Systems and methods for multi-echo steady-state sequences with improved fat suppression using adaptive DIXON techniques

Abstract

The present disclosure relates to systems and methods for multi-echo steady-state sequences with improved fat suppression using adaptive DIXON techniques. A computer-implemented method for suppressing fat in reconstructed magnetic resonance imaging data includes generating, via a processor, a first echo image and a second echo image from magnetic resonance scan data of a subject acquired by a magnetic resonance scanner using a multi-echo steady-state sequence; the computer-implemented method also includes estimating, via the processor, an echo phase difference between the first echo image and the second echo image to generate an echo phase difference image. The computer-implemented method also includes identifying, via the processor, voxels in the echo phase difference image having additional chemical shift phases. The computer-implemented method even further includes removing, via the processor, signals of voxels identified as having additional chemical shift phases in both the first echo image and the second echo image to generate a first fat removal image and a second fat removal image, respectively.

Description

BACKGROUND

[0001] The subject matter disclosed herein relates to medical imaging, and more particularly, to systems and methods for multi-echo steady state sequences with improved fat suppression using adaptive Dixon technique.

[0002] Non-invasive imaging techniques allow for obtaining images of a patient / subject's internal structure or features without performing an invasive procedure on the patient / subject. In particular, such non-invasive imaging techniques rely on various physical principles, such as differential transmission of X-rays through a target volume, reflection of sound waves within a volume, paramagnetism of different tissues and materials within a volume, decomposition of a target radioactive isotope within the body, etc., to collect data and construct images or otherwise represent the observed internal features of the patient / subject.

[0003] During MRI, when a material such as human tissue is subjected to a uniform magnetic field (polarizing field Bo), the individual magnetic moments of the spins in the tissue attempt to align with this polarizing field. If the material or tissue is also subjected to a magnetic field which lies in the x-y plane (excitation field Bi) that is near the Larmor frequency, the components of the magnetic moments in the x-y plane will be excited to precess around the z-axis in a circular or helical path at the Larmor frequency. z The magnetic moments can be rotated or "tilted" into the x-y plane to produce a net transverse magnetization M t After the excitation signal Bi is terminated, the excited spins emit a signal which can be received and processed to form an image.

[0004] When producing images with these signals, magnetic field gradients (G x , G y , and G z ) are employed. Typically, the region to be imaged is scanned with a series of measurement cycles in which the gradients are varied according to the particular localization method used. The resulting set of received nuclear magnetic resonance (NMR) signals is digitized and processed to reconstruct an image using one of the well-known reconstruction techniques.

[0005] MR neurography is very similar to conventional MRI. In MR neurography, thin- slice high-resolution sequences with T2 weighting and fat suppression are performed, striving to optimally increase the discernibility of nerve tissue signal. When the physician can visualize the nerve, they are more easily able to locate the site of nerve injury and diagnose the underlying cause. SUMMARY

[0006] The following presents a summary of certain embodiments disclosed herein. It should be understood that these aspects are merely examples and are not intended to limit the scope of the disclosure. Indeed, the disclosure can encompass a variety of aspects that can not be set forth below.

[0007] In one embodiment, a computer-implemented method for suppressing fat in reconstructed magnetic resonance imaging data is provided. The computer-implemented method includes generating, via a processor, first and second echo images from magnetic resonance scan data of a subject acquired by a magnetic resonance scanner with a multi-echo steady state sequence. The computer-implemented method also includes estimating, via the processor, an echo phase difference between the first and second echo images to generate an echo phase difference image. The computer-implemented method further includes identifying, via the processor, voxels in the echo phase difference image having an extra chemical shift phase. The computer-implemented method even further includes removing, via the processor, signal of voxels identified as having the extra chemical shift phase in both the first and second echo images to generate first and second fat removed images, respectively.

[0008] In another embodiment, a system for suppressing fat in reconstructed magnetic resonance imaging data is provided. The system includes a memory encoding processor executable routines. The system also includes a processor configured to access the memory and execute the processor executable routines, where the processor executable routines, when executed by the processor, cause the processor to perform actions. The actions include generating first and second echo images from magnetic resonance scan data of a subject acquired by a magnetic resonance scanner with a multi-echo steady state sequence. The actions also include estimating an echo phase difference between the first and second echo images to generate an echo phase difference image. The actions further include identifying voxels in the echo phase difference image having an extra chemical shift phase. The actions even further include removing signal of voxels identified as having the extra chemical shift phase in both the first and second echo images to generate first and second fat removed images, respectively.

