k-Space Sampling for Accelerated Stack-of-Stars Magnetic Resonance Imaging Using Compressive Sensing and AI

Abstract

Disclosed herein is a medical system 100, 300 comprising a memory 110 storing machine executable instructions 120 and a computing system 140. Execution of the machine executable instructions causes the computing system to receive (200) k-space data 122 and reconstruct (202) at least one three-dimensional magnetic resonance image 124 from the k-space data. The k-space data includes k-space data lines 510, 514 sampled according to a three-dimensional stack-of-stars sampling pattern 500. The three-dimensional stack-of-stars includes discrete sampling planes 508 periodically distributed along a Cartesian sampling direction axis 506. The k-space data lines pass through the Cartesian sampling direction axis. The k-space lines are perpendicular to the Cartesian sampling direction axis. At least a portion of the k-space data lines 514 are distributed between the discrete sampling planes.

Description

[Technical field]

[0001] The present invention relates to magnetic resonance imaging, and in particular to a stack-of-stars sampling pattern in k-space for magnetic resonance imaging. [Background technology]

[0002] Various tomographic medical imaging techniques, such as magnetic resonance imaging (MRI), computed tomography, positron emission tomography, and single photon emission tomography, allow detailed visualization of a subject's anatomy. A common feature of all of these imaging modalities is that it takes a significant amount of time to acquire the medical imaging data necessary to reconstruct a medical image. During magnetic resonance imaging, data is sampled in k-space. The measured k-space data is then reconstructed into a magnetic resonance image. One means of accelerating the acquisition of magnetic resonance images is to intentionally undersample k-space and reconstruct the magnetic resonance image using a compressed sensing (CS) algorithm. Undersampling can result in image artifacts. However, when undersampling is performed using random or pseudorandom sampling, the undersampling artifacts primarily manifest themselves as noise. Magnetic resonance images are sparsely represented from images in k-space. In MRI, the well-known CS algorithm is used to iteratively reconstruct the image.

[0003] A common difficulty encountered during magnetic resonance imaging is that k-space data acquisition can take several minutes and the subject may move. It is therefore desirable to have a means of correcting for subject motion. One technique that is less sensitive to subject motion is so-called radial sampling. Instead of acquiring k-space data lines using a Cartesian sampling pattern, k-space data lines or spokes are acquired in a plane at different rotation angles rotated around an axis. The acquired k-space data is less sensitive to motion because it was not acquired using a single frequency and phase encoding direction. The stack-of-stars sampling pattern is a three-dimensional k-space sampling pattern that is used to sample radial acquisitions that are repeated in many different planes.

[0004] European Patent Publication EP 3709042 A1 discloses a method for MR imaging of an object. The European Patent Publication discloses MR imaging using a 3D radial or spiral acquisition scheme that provides enhanced image quality in the presence of motion. The method comprises the steps of generating MR signals by subjecting the object to an imaging sequence with RF pulses and switching magnetic field gradients, acquiring MR signals using a 3D radial or spiral acquisition scheme with oversampling of the central part of k-space, detecting motion-induced displacements and / or deformations of the object during acquisition of the MR signals and assigning each of the acquired MR signals to a motion state, and reconstructing an MR image from the MR signals weighted in the central part of k-space, where stronger weights are applied to MR signals acquired in more frequent motion states, whereas weaker weights are applied to MR signals acquired in less frequent motion states. Furthermore, the invention relates to an MR device and a computer program for the MR device. Summary of the Invention

[0005] The invention provides a medical system, a computer program and a method according to the independent claims. Embodiments are set out in the dependent claims.

[0006] During magnetic resonance imaging, sampling of k-space is performed by repeatedly traversing and sampling paths in k-space. These paths are commonly referred to as k-space lines. Depending on the placement of the k-space lines and the sample points selected in each k-space line, different types of sampling patterns can be used.

[0007] One technique is the so-called stack of stars. In stack of stars, there is a Cartesian sampling direction. For the sake of example, the Cartesian sampling direction is arbitrarily set to k for some examples herein. z The Cartesian sampling direction has a Cartesian sampling direction axis. Also, in the examples herein, k z The axis is the Cartesian sampling direction k z In this case, Cartesian sampling direction axes can be used to define a set of cylindrical coordinates, where the Cartesian sampling direction axis is the height axis and k angle is used to represent the radial k-space angle used in the Cartesian sampling direction.

[0008] In a conventional stack-of-stars sampling pattern, k-space sampling is performed in discrete sampling planes perpendicular to a Cartesian sampling direction axis. Within each of these planes, k-space lines are sampled by rotating the k-space lines about the Cartesian sampling direction axis. The k-space lines pass through the Cartesian sampling direction axis. When a plot showing the positions of the k-space lines is observed, the pattern resembles a stack of stars with the stars centered in each of the discrete sampling planes.

[0009] The example provides improved image quality by relaxing the requirement that k-space data lines be confined to discrete sampling planes. In the example, at least some of the k-space data lines are displaced along a Cartesian sampling direction axis. These displaced k-space data lines are still perpendicular to and still pass through the Cartesian sampling direction axis, but are not limited to being in discrete sampling planes. This has the advantage that when performing three-dimensional magnetic resonance imaging reconstruction, the sampled points in k-space are better distributed. This results in better image quality.

[0010] Another advantageous possibility is when the k-space data is undersampled. Undersampling the k-space data can lead to artifacts in the resulting magnetic resonance image. A better distribution of sample points in k-space leads to the resulting undersampling artifacts being better noise approximation. This in turn leads to improved means of reconstructing images, for example using compressed sensing (CS) reconstruction algorithms.

[0011] In one aspect, the present invention provides a medical system comprising a memory storing machine executable instructions and a computing system. Execution of the machine executable instructions causes the computing system to receive k-space data. The k-space data includes k-space data lines sampled according to a three-dimensional stack-of-stars sampling pattern. The stack-of-stars sampling pattern samples the k-space data in a Cartesian sampling direction, and the k-space data is then rotationally sampled in a plane perpendicular to the Cartesian sampling direction. That is, there is a Cartesian sampling direction axis, and in a particular plane, the sampled k-space data lines rotate about this Cartesian sampling direction axis.

[0012] The three-dimensional stack of stars includes discrete sampling planes periodically distributed along the Cartesian sampling direction axis. Periodically may mean at set or regular intervals displaced along the Cartesian sampling axis. The k-space data lines pass through the Cartesian sampling direction axis and are perpendicular to the Cartesian sampling direction axis. At least some of the k-space data lines are distributed between the discrete sampling planes. Typically, in a stack of stars sampling pattern, the k-space lines are restricted to the discrete sampling planes. In this embodiment, this is not a strict requirement. At least some or some of the k-space data lines are offset from the discrete sampling planes to be between two of the planes. This has the technical advantage that the k-space data is distributed throughout the entire k-space.

