阅读说明：本技术 磁共振脊柱成像中的相位过采样数据的重建与再利用 (Reconstruction and reuse of phase oversampled data in magnetic resonance spine imaging ) 是由 B·D·约翰逊 I·迪米特洛夫 S·K·甘吉 于 2020-02-05 设计创作，主要内容包括：接收针对第一视场(FOV)中的对象的MRI检查的第一k空间数据,并且接收针对与所述第一视场相邻或交叠的第二视场的第二k空间数据。备选地,从非瞬态数据存储介质检索包括第一视场的相位和/或切片过采样k空间数据的第二k空间数据。通过使用第二k空间数据的至少部分作为相位和/或切片过采样来重建第一k空间数据以生成涵盖第一视场并且扩展到第二视场中的第一扩展视场的第一扩展图像。将所述第一扩展图像裁剪到所述第一视场,以生成针对所述第一MRI检查的所述第一视场的图像。(First k-space data for an MRI examination of an object in a first field of view (FOV) is received, and second k-space data for a second field of view adjacent or overlapping the first field of view is received. Alternatively, second k-space data comprising phase and/or slice oversampled k-space data of the first field of view is retrieved from the non-transitory data storage medium. The first k-space data is reconstructed by oversampling as phase and/or slice using at least part of the second k-space data to generate a first extended image covering the first field of view and extending into a first extended field of view in the second field of view. Crop the first extended image to the first field of view to generate an image of the first field of view for the first MRI exam.)

1. A non-transitory storage medium storing instructions readable and executable by an electronic processor (30, 42, 50) to perform an image reconstruction method comprising:

receiving first k-space data of an object (22) acquired for a first field of view (FOV);

retrieving second k-space data of the object acquired for a second field of view from a non-transitory data storage medium (40), wherein the second k-space data has the same angle (74) as the first k-space data and the second field of view is adjacent to or overlaps the first field of view;

reconstructing (62) the first and second k-space data to generate an extended image (64) covering and being larger than an extended field of view (EFOV) of the first field of view; and is

Cropping (66) the expanded image to the first field of view to generate an image (68) of the first field of view.

2. The non-transitory storage medium of claim 1, wherein:

the second k-space data comprises phase oversampled k-space data for a phase encoding direction (PE) of the first field of view (FOV), and

the combination of the reconstruction (62) and the cropping (66) of the first and second k-space data for generating the extended image (64) generates the image (68) of the first field of view without aliasing or warping in the phase encoding direction.

3. The non-transitory storage medium of any one of claims 1-2, wherein:

the second k-space data comprises oversampled k-space data for a slice selection direction (SS) of the first field of view (FOV), and

the combination of the reconstruction (62) and the cropping (66) of the first and second k-space data for generating the extended image (64) generates the image (88) of the first field of view without aliasing or warping in the slice selection direction.

4. The non-transitory storage medium of any one of claims 1-3 wherein there is one of:

(i) the first field of view is a cervical region, the second field of view is a thoracic region, and the image of the first field of view is a cervical image;

(ii) the first field of view is a thoracic region, the second field of view is a cervical or lumbar region, and the image of the first field of view is a thoracic image (100); or

(iii) The first field of view is a lumbar region, the second field of view is a thoracic region, and the image of the first field of view is a lumbar image.

5. The non-transitory storage medium of any one of claims 1-3 wherein:

the first field of view and the second field of view are two stations of a multi-station imaging examination that are adjacent or overlapping along an axial direction (24).

6. The non-transitory storage medium according to any one of claims 1-3, wherein the receiving of the first k-space data includes retrieving the first k-space data from the non-transitory data storage medium (40), and the image reconstruction method further includes:

displaying the image (68, 88, 112) of the first field of view (FOV) on a display (52, 54) with an indication (80, 90, 114) of the extended field of view (EFOV);

receive, via a user input device (56, 82, 84), a request (87) to increase a field of view, and in response to receiving the request, (i) retrieve the first and second k-space data from the non-transitory data storage medium, (ii) re-reconstruct (62) the first and second k-space data to regenerate the expanded image of the expanded field of view, and (iii) control the display to display the regenerated expanded image (64rr, 120) of the expanded field of view.