[0009] In another embodiment, a non-transitory computer readable medium is provided. The non-transitory computer readable medium includes processor executable code that, when executed by a processor, causes the processor to perform actions. The actions include generating first and second echo images from magnetic resonance scan data of a subject acquired by a magnetic resonance scanner with a multi-echo steady state sequence. The actions also include estimating an echo phase difference between the first and second echo images to generate an echo phase difference image. The actions further include identifying voxels in the echo phase difference image having an extra chemical shift phase. The actions even further include removing signal of voxels identified as having the extra chemical shift phase in both the first and second echo images to generate first and second fat removed images, respectively. BRIEF DESCRIPTION OF DRAWINGS

[0010] These and other features, aspects, and advantages of the present disclosure will become better understood when the following detailed description is read with reference to the accompanying drawings in which like characters represent like parts throughout the drawings, wherein:

[0011] Figure 1 Embodiments of a magnetic resonance imaging (MRI) system suitable for use with the disclosed technology are illustrated;

[0012] Figure 2 A pulse sequence diagram with water- excited multi-echo (e.g., dual-echo) steady-state pulse sequence in accordance with aspects of the present disclosure is illustrated;

[0013] Figure 3 A flowchart of a method for suppressing fat in reconstructed magnetic resonance imaging data in accordance with aspects of the present disclosure;

[0014] Figure 4 An example of an echo image (e.g., derived from a free induction decay readout) in accordance with aspects of the present disclosure is depicted;

[0015] Figure 5 An example of an echo image (e.g., derived from a time-reversed steady-state free precession echo readout) in accordance with aspects of the present disclosure is depicted;

[0016] Figure 6 An example of an echo phase difference image in accordance with aspects of the present disclosure is depicted;

[0017] Figure 7 An example of an echo phase difference image in accordance with aspects of the present disclosure is depicted, wherein voxels with an additional chemical shift phase are identified;

[0018] Figure 8 An example of a fat removal image (derived from an echo image in Figure 3 using the method in Figure 4 ) in accordance with aspects of the present disclosure is depicted;

[0019] Figure 9 An example of a fat removal image (derived from an echo image in Figure 3 using the method in Figure 5 ) in accordance with aspects of the present disclosure is depicted;

[0020] Figure 10 An example of an image of a subject's arm acquired with steady-state multi-echo acquisition and reconstructed with typical reconstruction or adaptive Dixon techniques as described in the method in Figure 3 in accordance with aspects of the present disclosure is depicted;

[0021] Figure 11Depicted are examples of reconstructed echo (or readout) images of a subject's arm acquired using steady-state multi-echo acquisition and using typical reconstruction or as described herein, in accordance with aspects of the present disclosure. Figure 3 Reconstruction is done using the adaptive Dixon technique described in the method;

[0022] Figure 12 Depicted are additional examples of reconstructed echo (or readout) images of a subject's arm acquired using steady-state multi-echo acquisition and using typical reconstruction or as described herein. Figure 3 Reconstruction is done using the adaptive Dixon technique described in the method;

[0023] Figure 13 Depicted are additional examples of reconstructed echo (or readout) images of a subject's forearm acquired using steady-state multi-echo acquisition and using typical reconstruction or as described herein, in accordance with aspects of the present disclosure. Figure 3 Reconstruction is done using the adaptive Dixon technique described in the method;

[0024] Figure 14 Depicts a box plot of the average contrast between subcutaneous fat and normal muscle for echo (or readout) images acquired using steady-state multi-echo acquisition and reconstructed using typical or Figure 3 Reconstructed using the adaptive Dixon technique described in the method in ; and

[0025] Figure 15 A table depicting average contrast between abnormal muscle and normal muscle for echo (or readout) images acquired using steady-state multi-echo acquisition and reconstructed using typical or conventional techniques according to aspects of the present disclosure is provided. Figure 3 The adaptive Dixon technique described in the method is used to reconstruct. DETAILED DESCRIPTION

[0026] One or more specific embodiments are described below. In order to provide a concise description of these embodiments, not all features of an actual implementation are necessarily described in the specification. It should be understood that in the development of any such actual implementation, as in any engineering or design project, many implementation-specific decisions must be made to achieve the developer's specific goals, such as complying with system-related and business-related constraints that may vary from implementation to implementation. Furthermore, it should be understood that such development efforts may be complex and time-consuming, but remain a routine task of design, fabrication, and manufacturing for those of ordinary skill having the benefit of this disclosure.

[0027] In introducing elements of various embodiments of the present inventive subject matter, the articles "a," "an," "the," and "said” are intended to mean that there are one or more of the elements. The terms "comprising," "including," and "having" are intended to be inclusive and mean that there can be additional elements other than the listed elements. Additionally, any numerical examples in the following discussion are intended to be non-limiting, and additional numerical values, ranges, and percentages are within the scope of the disclosed embodiments.

[0028] While various aspects of the following discussion are provided in the context of medical imaging, it should be understood that the disclosed technology is not limited to such medical contexts. Indeed, the examples and explanations provided in such medical contexts are merely for ease of explanation by providing examples of real-world implementations and applications. However, the disclosed technology can also be used in other contexts, such as image reconstruction for non-destructive inspection of manufactured parts or goods (i.e., quality control or quality review application scenarios) and / or non-invasive inspection of packages, boxes, luggage, etc. (i.e., security or screening application scenarios). Generally, the disclosed technology can be used in any imaging or screening context or image processing or photography field where a set or class of acquired data undergoes a reconstruction process to generate an image or volume.