[0013] This has a significant advantage when reconstructing a three-dimensional magnetic resonance image from k-space data. For example, when the k-space data is undersampled, algorithms such as compressed sensing or artificial intelligence-based reconstruction algorithms are used to reconstruct the image. These techniques typically rely on undersampling artifacts that appear as background noise. With samples of the k-space data better distributed throughout the k-space, there is a better approximation between the similar noise and the undersampling artifacts. In that case, the compressed sensing or artificial intelligence-based reconstruction algorithms perform better and higher acceleration factors can be achieved.

[0014] Execution of the machine-executable instructions further causes the computing system to reconstruct at least one three-dimensional magnetic resonance image from the k-space data.

[0015] In another embodiment, at least some of the k-space data lines distributed between the discrete sampling planes are irregularly distributed in a linear direction along the Cartesian sampling direction axis. In other words, the k-space data lines that are not within the discrete sampling planes are offset along the Cartesian sampling direction axis to be irregularly distributed. The k-space data lines do not follow a regular pattern. The k-space data lines can be distributed, for example, randomly or pseudo-randomly. This has the effect of better distributing the k-space data irregularly and improving, for example, the quality of compressed sensing image reconstruction.

[0016] In another embodiment, at least a portion of the k-space lines distributed between the discrete sampling planes are regularly distributed in a linear direction along a Cartesian sampling direction axis, For example, a particular sampling pattern can be used to distribute how the k-space data lines are offset from the discrete sampling planes.

[0017] In another embodiment, at least some of the k-space lines distributed between the discrete sampling planes can be distributed in a linear direction along the Cartesian sampling direction axis in an irregular or regular manner. In this embodiment, some of the k-space lines offset from the sampling planes can be linearly distributed along the Cartesian sampling direction axis or linearly offset from the Cartesian sampling direction axis. However, some of the k-space lines can be distributed irregularly. For example, it is beneficial to have a well-sampled k-space near the origin of the Cartesian sampling direction axis. Within the central region, the k-space data lines can be regularly distributed within this central region. The k-space data lines can be offset from the discrete sampling planes with a random or pseudo-random offset, since one is outside the central region. This also has the advantage of improving the quality of the reconstructed magnetic resonance image.

[0018] In another embodiment, there is a maximum distance between sample points of the three-dimensional stack of stars sampling pattern. This applies not only to the k-space data lines, but also to the sampling points between the various lines. Setting the maximum distance ensures that there is a certain minimum sampling of k-space.

[0019] In another embodiment, there is a minimum distance between sample points of a three-dimensional stack of stars sampling pattern. In this example, the sample points are prevented from being too close to each other. This is beneficial because it forces a more uniform or distributed k-space sampling. This is also particularly beneficial for undersampled data.

[0020] In another embodiment, the maximum distance between sample points is position and / or angle dependent in k-space and the minimum distance between sample points is position and / or angle dependent in k-space. The at least one three-dimensional magnetic resonance image can correspond to a rectangular field of view, where the maximum distance between sample points and the minimum distance between sample points define the rectangular field of view.

[0021] Prior to acquiring k-space data, an operator of a magnetic resonance imaging system typically defines a field of view for the image, which is essentially a volume in space of the imaged subject. In magnetic resonance imaging, the locations sampled in k-space determine the field of view. When performing Cartesian k-space sampling, it is straightforward to define the sample locations to obtain a rectangular field of view. In radial sampling, the radial angle is given by Larson et al., "Anisotropic Field-of-Views in Radial Imaging," IEEE Transactions on Medical Imaging, Vol. 27, No. 1, January 2008, pp. 47-57; or Wu et al., "Anisotropic Field-of-View support for Golden Angle Radial Imaging," Magn. Reson. Med. July 2016;76(1):229-236. doi:10.1002 / mrm.25898 This is defined to obtain a rectangular field of view as described in

[0022] In another embodiment, the minimum and maximum distances between k-space sample points are scaled according to a specified or predetermined rectangular field of view.

[0023] In another embodiment, execution of the machine executable instructions causes the computing system to receive a selection of a rectangular field of view. Execution of the machine executable instructions further causes the computing system to calculate a minimum and maximum distance between k-space sampling locations in the k-space data lines for a volume representing the rectangular field of view. Execution of the machine executable instructions further causes the computing system to modify the stack-of-stars sampling pattern to match the maximum distance between sample points and the minimum distance between sample points in k-space. Several portions of the k-space lines in the Cartesian direction can be spaced at equal distances. These portions can be measured in one shot followed by a pre-pulse such as a fat suppression pulse. In subsequent shots, irregularly shifted shots can be sampled. The pre-pulse can also be shared across several regularly shifted shots. Additionally, the radial k-space lines can be regularly or irregularly shifted in the Cartesian direction to compensate for the applied phase shift during reconstruction.

[0024] In another embodiment, the positions of the k-space samples in the k-space data lines are determined using a Poisson disk algorithm to provide a (pure) irregular sampling density.

[0025] In another embodiment, the three-dimensional stack of stars sampling pattern has a sample point density that decreases as the distance from the origin of the Cartesian sampling direction axis increases. In other words, the sampling of low k-space data near the k-space center in the Cartesian sampling direction is higher. This is particularly beneficial because the k-space information higher in space can be undersampled without excessive loss of image content. This principle generally applies to image compressibility as commonly applied in compressed sensing. Ensuring that the central region of the k-space data is better sampled provides a higher quality image when it is reconstructed. The sampling density attenuation can be adjusted to strong attenuation, medium attenuation or low attenuation, depending on the expected image content. In the case of angiography, where the image content is determined by high contrast of the vessels and low contrast of the background, a strong sampling intensity attenuation is used, resulting in a high acceleration factor, and for weaker contrast, a medium attenuation is appropriate.

[0026] In another embodiment, the sample points of at least some of the individual lines of k-space data are irregularly and / or regularly distributed along the length of each individual line. In this embodiment, there are three cases where the k-space data in a particular line is irregularly distributed, or regularly distributed, or a combination of the two. The fact that the k-space data is irregularly distributed essentially helps to ensure that any artifacts due to undersampling closely resemble the noise distribution.

[0027] In some examples, it is useful to sample, for example, a portion of the k-space data with full sampling and then undersampling in another region. For example, the central region of the k-space data can be sampled with a regular pattern, and then outside the central region, the sampling is irregular. In other options, a regular pattern is used in each case. This regular sampling can be combined, for example, with other types of sampling, for example, with a distribution of k-space lines in a linear direction along a Cartesian sampling direction axis. In one example, the sampling within the discrete sampling plane can be regular, and then the k-space lines offset from the discrete sampling plane can be irregular. This also results in a sampling pattern that better enables the reconstruction of at least one magnetic resonance image.

