 22-29 optionally includes that the speci fied projection direction is orthogonal to a present or a future radiation therapy beam direction.

In Example 3 1. the subject matter of any one or more of Examples 22-30 optionally incl udes that the speci fied projection direction is established without requiring a radiation therapy beam direction.

In Example 32. the subject matter of any one or more of Examples 22-3 1 optionally includes that the reference image comprises a second 2D projection image generated using earlier-acqui red imaging information.

In Example 33. the subject matter of Example 32 optionally includes that the second 2D projection image is generated using four-dimensional (4D) imaging information assembled from earlier-acqui red imaging information.

In Example 34. the subject matter of any one or more of Examples 22-33 optionally includes that the reference image comprises three-dimensional (3 D) imaging inf ormation corresponding to earlier-acquired imaging information.

In Example 35, the subject matter of any one or more of Examples 22-34 optionally includes that the reference image comprises 4D imaging information assembled from earlier-acquired imaging information.

In Example 36, the subject matter of any one or more of Examples 22-35 optionally includes that the reference image comprises a 3D image extracted from a portion of 4D imaging information, the 4D imaging information assembled from earlier-acquired imaging information.

In Example 37, the subject matter of Example 36 optionally includes that the selected portion of the 4D imaging information comprises a speci fied portion of a physiologic cycle. In Example 38, the subject matter of any one or more of Examples 22-37 optionally includes that determining the change includes using a series of two or more 2D proj ection images generated using imaging information obtained via nuclear magnetic resonance (MR) imaging.

In Example 39, the subject matter of any one or more of Examples 22-38 optionally includes that determining the change includes registering at least a portion of the 2D projection i mage w ith the ref erence image.

In Example 40, the subject matter of any one or more of Examples 22-39 optionally incl udes that determining the change comprises extracting a feature from the 2D projection image.

In Example 4 1. the subject matter of any one or more of Examples 22-40 optionally includes that determining the change incl udes segmenting a portion of the 2D projection image.

In Example 42. the subject matter of Example 4 1 optionally includes that the segmented portion of the 2D projection image comprises a perspective of a radiation therapy target.

In Example 43. the subject matter of any one or more of Examples 22-42 optionally includes that determining the change comprises triangulating between determined perspectives of the radiation therapy target segmented from two or more 2D projection images.

In Example 44. the subject matter of any one or more of Examples 22-43 optionally includes predicting a location of a radiation therapy target using the determined change and a prediction model .

In Example 45. the subject matter of Example 44 optionally includes that the prediction model incl udes using information indicative of target motion established at least in part using extracted perspectives of the radiation therapy target from a series of acqui red 2D projection images and the determined change betw een at least one 2D projection image and the reference image.

Example 46 can include, or can optionally be combined w ith the subj ect matter of one or any combination of Examples I -45 to include, subj ect matter (such as an apparatus, a method, a means for performing acts, or a machine readable medium including instructions that, w hen performed by the machine. that can cause the machine to perform acts ), such as can include an imaging system, comprising: at least one processor circuit and a processor-readable storage medium, the processor readable storage medium including instructions that, when performed by the processor circuit, cause the processor circuit to generate four-dimensional (4D) imaging information representative of a physiologic cycle of a subj ect, including: generating two or more tw o- dimensional (2D) images, the 2D images comprising projection images representative of di fferent projection angles, and the 2D i mages generated using imaging information obtained via nuclear magnetic resonance (MR) imaging; assigning the particular 2D images to bins at least in part using information indicative of temporal positions within the physiologic cycle corresponding to the particular 2D i mages; constructing three-dimensional (3 D) i mages using the binned 2D images; and constructing the 4D imaging information, comprising aggregating the 3 D images.