[0029] Currently, water excitation is used with three-dimensional (3D) dual echo steady state (DESS) sequences, such as steady-state multi-echo acquisition (MENSA) as a 3D gradient echo sequence, to suppress fat signal. For 3D DESS with water excitation, the first free induction decay (FID) readout provides high quality anatomical detail and the second time-reversed steady state free precession (SSFP) echo readout provides additional T2 weighting. This sequence has proven beneficial for 3D MR neurography. However, due to system imperfections (i.e., B0 and B1 inhomogeneity), fat signal can not be completely suppressed by the spectral spatial water excitation.

[0030] The present disclosure provides systems and methods for multi-echo steady state (e.g., with water excitation sequences, such as 3D dual echo steady state sequences with water excitation) such as MENSA with improved fat suppression using adaptive Dixon techniques. In particular, the phase behavior in the MENSA and region growing algorithm-like reconstruction is exploited to further suppress fat signal. The disclosed embodiments can be used for various applications, such as musculoskeletal MRI and neurography. In the case of neurography, better suppression of fat improves nerve delineation (e.g., nerves embedded inside fat pads). The disclosed embodiments improve diagnostic confidence. In certain embodiments, the disclosed embodiments can be used with other types of multi-echo sequences.

[0031] For example, disclosed embodiments include a computer-implemented method for suppressing fat in reconstructed magnetic resonance imaging data. The computer- implemented method includes generating, via a processor, first and second echo images from magnetic resonance scan data of a subject acquired by a magnetic resonance scanner with a multi-echo steady state sequence (e.g., a three-dimensional multi-echo steady state sequence with water excitation). The computer-implemented method also includes estimating, via the processor, an echo phase difference between the first and second echo images to generate an echo phase difference image. The computer-implemented method further includes identifying, via the processor, voxels in the echo phase difference image having an extra chemical shift phase. The computer-implemented method even further includes removing, via the processor, signal of voxels identified as having an extra chemical shift phase in both the first and second echo images to generate first and second fat removed images, respectively.

[0032] In another example, disclosed embodiments include a system for suppressing fat in reconstructed magnetic resonance imaging data. The system includes a memory encoding processor-executable routines. The system also includes a processor configured to access the memory and execute the processor-executable routines, where the processor-executable routines, when executed by the processor, cause the processor to perform actions. The actions include generating, from magnetic resonance scan data of a subject acquired by a magnetic resonance scanner with a multi-echo steady state sequence (e.g., a three-dimensional multi-echo steady state sequence with water excitation), first and second echo images. The actions also include estimating an echo phase difference between the first and second echo images to generate an echo phase difference image. The actions further include identifying voxels in the echo phase difference image having an extra chemical shift phase. The actions even further include removing signal of voxels identified as having an extra chemical shift phase in both the first and second echo images to generate first and second fat removed images, respectively.

[0033] In yet another example, a non-transitory computer-readable medium includes processor-executable code that, when executed by a processor, causes the processor to perform actions. The actions include generating, from magnetic resonance scan data of a subject acquired by a magnetic resonance scanner with a multi-echo steady state sequence (e.g., a three-dimensional multi-echo steady state sequence with water excitation), first and second echo images. The actions also include estimating an echo phase difference between the first and second echo images to generate an echo phase difference image. The actions further include identifying voxels in the echo phase difference image having an extra chemical shift phase. The actions even further include removing signal of voxels identified as having an extra chemical shift phase in both the first and second echo images to generate first and second fat removed images, respectively.

[0034] In certain embodiments, a region growing algorithm is used to identify voxels in the echo phase difference image having an extra chemical shift phase. In certain embodiments, the first echo image is derived from a spin echo readout in a repetition time, and the second echo image is derived from a time-reversed steady state free precession echo readout in a repetition time (i.e., the same repetition time). The first echo image provides clear anatomical delineation defined primarily by T1 / T2*. The second echo image has heavily T2-weighted contrast, which in neurography provides high contrast in peripheral nerves. In certain embodiments, the disclosed embodiments also include generating a reconstructed image by combining the first fat removal image and the second fat removal image. In certain embodiments, the extra chemical shift phase is from water / fat chemical shift. In certain embodiments, the identified voxels having an extra chemical shift phase are fat voxels.

[0035] In view of the above, Figure 1 is a magnetic resonance imaging (MRI) system 100, shown schematically as including a scanner 102, scanner control circuitry 104, and system control circuitry 106. According to embodiments described herein, the MRI system 100 is generally configured to perform MR imaging.

[0036] The system 100 also includes remote access and storage systems or devices, such as a picture archiving and communication system (PACS) 108, or other devices, such as remote radiology equipment, enabling on-site or off-site access to data acquired by the system 100. In this way, MR data can be acquired and then processed and evaluated on-site or off-site. While the MRI system 100 can include any suitable scanner or detector, in the illustrated embodiment, the system 100 includes a whole-body scanner 102 having a housing 120 through which a bore 122 is formed. A diagnostic table 124 is moveable into the bore 122 to allow a patient 126 to be positioned therein for imaging of selected anatomical structures within the patient.