[0028] In another embodiment, at least a portion of the k-space data is arranged according to a shear grid in k-space. For example, the sample points are arranged such that some of the sample points are in the discrete sampling planes and some of the k-space sampling points are outside or between the discrete sampling planes. In some examples, the entire k-space sampling pattern follows a shear structure for the k-space sampling points or in some examples.

[0029] The medical system further comprises a magnetic resonance imaging system, the memory further comprising pulse sequence commands configured to control the magnetic resonance imaging system to acquire k-space data using a three-dimensional stack-of-stars sampling pattern, execution of the machine-executable instructions further causes the computing system to acquire the k-space data by controlling the magnetic resonance imaging system with the pulse sequence commands.

[0030] It should be noted that the computing system may represent different types of systems in different examples. For example, in conjunction with the previous embodiment, the computing system may be incorporated into a control system for a magnetic resonance imaging system, where, for example, reconstruction may be performed. In another example, the computing system may represent one or more computing or computer systems. For example, the control of the magnetic resonance imaging system and the reconstruction of the magnetic resonance image may be separated. The k-space data may be acquired by the magnetic resonance imaging system, and then the k-space data may be transmitted to a server, for example, in a radiology department, or even to a cloud-based reconstruction system. For example, in this case, data may be acquired using a magnetic resonance imaging system, and then the k-space data may be transmitted to a remote or cloud-based system to reconstruct the magnetic resonance image, which then further transmits it to a radiologist's workstation or computing system, where the radiologist then diagnoses the image.

[0031] Prior to acquiring k-space data, the operator of the magnetic resonance imaging system typically defines a field of view for the image, which is essentially a volume in space of the imaging subject. In magnetic resonance imaging, the sampled locations in k-space determine the field of view. When performing Cartesian k-space sampling, it is straightforward to define the sample locations to obtain a rectangular field of view. In radial sampling, the sample locations are defined by the radial angle and a rectangular field of view scale factor as described in Wu et al., "Anisotropic Field-of-View support for Golden Angle Radial Imaging," Magn Reson Med. July 2016;76(1):229-236. doi:10.1002 / mrm.25898, or in Larson et al., "Anisotropic Field-of-Views in Radial Imaging," IEEE TRANACTIONS ON MEDICAL IMAGING, Vol. 27, No. 1, January 2008, 47-57, first paragraph of section 5.

[0032] In another embodiment, execution of the machine executable instructions causes the computing system to receive a selection of a rectangular field of view. Execution of the machine executable instructions further causes the computing system to calculate a minimum and maximum distance between k-space sampling locations in a k-space data line for a volume representing the rectangular field of view. Execution of the machine executable instructions further causes the computing system to modify the stack of stars sampling pattern to match the maximum distance between sample points in k-space and the minimum distance between sample points. In another embodiment, the attenuation of the k-space sampling density is adjusted for kz. Strong, medium or weak sampling density attenuation can be used. In another embodiment, the medical system includes a user interface with an acquisition acceleration control. This can be, for example, a control in a graphical user interface. This can have, for example, a slider knob or a location where an acceleration factor can be entered or selected. The acquisition acceleration control is configured to provide a selected acceleration factor. The acceleration factor is often also called a reduction factor and is defined as the ratio of the amount of k-space data that needs to be fully sampled to the amount that is actually collected or sampled. For example, if only half of the k-space data that needs to be fully sampled is sampled, the acceleration factor is 2.

[0033] Execution of the machine executable instructions further causes the computing system to receive a selected acceleration factor from the acquisition acceleration control. Execution of the machine executable instructions further causes the computing system to calculate a sample point distribution from the three-dimensional stack of stars sampling pattern such that a current acceleration factor of the three-dimensional stack of stars sampling pattern matches the selected acceleration factor before acquiring the k-space data. In other words, after receiving the selected acceleration factor, the pulse sequence command can be modified such that fewer k-space samples are acquired. This allows the operator to have a trade-off between acquisition time and quality of the reconstructed image.

[0034] The sample points can be shifted from the three-dimensional stack of stars sampling pattern using a variety of techniques. The sample points can be shifted, for example, randomly or pseudo-randomly. In other cases, the sample points to be shifted ahead may be pre-programmed or included in a list. In another example, a particular selected acceleration factor is selected based on the rectangular FOV scale factor and the strong, medium and low signal density attenuation, k z and k angle The 3D stack of stars sampling pattern is programmed by distributing the sample points evenly across the directions. There are various ways to implement this and accomplish shifting the sample points from the 3D stack of stars sampling pattern.

[0035] In another embodiment, the three-dimensional stack of stars sampling pattern is arranged such that adjacent k-space data lines are rotated using a golden angle rotation scheme or a pseudo-golden angle rotation scheme during k-space data acquisition. The use of a golden angle rotation scheme is useful in several situations. In many cases, the use of a golden angle rotation scheme is useful for reducing alias patterns in the reconstructed image, and the use of a golden angle rotation scheme is useful when binning, sorting, or soft gating of magnetic resonance imaging k-space data is performed during motion. The golden angle is approximately 111.25°. The golden angle may be used in schemes where there is rotation in the sampling pattern. The pseudo-golden angle rotation scheme also uses the golden angle, but it is slightly adjusted to fit on an equidistant radial grid. In addition to the golden angle, small golden angles, such as those described in [Magn Reson Med. 2016 June;75(6):2372-8.Golden ratio sparse MRI using tiny golden angles, Wundrak et al.], can also be used, for example, for balanced sequences to avoid large k-space jumps.

[0036] In another embodiment, the k-space data is a four-dimensional data set. That is, the k-space data is three-dimensional but also time-resolved. Execution of the machine-executable instructions further causes the computing system to assemble the k-space data into time-resolved k-space data on different grids. The term "time-resolved k-space data" is referred to herein as k-space data that is binned or sorted according to either a particular time interval or a subject's motion phase. For example, the time units may be measured relative to the subject's dynamic phase after contrast administration, the subject's cardiac phase, and / or the respiratory phase. Assembly of the k-space data can be accomplished in various ways. As mentioned, the k-space lines can be binned (separately sorted) or soft gating can be used. In soft gating, the k-space data is weighted according to the degree to which it contributes to a particular time or motion phase and fits into a particular time interval or phase of the subject's motion.

[0037] The assembly of the k-space data to different grids refers to a resampling operation. For example, k-space data was acquired at different rotational positions and at different locations along the Cartesian sampling direction. The k-space data for a particular time interval is then resampled, which serves for reconstruction. Execution of the machine executable instructions further causes the computing system to reconstruct a four-dimensional image sequence from the time-resolved k-space data. As mentioned above, this can be a four-dimensional image showing an evolution through time, or the time is measured relative to the motion phase of the subject. For example, the four-dimensional image sequence can be dynamic phase images, which are cardiac cine images that are compensated for both cardiac and respiratory phases.