7. The non-transitory storage medium according to any one of claims 1-3, wherein the receiving of the first k-space data includes controlling a Magnetic Resonance Imaging (MRI) scanner (10) to acquire the first k-space data.

8. An image reconstruction apparatus comprising:

a display (52, 54);

an electronic processor (30, 42, 50); and

a non-transitory storage medium storing instructions readable and executable to perform an image reconstruction method, the image reconstruction method comprising:

receiving k-space data (60) of an object (22) acquired for a field of view (FOV), wherein the received k-space data comprises oversampled k-space data;

reconstructing (62) the k-space data including the oversampled k-space data to generate an extended image (64) of an extended field of view (EFOV) encompassing and being larger than the field of view;

cropping (66) the expanded image to the field of view to generate an image (68) of the field of view;

displaying the image of the field of view on the display; and is

Storing the k-space data comprising the oversampled k-space data in a non-transitory data storage medium (40).

9. The image reconstruction apparatus according to claim 8, wherein:

the field of view (FOV) is a cervical spine examination field of view, and the k-space data includes k-space data for a cervical spine Magnetic Resonance Imaging (MRI) examination (104) and k-space data for a thoracic spine MRI examination (102);

the reconstructing (62) includes reconstructing the k-space data including the oversampled k-space data including the k-space data of the thoracic spine MRI exam to generate the extended image encompassing the cervical spine exam field of view and extending into the extended field of view (EFOV) of thoracic spine exam field of view; and is

Cropping (66) the extended image to the cervical spine examination field of view generates a cervical spine image without aliasing or warping.

10. The image reconstruction apparatus according to claim 8, wherein:

the field of view (FOV) is a lumbar examination field of view and the k-space data includes k-space data of a lumbar Magnetic Resonance Imaging (MRI) examination (106) and k-space data of a thoracic MRI examination (102);

the reconstructing (62) comprises reconstructing the k-space data comprising the oversampled k-space data comprising the k-space data of the thoracic MRI exam to generate the extended image encompassing the lumbar exam field of view and extending into the extended field of view (EFOV) of a thoracic exam field of view; and is

Cropping (66) the extended image to the lumbar exam field of view generates a lumbar image without aliasing or warping.

11. The image reconstruction apparatus according to claim 8, wherein:

the field of view (FOV) is a thoracic examination field of view and the k-space data comprises k-space data of a thoracic Magnetic Resonance Imaging (MRI) examination (102) and k-space data of a cervical and/or lumbar MRI examination (104, 106);

the reconstructing (62) comprises reconstructing the k-space data comprising the oversampled k-space data comprising the k-space data of the cervical and/or lumbar MRI exam to generate the extended image encompassing the thoracic exam field of view and extending into the extended field of view (EFOV) in the cervical and/or lumbar exam field of view; and is

Cropping (66) the extended image to the thoracic exam field of view generates a thoracic image (100) without aliasing or warping.

12. The image reconstruction apparatus according to claim 8, wherein:

the field of view (FOV) is a station field of view of a multi-station Magnetic Resonance Imaging (MRI) examination, and the k-space data includes k-space data of the station field of view and k-space data of an adjacent or overlapping station field of view of the multi-station MRI examination that is adjacent to or overlaps the station field of view;

said reconstructing (62) comprises reconstructing said k-space data comprising said oversampled k-space data including said k-space data of said adjacent or overlapping station fields of view to generate said extended image covering said station fields of view and extending into said extended field of view (EFOV) in said adjacent or overlapping station fields of view; and is

Cropping (66) the expanded image to the station field of view generates an image of the station field of view without aliasing or warping.

13. Image reconstruction apparatus as claimed in claim 8, wherein the displaying comprises displaying the image (68, 88, 112) of the field of view (FOV) together with an indication (80, 90, 114) of the extended field of view (EFOV).

14. The image reconstruction device of claim 13, further comprising:

a user input device (56, 82, 84);

wherein the image reconstruction method further comprises: receiving a request (87) via the user input device to increase a field of view; and in response to receiving the request, (i) retrieving the k-space data comprising the oversampled k-space data from the non-transitory data storage medium (40), (ii) reconstructing (62) the k-space data comprising the oversampled k-space data to regenerate the extended image of the extended field of view (EFOV), and (iii) displaying the regenerated extended image (64rr, 120) of the extended field of view on the display (52, 54).