[0037] The scanner 102 includes a series of associated coils for generating controlled magnetic fields for exciting spin magnetic material within the anatomy of a subject being imaged. In particular, a primary magnetic coil 128 is provided for generating a primary magnetic field B0, which is generally aligned with the bore 122. A series of gradient coils 130, 132, and 134 allow for the generation of controlled gradient magnetic fields during an examination sequence for positionally encoding certain spin magnetic nuclei within the patient 126. A radio frequency (RF) coil 136 is configured to generate RF pulses for exciting certain spin magnetic nuclei within the patient. In addition to the coils that can be local to the scanner 102, the system 100 includes a set of receive coils 138 (e.g., a coil array) configured for placement proximal to the patient 126 (e.g., against the patient). For example, the receive coils 138 can include a cervical / thoracic / lumbar (CTL) coil, a head coil, a single-sided spine coil, etc. Generally, the receive coils 138 are placed proximal to or over the head of the patient 126 in order to receive weak RF signals (weak relative to the transmit pulses generated by the scanner coils) generated by certain spin magnetic nuclei within the patient 126 as the patient returns to their relaxed state.

[0038] The various coils of the system 100 are controlled by external circuitry to generate the desired fields and pulses and to read the emissions from the spin magnetic material in a controlled manner. In the illustrated embodiment, a main power supply 140 provides power to the primary field coil 128 to generate the primary magnetic field B0. A power input 44 (e.g., power from a utility or power grid), a power distribution unit (PDU), a power supply (PS), and a driver circuit 150 together provide power to cause the gradient field coils 130, 132, and 134 to pulse. The driver circuit 150 can include amplification and control circuitry for supplying current to the coils in accordance with a defined pulse sequence output by the scanner control circuit 104.

[0039] Another control circuit 152 is provided for regulating the operation of the RF coil 136. The circuit 152 includes a switching device for alternating between an active mode of operation and a passive mode of operation in which the RF coil 136 respectively emits and does not emit a signal. The circuit 152 also includes an amplification circuit configured to generate RF pulses. Similarly, the receive coils 138 are connected to a switch 154 that is capable of switching the receive coils 138 between a receive mode and a non-receive mode. Thus, in the receive mode, the receive coils 138 resonate with RF signals produced by the spin magnetic nuclei within the patient 126, and in the non-receive mode, they do not resonate with RF energy from the transmit coil (i.e., coil 136) in order to prevent unwanted operation. Additionally, a receive circuit 156 is configured to receive data detected by the receive coils 138 and can include one or more multiplexing and / or amplification circuits.

[0040] It should be noted that while the scanner 102 and control / amplification circuitry are illustrated as being coupled by a single wire, in actual instances there can be many such wires. For example, separate wires can be used for control, data communication, power transmission, etc. In addition, appropriate hardware can be provided along each type of wire for proper handling of data and current / voltage. Indeed, various filters, digitizers, and processors can be provided between the scanner and either or both of the scanner control circuitry 104 and the system control circuitry 106.

[0041] As shown, the scanner control circuitry 104 includes interface circuitry 158 that outputs signals used to drive the gradient field coils and RF coils and to receive data representative of magnetic resonance signals produced in an examination. The interface circuitry 158 is coupled to control and analysis circuitry 160. Based on a protocol selected via the system control circuitry 106, the control and analysis circuitry 160 performs commands for driving the circuitry 150 and 152.

[0042] The control and analysis circuitry 160 is also used to receive magnetic resonance signals and perform subsequent processing prior to transmission of data to the system control circuitry 106. The scanner control circuitry 104 also includes one or more memory circuits 162 that store configuration parameters, pulse sequence descriptions, examination results, etc. during operation.

[0043] Interface circuitry 164 is coupled to the control and analysis circuitry 160 for exchanging data between the scanner control circuitry 104 and the system control circuitry 106. In certain embodiments, the control and analysis circuitry 160, while illustrated as a single unit, can include one or more hardware devices. The system control circuitry 106 includes interface circuitry 166 that receives data from the scanner control circuitry 104 and transmits data and commands back to the scanner control circuitry 104. The control and analysis circuitry 168 can include a CPU in a general purpose or special purpose computer or workstation. The control and analysis circuitry 168 is coupled to a memory circuit 170 to store programming code for operating the MRI system 100, as well as to store processed image data for later reconstruction, display, and transmission. The programming code can execute one or more algorithms configured to perform reconstruction of acquired data as described below when executed by the processor. In certain embodiments, image reconstruction can occur on a separate computing device having processing circuitry and memory circuitry.

[0044] In certain embodiments, the programming code is configured to generate a first echo image and a second echo image from magnetic resonance scan data of a subject acquired by a magnetic resonance scanner with a three-dimensional dual-echo steady-state sequence with water excitation. The programming code is further configured to estimate an echo phase difference between the first echo image and the second echo image to generate an echo phase difference image. The programming code is further configured to identify voxels in the echo phase difference image having an extra chemical shift phase. The program code is even configured to remove signal of voxels identified as having an extra chemical shift phase in both the first echo image and the second echo image to generate a first fat removed image and a second fat removed image, respectively.

