Acquisition of diffusion-weighted measurement data with non-trapezoidal gradient pulse shapes for diffusion coding

Abstract

An inventive method for acquiring diffusion-weighted measurement data of a test object using a magnetic resonance system with a measurement protocol with non-trapezoidal gradient pulse shapes for diffusion coding comprises the steps a) Loading a measurement protocol to be used with measurement parameters to be set, which include a desired non-trapezoidal gradient pulse shape, b) Loading pre-configured characteristics for at least the non-trapezoidal gradient pulse shape, c) Assigning possible parameter values ​​of at least one measurement parameter to be set in the measurement protocol to at least one category indicating the feasibility of the measurement protocol on the magnetic resonance system based on the charged characteristics, d) Entering at least one desired parameter value from at least one measurement parameter of the measurement protocol, taking into account the assignment made, e) If a parameter value has not yet been entered for each measurement parameter to be set in the measurement protocol, with which the measurement protocol can be executed, repeat steps c) and d), at least for measurement parameters for which no parameter value has yet been set, until parameter values ​​have been entered for all measurement parameters to be set in the measurement protocol, with which the measurement protocol can be executed. g) Acquisition of diffusion-weighted measurement data using the measurement protocol with the entered parameter values.

Description

[0001] The invention relates to an improved acquisition of diffusion-weighted measurement data with non-trapezoidal gradient pulse shapes for diffusion coding, in particular taking into account limit values ​​specified by the hardware of a magnetic resonance system used.

[0002] Magnetic resonance imaging (MRI) is a well-known technique used to generate images of the interior of an object. In simplified terms, the object is positioned in a magnetic resonance imaging (MRI) scanner within a relatively strong, static, homogeneous background magnetic field, also known as the B0 field, with field strengths ranging from 0.2 Tesla to 7 Tesla and higher. This causes the object's nuclear spins to align with the background magnetic field. To trigger measurable nuclear spin resonances, high-frequency excitation pulses (RF pulses) are applied to the object. The resulting nuclear spin resonances are measured as so-called k-space data using specially designed coils, and MR images or spectroscopic data are then derived from these measurements. The alternating magnetic field generated by the excitation pulses applied via at least one transmitting coil is also referred to as the B1 field.To spatially encode the measurement data, rapidly switched magnetic gradient fields, or gradients for short, are superimposed on the underlying magnetic field. The temporal evolution of such a gradient field, e.g., along a gradient axis, can also be described as a gradient pulse shape. A scheme used that describes a temporal sequence of applied RF pulses and switched gradients is called a pulse sequence, or simply a sequence. The recorded measurement data are digitized and stored as complex numerical values ​​in a k-space matrix. A corresponding MR image can be reconstructed from the k-space matrix containing these values, for example, using a multidimensional Fourier transform.

[0003] A magnetic resonance imaging scan typically consists of a large number of individual partial measurements, in which raw data are recorded from different layers of the object under investigation, from which volume image data can then be reconstructed.

[0004] Furthermore, many examinations require multiple, i.e., a whole series of magnetic resonance imaging (MRI) scans of the subject, during which a specific measurement parameter is varied. The effect of this parameter on the subject is then observed based on these measurements, allowing for subsequent diagnostic conclusions. A series of MRI scans typically consists of at least two, but usually more than two. Ideally, the measurement parameter is varied in such a way that the contrast of a specific material type excited during the measurements—for example, a tissue type within the subject or a chemical substance that is significant for most or certain tissue types, such as water—is influenced as strongly as possible by the variation of the measurement parameter. This ensures that the effect of the measurement parameter on the subject is particularly clearly visible.

[0005] A typical example of a series of magnetic resonance images acquired while varying a measurement parameter that strongly influences contrast is diffusion-weighted imaging (DWI). Diffusion refers to the Brownian motion of molecules in a medium. In DWI, multiple images with different diffusion directions and weightings are typically acquired and combined. The strength of the diffusion weighting is usually defined by the so-called "b-value." The diffusion images with different diffusion directions and weightings, or the images combined from them, can then be used for diagnostic purposes.Thus, by appropriately combining the recorded diffusion-weighted images, parameter maps with special diagnostic significance can be generated, such as maps that represent the "Apparent Diffusion Coefficient (ADC)" or the "Fractional Anisotropy (FA)".

[0006] Diffusion imaging is often based on echoplanar imaging (EPI) because of the short acquisition time of EPI sequences per image and their robustness to motion.

[0007] In diffusion-weighted imaging, additional gradients are inserted into a pulse sequence to visualize or measure the diffusion properties of tissue. These gradients cause tissues with rapid diffusion (e.g., cerebrospinal fluid, CSF) to experience greater signal loss than tissues with slower diffusion (e.g., grey matter in the brain). The resulting diffusion contrast is becoming increasingly important clinically, and applications now extend far beyond the classic early detection of ischemic strokes.

[0008] For many years, measurements of diffusion properties in different tissue types using magnetic resonance imaging have been an indispensable tool in clinical diagnostics. Typically, diffusion coding with two or more diffusion gradients with trapezoidal gradient pulse shapes, as described by Stejskal and Tanner in "Spin Diffusion Measurements: Spin Echoes in the Presence of a Time-Dependent Field Gradient," J. Chem. Phys. 42: pp. 288-292 (1965), is employed because it is a) efficient in utilizing the performance limits of a gradient system and b) easy to describe both within the framework of diffusion models and for technical calculations (e.g., regarding stimulation or limitations of the gradient system).

[0009] Recently, however, there has been an increased interest in more complex gradient pulse shapes in the research and diagnostics sector, particularly for tensor-weighted diffusion measurements, which promise access to new contrast properties "by design" as described, for example, in the article by Szczepankiewicz et al., "Gradient Waveform Design for Tensor-Valued Encoding in Diffusion MRI", J. Neurosc. Methods 348: p. 109007 (2021).

[0010] However, the increasing complexity of the gradient pulse shapes used here poses major challenges to the preparation, planning and execution of such measurements.

[0011] When a user processes measurement protocols, for example, in the context of tensor-weighted diffusion measurements with complex gradient pulse shapes, the user is typically presented with a possible value range for each measurement parameter within a group of parameters to be processed. Each value range is categorized, for example, into at least one of the following categories: "non-adjustable parameter values," "conditionally adjustable parameter values," and "unrestricted adjustable parameter values." This categorization can be displayed to the user in a suitable format, such as color-coded, to facilitate the selection of appropriate parameter values.

[0012] Parameters can be assigned to category A ("non-adjustable parameter value") if a parameter value assigned to this category A for a modified measurement parameter would result in an unexecutable measurement protocol. Such parameter values ​​can, for example, be locked for the user, preventing them from being selected for the corresponding measurement parameter. However, parameter values ​​assigned to this category A can still be displayed or otherwise brought to the user's attention, for instance, if they could be adjusted if the measurement protocol being processed undergoes a corresponding change elsewhere, such as a change to at least one other measurement parameter within the protocol, particularly one that is dependent on the modified measurement parameter. Such dependencies between different measurement parameters further increase the complexity of processing measurement protocols.

[0013] An assignment to category B ("conditionally adjustable parameter value") can occur if a parameter value assigned to this category B for a modified measurement parameter requires an automated, and in particular clearly defined, change to the respective parameter value of at least one other measurement parameter besides the modified measurement parameter in order to make the measurement protocol executable. The required automated change can then occur, for example, as soon as the user sets a parameter value assigned to category B for a modified measurement parameter, so that the adjustments necessary for the measurement protocol to be executable are carried out automatically.

[0014] An assignment to category C ("unrestricted") is possible if a parameter value assigned to this category C for a processed measurement parameter has no impact on the executability of the processed measurement protocol. A parameter value assigned to this category C for a processed measurement parameter can be selected without the need for further adjustments.

[0015] To enable users to assign a measurement protocol to at least one of categories A, B, or C while it is being processed, complex calculations are necessary. This is particularly important because every change to the value of a measurement parameter can alter the assignment of possible parameter values ​​to categories A ("not adjustable"), B ("conditionally adjustable"), and C ("fully adjustable") for at least one other measurement parameter, and potentially for many others.