15. The image reconstruction device according to any one of claims 8 to 14, wherein:

the oversampled k-space data comprising phase oversampled k-space data for a phase encoding direction (PE) of the field of view (FOV); and is

The combination of the reconstructing (62) and the cropping (66) generates the image (68) of the field of view without aliasing or warping in the phase encoding direction.

16. The image reconstruction device according to any one of claims 8 to 15, wherein:

the oversampled k-space data comprising oversampled k-space data for a slice selection direction (SS) of the field of view (FOV); and is

The combination of the reconstructing (62) and the cropping (66) generates the image (88) of the field of view without aliasing or warping in the slice selection direction.

17. An image reconstruction method, comprising:

receiving first k-space data acquired from a subject for a first Magnetic Resonance Imaging (MRI) examination in a first field of view;

receiving second k-space data acquired from the subject for a second MRI examination in a second field of view, wherein the second k-space data has the same angle as the first k-space data and the second field of view is adjacent to or overlaps the first field of view;

reconstructing at least portions of the first and second k-space data to generate a first extended image encompassing the first field of view and extending into a first extended field of view of the second field of view;

crop the first extended image to the first field of view to generate an image of the first field of view for the first MRI exam;

reconstructing at least portions of the second k-space data and the first k-space data to generate a second expanded image encompassing the second field of view and expanded into a second expanded field of view in the first field of view; and is

Crop the second extended image to the second field of view to generate an image of the second field of view for the second MRI exam;

wherein the reconstructing and the cropping are performed by an electronic processor.

18. The image reconstruction method according to claim 17, comprising:

receiving third k-space data acquired from the subject for a third MRI examination in a third field of view, wherein the third k-space data has the same angle as the first and second k-space data, and the third field of view is adjacent to or overlaps the second field of view;

reconstructing at least portions of the third k-space data and the second k-space data to generate a third extended image encompassing the third field of view and extending into a third extended field of view in the second field of view; and is

Crop the third extended image to the third field of view to generate an image of the third field of view for the third MRI exam;

wherein the reconstruction to generate the second expanded image comprises reconstructing the second k-space data and at least part of the first k-space data and at least part of the third k-space data to generate the second expanded image encompassing the second field of view and expanding into the first field of view and into the second expanded field of view in the third field of view.

19. The image reconstruction method according to claim 18, wherein:

the first MRI examination is a cervical MRI examination and the first field of view is a cervical field of view;

the second MRI examination is a thoracic MRI examination and the second field of view is a thoracic field of view; and is

The third MRI examination is a lumbar MRI examination and the third field of view is a lumbar field of view.

20. The image reconstruction method according to any one of claims 17-19, wherein:

the reconstruction for generating the first extended image of the first extended field of view uses the at least part of the second k-space data as phase and/or slice oversampling, and the combination of the reconstruction for generating the first extended image and the cropping the first extended image into the first field of view generates the image of the first field of view for the first MRI examination without aliasing or warping in the direction of the phase and/or slice oversampling; and is

The reconstruction of the second expanded image for generating the second expanded field of view uses the at least part of the first k-space data as phase and/or slice oversampling, and the combination of the reconstruction for generating the second expanded image and the cropping of the second expanded image to the second field of view generates the image of the second field of view for the second MRI examination without aliasing or warping in the direction of the phase and/or slice oversampling.

Technical Field

The following generally relates to the field of Magnetic Resonance Imaging (MRI), MRI image reconstruction, spine MRI examination, multi-station MRI examination, and related fields.

Background

In MRI, in order to avoid aliasing or warping in images defining a field of view (FOV), oversampling in the readout, phase encoding and slice directions is known. For readout, oversampling can be performed without the expense of increased acquisition time. However, phase-encoded oversampling increases acquisition time, as does slice oversampling, which requires acquisition of additional slices that extend beyond the boundaries of the prescribed FOV. Acquired k-space data, including oversampled data, is reconstructed to generate an image that is larger in spatial extent than the prescribed FOV due to the oversampling. The image is then cropped to preserve only the specified FOV without aliasing or wrap-around. The images are uploaded to a Picture Archiving and Communication System (PACS) and form the final clinical image that is retrieved from the PACS and viewed by a clinician.