[0045] In certain embodiments, the programming code is configured to identify voxels in the echo phase difference image having an extra chemical shift phase with a region growing algorithm. In certain embodiments, the first echo image is derived from a free induction decay readout in a repetition time, and the second echo image is derived from a time-reversed steady- state free precession echo readout in a repetition time (i.e., the same repetition time). The first echo image provides clear anatomical demarcations defined primarily by T1 / T2*. The second echo image has heavily T2-weighted contrast, which provides higher contrast in peripheral nerves in neurography. In certain embodiments, the programming code is further configured to generate a reconstructed image by combining the first fat removed image and the second fat removed image. In certain embodiments, the extra chemical shift phase is from a water / fat chemical shift. In certain embodiments, the voxels identified as having an extra chemical shift phase are fat voxels.

[0046] Additional interface circuits 172 can be provided to facilitate exchange of information between the external system components and the image data, configuration parameters, etc. Finally, system control and analysis circuitry 168 can be communicably coupled to various peripheral devices to facilitate operator interface and production of hard copies of reconstructed images. In the illustrated embodiment, these peripheral devices include a printer 174, a monitor 176, and a user interface 178, including devices such as a keyboard, mouse, touch screen (e.g., integral with monitor 176), etc.

[0047] Figure 2A pulse sequence diagram 180 is illustrated with a multi-echo (e.g., dual-echo) steady-state pulse sequence with water excitation. A first row 182 (e.g., top row) of the pulse sequence diagram 180 illustrates RF over time. As shown, RF pulses 184, 186 are produced. Reference sign 187 represents a repetition time (TR) between the RF pulses 184, 186. A second row 188 of the pulse sequence diagram 180 illustrates a gradient (e.g., slice-selective gradient) applied along a slice direction over time. As shown, the slice-selective gradient is applied concurrently with the RF pulses 184, 186. In particular, the slice-selective gradient is initially applied in a positive direction during the RF pulse 184 and then applied in a negative direction to refocus. Just prior to the RF pulse 186, the slice-selective gradient is applied in a negative direction during the RF pulse 186 and then applied in a positive direction. A third row 190 of the pulse sequence diagram 180 illustrates a gradient (e.g., a read or frequency-encoding gradient) applied along a frequency direction over time. A fourth row 192 of the pulse sequence diagram 180 illustrates a gradient applied along a phase direction over time. As shown, a fifth row 194 of the pulse sequence diagram 180 illustrates an MRI signal received or acquired during the read. As depicted in the MRI signal, there is a first echo 196 and a second echo 198. The first echo 196 is associated with a free induction decay. The second echo 198 is associated with a time-reversed steady-state free precession. Both the first echo 196 and the second echo 198 occur in the same repetition time 187. The MRI signal is digitized during data acquisition. The vertical axis of each row represents amplitude.

[0048] Figure 3 is a flowchart of a method 200 for suppressing fat in reconstructed magnetic resonance imaging data. One or more steps of the method 200 can be performed by processing circuitry of the magnetic resonance imaging system 100 in Figure 1 or a remote computing system. One or more steps of the method 200 can be performed concurrently or in a different order than depicted. Figure 3

[0049] ​Method 200 includes generating first and second echo images from magnetic resonance scan data of a subject acquired by a magnetic resonance scanner with a three-dimensional multi-echo steady state sequence with water excitation (block 202). In certain embodiments, the three-dimensional multi-echo steady state sequence with water excitation is a MENSA sequence. In certain embodiments, the first echo image is derived from a free induction decay (FID) readout in a repetition time, and the second echo image is derived from a time-reversed steady state free precession (SSFP) echo readout in a repetition time (i.e., the same repetition time). The contribution of residual fat signal from both readouts (e.g., SSFP-FID and SSFP-echo) behaves similarly to fat described in Dixon techniques. Adaptation of Dixon techniques (referred to herein as aDixon) as described below enables identification and removal of residual fat signal. Figure 4 An example of a first echo image 204 (e.g., derived from a free induction decay readout) is depicted. Figure 5 An example of a second echo image 206 (e.g., a time-reversed steady state free precession echo readout) is depicted. The first echo image 204 and the second echo image 206 are acquired during the same repetition time. The first echo image 204 provides clear anatomical delineation defined primarily by T1 / T2*. The second echo image 206 has heavy T2-weighted contrast, which in neurography adds high signal to peripheral nerves.

[0050] Returning to Figure 3 , method 200 also includes estimating an echo phase difference between the first echo image and the second echo image to generate an echo phase difference image (block 208). Figure 6 An example of an echo phase difference image 210 is depicted. The following equations illustrate how the echo phase difference is estimated. For MENSA, S W1 and S W2 represent the two echo signals in a water region voxel in the first and second echo images, respectively. S F1 and S F2 represent the two echo signals in a fat voxel under the first and second echo images, respectively. These equations include:

[0051]

[0052] where φ0is the initial phase, φ B0 is the B0phase, and φ is the phase from water / fat chemical shift. The echo phase difference for a given image is composed of the phase difference dφ W in water voxels and the phase difference dφ F in fat voxels, as shown in the following equations.