[0016] The calculations include, in particular, a feasibility study of a measurement protocol intended for a given measurement. To use a desired non-trapezoidal gradient pulse shape, several hurdles due to the more complex gradient amplitude profiles must be overcome during its description to ensure that the desired gradient pulse shape can actually be used later. Converting imported, described non-trapezoidal gradient pulse shapes is also not straightforward. The complex profiles also present new challenges in ensuring the correct conversion of the desired gradient pulse shape's specified moments, especially the zeroth moment, as well as in compensating for accompanying Maxwell field terms and / or avoiding signal interference.

[0017] Only prototypical implementations are currently known. A user can usually choose from a number of predefined non-trapezoidal gradient pulse shapes and modify them to a certain extent. However, in the known methods, dependencies on other protocol parameters are not automatically taken into account when selecting a gradient pulse shape.

[0018] Some non-trapezoidal (but more complex) gradient pulse shapes can be approximated for the necessary calculations by a certain number of trapezoidal gradient pulse shapes. This applies, for example, to sinusoidal or cosine-like gradient pulse shapes. The necessary calculations can be performed sufficiently quickly as long as the number of half-waves of the gradient pulse shapes to be approximated is sufficiently small (for example, less than 100 half-waves). However, this approach is no longer practical for more complex gradient pulse shapes, which can be characterized by several hundred or thousand half-waves.

[0019] In principle, a subsequent (e.g., one-time) feasibility check of a modified measurement protocol immediately before starting the measurement would also be conceivable. However, this would often force a user to modify their measurement plan at very short notice (and possibly necessitating significant changes to parameter values). This is hardly practical in a clinical setting.

[0020] Another possibility would be to provide pre-tested measurement protocols (with defined measurement parameters) before any measurement, for example, once during the installation of a magnetic resonance imaging (MRI) system. However, this approach lacks the flexibility required in clinical applications for customizing measurement protocols (e.g., regarding spatial resolution, number of slices, or contrast-determining parameters such as echo time (TE) or repetition time (TR)) for individually performed measurements.

[0021] Another possibility is the use of computers with such high processing power that they can perform the necessary calculations described with sufficient speed during the planning of a measurement. Disadvantages of such a solution include both high energy consumption during operation and the costs associated with such high-performance computers.

[0022] The purpose of the described method is to enable the user to plan measurements with non-trapezoidal gradient pulse shapes for diffusion coding in a simplified, possibly interactive, manner, taking hardware limitations into account.

[0023] The problem is solved by a method for acquiring diffusion-weighted measurement data of a test object using a magnetic resonance system, employing a measurement protocol with non-trapezoidal gradient pulse shapes for diffusion coding according to claim 1, a magnetic resonance system according to claim 9, a computer program according to claim 10, and an electronically readable data carrier according to claim 11.

[0024] An inventive method for acquiring diffusion-weighted measurement data of a test object using a magnetic resonance system with a measurement protocol with non-trapezoidal gradient pulse shapes for diffusion coding comprises the steps a) Loading a measurement protocol to be used with measurement parameters to be set, which include a desired non-trapezoidal gradient pulse shape, b) Loading pre-configured characteristics for at least the non-trapezoidal gradient pulse shape, c) Assigning possible parameter values ​​of at least one measurement parameter to be set in the measurement protocol to at least one category indicating the feasibility of the measurement protocol on the magnetic resonance system based on the charged characteristics, d) Entering at least one desired parameter value from at least one measurement parameter of the measurement protocol, taking into account the assignment made, e) If a parameter value has not yet been entered for each measurement parameter to be set in the measurement protocol, with which the measurement protocol can be executed, repeat steps c) and d), at least for measurement parameters for which no parameter value has yet been set, until parameter values ​​have been entered for all measurement parameters to be set in the measurement protocol, with which the measurement protocol can be executed. g) Acquisition of diffusion-weighted measurement data using the measurement protocol with the entered parameter values.

[0025] The inventive loading of prepared characteristics for at least one non-trapezoidal gradient pulse shape of the measurement protocol allows for the rapid assignment of possible parameter values ​​to a category indicating the feasibility of executing the measurement protocol on the magnetic resonance system while adhering to the loaded limit values. This assignment to the at least one category makes it easier for a user to find parameter values ​​for the measurement parameters to be set in a measurement protocol to be used during the planning of a measurement.

[0026] A magnetic resonance system according to the invention comprises a magnet unit, a gradient unit, a radio frequency unit and a control device designed for carrying out a method according to the invention with a monitoring unit.

[0027] A computer program according to the invention implements a method according to the invention on a control device when it is executed on the control device. For example, the computer program includes instructions that, when the program is executed by a control device, e.g., a control device of a magnetic resonance system, cause this control device to execute a method according to the invention. The control device can be in the form of a computer.

[0028] The computer program can also be in the form of a computer program product that can be directly loaded into a memory of a control device, with program code means to execute a method according to the invention when the computer program product is executed in a computing unit of the control device.

[0029] A computer-readable storage medium according to the invention comprises instructions which, when executed by a control device, e.g. a control device of a magnetic resonance system, cause it to execute a method according to the invention.

[0030] The computer-readable storage medium can be designed as an electronically readable data carrier which includes electronically readable control information stored on it, which includes at least one computer program according to the invention and is designed in such a way that, when the data carrier is used in a control unit of a magnetic resonance system, it carries out a method according to the invention.

[0031] The advantages and explanations given regarding the procedure also apply analogously to the magnetic resonance system, the computer program product and the electronically readable data carrier.

[0032] Further advantages and details of the present invention will become apparent from the exemplary embodiments described below and from the drawings. The examples shown do not constitute a limitation of the invention. They show: Fig. 1 a schematic flowchart of a method according to the invention for acquiring diffusion-weighted measurement data of a test object with a magnetic resonance system using a measurement protocol with non-trapezoidal gradient pulse shapes for diffusion coding, Fig. 2 a schematic comparison of different possible excitation and recording methods of echo signals with diffusion gradients for diffusion weighting, Fig. 3 a schematically illustrated magnetic resonance system according to the invention.

[0033] Fig.Figure 1 is a schematic flowchart of a method according to the invention for acquiring diffusion-weighted measurement data of an object under investigation using a magnetic resonance system and a measurement protocol with non-trapezoidal gradient pulse shapes for diffusion coding.

[0034] A measurement protocol MP to be used with measurement parameters mpj to be set, which include a desired non-trapezoidal gradient pulse shape GF, is loaded (Block 101).

[0035] Pre-configured characteristics CGF for at least the desired non-trapezoidal gradient pulse shape GF are loaded (Block 103).

[0036] Possible parameter values ​​PWij of at least one measurement parameter mpj of the measurement protocol MP are assigned to at least one category indicating the executability of the measurement protocol MP on the magnetic resonance system 1 based on the loaded characteristics (Block 105).

[0037] Taking into account the assigned values ​​of the possible parameter values ​​PWij of the measurement parameters mpj of the measurement protocol MP, at least one desired parameter value vPWij is entered (Block 107). This can be done manually or automatically. By considering the assigned values ​​of the possible parameter values ​​PWij of the measurement parameters mpj of the measurement protocol MP to at least one category K1, K2, K3, which indicates the executability of the measurement protocol MP on the magnetic resonance system 1, it can be ensured that the entered parameter values ​​vPWij of the measurement parameters mpj of the measurement protocol are executable on the magnetic resonance system 1.

[0038] A query (100) determines whether a parameter value (vPWij) has already been set for each measurement parameter (mpj) of the measurement protocol (MP) that allows the measurement protocol to be executed. If this is not the case (query 100, n), then, at least for measurement parameters (mpj) for which no parameter value (vPWij) has yet been set (j≠j), possible parameter values ​​(PWij) of at least one measurement parameter (mpj) of the measurement protocol (MP) are assigned to at least one category (K1, K2, K3) indicating the executability of the measurement protocol (MP) on the magnetic resonance system (MRI) 1, based on the loaded characteristics (CGF) (block 105). Entered parameter values ​​(vPWij) can be taken into account, so the assignment may change.Taking into account the (re)assignment of the possible parameter values ​​PWij of the measurement parameters mpj of the measurement protocol MP, at least one desired parameter value vPWij is entered (block 107) until parameter values ​​vPWij have been entered for all measurement parameters mpj of the measurement protocol MP to be set, with which the measurement protocol MP can be executed. Characteristics CGF can include various terms and values ​​relevant to the gradient pulse shape GF, which can be determined in advance, for example, in preliminary calculations, and used in the planning and / or execution of a measurement, which can significantly reduce the computational effort. Examples of such terms and values ​​are described below.