[0038] In another embodiment, the k-space data is combined using a k-space weighted reconstruction algorithm or using a certainty map reconstruction. One example of this is that the k-space data is combined using a k-space weighted image contrast (KWIC) weighted reconstruction. KWIC reconstruction is also known as k-space weighted image contrast reconstruction. In this particular embodiment, the radial center region of k-space is highly oversampled. During k-space data acquisition, k-space data ions are acquired in groups or passes that collect k-space data from various dynamic, respiratory and cardiac phases. The k-space is weighted during reconstruction with the goal of providing optimal image contrast for the required time phase. This can be achieved by providing weights or certainty maps to the compressed sensing and AI algorithms. This embodiment has the advantage of providing high quality images with sufficient SNR and resolution to eliminate streaking artifacts in the resulting magnetic resonance images.

[0039] In another embodiment, the time-resolved k-space data is motion-resolved k-space data and is assembled using autonomous navigation or subject motion sensor data. For example, oversampled radial k-space data can be used for autonomous navigation taking into account coil element information. In another example, for example, an ECG or respiration sensor (e.g., a camera) is present on the subject to record subject motion data. This is used to bin or soft-gate the data. The k-space data is further assembled into motion-resolved k-space data using binning or soft-gating. A computing system is configured or programmed to assemble the time-resolved k-space data using autonomous navigation or subject motion sensor data. The computing system is further configured to execute the machine-executable instructions to assemble the k-space data into motion-resolved k-space data using binning or soft-gating.

[0040] In another embodiment, the k-space data is undersampled, for example where lines of k-space data are sampled that are offset from the discrete sampling plane, providing better reconstruction of the undersampled image.

[0041] In another embodiment, the at least one three-dimensional magnetic resonance image is reconstructed using a compressed sensing reconstruction algorithm.

[0042] In another embodiment, at least one three-dimensional magnetic resonance image is reconstructed using a neural network reconstruction algorithm. This embodiment may take a variety of forms. For example, there may be a neural network trained to completely reconstruct an image from undersampled k-space data. However, in many cases, the neural network reconstruction algorithm is a compressed sensing reconstruction algorithm that incorporates a neural network into a compressed sensing algorithm.

[0043] Within the compressed sensing algorithm, an image is reconstructed from the current k-space data, and then a denoising filter is applied, which is then used iteratively in a routine consistency step to improve the quality of the image. A neural network can be used in any of these various steps of the compressed sensing algorithm. Using a neural network as a denoising filter is often very effective in improving the quality of the compressed sensing algorithm. For example, the denoising filter can be trained on a particular view or type of image. In this case, the denoising filter can not only be a straight noise filter, but can also essentially incorporate prior knowledge about the type of magnetic resonance imaging scan being performed. Prior knowledge about time and motion information with respect to the k-space weights / certainty map can be provided to the compressed sensing algorithm and / or the neural network. This can be trained by using a fully sampled image or a fully sampled k-space data set (possibly KWIC filtered) used to generate the image, and simulated undersampled data can be created by taking a portion of this fully sampled k-space data. This simulated undersampled k-space data can then be used in a compressed sensing algorithm and the resulting image can be input to a neural network, which can then be compared to an image constructed from a fully sampled data set.

[0044] In another aspect, the invention provides a computer program product including machine executable instructions for execution by a computing system for controlling a medical system. Execution of the machine executable instructions causes the computing system to receive k-space data. The k-space data includes k-space data lines sampled according to a three-dimensional stack of stars sampling pattern. The three-dimensional stack of stars includes discrete sampling planes periodically distributed along a Cartesian sampling direction axis. The k-space data lines pass through and are also perpendicular to the Cartesian sampling direction axis. At least a portion of the k-space data lines are separated between the discrete sampling planes. Execution of the machine executable instructions further causes the computing system to reconstruct at least one three-dimensional magnetic resonance image from the k-space data.

[0045] In another aspect, the present invention provides a method of medical imaging. The method includes receiving k-space data. The k-space data includes lines of k-space data sampled according to a three-dimensional stack of stars sampling pattern. The three-dimensional stack of stars includes discrete sampling planes periodically distributed along a Cartesian sampling direction axis. The k-space data lines pass through the Cartesian sampling direction axis. The k-space data lines are also perpendicular to the Cartesian sampling direction axis. At least a portion of the k-space data lines are distributed between the discrete sampling planes. The method further includes reconstructing at least one three-dimensional magnetic resonance image from the k-space data.

[0046] It should be understood that one or more of the above-described embodiments of the present invention may be combined, unless the combined embodiments are mutually exclusive.

[0047] As will be appreciated by one of ordinary skill in the art, aspects of the present invention may be embodied as an apparatus, a method, or a computer program product. Accordingly, aspects of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, microcode, etc.), or an embodiment combining software and hardware aspects, all of which may be referred to generally herein as a "circuit," "module," or "system." Additionally, aspects of the present invention may take the form of a computer program product embodied in one or more computer-readable medium(s) having computer-executable code embodied therein.

[0048] Any combination of one or more computer readable media may be utilized. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. As used herein, a "computer readable storage medium" encompasses any tangible storage medium that stores instructions that are executable by a processor or a computing system of a computing device. A computer readable storage medium may be referred to as a computer readable non-transitory storage medium. A computer readable storage medium may also be referred to as a tangible computer readable medium. In some embodiments, a computer readable storage medium may also store data that can be accessed by a computing system of a computing device. Examples of computer readable storage media include, but are not limited to, floppy disks, magnetic hard disk drives, solid state hard disks, flash memory, USB thumb drives, random access memory (RAM), read only memory (ROM), optical disks, magneto-optical disks, and register files of a computing system. Examples of optical disks include compact disks (CDs) and digital versatile disks (DVDs), e.g., CD-ROM, CD-RW, CD-R, DVD-ROM, DVD-RW, or DVD-R disks. The term computer-readable storage medium also refers to various types of recording media that can be accessed by a computer device over a network or communication link. For example, data may be read by a modem, by the Internet, or by a local area network. Computer executable code embodied in a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, fiber optic cable, RF, etc., or any suitable combination of the above.

[0049] A computer-readable signal medium may include a propagated data signal in which computer-executable code is embodied, for example in baseband or as part of a carrier wave. Such a propagated signal may take any of a variety of forms, including but not limited to electromagnetic, optical, or any suitable combination thereof. A computer-readable signal medium may be any computer-readable medium that is not a computer-readable storage medium but is capable of communicating, propagating, or transporting a program for use by or in connection with an instruction execution system, apparatus, or device.

[0050] "Computer memory" or "memory" is one example of a computer-readable storage medium. Computer memory is any memory directly accessible to a computing system. "Computer storage" or "storage" is a further example of a computer-readable storage medium. Computer storage is any non-volatile computer-readable storage medium. In some embodiments, computer storage may also be computer memory, or vice versa.