[0053]

[0054] Thus, estimating the echo phase difference between the first echo image and the second echo image is represented by the following equation:

[0055]

[0056] where I dφ is a complex unitary matrix whose phase represents the echo phase difference image, I1represents the first echo image, I2represents the second echo image, and H represents the Hermitian operator.

[0057] Returning to Figure 3 , the method 200 includes identifying voxels in the echo phase difference image having an extra chemical shift phase (φ) (block 212). In certain embodiments, since φ B0 and φ are both smooth, a region growing algorithm is utilized to identify voxels in the echo phase difference image having an extra chemical shift phase (i.e., fat voxels). Figure 7 An example of the echo phase difference image 214 is depicted, with voxels having an extra chemical shift phase identified via the growing algorithm (e.g., highlighted via a segmentation mask).

[0058] Returning to Figure 3 , the method 200 includes removing the signal of voxels identified as having an extra chemical shift phase in both the first echo image and the second echo image to generate a first fat removed image and a second fat removed image, respectively (block 216). Figure 8 An example of the first fat removed image 218 (derived from the first echo image 204) is depicted, with the signal of the identified fat voxels removed. Figure 9 An example of the second fat removed image 220 (derived from the second echo image 206) is depicted, with the signal of the identified fat voxels removed. Blocks 208, 212, and 216 together represent an adaptive Dixon (aDixon) technique. The method 200 further includes generating a reconstructed image by combining the first fat removed image and the second fat removed image (block 222).

[0059] Figure 10Examples of reconstructed images 224, 226 of a subject's arm are depicted. Both images 224, 226 are reconstructed from the same magnetic resonance imaging data acquired with MENSA. Image 224 is reconstructed with a typical reconstruction. Image 226 is reconstructed with the adaptive Dixon technique as described in method 200. In comparing images 224 and 226, the medial antebrachial cutaneous nerve is better depicted in image 226 (as indicated by arrow 227). In image 226, the fat signal surrounding the medial antebrachial cutaneous nerve is better suppressed. As shown in image 224, the spectral spatial water alone cannot completely suppress the fat signal due to system imperfections.

[0060] Figure 11 Examples of reconstructed echo (or read) images 228, 230, 232, and 234 of a subject's arm are depicted. Echo images 228, 230, 232, and 234 are reconstructed from the same magnetic resonance imaging data acquired with MENSA. Echo images 228 and 230 are derived from inductive decay reads. Echo images 232 and 234 are derived from time-reversed steady- state free precession echo reads acquired during the same repetition time as the inductive decay reads. Echo images 228 and 230 provide clear anatomical demarcations defined primarily by T1 / T2*. Echo images 232 and 234 have heavy T2-weighted contrast, which in nerve contrast adds high signal to peripheral nerves. Echo images 228 and 232 are reconstructed with a typical reconstruction. Echo images 230 and 234 are reconstructed with the adaptive Dixon technique as described in method 200. In comparing echo images 228 and 230 and comparing echo images 232 and 234, the fat suppression is superior in echo images 230 and 234. As shown in echo images 228 and 232, the spectral spatial water alone cannot completely suppress the fat signal due to system imperfections. As depicted in echo images 230 and 234, the relative signal levels of the nerves with respect to the muscle are unchanged. The lack of significant change in muscle contrast in the adaptive Dixon technique indicates that it does not adversely affect the interpretation of nerve and muscle signals.

[0061] Figure 12Additional examples of reconstructed echo (or read) images 236, 238, 240, and 242 of a subject's arm are depicted. Echo images 236, 238, 240, and 242 are reconstructed from the same magnetic resonance imaging data acquired with MENSA. Echo images 236 and 238 are derived from inductive decay reads. Echo images 240 and 242 are derived from time-reversed steady- state free precession echo reads acquired during the same repetition time as the inductive decay reads. Echo images 236 and 238 provide clear anatomical demarcations defined primarily by T1 / T2*. Echo images 240 and 242 have heavy T2-weighted contrast, which in neurography would add high signal to peripheral nerves. Echo images 236 and 240 are reconstructed with typical reconstruction. Echo images 238 and 242 are reconstructed with adaptive Dixon techniques as described in method 200. In comparing echo images 236 and 238 and comparing echo images 240 and 242, fat signal is better suppressed in echo images 238 and 242. As shown in echo images 236 and 240, spectral spatial water alone cannot completely suppress fat signal due to system imperfections. The relative signal levels of nerves versus muscle are also unchanged as depicted in echo images 238 and 242. The lack of significant change in muscle contrast in the adaptive Dixon techniques indicates that it does not adversely affect nerve and muscle signal interpretation.