[0039] Using the measurement protocol MP with the entered parameter values ​​vPWij, diffusion-weighted measurement data DWMD are recorded and, if necessary, further processed (Block 109).

[0040] The diffusion-weighted measurement data DWMD to be acquired are in particular non-linear diffusion-weighted measurement data DWMD, for example track-weighted and / or spherical diffusion-weighted measurement data DWMD.

[0041] The at least one non-trapezoidal gradient pulse shape GF can be defined along at least one axis running in a direction of a coordinate system, for example a logical coordinate system of the gradient directions Gx, Gy, Gz, and wherein the at least one non-trapezoidal gradient pulse shape GF is optionally defined with a predetermined scaling factor associated with the direction of the logical coordinate system.

[0042] As explained below, measurement parameters mpj of the measurement protocol MP that can be set may include spoiler gradients, compensation gradients and / or corrected gradient amplitudes.

[0043] A desired non-trapezoidal gradient pulse shape (GF) can be selected from a range of non-trapezoidal gradient pulse shapes and loaded as a description of the respective non-trapezoidal gradient pulse shape, which is decomposed into at least one section. These sections are stored individually or together in separate files for a given gradient pulse shape. The advantages of such a decomposition into sections are described below.

[0044] A zeroth moment of a desired non-trapezoidal gradient pulse shape GF can be checked before and / or during the acquisition of diffusion-weighted measurement data, and if the check reveals a deviation from a target value, a gradient amplitude of the non-trapezoidal gradient pulse shape can be corrected to compensate for the deviation.

[0045] To enable the use of non-trapezoidal gradient pulse shapes (hereinafter also referred to simply as pulse shapes), it may be useful, depending on the measurement protocol used, to allow a user to specify pulse shapes with one or more sections DW1, DW2, DW3. For example, diffusion coding can be interrupted by one or more RF pulses, independent of the actual readout module. For instance, spin-echo coding (SE) can be interrupted by one RF refocusing pulse (RF2), and double-refocused spin-echo coding (DSE) by two RF refocusing pulses (RF2). Stimulated-echo coding (STE) can be interrupted by a "storage" RF pulse (RF3) and a "recovery" RF pulse (RF4). More complex and combined codings with (multiple) spin echoes and (multiple) stimulated echoes are possible.

[0046] In Fig.Figure 2 outlines different possible scenarios, in each case where, after an RF excitation pulse RF1, the generated signals are recorded using a readout module RO. Various well-known recording techniques are possible as the readout module RO. For example, single-shot (ss) EPI, multi-shot (ms) EPI, ss spiral, ms spiral, HASTE, turbo spin echo (TSE), turbo gradient spin echo (TGSE), simple spin echo (SE), or even a simple gradient echo (GRE) recording technique are all possibilities.

[0047] For the sake of simplicity, the following are omitted from the presentation: Fig. Two additional imaging gradients. In one STE variant, gradient pulses between a storage RF pulse RF3 and a recovery RF pulse RF4 have no effect on diffusion coding.

[0048] Depending on the coding scheme, the following mappings regarding the number of sections can be supported, for example: One section: • GRE coding • SE coding (Section #2 is omitted) • DSE encoding (Sections #1 and #3 are omitted) • STE coding (Section #2 is omitted) Two sections: • SE coding • DSE encoding (Section #3 is omitted) • STE coding

[0049] Three sections: • DSE encoding

[0050] The definition of the individual sections can be done – as already known – in the form of a text description. The advantage of such a description is the intuitive understanding of the content and the ease of creation and modification by the user. Ideally, a standardized format (for example, XML) is used for this purpose. However, binary formats are also conceivable, which can be created, for example, with a suitable generation or conversion program.

[0051] It is possible to store each section individually: in this case, for example, a DSE encoding would have three separate descriptions for each section DW1, DW2, DW3. However, the sections can also be contained in a single description that allows for a corresponding assignment.

[0052] The section descriptions can be hard-coded in the program code. Preferably, users should be able to read pulse shapes from an external description. For this purpose, one or more descriptions of related sections can be provided in file form.

[0053] It can be stipulated that the description of a pulse shape includes a defined distance P between each pair of sections DW1, DW2, DW3. This becomes particularly important when the accumulated zero gradient moment of preceding sections is non-zero. In this case, a distance P between two sections DW1, DW2, DW3 influences the shape of the diffusion encoding, so that this can be influenced by specifying the distance P.

[0054] The actual description of the pulse shape can be performed piecewise and constantly. In this case, the start and end times of a segment are specified, as well as the constant amplitude of the three gradient axes within that segment. The times can be defined in a fixed grid. This grid can be defined by the measurement system, for example, by a gradient control unit 5'.

[0055] Alternatively, as already known, a piecewise linear description of the pulse shape can be used. In this case, the times for the beginning and end of a linear segment are specified, as well as the respective amplitudes.

[0056] The amplitudes G(t) of the pulse waveform can be specified in absolute values ​​in both cases (typically in units of mT / m); however, this already predetermines the diffusion weighting (b-value). Preferably, the amplitudes are defined with relative, e.g., normalized values ​​(for example, in an interval from -1 to +1), so that the actual scaling to the required absolute amplitudes can later be performed based on the requested b-value.

[0057] The amplitudes G(t) can refer – as already known – to "physical" or "logical" coordinates. "Physical" coordinates refer to the three axes of the gradient coil (x, y, z) and have the advantage that, regardless of the acquisition orientation, maximum performance (especially maximum amplitude) can be used on each gradient axis. "Logical" coordinates refer to the three axes of image coding (readout coding, phase coding, layer coding) and have the advantage that, regardless of the acquisition orientation, a unique relationship between diffusion coding and the acquired layer is ensured.

[0058] The description can contain a comment field that will be displayed to the user as a hint during later measurement planning.

[0059] The following are some exemplary descriptions of possible non-trapezoidal pulse shapes:

[0060] At the start of measurement planning, a complete set of descriptions of available pulse shapes can be imported. These can be hardcoded in the program, as mentioned, or stored in specific files (or in a specific directory). During measurement planning, the various pulse shapes can then be offered to the user for selection in a "selection menu".

[0061] Ideally, during measurement planning, the user should have the option to import the desired pulse shape(s) from a description file using standard procedures (e.g., an "Open File" dialog). Advantageously, at this stage, for example with query 100, the validity of the pulse shape description is checked (and any formatting errors are displayed), the compatibility of the pulse shape with the current measurement protocol MP is verified, dependent measurement parameters are adjusted if necessary or useful, and / or pre-calculations are performed for rapid measurement planning and execution.

[0062] When reading a pulse shape from a description, it's important to note that the accuracy of setting desired amplitudes may be limited. For example, a DAC converter in a magnetic resonance system can only represent a limited number of values. It's also conceivable that a data transport layer in the measurement system, for technical reasons (such as limited data packet size and propagation time performance), supports lower accuracy in representing amplitude values. If the amplitude values ​​of the pulse shapes are defined with high accuracy (e.g., double floating-point precision) but applied with reduced accuracy (e.g., single floating-point or fixed-point precision), the resulting gradient moment M can be inaccurate. x / y / z = G ∫dt f x / y / z (t) deviates from the target value, which immediately leads to a loss of signal.

[0063] Currently, such deviations are calculated and compensated for as much as possible using a separate so-called "balance" gradient. The disadvantage of this approach is that this additional "balance" gradient takes time, and the actual diffusion coding deviates from the desired value.