[0051] As used herein, a "computing system" encompasses electronic components capable of executing programs or machine-executable instructions or computer-executable code. References to a computing system, including examples of a "computing system," should be interpreted as including more than one computing system or processing core, as the case may be. A computing system may be, for example, a multi-core processor. A computing system may also refer to a collection of computing systems within a single computer system, or a collection of computing systems distributed among multiple computer systems. The term computing system should also be interpreted as referring to a collection or network of computing devices, each of which may include a processor or computing system, as the case may be. Machine-executable code or instructions may be executed by multiple computing systems or processors, which may be within the same computing device, or even distributed across multiple computing devices, such as multiple computing cores of a graphical processing unit (GPU).

[0052] Machine executable instructions or computer executable code may include instructions or programs that cause a processor or other computing system to perform aspects of the invention. Computer executable code for performing operations related to aspects of the invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, Smalltalk, or C++, and traditional procedural programming languages ​​such as the "C" programming language or similar programming languages, and may be compiled into machine executable instructions. In some examples, the computer executable code may be in the form of a high-level language or in pre-compiled form, and may be used in conjunction with an interpreter that generates machine executable instructions on the fly. In other examples, the machine executable instructions or computer executable code may be in the form of programming for a programmable logic gate array.

[0053] The computer executable code may run entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be to an external computer (e.g., through the Internet using an Internet Service Provider).

[0054] Aspects of the present invention are described with reference to flowcharts, examples and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the present invention. It is understood that each block or part of the blocks of the flowcharts, examples and / or block diagrams, where applicable, can be implemented by computer program instructions in the form of computer executable code. It is further understood that combinations of blocks in different flowcharts, examples and / or block diagrams may be combined if not mutually exclusive. These computer program instructions may be provided to a general purpose computer, a special purpose computer, or other programmable data processing device computing system to create a machine such that the instructions, which execute via the computer or other programmable data processing device computing system, cause the means to perform the functions / acts specified in one or more blocks of the flowcharts and / or block diagrams.

[0055] These machine-executable instructions or computer program instructions may also be stored on a computer-readable medium that can instruct a computer, other programmable data processing apparatus, or other device to function in a certain manner, such that the instructions stored on the computer-readable medium produce an article of manufacture including instructions that perform the functions / acts specified in one or more blocks of the flowcharts and / or block diagrams.

[0056] The machine-executable instructions or computer program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to generate a computer-implemented process such that a series of operational steps are performed on the computer, other programmable apparatus, or other device, such that the instructions executing on the computer or other programmable apparatus provide a process for performing the functions / acts specified in one or more blocks of the flowcharts and / or block diagrams.

[0057] A "user interface" as used herein is an interface that allows a user or operator to interact with a computer or computer system. A "user interface" is sometimes referred to as a "human interface device." A user interface can provide information or data to an operator and / or receive information or data from an operator. A user interface may allow input from an operator to be received by the computer and may provide output from the computer to a user. In other words, a user interface may allow an operator to control or manipulate a computer and an interface may allow a computer to show the results of the operator's control or manipulation. Displaying data or information on a display or graphical user interface is an example of providing information to an operator. Receiving data via a keyboard, mouse, trackball, touchpad, wand, graphic tablet, joystick, gamepad, webcom, headset, pedals, wired gloves, remote control, and accelerometer are all examples of user interface elements that allow for the reception of information or data from an operator.

[0058] As used herein, a "hardware interface" encompasses an interface that allows a computing system of a computer system to interact with and / or control external computing devices and / or devices. A hardware interface may allow a computing system to send control signals or commands to external computing devices and / or devices. A hardware interface may also allow a computing system to exchange data with external computing devices and / or devices. Examples of hardware interfaces include, but are not limited to, a universal serial bus, an IEEE 1394 port, a parallel port, an IEEE 1284 port, a serial port, an RS-232 port, an IEEE 488 port, a Bluetooth connection, a wireless local area network connection, a TCP / IP connection, an Ethernet connection, a control voltage interface, a MIDI interface, an analog input interface, and a digital input interface.

[0059] As used herein, a "display" or "display device" encompasses an output device or user interface adapted to display images or data. A display outputs visual, audio, and / or tactile data. Examples of displays include, but are not limited to, computer monitors, television screens, touch screens, tactile electronic displays, Braille screens, cathode ray tubes (CRTs), storage tubes, bi-stable displays, electronic paper, vector displays, flat panel displays, vacuum fluorescent displays (VFs), light emitting diode (LED) displays, electroluminescent displays (ELDs), plasma display panels (PDPs), liquid crystal displays (LCDs), organic light emitting diode displays (OLEDs), projectors, and head mounted displays.

[0060] K-space data is defined herein as the recorded measurements of radio frequency signals emitted by atomic spins using the antenna of a magnetic resonance apparatus during a magnetic resonance imaging scan. Magnetic resonance data is an example of tomographic medical image data.

[0061] A magnetic resonance imaging (MRI) image or MR image is defined herein as a reconstructed two- or three-dimensional visualization of anatomical data contained within magnetic resonance imaging data, which visualization can be performed using a computer.

[0062] Preferred embodiments of the invention will now be described, by way of example only, with reference to the drawings in which: [Brief description of the drawings]

[0063] [Figure 1] An example of a medical system is shown below. [Diagram 2] 2 shows a flow chart illustrating a method of using the medical system of FIG. 1. [Diagram 3] An example of a medical system is shown below. [Figure 4] 4 shows a flow chart illustrating a method of using the medical system of FIG. 3. [Diagram 5] Indicates the coordinate system. [Figure 6] 4 shows an example of an acquired acceleration control unit. [Figure 7] 1 shows an example of a k-space sampling pattern. [Figure 8] Some examples of signal density functions are shown. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0064] Like numbered elements in these figures are equivalent elements or perform the same function. An element previously described is not necessarily mentioned in a subsequent figure if its function is equivalent.

[0065] FIG. 1 illustrates an example of a medical system 100. In this example, the medical system 100 is shown to include a computer 102. The computer 102 represents one or more computer systems distributed in one or more locations. The computer 102 is shown to include a computing system 104. The computing system 104 is also intended to represent one or more computing systems located in one or more machines. The computing system 104 is shown to communicate with an optional hardware interface 106 and an optional user interface 108. The hardware interface 106 may be, for example, the computing system 104 for communicating with other components of the medical system 100, if present. The user interface 108 allows an operator to control operations and functions and to interact with the medical system 100.

[0066] Computing system 104 is further shown as being in communication with memory 110. Memory 110 is intended to represent various types of memory that computing system 104 can access.