Certain improvements are disclosed below.

Disclosure of Invention

In some non-limiting illustrative embodiments disclosed herein, a non-transitory storage medium storing instructions readable and executable by an electronic processor to perform an image reconstruction method, the image reconstruction method comprising: receiving first k-space data acquired from an object for a first field of view; retrieving second k-space data acquired from the object for a second field of view from a non-transitory data storage medium, wherein the second k-space data has the same angle as the first k-space data and the second field of view is adjacent to or overlaps the first field of view; reconstructing the first k-space data and the second k-space data to generate an expanded image covering and larger than an expanded field of view of the first field of view; and crop the expanded image to the first field of view to generate an image of the first field of view.

In some non-limiting illustrative embodiments disclosed herein, an image reconstruction device includes a display, an electronic processor, and a non-transitory storage medium storing instructions readable and executable to perform an image reconstruction method. The method comprises the following steps: receiving k-space data acquired from an object for a field of view, wherein the received k-space data comprises oversampled k-space data; reconstructing the k-space data including the oversampled k-space data to generate an expanded image of an expanded field of view encompassing and being larger than the field of view; cropping the extended image to the field of view to generate an image of the field of view; displaying the image of the field of view on the display; and storing the k-space data including the oversampled k-space data in a non-transitory data storage medium.

In some non-limiting illustrative embodiments disclosed herein, an image reconstruction method comprises: receiving first k-space data acquired from the subject for a first MRI examination in a first field of view; receiving second k-space data acquired from the subject for a second MRI examination in a second field of view, wherein the second k-space data has the same angle as the first k-space data and the second field of view is adjacent to or overlaps the first field of view; reconstructing at least portions of the first and second k-space data to generate a first extended image encompassing the first field of view and extending into a first extended field of view of the second field of view; crop the first extended image to the first field of view to generate an image of the first field of view for the first MRI exam; reconstructing at least portions of the second k-space data and the first k-space data to generate a second expanded image encompassing the second field of view and expanded into a second expanded field of view in the first field of view; and crop the second extended image to the second field of view to generate an image of the second field of view for the second MRI examination. The reconstruction and cropping are suitably performed by an electronic processor.

One advantage resides in reduced Magnetic Resonance Imaging (MRI) examination data acquisition time and improved patient workflow efficiency.

Another advantage resides in reduced instances of patient recall checks.

Another advantage resides in improving diagnostic performance of MRI examinations by providing additional fields of view when clinically beneficial without an accompanying increase in MRI examination data acquisition time.

Another advantage resides in providing an MRI Information Technology (IT) infrastructure that facilitates one or more of the foregoing benefits and/or other benefits.

A given embodiment may provide none, one, two, or all of the aforementioned advantages, and/or may provide other advantages as will become apparent to those skilled in the art upon reading and understanding the present disclosure.

Drawings

The invention may take form in various components and arrangements of components, and in various steps and arrangements of steps. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention.

Figure 1 diagrammatically illustrates a Magnetic Resonance Imaging (MRI) Information Technology (IT) infrastructure including aspects that facilitate reusing acquired k-space data as oversampled data and/or increasing the field of view (FOV) of the image.

Fig. 2 diagrammatically illustrates a display shown on a workstation display or the like of a brain image specifying the FOV together with an indication of the extended field of view available through the reuse of oversampled k-space data in the phase encoding direction.

Fig. 3 diagrammatically illustrates a display shown on a workstation display or the like of a brain image specifying the FOV together with an indication of the extended field of view available through the reuse of oversampled k-space data in the slice selection direction.

Fig. 4 diagrammatically illustrates reusing data from a cervical MRI examination, a thoracic MRI examination, and a lumbar MRI examination to provide oversampled k-space data in the respective MRI examinations.

Figure 5 diagrammatically illustrates reusing oversampled k-space data to extend the FOV of an image to provide additional/improved imaging of clinically relevant features.