[0064] In one embodiment, to ensure the preservation of the axis-specific zero moments after receiving (Block 102) a description B(GF) of a non-trapezoidal pulse shape, the following basic idea can be used: 1. Read current target amplitudes A(GF) with high precision (Block 104). 2. From this, determine the current target torques M(GF) for each axis with high precision (Block 106). 3. Convert target amplitudes A(GF) into representable actual values ​​a(GF) with low precision (Block 108). 4. Determine the resulting actual moments m(GF) with high precision (Block 110). 5. Determine the deviation D(GF) from target moments M(GF) and actual moments m(GF) with high precision (Block 112). 6. Take into account the accumulated deviation D(GF) and, if necessary, correct actual values ​​m(GF), a(GF) so that the actual moments m(GF) correspond as closely as possible to the target moments M(GF) (Block 114).

[0065] When reading in the pulse shape, it can be simultaneously taken into account that some measurement systems require a description of the gradient pulse on a defined time grid for operation. For example, it may be necessary to convert the pulse shape to a fixed grid of 1 µs, 10 µs, or 50 µs. According to the invention, even during this conversion—which can optionally be performed in a single step with the reading of the piecewise linear or piecewise constant description—axis-specific zero moments are preserved with high precision. In practice, this can, for example, lead to the actual amplitude fluctuating over time in a section with a constant target amplitude (a behavior also known as "jitter"). In this way, the target moment is realized with high precision in the time integral.

[0066] The following pseudo-code demonstrates, as an example for one of the three axes, a reading and a grid R used for piecewise constant sections: T[i] are the support points (i.e., the start and end times of an interval; in this example, T[i] = k i * T Raster with k i ∈ N0), G[i] are the target amplitudes with high precision, g[i] are the actual values ​​with low precision, T Raster This is the time grid defined by the measurement system. The first for loop iterates over the piecewise constant sections of the description, and the second for loop iterates over the grid points within the currently processed section:

[0067] Provided that the piecewise constant description is consistently available on the fixed hardware raster R (e.g. with a raster time of 10µs as in the xml example above), the pseudocode (again using one of the three axes as an example) simplifies as follows:

[0068] Such an algorithm can not only be used, as just described, for moment-preserving reading of non-trapezoidal pulse shapes in measurement preparation, but can also be used to preserve the zero moments of the pulse shapes, e.g. after applying corrections and transformations, during the measurement process.

[0069] Piecewise linear sections (G[i] are the support setpoints with high precision, g[i] the actual values ​​with low precision; predefined grid for g[i]) can be converted as follows (again as an example for one of the three axes):

[0070] For control purposes, it may be provided that the user can export the converted raster data in a suitable form (for example, as an XML file in raster format).

[0071] Furthermore, it can be implemented to check the validity of the input pulse shapes, e.g., using query 100. In addition to purely semantic checks (text and number formatting) and a validation of parameter ranges (amplitude and time data), the usability with at least one of the encodings (GRE, SE, DSE, STE) can be verified. Usability is possible if the time integral of the zero gradient moment on each axis, taking into account the respective signal path, is nearly zero. - GRE: ∫dt f #1 (t) = 0 - SE, STE: ∫dt f #1 (t) - ∫dt f #2 (t) = 0 - DSE: ∫dt f #1 (t) - ∫dt f #2 (t) + ∫dt f #3 (t) = 0

[0072] To uniquely link a measurement protocol (MP) with a pulse shape description, a checksum can be calculated and saved with the measurement protocol. This allows for a simple way to ensure the validity of a created measurement protocol: if the corresponding description is changed after the pulse shape has been read (for example, by modifying the associated text file), this deviation is detected by comparing the checksums. The execution of the now inconsistently parameterized measurement can then be refused. While it would theoretically be possible to achieve a consistency check by storing the complete pulse shape in the measurement protocol, the required data volume becomes immense for extensive pulse shapes.

[0073] The checksum can, for example, take into account the number of sections, the time and amplitude data for the three axes, and any planned pauses between sections. A standardized checksum calculation can be performed (e.g., according to CRC-32).

[0074] Depending on the selected pulse shape (GF), it may be useful or necessary to adjust or modify other measurement parameters (mpj) to enable the measurement or improve image quality. The selection of this pulse shape (GF) would then be a "conditionally adjustable parameter value".

[0075] After reading in a pulse shape GF, a suitable coding scheme can be automatically set. "Suitable" means that a) the number of sections is compatible with the scheme and b) the time integral of the zeroth moment is compatible with the scheme. For example, for a description of a pulse shape with two sections DW1, DW2 and ∫dt f#1 (t) - ∫dt f #2 If (t) = 0, spin echo (SE) encoding can be automatically activated.

[0076] After reading in a pulse shape GF, a suitable direction assignment for a desired diffusion direction can be set automatically, either additionally or alternatively. For example, the direction assignment can be set that applies the pulse shape in its native form G'(t) = RG(t) with the rotation matrix R = 1.

[0077] After reading in a pulse shape GF with a predefined distance P between two sections DW1, DW2, DW3, it can be checked whether this distance can actually be achieved. If not, a mode can be automatically activated in which the distances P of sections DW1, DW2, DW3 can be freely set by a user, deviating from the originally predefined value. The distances P can, for example, be automatically determined in such a way that they are compatible with the current measurement protocol MP.

[0078] For example, in spin-echo (SE) coding, the distance P between section DW1 and section DW2 may be limited by the duration P of the refocusing module in which the RF refocusing pulse RF2 is injected.

[0079] After reading in a pulse shape GF, it can be further checked whether the zero moment ("implicit spoil moment") realized by the diffusion gradients of individual sections DW1, DW2, DW3 lies above a threshold value for all b-values ​​≠ 0 set in the protocol. If this is not the case, additional spoiler gradients can be automatically activated in the measurement protocol to reduce signal contributions from unwanted signal paths (e.g., signal paths of a free induction decay (FID)). The determination of such spoiler gradients is described, for example, in US10557909B2.

[0080] In Fig. Figure 2 shows spoiler gradients Sp for SE coding SE* without diffusion coding, i.e., for b=0, which can be switched before and after the RF refocusing pulse RF2. Similarly, spoiler gradients Sp for STE coding STE* without diffusion coding, i.e., for b=0, are also shown.

[0081] Alternatively, from the measurement protocol MP, those b-values ​​≠ 0 can be discarded where the zero moment realized by the diffusion gradients lies below the threshold value. For example, measurements with b = 0, 200, 400, 700, 1000 s / mm 2 provided for, and the threshold value only at b = 500s / mm 2 Once this value is reached, the number of b-values ​​can be automatically reduced to 3, so that only b = 0, 700, 1000s / mm 2 to be preserved.

[0082] The threshold can, for example, be defined as a multiple of a spoiler moment M. SP It will be determined which leads to a dephasing of the magnetization of 2π within a voxel.

[0083] The determination and use of spoiler moments can be used not only to determine and describe parameter dependencies in measurement planning, but also for targeted (de)activation of spoiler gradients during measurement execution.

[0084] Ideally, the system automatically informs the user about the changes made.

[0085] During subsequent measurement planning (i.e., processing the measurement protocol to find usable parameter values), the dependencies can be further checked and taken into account. In this process, "non-adjustable parameter values," "conditionally adjustable parameter values," and "unrestricted adjustable parameter values" can be displayed to the user in a suitable form (e.g., color-coded or with other labeling).

[0086] For example, a user can switch between SE coding and STE coding during measurement planning. The necessary requirements for the pulse shape GF are identical (one or two sections, ∫dt f). #1 (t) - ∫dt f #2(t) = 0). As another example, the user can switch between "flexible" and "predefined" distances P between sections DW1, DW2, DW3, provided the current measurement protocol MP allows it (e.g., after the user has selected a shorter refocusing module). Furthermore, the user could deactivate additional spoil gradients once the b-values ​​set in the measurement protocol no longer necessitate them.

[0087] Furthermore, it may be possible to provide the user with information about current time intervals during measurement planning – for example, in the form of a help text. This enables the user to optimize pulse shapes for a specific measurement protocol. The time intervals may include, for example, a maximum available duration for each section DW1, DW2, DW3, an actual duration for each section DW1, DW2, DW3, a minimum required duration of the distance P between two sections DW1, DW2, DW3, and / or an actual duration of the distance P between two sections DW1, DW2, DW3.