[0067] The memory 110 is shown as including machine-executable instructions 120. The machine-executable instructions 120 enable the computing system 104 to perform various data tasks, such as numerical calculations and image processing. In cases where the medical system 100 includes additional components, the machine-executable instructions 120 may also be configured to enable the computing system 104 to control these additional components. For example, if the medical system 100 includes a magnetic resonance imaging system, the machine-executable instructions 120 would enable the computing system 104 to control the magnetic resonance imaging system.

[0068] The memory 110 is further shown as including k-space data 122. The k-space data includes lines of k-space data sampled according to a three-dimensional stack of stars sampling pattern. The three-dimensional stack of stars includes discrete sampling planes periodically distributed along a Cartesian sampling direction axis. The k-space data lines pass through the Cartesian sampling direction axis. The k-space data lines are perpendicular to the Cartesian sampling direction axis. At least a portion of the k-space data lines are distributed between the discrete sampling planes. The memory 110 is further shown as including a three-dimensional magnetic resonance image 124 reconstructed from the k-space data 122. The machine-executable instructions 120 are configured to effect reconstruction of the three-dimensional magnetic resonance image 124 from the k-space data 122.

[0069] Figure 2 shows a flow chart illustrating a method of operating the medical system 100 of Figure 1. First, in step 200, k-space data 122 is received. Then, in step 202, a three-dimensional magnetic resonance image 124 is reconstructed from the k-space data 122.

[0070] Figure 3 illustrates a further example of a medical system 300. The medical system 300 illustrated in Figure 3 is similar to the medical system 100 of Figure 1, except that the medical system 300 further comprises a magnetic resonance imaging system 302 controlled by the computing system 104.

[0071] The magnetic resonance imaging system 302 comprises a magnet 304. The magnet 304 is a superconducting cylindrical magnet with a bore 306 running through it. Different types of magnets can be used, for example both segmented cylindrical magnets and so-called open magnets. A segmented cylindrical magnet is similar to a standard cylindrical magnet except that it is split into two sections to allow the cryostat access to the iso-face of the magnet, such magnets may be used, for example, in conjunction with charged particle beam therapy. An open magnet has two magnet sections, one above the other, with a space between them large enough to accommodate the subject, the arrangement of the two section areas being similar to that of a Helmholtz coil. Open magnets are common because the subject is relatively unconfined. Inside the cryostat of the cylindrical magnet there is a collection of superconducting coils.

[0072] Within the bore 306 of the cylindrical magnet 304 is an imaging zone 308, where the magnetic field is sufficiently strong and uniform to perform magnetic resonance imaging. A field of view 309 is shown within the imaging zone 308. Acquired magnetic resonance data is typically acquired for the field of view 309. A subject 318 is shown as supported by a subject support 320 such that at least a portion of the subject 318 is within the imaging zone 308 and field of view 309.

[0073] Also present within the magnet bore 306 are a set of magnetic field gradient coils 310, which are used for preliminary magnetic resonance data acquisition to spatially encode magnetic spins within an imaging zone 308 of the magnet 304. The magnetic field gradient coils 310 are connected to a magnetic field gradient coil power supply 312. The magnetic field gradient coils 310 are intended to be exemplary. Typically, the magnetic field gradient coils 310 include three separate sets of coils for spatial encoding in three orthogonal spatial directions. The magnetic field gradient power supply supplies electrical current to the magnetic field gradient coils 310. The current supplied to the magnetic field gradient coils 310 is controlled as a function of time and can be ramped or pulsed.

[0074] Proximate to the imaging zone 308 is a radio frequency coil 314 for manipulating the orientation of magnetic spins in the imaging zone 308 and for receiving radio transmissions from spins also in the imaging zone 308. A radio frequency antenna includes multiple coil elements. A radio frequency antenna may also be referred to as a channel or an antenna. The radio frequency coil 314 is connected to a radio frequency transceiver 316. The radio frequency coil 314 and the radio frequency transceiver 316 may be replaced by separate transmit and receive coils and separate transmitters and receivers. It is understood that the radio frequency coil 314 and the radio frequency transceiver 316 are exemplary. The radio frequency coil 314 is also intended to represent a special transmit antenna and a special receive antenna. Similarly, the transceiver 316 may also represent a separate transmitter and receiver. The radio frequency coil 314 may also have multiple receive / transmit elements, and the radio frequency transceiver 316 may have multiple receive / transmit channels.

[0075] The transceiver 316 and the gradient controller 312 are shown as being connected to the hardware interface 106 of the computer system 102 .

[0076] The memory 110 is shown as including pulse sequence commands 330. The pulse sequence commands are commands or data that can be converted into commands that can be used to control the magnetic resonance imaging system 302 to acquire k-space data 332. The pulse sequence commands 330 are configured to acquire the k-space data 122 such that the k-space data lines are acquired according to a three-dimensional stack-of-stars sampling pattern. The sampling pattern has discrete sampling planes periodically distributed along a Cartesian sampling direction axis. The k-space data lines pass through and are perpendicular to the Cartesian sampling direction axis. At least a portion of the k-space data lines are distributed between the discrete sampling planes.

[0077] Figure 4 shows a flow chart illustrating a method of operating the medical system 300 of Figure 3. Initially, in step 400, the computing system 104 uses pulse sequence commands 330 to control the magnetic resonance imaging system 302 to acquire k-space data 122. The method then proceeds to steps 200 and 202 as shown in Figure 2.

[0078] FIG. 5 shows an example of a stack of stars sampling pattern 500 in k-space. x Axis 502, k y Axis 504 and k z You can see the axis 506. In this example, k zAxis 506 is considered to be the Cartesian sampling direction axis 506. In this particular example, there are three discrete sampling planes 508 visible. Typically in a stack of stars sampling pattern, the k-space lines are confined to the planes 508. Shown are two in-plane sampled k-space lines 510, shown as rotating about the Cartesian sampling direction axis 506, in one of the planes 508. A circle 512 indicates the rotation about the axis 506. In this example, at least a portion of the k-space data lines are distributed between the discrete sampling planes 508. In FIG. 5, one line 514 of three k-space data lines is shown. The out-of-plane sampled k-space data line 514 is offset 516 from the plane 508 along the axis 506.

[0079] FIG. 6 shows an example of an acquisition acceleration control 600, which may be implemented, for example, as a graphical user interface for a user interface. The acquisition acceleration control 600 is displayed as a window 600. The window 600 has a displayed and / or editable acquisition acceleration coefficient 602. There is also a slider control 604 to change the acceleration coefficient. Other controls such as buttons, dials or other graphical user interface controls may also be used. Once the correct acceleration coefficient is selected, there is a button 606 to accept or select the displayed acceleration coefficient 602. After the button 606 is pressed, the system shown in FIG. 3 then operates to edit the pulse sequence command 330 so that the k-space data acquired has an acceleration coefficient that matches the value of 602. One straightforward way to do this is to calculate a sampling pattern for the k-space data given a specified (rectangular) FOV and sampling density function, and then to calculate the k-space acceleration coefficient (as specified by the input criteria) so that the k-space data has the appropriate acceleration coefficient. z and k angle For a uniform distribution, the distance to adjacent profiles is calculated by dividing the radial k-space lines by the mid-k zShifting the k-space lines in an irregular grid using dithering further distributes the k-space sampling with improved noise characteristics. z and k angle The coil sensitivity profile may further be considered to optimize the distribution of the sample in the direction.