Detailed Description

As previously described, in conventional Magnetic Resonance Imaging (MRI) Information Technology (IT) infrastructure, acquired k-space data, including oversampled data, is reconstructed to generate an image in an extended field of view (FOV) that encompasses and is larger than a prescribed FOV due to the oversampling. The image is then cropped to the prescribed FOV (in a cropping operation, the portion of the image outside the prescribed FOV is removed, leaving only the portion of the image in the prescribed FOV) without aliasing or warping, which is uploaded to the PACS and forms the final clinical image viewed by the clinician. In conventional MRI IT infrastructure, the acquired k-space data (with or without over-sampled data) is typically not stored on the PAC, and is also typically not stored at the MRI technician's workstation.

It is herein appreciated that there are certain situations in which the acquired k-space data, and in particular oversampled data, may advantageously be stored and subsequently reused or repurposed. In one such case, the clinician may notice a feature (or portion of a feature) at the periphery of the image that specifies the FOV. Since the oversampled data is typically not stored in the PACS or at the MRI technician's workstation, the clinician must attempt to determine clinical findings from existing images of the prescribed FOV or may decide to call back an MRI exam that requires the prescribed FOV with a larger or shifted in order to better capture the noted features or feature portions. It should be appreciated herein that these unsatisfactory options may be avoided by storing the acquired k-space data including the oversampled data in the PACS and/or at the MRI technician's workstation and repeatedly extending the reconstruction of the FOV without subsequent cropping to the original prescribed FOV. The resulting image will have an extended FOV and can be transmitted to a clinician for clinical reading of MRI images. While portions of the image in the additional FOV (i.e., outside the originally prescribed FOV) may have some aliasing or warping due to lack of oversampling for the extended FOV, it nevertheless provides useful image content for evaluating the noted features or portions of features without the cost and inconvenience to the patient that would be introduced by requiring a back-up MRI examination. Even with some possible aliasing or warping, the extra image content may provide a useful background.

In another case, the clinician may require two or more MRI examinations of different but adjacent or overlapping anatomical regions. For example, one such condition that commonly occurs is an MRI spine examination. Spinal MRI examinations are the most common required MRI examinations, and by considering some estimates of about one-fourth of all MRI examinations. Often, examination of more than one section of the spine is required, for example, cervical, thoracic and/or lumbar MRI examination. This is particularly true when certain pathologies (such as bone metastases, multiple sclerosis, spinal cord compression, etc.) are involved. These spinal MRI examinations typically employ the administration of a contrast agent, and therefore perform post-contrast imaging, which prolongs the patient's time on a table. A full spine examination requires the patient to lie still and scan for a long time, which results in a higher percentage of patient movement and therefore reduced image quality and potentially undiagnosed scanning. These examinations also result in a large amount of Radio Frequency (RF) exposure of the patient, resulting in a high Specific Energy Dose (SED). Furthermore, full spine imaging takes a long time interval of the MRI scanner, thus reducing the productivity of the MRI scanner, and also making it difficult to arrange a long time block for a spine MRI examination.

In embodiments disclosed herein, when performing spinal MRI examinations (e.g., cervical MRI and thoracic MRI; or thoracic MRI and lumbar MRI) of two adjacent or overlapping anatomical regions, then the k-space data acquired for one MRI examination (e.g., thoracic MRI) may be reused to provide oversampled k-space data for another MRI examination (e.g., cervical MRI in this example). In the case of a complete spine examination, including cervical, thoracic, and lumbar MRI examinations, this reuse of k-space data may reduce the scan time of the full spine by approximately two-thirds, thereby enhancing patient comfort and improving compliance, safety, and increasing workflow efficiency of the MRI scanner. This reuse of k-space data is facilitated because the spine is a piece of anatomy with long, relatively straight, and continuous structures. The cervical vertebrae are located above and adjacent to the thoracic vertebrae, and the lumbar vertebrae are located below and adjacent to the thoracic vertebrae. Since the cervical, thoracic and lumbar portions are structurally similar and adjacent to each other, k-space data from MRI examination of adjacent anatomical regions is reused to provide oversampled k-space data to correct aliasing/warping for speeding up the MRI examination.