[0088] In general, it can also be planned to apply different gradient pulse shapes (GF) sequentially (or interleaved) in a single measurement. This is relevant, for example, when differences in diffusion contrast that depend on the pulse shape are to be determined and displayed. For instance, a first pulse shape with several diffusion weights (b-values) can be measured, and an initial ADC can be determined from this. This process is then repeated within the same measurement for a second pulse shape. An absolute (aADC = (ADC2 - ADC1)) or relative (rADC = (ADC2 - ADC1) / ADC1) deviation can thus be determined from the data and displayed as a map. Such a procedure is described, for example, in US10006979B2. The advantage of recording both pulse shapes in one measurement is the high consistency of the data (e.g., with regard to patient movement or adjustment or calibration data).

[0089] Further gradient pulse shapes GF, with their sections DW1, DW2, DW3, can be stored in separate files or – preferably – in a single file with a correspondingly extended format. For example, (in both piecewise constant and piecewise linear definitions) the data lines can be extended accordingly, e.g., in the following form:

[0090] It is possible to apply different rotations and / or scalings of the gradient pulse shape sequentially (or in a nested fashion) within a single measurement. This is relevant, for example, when a diffusion tensor is to be determined and displayed from the measurement.

[0091] Scaling the pulse shape GF, which is accompanied by a modified b-value for this implementation, is possible, for example, with a scaling matrix S = ((S, 0, 0), (0, S, 0), (0, 0, S)), such that: f'(t)=S_f(t)f(t)=(fx(t),fy(t),fz(t))

[0092] A rotation of the pulse shape GF, which is accompanied by a rotation of the b-matrix that is modified for this implementation, is possible with a rotation matrix R such that: f'(t)=R_f(t)f(t)=(fx(t),fy(t),fz(t)).

[0093] A rotation in three-dimensional space can be uniquely defined using three values ​​– for example, the Euler angles. Experts know how to create a rotation matrix from these values. Scaling and rotation can be specified together using a transformation matrix T = RS = SR.

[0094] For example, the elements of all transformation matrices to be used during the measurement can be specified by the user. Here again, fixed encodings (e.g., with user selection) or specification and import via a suitable file format, such as XML, are conceivable.

[0095] The specification can be limited to the minimum necessary number of parameters, for example, three Euler angles (and possibly a scaling parameter). Alternatively, the specification can include all nine (partially redundant) elements of the transformation matrix: this has the advantage of an intuitive understanding of the rotation. From the specified direction vectors of the rotated axes in the original coordinate system (x' = (x x , x y , x z ), y' = (y x , y y , y z ), z' = (z x , z y , z z )) and the additional scaling S results in the transformation matrix T = S ((x x , y x , z x ), (x y , y y , z y ), (x z , y z , z z )). Here, |x'| = |y'| = |z'| = 1 (normalization of the axes) and x' y' = x' z' = y' z' = 0 (pairwise perpendicular axes).

[0096] Instead of describing different non-trapezoidal pulse shapes on the three axes, it may be possible to use the same non-trapezoidal pulse shapes with different scales G(t) = G f(t) (f x , f y , f z ) or (preferably) a non-trapezoidal pulse shape on one axis, for example on the x-axis G(t) = G (f x (t), 0, 0). While this approach does not allow for the realization of the complex diffusion codings described at the beginning—for example, a spherical tensor coding—it nevertheless offers access to new (e.g., time-dependent) diffusion properties and, in particular, permits the combination of non-trapezoidal pulse shapes with "diffusion directions" in the conventional sense. The diffusion directions are defined by a set of N vectors V. n = (v n,x , v n,y , v n,z ) described, and the pulse shapes to be applied successively in the measurement result from Gn (t) = G f(t) V n . In this way, for example, conventional diffusion measurements with the determination of trace images or diffusion tensor imaging (DTI) with the determination of maps of the tensor properties (in addition to the trace, for example anisotropy measures such as the fractional anisotropy FA or the radial anisotropy RA) can be carried out as is known in the prior art.

[0097] The definition of a gradient pulse shape (GF) can then be limited to listing the support points for one axis. The description of the pulse shape (GF) can contain information that allows the import procedure to recognize a "single-axis mode" and process the data accordingly.

[0098] As mentioned, when performing diffusion measurements with more than one RF pulse, it is important to note that in addition to a desired signal path (e.g., a spin echo, SE), a multitude of unwanted signal paths (e.g., free induction decays, FID) are generated. Complex diffusion encodings with three or more RF pulses (e.g., double-refocused spin echoes (DSE)) therefore rarely work without dedicated sequences of additional spoiler gradients, as described, for example, in US9952302B2. However, with simple diffusion encodings—frequently used in clinical practice—unwanted signal paths can be suppressed so effectively by the diffusion gradients themselves that no additional spoiler gradients are necessary, thus making the measurement more efficient. Only below a certain amplitude of the diffusion gradients—for example, in a measurement with a b-value of 0—is the implicit suppression insufficient.In this case, spoiler gradients Sp are used instead of diffusion gradients.

[0099] These simple experiments include spin-echo (SE) and stimulated-echo (STE) codings, as in the Fig. 2, in particular represented by the SE* and STE* lines without diffusion coding.

[0100] In SE coding, the spin echo signal path is desired, but the two FID signal paths generated after RF pulses RF1 and RF2 are not. In STE coding, the stimulated echo signal is desired, but the three FID signal paths generated after RF pulses RF1, RF3, and RF4, the three spin echo signal paths generated by combinations of any two of the three RF pulses RF1, RF2, and RF3, and the double spin echo signal path generated by the three RF pulses RF1, RF2, and RF3 are not.

[0101] As previously described, a spoil moment M above a predefined limit M can be used as a measure of "sufficient" suppression of unwanted signal paths. SP required. For an example with M SP = a / (γP) with the gyromagnetic ratio γ, the pixel size P and a factor a, with a = 2, γ = 2π * 42.575MHz / T and P = 1mm, for example, M SP ≈ 7.5mT / m ms.

[0102] It is known that for an SE experiment, the implicit suppression of unwanted signal pathways is sufficient if the zero moment of the diffusion gradients is both in section #1 (M D,1 ) as well as in section #2 (M D,2 ) above the limit: |M D,1 | ≥ M SP , |M D,2 | ≥ M SP According to the invention, normalized zero moments of the non-trapezoidal pulse shapes G are used in the measurement preparation. i (t) = G f i (t) pre-calculated and stored: m D,i = ∫dt f i(t). During the measurement process, it can be checked for the diffusion weighting (b-value) currently being measured, e.g. by means of query 100, whether the condition |G m is met for each section DW1, DW2. D,i | ≥ M SP The condition is met. If the condition is met, the intended non-trapezoidal pulse shape GF can be used. If the condition is not met, a pair of spoiler gradients Sp with moment M can be used instead. SP be used.

[0103] The magnitude of the implicit spoil moment on the three gradient axes |m is preferred when testing the condition. D,i | = √(m T D,i m D,i ) is used because - depending on the pulse shape - one or two components of f i (t) = (f x,i (t), f y,i (t), f z,j (t)) can be zero.

[0104] With a sufficiently dense grid of support points R, the integrals simplify to summations over values ​​that are constant during a grid interval.

[0105] It is also known that for an STE experiment, the implicit suppression of unwanted signal pathways is sufficient if the zero moment of the diffusion gradients is both in section #1 (M D,1 ) as well as in section #2 (M D,2 ) above a limit: |M D,1 | ≥ 2 M SP , |M D,2 | ≥ 2 M SP and simultaneously between RF storage and RF recovery with an additional spoiler gradient, a 0 moment |M Z | ≥ M SP is being implemented.

[0106] Here too, normalized zero moments of the non-trapezoidal pulse shapes G are used in the measurement preparation. i (t) = G f i (t) pre-calculated and stored: m D,i = ∫dt f i(t). During the measurement process, it can be checked for the diffusion weighting (b-value) currently being measured whether the condition |G m is met for each section. D,i | ≥ 2 M SP The condition is met. If this is the case, the intended non-trapezoidal pulse shape GF can be used. However, if the condition is not met, a pair of spoiler gradients Sp with a moment M can be used instead. SP be used.