[0062] Figure 13 Further examples of reconstructed echo (or read) images 244, 246, 248, and 250 of a subject's forearm are depicted. Echo images 244, 246, 248, and 250 are reconstructed from the same magnetic resonance imaging data acquired with MENSA. Echo images 244 and 246 are derived from inductive decay reads. Echo images 248 and 250 are derived from time-reversed steady-state free precession echo reads acquired during the same repetition time as the inductive decay reads. Echo images 244 and 246 provide clear anatomical demarcations defined primarily by T1 / T2*. Echo images 248 and 250 have heavy T2-weighted contrast, which in neurography would add high signal to peripheral nerves. Echo images 244 and 248 are reconstructed with typical reconstruction. Echo images 246 and 250 are reconstructed with adaptive Dixon techniques as described in method 200. In comparing echo images 244 and 246 and comparing echo images 248 and 250, regions of muscle edema are better depicted in echo images 246 and 250. This is because fat signal is better suppressed in echo images 246 and 250. As shown in echo images 244 and 248, spectral spatial water alone cannot completely suppress fat signal due to system imperfections.

[0063] Figure 14A box plot 252 depicting the mean contrast between subcutaneous fat and normal muscle (e.g., without muscle edema or fat infiltration) in echo (or read) images of the upper extremity of a subject acquired with MENSA and reconstructed with typical reconstruction or adaptive Dixon techniques as described in the methods of Figure 3 A region of interest was analyzed in the echo images for subcutaneous fat and normal appearing muscle. The fat to normal muscle contrast was obtained as the ratio between the corresponding signals. A non-parametric Wilcoxon signed rank test was used to compare the contrast between the regular reconstruction and the reconstruction with adaptive Dixon techniques, where p < 0.05 was considered statistically significant. The echo or read images designated as MENSA read 1 were derived from the inductive decay read and reconstructed with typical reconstruction. The echo or read images designated as MENSA read 2 were derived from the time-reversal steady-state free precession echo read (which was acquired during the same repetition time as the inductive decay read) and reconstructed with typical reconstruction. The echo or read images designated as MENSA aDixon read 1 were derived from the inductive decay read and reconstructed with adaptive Dixon techniques as described in the methods of Figure 3 The echo or read images designated as MENSA aDixon read 2 were derived from the time-reversal steady-state free precession echo read (which was acquired during the same repetition time as the inductive decay read) and reconstructed with adaptive Dixon techniques as described in the methods of Figure 3 The y-axis 254 represents the signal ratio. As shown in the box plot 252, the mean contrast between subcutaneous fat and normal muscle in the MENSA aDixon read 1 was reduced by 52% (p = 0.010) compared to the MENSA read 1. As shown in the box plot 252, the mean contrast between subcutaneous fat and normal muscle in the MENSA aDixon read 2 was reduced by 56% (p < 0.001) compared to the MENSA read 2.

[0064] Figure 15 A box plot 256 depicting the mean contrast between abnormal muscle (e.g., as determined by electromyography) and normal muscle (e.g., without muscle edema or fat infiltration) in echo (or read) images of the upper extremity of a subject acquired with MENSA and reconstructed with typical reconstruction or adaptive Dixon techniques as described in the methods of Figure 3The echo or readout images designated as MENSA readout 1 are derived from the Induced Attenuation readout and reconstructed with typical reconstruction. The echo or readout images designated as MENSA readout 2 are derived from the Time Reversal Steady State Free Precession echo readout (acquired during the same repetition time as the Induced Attenuation readout) and reconstructed with typical reconstruction. The echo or readout images designated as MENSA aDixon readout 1 are derived from the Induced Attenuation readout and reconstructed with the adaptive Dixon technique as described in the methods in Figure 3 The echo or readout images designated as MENSA aDixon readout 2 are derived from the Time Reversal Steady State Free Precession echo readout (acquired during the same repetition time as the Induced Attenuation readout) and reconstructed with the adaptive Dixon technique as described in the methods in Figure 3 The y-axis 258 represents signal ratio. As shown in the box plot 256, the average contrast between abnormal muscle and normal muscle does not show a significant change between MENSA aDixon readout 1 and MENSA readout 1 (1.116 vs. 1.105, p = 0.94). As shown in the box plot 256, the average contrast between abnormal muscle and normal muscle does not show a significant change between MENSA aDixon readout 2 and MENSA readout 2 (1.298 vs. 1.296, p = 0.99).

[0065] Technical effects of the disclosed subject matter include providing a dual echo steady state sequence (e.g., 3D dual echo steady state sequence) with motivation, such as MENSA, with improved fat suppression using adaptive Dixon technique. In particular, the phase behavior in the reconstruction with MENSA and region growing algorithm-like reconstruction is utilized to further suppress fat signal. The present subject matter can be used for various applications, such as musculoskeletal MRI and neurography. In the case of neurography, better suppression of fat improves nerve delineation (e.g., nerves embedded inside fat pads). Technical effects of the disclosed subject matter include improving diagnostic confidence.

[0066] With reference to the technology presented herein and claimed by the appended claims, and applying it in practice, the practical nature of the subject matter is apparent, and it clearly improves the art, and therefore, is not abstract, intangible, or purely theoretical. Furthermore, if any of the claims appended to the end of this specification contain one or more elements specified as "means for" or "steps for" performing a function, such elements are intended to be interpreted under 35 U.S.C. 112(f). However, for any claim containing elements specified in any other manner, such elements are not intended to be interpreted under 35 U.S.C. 112(f).