[0080] 4D radial stacks of stars offer several promising advantages for clinical scanning, such as significantly higher motion robustness and benign aliasing artifacts. Despite its advantages, current techniques still do not provide the temporal and spatial fidelity required for T1 contrast-weighted, free-breathing body imaging, and in contrast exhibit motion blur. The present invention suggests irregular 3D and 4D radial sampling schemes optimized for CSENSE and AI.

[0081] 3D radial stacks of stars offer several promising advantages for clinical scanning, such as significantly higher motion robustness, absence of aliasing artifacts, and benign undersampling behavior. Despite its advantages, current techniques still do not provide the temporal and spatial fidelity required for T1 contrast-enhanced, free-breathing body imaging, and in contrast exhibit motion blur. Some examples provide irregular 3D and 4D radial sampling schemes that can be optimized for CSENSE (compressive sensing reconstruction) and AI (e.g., reconstruction incorporating neural networks into its algorithms).

[0082] Some examples may include one or more of the following features: - Variable density radial sampling in k-space, where the central part of k-space is oversampled. -K angle -k z The space is an irregular golden angle radial undersampled into higher k-space. - Coil sensitivity may vary in different directions angle -k z An optimized distribution in space is considered. - A single acceleration user interface parameter, k angle -k z The optimal distribution in space is calculated and provided. - CSENSE or AI is used to reconstruct 3D magnetic resonance images. - For 4D imaging, continuous golden angle sampling is performed and view sharing is used to reconstruct the subsequent dynamic scan. The central part of k-space is sampled more frequently in Kt-space than the peripheral k-space by using a stronger CSENSE sampling attenuation.

[0083] Some examples are shown in Figure 7 below. angle -k z A radial sampling scheme is provided that ensures a high acceleration factor by irregular undersampling in space.

[0084] FIG. 7 shows various types of k-space sampling patterns. In the first column, various stack-of-stars sampling patterns 500 are shown. Each stack-of-stars sampling pattern is represented by a discrete sampling plane 700, 702, and 704. Within each of the discrete sampling planes 700, 702, and 704, there is a sampling line 706 of k-space. In each example, only one sampling line 706 is labeled. In these examples, the sampling lines 706 are depicted solely as being within the discrete sampling planes 700, 702, and 704. In all of the examples shown in FIG. 7, the individual sampling lines 706 may be displaced out of plane.

[0085] Next to each stack of stars sampling pattern 500 is a two-dimensional plot 700 showing the locations of the sampling points, represented as an oversampled grid. The vertical axis represents the Cartesian sampling direction 506, and the horizontal axis represents the radial angular k-space direction 702.

[0086] The dots 704 in the grid represent sample points in k-space where the kz-space points are represented in an oversampled grid, e.g., the dots 704 are singly labelled.

[0087] The stack of stars sampling pattern 706 and oversample grid 708 represent a conventional regular 3D radial stack of stars sampling pattern.

[0088] The stack of stars sampling pattern 710 and oversampled grid 712 represent an example of a shared k-space sampling pattern 1002. In this example, every odd radial k-space line is shifted by half a voxel. This simple shift results in an improved k-space distribution.

[0089] The stack of stars sampling pattern 714 and oversampled grid 716 represent an example of a variable k-space sampling pattern in which there is a slower falloff in the density of k-space sampling points (weaker signal density falloff) as one k-space sampling point moves further away from the origin.

[0090] The stack of stars sampling pattern 718 and the oversampled grid 720 represent further examples of variable k-space sampling patterns. The parameter sampling density is used to describe a second main characteristic of the sampling pattern. That characteristic can be constant or variable. It describes the degree to which the local average sampling density depends on the (radial) position in k-space. In this case, there is a very steep falloff in the density of k-space sample points as one k-space sample point moves further away from the origin. As mentioned above, examples 714 and 716 represent a slow falloff.

[0091] The stack of stars pattern 722 and oversampled grid 724 represent irregular sampling where the central k-space is oversampled with increased k-space angular density, while the peripheral k-space is undersampled. The samples are irregularly distributed in both the Cartesian direction 506 and the radial direction 702. angle The points are sorted golden angle. Note that to further reduce scan time, k-space shutters can be applied rather than sampling those outer and outer k-space points.

[0092] Variable density undersampling ensures high undersampling of higher k-space frequencies, whereby the sampling density depends on the position in kz space: the central k-space is more densely sampled; towards higher k-space the radial density decreases.

[0093] FIG. 8 shows an example of a weak signal density decay (SDC) function 800 and a strong signal density decay function 802 according to a 3D CS sampling scheme.

[0094] Typically, Cartesian k zThe lines are sampled ahead to avoid motion-related k-space discontinuities and are followed by increasing the golden angle for optimal distribution. The irregular sampling pattern helps to distribute radial undersampling artifacts offering the potential for higher acceleration coefficients.

[0095] A single acceleration coefficient in the user interface can be used to calculate the optimal acceleration distribution in radial and Cartesian functions, where the coil sensitivity profile is taken into account. CSENSE or AI is used to reconstruct the 3D magnetic resonance image.

[0096] For 4D imaging, continuous golden angle sampling is performed, and then view sharing or more advanced kt techniques or compressed sensing are used in the time direction to reconstruct the subsequent dynamic scan. The central part of k-space is more dense than the peripheral k-space by using stronger CSENSE sampling attenuation (SDC). t More frequent sampling in space.

[0097] For purely irregular sampling density, the Poisson-Disk algorithm (Randomness, Gap, Clustering) "dithering" can be used, which is used to generate an incomplete (subsampled) set of sample points in the underlying Cartesian grid in k-space, within the following constraints: 1. Samples are randomly distributed. 2. The (locally averaged) sampling density must follow some prescribed sampling density function, typically a function of radial distance from the k-space origin. 3. The samples must not be clustered, which means that the distance between any pair of two samples must not be smaller than some minimum value. In the case of varying sampling density or variable angular density sampling, such as used in rectangular FOV sampling, the minimum distance corresponds to the sampling density. 4. There are no large gaps, meaning there are no unsampled lattice points that could have been sampled while still meeting the previous minimum distance requirement.

[0098] While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description is to be considered illustrative or exemplary and not restrictive, and the invention is not limited to the disclosed embodiments.