[0107] Additionally, it may be possible to include a previously described additional spoiler gradient with a moment |M in any case. Z | = M SPto apply. As in the SE experiment, the magnitude of the implicit spoil moments can be used when checking the condition. Such pre-calculated normalized spoil moments can be encompassed by loaded characteristics, so that during measurement planning and / or execution, a determination of the implicit spoil moments based on the normalized pre-calculation can be carried out particularly quickly.

[0108] “Maxwell terms” are those that arise when switching “linear” gradient fields along the main field axis B. z (r) = G x x + G y y + G z z inevitably occurring transverse components B x (r) = G x (z + z 0x ) - α G z (x + x0) and B y (r) = G y (z + z 0y ) - (1-α) G z(y + y0). Depending on the position r, these can impose an additional phase on the signal, leading, for example, to unwanted signal loss due to dephasing within a voxel. The values ​​z 0x , z 0y x0, y0, and α depend on the design of the gradient coils. For "symmetrical" gradient coils commonly used in whole-body scanners, x0 = y0 = z. 0x = z 0y = 0 and α = ½.

[0109] It is known and common practice to approximate a quantity of the magnetic field relevant for MR physics by series expansion in order to make corrections on this basis, where: |B|≈B00th+B01st+B02nd with zero-order terms (“B0-like”), first-order terms (“gradient-like”) and second-order terms: B00th=B0+Gx2z0x2 / 2B0+Gy2z0y2 / 2B0+α2Gz2x02 / 2B0+(1−α)2Gz2y02 / 2B0 −α GxGzx0z0x / B0−(1−α)GyGzy0z0y / B0 B01st=Gxx+Gyy+Gzz+Gx2z z0x / B0+Gy2z z0y / B0+α2Gz2x x0 / B0+(1−α)2Gz2y y0 / B0 −α GxGz(x z0x+z x0) / B0−(1−α)GyGz(y z0y+z y0) / B0 B02nd=Gx2z2 / 2B0+Gy2z2 / 2B0+α2Gz2x2 / 2B0+(1−α)2Gz2y y0 / B0 −2α GxGz×z / 2B0−2(1−α)GyGzyz / 2B0

[0110] For the special case of a symmetrical gradient coil, this simplifies to: B00th=B0 B01st=Gxx+Gyy+Gzz B02nd=Gx2z2 / 2B0+Gy2z2 / 2B0+Gz2x2 / 8B0+GxGzx z / 2B0−GyGzy z / 2B0.

[0111] It is known how deviations from an idealized magnetic field without Maxwell terms (|B) ideal | = B0 + G x x + G y y + G z z) can be compensated. For asymmetric gradient coils, for example, additional gradient fields are switched on that specifically compensate for the first-order deviations: ΔGx=α2Gz2x0 / B0−αGxGzz0x / B0 ΔGy=(1−α)2Gz2y y0 / B0−(1−α)GyGzz0y / B0 ΔGz=Gx2z0x / B0+Gy2z0y / B0−αGxGzx0 / B0−(1−α)GyGzy0 / B0

[0112] For both symmetrical and asymmetrical gradient coils, the second-order deviations can also be at least partially compensated by adjusting them locally (e.g., around the center of a layer with coordinates r). M = (x m , y m , zM )) developed up to first order and then compensated again with additional gradient fields: ΔGx(rM)=α2Gz2xM / B0−αGxGzzM / B0 ΔGy(rM)=(1−α)2Gz2yM / B0−(1−α)GyGzzM / B0 ΔGz(rM)=Gx2zM / B0+Gy2zM / B0−αGxGzxM / B0−(1−α)GyGzyM / B0

[0113] These compensations are not limited to trapezoidal pulse shapes: for non-trapezoidal pulse shapes, the course of the compensation gradient G is calculated C for imaging around a reference point r M train c (t, r M ) = G 2 f c (t, rM ) = G 2 (f C,x (t, r M ), f C,y (t, r M ), f C,z (t, r M )) with fC,x(t,rM)=−(α2fz(t)2(x0+xM) / B0−αfx(t)fz(t)(z0x+zM) / B0) fC,y(t,rM)=−((1−α)2fz(t)2(y0+yM) / B0−(1−α)fy(t)fz(t)(z0y+zM) / B0) fC,z(t,rM)=−(fx(t)2(z0x+zM) / B0+fy(t)2(z0y+zM) / B0−αfx(t)fz(t)(x0+xM) / B0−(1−α)fy(t)fz(t)(y0+yM) / B0)

[0114] Since these curves depend on the location of the reference point (and possibly on additional rotations of the pulse shape), pre-calculations are only possible to a limited extent. In principle, all locations (and rotations) would have to be known at the time of measurement preparation, and the complete range of compensation pulse shapes would have to be calculated and stored. Alternatively, the compensation pulse shapes are calculated in real time during the measurement.

[0115] However, the accuracy of adjusting desired amplitudes may be limited for non-trapezoidal pulse shapes. For example, the DAC converter can only represent a limited number of values. It is also conceivable that a data transport layer in the measurement system, for technical reasons (e.g., limited data packet size and propagation time performance), only supports lower accuracy in representing amplitude values. If the amplitude values ​​of the compensation gradients are defined with high accuracy (e.g., double floating-point precision) but applied with reduced accuracy (e.g., single floating-point or fixed-point precision), the resulting gradient moment M C,x / y / z = G ∫dt f C,x / y / z (t) deviates from the target value, which immediately leads to a loss of signal. A solution to this problem is described below.

[0116] First, however, a new approach will be described that enables rapid pre-calculation for performing additional layer-specific Maxwell corrections for non-trapezoidal pulse shapes in simultaneous multi-layer imaging (SMS).

[0117] For example, US10613175B2 describes methods using SMS for layer-specific Maxwell corrections for trapezoidal pulse shapes, which proceed according to the following scheme: 1. Determining a reference point (for example, the center point r) SMS all simultaneously excited layers) 2. Determination of compensation gradients for this reference point (for example, G) C (t, r SMS ) = G 2 (f C,x (t, r SMS ), f C,y (t, r SMS ), f C,Z (t, r SMS ))) 3. Determination of a layer-specific additional compensation moment along the layer axis for each of the simultaneously excited layers with a reference point at r s : MC,s=(T_ G2∫dt(fC(t,rs)−fC(t,rSMS)))eS Here, the rotation matrix T is rotated from physical coordinates (x, y, z) into logical imaging coordinates (frequency coding, phase coding, layer coding), and e s is the unit vector along the direction of the layer coding. 4. Determination of time shifts of simultaneously applied RF pulses for the individual layer excitations such that the compensation moment M is determined for layer s. C,s is generated 5. Performing the measurement with the compensation gradients G C (t, r SMS ) and the calculated time shifts.

[0118] The necessary calculations are time-consuming. It is proposed to accelerate this procedure by including in step 3 (determining a layer-specific additional compensation moment along the layer axis for each of the simultaneously excited layers with a reference point at r) s Pre-calculated values ​​are used. This makes it easier to meet the real-time requirements of the measurement process.