Example 47 can include, or can optionally be combined with the subject matter of one or any combination of Examples 1-46 to incl ude, subject matter (such as an apparatus, a method, a means for performing acts, or a machine readable medium incl uding instructions that, when performed by the machine, that can cause the machine to perform acts ), such as can include a radiation therapy treatment system, comprising: a therapy generator; and a therapy output; a therapy controller system coupled to the radiation therapy generator and the radiation therapy output, the radiation therapy controller system comprising an imaging input, the i maging input configured to recei ve reference i maging information, the therapy control ler system configured to: generate a two- dimensional ( 2D) projection image using imaging information obtained via nuclear magnetic resonance (MR) imaging, the 2D proj ection image

corresponding to a speci ied projection direction, the speci fied projection direction including a path traversing at least a portion of an imaging subject; determine a change betw een the generated 2D projection image and the reference imaging information; and control delivery of the radiation therapy at radiation therapy output least in part using the determined change betw een the obtained 2D projection i mage and the reference i maging information. Each of the non-limiting examples described in this document can stand on its own, or can be combined in various permutations or combinations with one or more of the other examples.

The above detailed description includes references to the accompanyin drawings, which form a part of the detailed description. The drawings show, by way of illustration, specific embodiments in which the invention can be practiced. These embodiments are also referred to herein as "examples. ^" Such examples can include elements in addition to those shown or described.

However, the present inventors also contemplate examples in which only those elements shown or described are provided. Moreover, the present inventors also contemplate examples using any combination or permutation of those elements shown or described (or one or more aspects thereof), either with respect to a particular example (or one or more aspects thereof), or with respect to other examples (or one or more aspects thereof) shown or described herein.

In the event of inconsistent usages between this document and any documents so incorporated by reference, the usage in this document controls.

In this document, the terms "a" or "an ^" are used, as is common in patent documents, to include one or more than one. independent of any other instances or usages of "at least one ^" or "one or more. ^" In this document, the term "or" is used to refer to a nonexcl usi ve or. such that "A or B ^" includes "A but not B," "B but not A. ^" and "A and B. ^" unless otherwise indicated. In this document, the terms ^" including ^" and "in which" are used as the plain-English equivalents of the respective terms "comprising ^" and "wherein." Also, in the following claims, the terms "including ^" and "comprising ^" are open-ended, that is, a system, device, article, composition, formulation, or process that includes elements in addition to those listed after such a term in a claim are still deemed to fall within the scope of that claim. Moreover, in the following claims, the terms "first. ^" "second. ^" and "third. ^" etc. are used merely as labels, and are not intended to impose numerical requirements on their objects.

Method examples described herein can be machine or computer- implemented at least in part. Some examples can include a computer-readable medium or machine-readable medium encoded with instructions operable to configure an electronic device to perform methods as described in the above examples. An implementation of such methods can include code, such as microcode, assembly language code, a higher-level language code, or the like. Such code can include computer readable instructions for performing various methods. The code may form portions of computer program products. Further, in an example, the code can be tangibly stored on one or more volatile, non- transitory, or non-volatile tangible computer-readable media, such as during execution or at other times. Examples of these tangible computer-readable media can incl ude, but are not limited to, hard disks, removable magnetic disks. removable optical disks ( e.g.. compact disks and digital video disks), magnetic cassettes, memory cards or sticks, random access memories ( RAMs ), read only memories ( ROMs ), and the l ike.

The above description is intended to be illustrative, and not restrictive. For example, the above-described examples (or one or more aspects thereof) may be used in combination with each other. Other embodiments can be used, such as by one of ordinary skill in the art upon reviewing the above descri ption. The Abstract is provided to allow the reader to quickly ascertain the nature of the technical disclosure. It is submitted with the understanding that it will not be used to interpret or limit the scope or meaning of the claims. Also, in the above Detailed Description, various features may be grouped together to streamline the disclosure. This should not be interpreted as intending that an unclaimed disclosed f eature is essential to any claim. Rather, inventive subject matter may lie in less than all features of a particular disclosed embodi ment. Thus, the follow ing clai ms are hereby incorporated into the Detailed Description as examples or embodiments, with each claim standing on its own as a separate embodiment, and it is contemplated that such embodi ments can be combined with each other in various combinations or permutations. The scope of the invention should be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.