[0099] Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, "having, comprising, including" does not exclude other elements or steps, and singular elements do not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain means are recited in mutually different dependent claims does not indicate that a combination of these means cannot be used to advantage. A computer program may be stored / distributed on a suitable medium, such as an optical storage medium or a solid-state medium, provided together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems. Any reference signs in the claims should not be interpreted as limiting the scope. [Explanation of symbols]

[0100] 100 Medical Systems 102 Computer 104 Computing Systems 106 Optional Hardware Interface 108 Optional User Interface 110 Memory 120 Machine Executable Instructions 122 k-space data 124 Three-Dimensional Magnetic Resonance Imaging 200 k-space data received Reconstruct at least one three-dimensional magnetic resonance image from 202 k-space data 300 Medical Systems 302 Magnetic Resonance Imaging System 304 Magnet 306 Magnet Bore 308 Imaging Zone 309 Field of view 310 Magnetic field gradient coil 312 Magnetic field gradient coil power supply 314 Radio Frequency Coil 316 Transceiver 318 Target 320 Subject Support 330 Pulse Sequence Commands 400 Pulse sequence commands are used to control the magnetic resonance imaging system to acquire k-space data. Stack of Stars Sampling Pattern in 500 k Space 502k x shaft 504k y shaft 506k z Axis (Cartesian sampling direction axis) 508 Discrete Sampling Surface 510 in-plane sampling k-space lines 512 indicates rotation around the Cartesian sampling direction axis 514 out-of-plane sampled k-space data lines 516 Linear offset from the discrete sampling surface along the Cartesian sampling direction axis 600 Acquired acceleration control unit 602 Displayed and / or editable acceleration coefficients 604 Slider control section 606 Select Acceleration Factor Button 700 An oversampled grid showing the locations of sampling points 702 Radial angle k-space direction 704 Sampling Location 800 Weak SDC Functions 802 Strong SDC Functions

Claims

1. 1. A medical system comprising: a memory storing machine-executable instructions; and a computing system, Execution of the machine-executable instructions causes the computing system to: receiving k-space data, the k-space data comprising lines of k-space data sampled according to a three-dimensional stack-of-stars sampling pattern, the three-dimensional stack-of-stars sampling pattern comprising periodically distributed discrete sampling planes displaced at regular intervals along a Cartesian sampling direction axis, the k-space data lines passing through the Cartesian sampling direction axis, the k-space data lines being perpendicular to the Cartesian sampling direction axis, and at least a portion of the k-space data lines being distributed between the discrete sampling planes; reconstructing at least one three-dimensional magnetic resonance image from the k-space data; The at least some of the k-space data lines distributed among the discrete sampling planes are irregularly distributed in a linear direction aligned with the Cartesian sampling direction axis.

2. 2. The medical system of claim 1, wherein a maximum distance exists between sample points of the three-dimensional stack of stars sampling pattern, a minimum distance exists between the sample points of the three-dimensional stack of stars sampling pattern, and / or a Poisson disk algorithm is used to determine sample point locations in the k-space data lines.

3. The method of claim 1 or 2, wherein the distance between k-space sample points is scaled according to a predetermined rectangular field of view.

4. 4. The medical system of claim 1, wherein the three-dimensional stack of stars sampling pattern has a sample point density that decreases with increasing distance from an origin of the Cartesian sampling direction axis.

5. 5. The medical system of claim 1, wherein the sample points of at least some of the individual lines of k-space data are irregularly and / or regularly distributed in a longitudinal direction along each of the individual lines.

6. The medical system of claim 1 , wherein at least a portion of the k-space data is arranged according to a shear grid in k-space.

7. 7. The medical system of claim 1, further comprising a magnetic resonance imaging system, the memory further comprising pulse sequence commands configured to control the magnetic resonance imaging system to acquire the k-space data using the three-dimensional stack-of-stars sampling pattern, and execution of the machine-executable instructions further causes the computing system to acquire the k-space data by controlling the magnetic resonance imaging system with the pulse sequence commands.

8. The medical system includes a user interface having a captured acceleration control, the captured acceleration control configured to provide a selected acceleration coefficient, and execution of the machine-executable instructions further causes the computing system to: receiving the selected acceleration coefficient from the acquired acceleration control unit; 8. The medical system of claim 7, further comprising: a sampling point distribution calculated from the three-dimensional stack of stars sampling pattern such that a current acceleration factor of the three-dimensional stack of stars sampling pattern matches the selected acceleration factor before acquiring the k-space data.

9. 4. The medical system of claim 1, 2, or 3, wherein the three-dimensional stack of stars sampling pattern is arranged such that adjacent k-space data lines are rotated using a golden angle rotation scheme or a pseudo-golden angle rotation scheme during acquisition of the k-space data.

10. The k-space data is a four-dimensional data set, and execution of the machine-executable instructions further causes the computing system to: Assembling the k-space data into time-resolved k-space data at different grids; and The medical system of claim 1 , further comprising a step of reconstructing a four-dimensional image sequence from the time-resolved k-space data.

11. The medical system of claim 10 , wherein the k-space data is combined using a k-space weighted reconstruction algorithm or using a certainty map.

12. The k-space data is undersampled to: said at least one three-dimensional magnetic resonance image being reconstructed using a compressed sensing reconstruction algorithm; said at least one three-dimensional magnetic resonance image being reconstructed using a neural network reconstruction algorithm; The medical system according to claim 1 , wherein any one of the following applies:

13. 1. A computer program comprising machine executable instructions for execution by a computing system, the execution of said machine executable instructions causing said computing system to: receiving k-space data, the k-space data comprising lines of k-space data sampled according to a three-dimensional stack-of-stars sampling pattern, the three-dimensional stack-of-stars sampling pattern comprising periodically distributed discrete sampling planes displaced at regular intervals along a Cartesian sampling direction axis, the k-space data lines passing through the Cartesian sampling direction axis, the k-space data lines being perpendicular to the Cartesian sampling direction axis, and at least a portion of the k-space data lines being distributed between the discrete sampling planes; reconstructing at least one three-dimensional magnetic resonance image from the k-space data; the at least some of the k-space data lines distributed among the discrete sampling planes are irregularly distributed in a linear direction aligned with the Cartesian sampling direction axis. Computer program.

14. 1. A method of medical imaging, the method comprising: receiving k-space data, the k-space data comprising lines of k-space data sampled according to a three-dimensional stack-of-stars sampling pattern, the three-dimensional stack-of-stars sampling pattern comprising periodically distributed discrete sampling planes displaced at regular intervals along a Cartesian sampling direction axis, the k-space data lines passing through the Cartesian sampling direction axis, the k-space data lines being perpendicular to the Cartesian sampling direction axis, and at least a portion of the k-space data lines being distributed between the discrete sampling planes; and reconstructing at least one three-dimensional magnetic resonance image from the k-space data. the at least some of the k-space data lines distributed among the discrete sampling planes are irregularly distributed in a linear direction aligned with the Cartesian sampling direction axis. method.

Images