[0119] The crucial insight here is that all compensation gradients can be represented as a summation of terms with the same structure (characteristics): GC(t,r)=G2(fC,x(t,r),fC,y(t,r),fC,z(t,r)) fC,x / y / z(t,r)=∑kak,x / y / z(rl(k,x / y / z)+vk,x / y / z)fi(k,x / y / z(t) fj(k,x / y / z)(t)

[0120] The sum is calculated over all terms k = 1 ... K, a k,x / y / zare index- and axis-dependent constants, and I(k,x / y / z), i(k,x / y / z), j(k,x / y / z) assign a direction to each index for each axis of the compensation gradient. k,x / y / z represents a shift (relevant only for asymmetric gradient coils). For example, for f C,x (t, r): k=1:a1,x=−α2 / B0v1,x=x0l(1,x)=xi(1,x)=zj(1,x)=z→−α2 / B0(x+x0)fz(t)fz(t) k=2:a2,x=α / B0v2,x=z0xl(2,x)=zi(2,x)=xj(2,x)=z→α / B0(z+z0x)fx(t)fz(t)

[0121] These considerations allow for the one-time calculation of the integrals of normalized products of form F during measurement preparation for non-trapezoidal pulse shapes. ij = ∫dt f i (t) f j (t) with i, j ∈ {x, y, z} to determine and, for example, encompassed by the loaded characteristics, to be available during a measurement procedure for a fast calculation of the layer-specific compensation moments, where one can write: MC,s=(T_ G2∫dt(fC(t,rs)−fC(t,rSMS)))eS=(T_ G∫dt fC(t,rs−rSMS))eS∑kak,x((r−rSMS)l(k,x)+vk,x)fi(k,x)(t)fj(k,x)(t))=(T_ G2∫dt (∑kak,y((r−rSMS)l(k,y)+vk,y)fi(k,y)(t)fj(k,y)(t)))eS∑kak,z((r−rSMS)l(k,z )+vk,z)fi(k,z)(t)fj(k,z)(t))∑kak,x((r−rSMS)l(k,x)+vk,x)Fi(k,x)j(k,x))=(T_ G2(∑kak,y((r−rSMS)l(k,y)+vk,y)Fi(k,y),j(k,y)))eS∑kak,z((r−rSMS)l(k,z)+vk,z)Fi(k,z)j(k,z)

[0122] The crucial point is replacing the time integrals with the pre-calculated values ​​F, e.g., those encompassed by the characteristics CGF. ijThe time-consuming pre-calculation can be performed without any time pressure during measurement preparation. In step 4 (determination of time shifts of simultaneously applied RF pulses for the individual layer excitations), the compensation moments determined in this way along the layer axis can be realized in a known manner by individual time shifts of the simultaneously applied RF pulses for the individual layer excitations.

[0123] As already explained, when implementing non-trapezoidal diffusion gradients, it is crucial that the specified zero gradient moment M = G ∫dt f(t) (possibly taking into account compensation gradients, i.e., M(r)) is present on each axis. M ) = G ∫dt f(t) - G 2 ∫dt f C (t, r M)) is implemented with high accuracy. It should be noted, however, that the accuracy of setting desired amplitudes may be limited. For example, the DAC converter can only represent a limited number of values. It is also conceivable that a data transport layer in the measurement system, for technical reasons (e.g., limited data packet size and propagation time performance), only supports lower accuracy in representing amplitude values. If the amplitude values ​​of the non-trapezoidal diffusion or compensation gradients are defined with high accuracy (e.g., with double floating-point precision) but applied with reduced accuracy (e.g., single floating-point or fixed-point precision), the resulting zero gradient moment may deviate from the target value, which directly leads to signal loss.

[0124] To effectively counteract this problem, the following preliminary calculations are proposed: 1. If the non-trapezoidal pulse shapes are defined in logical imaging coordinates (frequency coding, phase coding, slice coding): Rotation into physical coordinates (x, y, z) with high precision, G'(t) = TG(t). 2. If an additional rotation according to a direction assignment (analogous to direction vectors in conventional diffusion imaging) and / or a measurement with a specific b-value is planned: Rotation and scaling of the pulse shapes according to direction assignment and b-value with high precision, G"(t) = SR G'(t). 3. If Maxwell corrections are to be performed: Calculation of compensation gradients in physical coordinates G C (t) with high precision (based on G"(t)). 4. Determination of a corrected pulse shape G'''(t) = G"(t) - G C (t) with high precision 5. If the corrected pulse shape is applied by the measurement system in logical imaging coordinates: high-precision rotation, G'''(t) = T -1 G'''(t). 6. Moment-preserving transformation into a low-precision representation g'''(t), for example according to an algorithm analogous to one already mentioned above with reference to moment-preserving reading of non-trapezoidal pulse shapes, which can be expressed in pseudo-code as follows: (pseudo-code; G[i] are the target amplitudes with high precision, g[i] are the actual values ​​with low precision), assuming a fixed hardware time grid T Raster :

[0125] The order of the steps in the pseudocode can vary; for example, step 2 can be executed before step 1. Steps can be combined; for example, steps 1 and 2 can be combined (M = SRT or M' = STR).

[0126] In summary, the presented method offers various advantages, allowing different problems, such as "spoiling" (relevant for SE and STE encodings) and "Maxwell compensation" and "moment-preserving transformation" (both relevant for arbitrary encodings), to be considered independently.

[0127] In particular, preliminary calculations can be performed before a measurement is carried out to acquire diffusion-weighted measurement data, especially to determine the values ​​described above. m D,i = ∫dt f i (t) for each section DW1, DW2, DW3 of a pulse shape GF, and F D,i = ∫dt f i (t) f T i (t) for each section DW1, DW2, DW3 of a pulse shape GF, which can each be encompassed by charged characteristics of the pulse shape GF.

[0128] During a measurement procedure, the characteristics can be used to determine the results. a. The values ​​of |M D,i | = |G m D,i| the current amplitude G, which correlates with the achieved b-value, possibly including consideration of the current rotation, can be calculated quickly and easily, b. A spoiler moment |M D,1 | ≥ c M SP , |M D,2 | ≥ c M SP (SE: c=1, STE: c=2) can be checked. If the checked condition is not met, the following can happen: ▪ Deactivation of non-trapezoidal diffusion gradients, ▪ (SMS: Deactivation of temporal HF pulse shifts), ▪ Activation of spoiler gradient pairs, ▪ (STE: Activation of the additional spoiler gradient), ▪ Continue at d (applying the gradients). If the condition is met: • Disabling spoiler gradient pairs • Activation of non-trapezoidal diffusion gradients (possibly including consideration of the current rotation) • Determination of Maxwell compensation gradients ◯ Without SMS: Reference position = Shift position ◯ With SMS: Reference position = middle position of the SMS layers ◯ Calculation of the non-trapezoidal compensation gradients (possibly including consideration of the current rotation) • (SMS: Determination of Maxwell compensation moments) • (SMS: Calculation of time-dependent RF pulse shifts for the shifts) • (STE: Activation of the additional spoiler gradient) • Transformation of the non-trapezoidal pulse shape ◯ Addition of the compensation gradients to the diffusion gradients ◯ Moment-preserving transformation of the corrected diffusion gradients (possibly after taking a coordinate transformation into account) c. (SMS: Calculation of HR pulses with or without time shifts) d. Applying the activated gradients and the RF pulses

[0129] When determining implicit spoil moments, rotations of the non-trapezoidal gradient pulses (coordinate systems and / or direction assignments) can be taken into account. The rotated pulse shapes are obtained with a rotation matrix R as G' i (t) =G f' i (t) = RG' i (t) = GR f i (t). The normalized zero moments in rotated coordinates result from the pre-calculated moments of the original coordinates m. D,i = ∫dt f i (t) according to m' D,i = ∫dt f' i (t) = ∫dt R f i (t) = R ∫dt f i (t) = R m D,i . If the magnitude of the implicit spoil moment is used to check the spoil condition, then |m' D,i | = √(m' T D,i m' D,i ) = √(RM D,i ) T (R m D,i ) = √m T D,i R T R m D,i = |m D,iThe magnitude of the spoil moment is invariant under rotations (for rotation matrices, R holds). T = R -1 , also R T R = 1).

[0130] Maxwell field terms can be calculated and locally compensated in the described manner without restriction for symmetrical or asymmetrical gradient coils. The calculations and compensations can also be performed for magnets with a horizontal orientation of the main field (parallel to the patient's body axis) and with a vertical orientation of the main field (perpendicular to the patient's body axis). For vertical field magnets (where, for example, by convention the y-axis of the gradient coil runs along the direction of the main field), only the z- and y-coordinates need to be interchanged in the calculations.