[0067] This written description uses examples to disclose the subject matter, including the best mode, and also to enable any person skilled in the art to practice the subject matter, including making and using any devices or systems and performing any incorporated methods. The patentable scope of the subject matter is defined by the claims, and can include other examples that occur to those skilled in the art. Such other examples are intended to fall within the scope of the claims if they have structural elements that do not differ from the literal language of the claims, or if they include equivalent structural elements with insubstantial differences from the literal languages of the claims.

Claims

1. A computer-implemented method for suppressing fat in reconstructed magnetic resonance imaging data, the method comprising: generating, via a processor, a first echo image and a second echo image from magnetic resonance scan data of a subject acquired by a magnetic resonance scanner using a multi-echo steady-state sequence; estimating, via the processor, an echo phase difference between the first echo image and the second echo image to generate an echo phase difference image; identifying, via the processor, voxels in the echo phase difference image having an additional chemical shift phase; as well as Signals of the voxels identified as having the additional chemical shift phase in both the first echo image and the second echo image are removed via the processor to generate a first fat-removed image and a second fat-removed image, respectively. 2 . The computer-implemented method of claim 1 , wherein a region growing algorithm is used to identify the voxels in the echo phase difference image having the additional chemical shift phase.

3. The computer-implemented method of claim 1 , wherein the first echo image is derived from free induction decay readouts in a repetition time, and the second echo image is derived from time-reversed steady-state free precession echo readouts in the repetition time. 4 . The computer-implemented method of claim 1 , further comprising generating a reconstructed image by combining the first fat-removed image and the second fat-removed image.

5. The computer-implemented method of claim 1 , wherein the echo phase difference is estimated using the following equation: Among them I dφ is a complex unitary matrix, its phase represents the echo phase difference image, I1 represents the first echo image, I2 represents the second echo image, and H represents a Hermitian operator. The computer-implemented method of claim 1 , wherein the additional chemical shift phase is from a water / fat chemical shift. 7 . The computer-implemented method of claim 6 , wherein the identified voxels having the additional chemical shift phase are fat voxels.

8. The computer-implemented method of claim 1, wherein the multi-echo steady-state sequence is used in conjunction with a water excitation pulse.

9. A system for suppressing fat in reconstructed magnetic resonance imaging data, the system comprising: a memory encoding processor-executable routines; a processor configured to access the memory and to execute the processor-executable routine, wherein the processor-executable routine, when executed by the processor, causes the processor to: generating a first echo image and a second echo image from magnetic resonance scan data of the subject acquired by a magnetic resonance scanner using a multi-echo steady-state sequence; estimating an echo phase difference between the first echo image and the second echo image to generate an echo phase difference image; identifying voxels in the echo phase difference image having an additional chemical shift phase; as well as Signals of the voxels identified as having the additional chemical shift phase are removed from both the first echo image and the second echo image to generate a first fat-removed image and a second fat-removed image, respectively. 10 . The system of claim 9 , wherein a region growing algorithm is used to identify the voxels in the echo phase difference image having the additional chemical shift phase.

11. The system of claim 9, wherein the first echo image is derived from a free induction decay readout and the second echo image is derived from a time-reversed steady-state free precession echo readout.

12. The system of claim 9, wherein the processor-executable routine, when executed by the processor, further causes the processor to generate a reconstructed image by combining the first fat-removed image and the second fat-removed image.

13. The system of claim 9, wherein the echo phase difference is estimated using the following equation: Among them I dφ is a complex unitary matrix, its phase represents the echo phase difference image, I1 represents the first echo image, I2 represents the second echo image, and H represents a Hermitian operator.

14. The system of claim 9, wherein the additional chemical shift phase is from a water / fat chemical shift.

15. The system of claim 14, wherein the identified voxels having the additional chemical shift phase are fat voxels.

16. A non-transitory computer-readable medium comprising processor-executable code that, when executed by a processor, causes the processor to: generating a first echo image and a second echo image from magnetic resonance scan data of the subject acquired by a magnetic resonance scanner using a multi-echo steady-state sequence; estimating an echo phase difference between the first echo image and the second echo image to generate an echo phase difference image; identifying voxels in the echo phase difference image having an additional chemical shift phase; as well as Signals of the voxels identified as having the additional chemical shift phase are removed from both the first echo image and the second echo image to generate a first fat-removed image and a second fat-removed image, respectively. 17 . The non-transitory computer readable medium of claim 16 , wherein a region growing algorithm is used to identify the voxels in the echo phase difference image having the additional chemical shift phase.

18. The non-transitory computer-readable medium of claim 16, wherein the first echo image is derived from free induction decay readings in a repetition time, and the second echo image is derived from time-reversed steady-state free precession echo readings in the repetition time.

19. The non-transitory computer readable medium of claim 16, wherein the processor executable code, when executed by the processor, further causes the processor to generate a reconstructed image by combining the first fat removed image and the second fat removed image.

20. The non-transitory computer readable medium of claim 15, wherein the additional chemical shift phase is from a water / fat chemical shift, and wherein the identified voxels having the additional chemical shift phase are fat voxels.

Images