[0131] Rotations of the non-trapezoidal gradient pulses (for different directional assignments, analogous to the diffusion directions of conventional diffusion imaging) can be taken into account when determining layer-specific compensation moments as follows. The pulse shape rotated with a rotation matrix R is given by G'(t) = G f'(t) = R G'(t) = GR f(t). The calculation of the corresponding compensation moments M' C,s = (TG 2 ∫dt (f' C (t, r s ) - f' C (t, r SMS ))) e s Ultimately, this requires the integrals of normalized products of the form F' ij = ∫dt f' i (t) f' j (t) with i, j ∈ {x, y, z}. The auxiliary matrix H'(t) = f'(t) f' T (t) contains in its nine elements H' ij (t) all relevant products f' i (t) f' j (t) and allows the calculation of F' = ∫dt H'(t). With H'(t) = f'(t) f' T (t) = (R f(t))( R f(t)) T = R f(t) fT (t) R T = RH(t) R T it follows that F' = R ∫dt H(t) R T = RFR T It is therefore sufficient to use the pre-calculated values ​​F during the measurement process. ij to rotate in a matrix representation to obtain the required values ​​F' ij to determine the rotated pulse shapes.

[0132] For identical, non-trapezoidal gradient pulse shapes GF on all axes G n (t) = G f(t) V n with the direction vectors V n = (v n,x , v n,y , v n,z The elements of the b-matrix can be determined in the same way. A rotation matrix R can be used. n are determined in such a way that they form a single-axis gradient pulse shape assumed in the pre-calculation, for example on the Gx-axis: G x (t) = G f(t) (1, 0, 0), just transformed into the desired direction assignment of the diffusion direction: Gn(t)=R_nGx(t),with Rn=(vn,x,vn,y,vn,x0,0,00,0,0)

[0133] Fig. Figure 3 schematically represents a magnetic resonance system 1 according to the invention. This comprises a magnet unit 3 for generating the basic magnetic field, a gradient unit 5 for generating the gradient fields, a radio frequency unit 7 for irradiating and receiving radio frequency signals, and a control device 9 designed for carrying out a method according to the invention.

[0134] In the Fig.Figure 3 shows only a rough schematic representation of these subunits of the magnetic resonance system 1. The high-frequency unit 7 can consist of several subunits and, for example, comprise several coils. In particular, the high-frequency unit 7 can comprise a body coil that is permanently integrated into the magnetic resonance system 1 and, in turn, can comprise, for example, two antenna elements 7.1 and 7.2. Furthermore, the high-frequency unit 7 can comprise one or more different local coils 7*, which can be designed either only for transmitting high-frequency signals or only for receiving the triggered high-frequency signals, or for both, and which themselves can comprise several antenna elements and associated coil channels.

[0135] To examine a test object U, for example a patient or a phantom, it can be placed on a table L in the magnetic resonance imaging (MRI) system 1 within its measurement volume. Layers S1 or S2 represent exemplary target volumes of the test object, from which echo signals can be recorded and acquired as measurement data.

[0136] The control unit 9 serves to control the magnetic resonance system 1 and can, in particular, control the gradient unit 5 by means of a gradient controller 5' and the radio frequency unit 7 by means of a radio frequency transmit / receive controller 7'. The radio frequency unit 7 can comprise several channels on which signals can be transmitted or received.

[0137] The high-frequency unit 7, together with its high-frequency transmit / receive control 7', is responsible for generating and transmitting a high-frequency alternating field to manipulate the spins in a region to be manipulated (for example, in layers S to be measured) of the object under investigation U. The center frequency of the high-frequency alternating field, also referred to as the B1 field, is generally set as close as possible to the resonance frequency of the spins to be manipulated. Deviations from the center frequency to the resonance frequency are referred to as off-resonance. To generate the B1 field, controlled currents are applied to the RF coils in the high-frequency unit 7 by means of the high-frequency transmit / receive control 7'.

[0138] Furthermore, the control unit 9 comprises a monitoring unit 15 for monitoring the feasibility of acquiring diffusion-weighted measured values ​​according to the invention. The control unit 9 is designed overall to carry out a method according to the invention.

[0139] A computing unit 13, encompassed by the control unit 9, is designed to perform all the necessary calculations for the required measurements and determinations. Intermediate results and final results required for this purpose, or determined in the process, can be stored in a storage unit S of the control unit 9. The units shown here are not necessarily to be understood as physically separate units, but merely represent a subdivision into conceptual units, which can also be realized, for example, in fewer or even just a single physical unit.

[0140] Via an input / output device (I / O) of the magnetic resonance system 1, control commands can be sent to the magnetic resonance system by a user, for example, and / or results from the control device 9, such as image data, can be displayed.

[0141] The method described herein may also be in the form of a computer program comprising instructions that execute the described method on a control unit 9. Likewise, a computer-readable storage medium may be present, comprising instructions that, when executed by a control unit 9 of a magnetic resonance system 1, cause it to execute the described method.

[0142] Regardless of the grammatical gender of a particular term, persons with male, female or other gender identities are included. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] US 10557909B2

[0079] US 10006979B2

[0088] US 9952302B2

[0098] US 10613175B2

[0117] Cited non-patent literature

[0000] Stejskal and Tanner in “Spin Diffusion Measurements: Spin Echoes in the Presence of a Time-Dependent Field Gradient,” J. Chem. Phys. 42: pp. 288-292 (1965)

[0008] Szczepankiewicz et al., “Gradient Waveform Design for Tensor-Valued Encoding in Diffusion MRI,” J. Neurosc. Methods 348: p. 109007 (2021

[0009]

Claims

[1] Method for acquiring diffusion-weighted measurement data of a test object using a magnetic resonance system with a measurement protocol with non-trapezoidal gradient pulse shapes for diffusion coding comprising the steps: a) Loading a measurement protocol to be used with measurement parameters to be set, which include a desired non-trapezoidal gradient pulse shape, b) Loading pre-configured characteristics for at least the non-trapezoidal gradient pulse shape, c) Assigning possible parameter values ​​of at least one measurement parameter to be set in the measurement protocol to at least one category indicating the feasibility of the measurement protocol on the magnetic resonance system based on the charged characteristics, d) Entering at least one desired parameter value from at least one measurement parameter of the measurement protocol, taking into account the assignment made, e) If a parameter value has not yet been entered for each measurement parameter to be set in the measurement protocol, with which the measurement protocol can be executed, repeat steps c) and d), at least for measurement parameters for which no parameter value has yet been set, until parameter values ​​have been entered for all measurement parameters to be set in the measurement protocol, with which the measurement protocol can be executed. g) Acquisition of diffusion-weighted measurement data using the measurement protocol with the entered parameter values. [2] Method according to claim 1, wherein the at least one non-trapezoidal gradient pulse shape is defined along at least one axis extending in a direction of a coordinate system, and wherein the at least one non-trapezoidal gradient pulse shape is optionally defined with a predetermined scaling factor associated with the direction of the logical coordinate system. [3] Method according to one of the preceding claims, wherein measurement parameters to be set include spoiler gradients. [4] Method according to any of the preceding claims, wherein the measurement parameters to be set include compensation gradients. [5] Method according to any of the preceding claims, wherein the measurement parameters to be set include corrected gradient amplitudes. [6] Method according to one of the preceding claims, wherein a desired non-trapezoidal gradient pulse shape can be selected from a selection of non-trapezoidal gradient pulse shapes and can be loaded as a description of the respective non-trapezoidal gradient pulse shape, which is divided into at least one section, which is stored individually or together for a gradient pulse shape in a separate file. [7] Method according to claim 6, wherein the description describes the non-trapezoidal gradient pulse shape piecewise as constant. [8] Method according to one of the preceding claims, wherein a zeroth moment of a desired non-trapezoidal gradient pulse shape is checked before and / or during the acquisition of diffusion-weighted measurement data and, if the check reveals a deviation from a target value, a gradient amplitude of the non-trapezoidal gradient pulse shape is corrected to compensate for the deviation. [9] Magnetic resonance system (1) comprising a magnet unit (3), a gradient unit (5), a radio frequency unit (7) and a control unit (9) with a radio frequency transmit / receive control (7') and with a monitoring unit (15), wherein the control unit (9) is configured to perform a method according to any one of claims 1 to 8 on the magnetic resonance system (1). [10] Computer program comprising commands which, when the program is executed by a control device (9) of a magnetic resonance system (1), cause it to execute the method according to any one of claims 1 to 8. [11] Computer-readable storage medium comprising instructions which, when executed by a control device (9) of a magnetic resonance system (1), cause it to execute the method according to any one of claims 1 to 8.

Images