Re-use of MRI k-space data for providing oversampled k-space data is generally more feasible in any case where two MRI examinations of adjacent or overlapping fields of view are performed and k-space data of the two MRI examinations are performed at the same angle, i.e. the same angular orientation of the slice selection, phase encoding and readout directions. Consistency of other acquisition parameters, such as slice thickness and resolution, is also helpful, but the conversion may be adjusted for differences in some acquisition parameters (e.g., the resampling may compensate for differences in slice thickness and/or resolution).

Referring to fig. 1, an illustrative MRI IT infrastructure is shown that facilitates the methods of k-space data reuse disclosed herein. A Magnetic Resonance Imaging (MRI) scanner 10 includes typical components such as a housing 12 containing a main magnet 14 (typically including superconducting windings, but resistive magnets are also contemplated), a set of magnetic field gradient coils 16, an optional whole-body Radio Frequency (RF) coil 18 (additionally or alternatively, various local RF coils, RF coil arrays, etc. may be used for RF excitation and/or magnetic resonance readout), a patient support 20 (preferably but not necessarily including robotic actuators for patient positioning) for moving a patient 22 in and out of an examination region along an axial direction 24 (in the illustrative case, the examination region is within a bore 26 of the housing 12, although other MRI configurations (such as vertical magnet MRI, open-bore MRI, etc.) are also contemplated), and so forth. Although a single MRI scanner 10 is illustrated, it will be appreciated that some MRI laboratories may include one, two, three, or more of the various types of MRI scanners.

The illustrative MRI IT infrastructure also includes an MRI controller 30, the MRI controller 30 providing a user interface for an MRI technician to the MRI scanner 10 in order to program and/or control the MRI scanner 10 to perform a desired MRI examination, such as a brain MRI examination, a spinal MRI examination, and the like. The MRI controller 30 includes an electronic processor (not shown, e.g., a microprocessor, possibly multi-core or otherwise configured as known in the digital electronics arts), a display 32, and one or more user input devices, such as an illustrative keyboard 34 (and/or a mouse, touchpad, touch-sensitive overlay for the display 32, etc.). The MRI controller 30 also typically includes a local non-transitory data storage medium 36, such as a hard disk drive, Solid State Drive (SSD), flash memory, or the like, for local storage of k-space data acquired by the MRI scanner 10 and/or for other purposes. The illustrative MRI controller 30 is implemented as a computer having suitable peripherals 32, 34, 36. Although a single MRI controller 30 is illustrated, it will be appreciated that in some embodiments where the MRI laboratory includes multiple MRI scanners, there may be multiple MRI controllers for the MRI scanners.

The illustrative MRI IT infrastructure also includes a Picture Archiving and Communication System (PACS) that includes a non-transitory data storage medium 40 (e.g., a hard disk drive, solid state drive or SSD, a redundant array of independent disks or RAID, various combinations thereof, etc.) and associated electronic processor(s) (e.g., implemented by a network-based server computer 42 in the illustrative example). The PACS 40, 42 stores MRI images and possibly images of other medical imaging modalities (e.g., PET, CT, etc.), and may optionally be integrated with one or more other medical information systems, such as a radiology information system (i.e., PACS/RIS system). The PACS 40, 42 provide a network-based storage system for storing medical images, including MRI images, and related metadata (e.g., imaging parameters indicating use in acquiring the images, metadata identifying the patient, the examination date, the reason for the examination, the anatomical region, the imaging modality, etc.). Accordingly, the MRI controller 30 is connected with the PACS 40, 42 via an electronic data network (e.g., a wired or wireless local area network or LAN, the internet, various combinations thereof, etc.) to upload images of MRI examinations along with associated metadata to the PACS 40, 42 for subsequent viewing by a radiologist, physician, and/or other clinician(s).

The illustrative MRI IT infrastructure also includes a radiology workstation 50 at which radiology workstation 50 a radiologist may view images of an MRI examination and dictate (or type in or otherwise input) a radiology report summarizing the radiologist's findings. A radiology workstation 50 is connected with the PACS 40, 42 via an electronic data network to retrieve images and metadata of MRI examinations from the PACS 40, 42 for review by radiologists. The illustrative radiology workstation 50 includes two displays 52, 54 that may be useful, for example, to allow a radiologist to have images displayed on one display and radiology draft reports displayed on the other display; however, a single display is contemplated, such as having three or even more displays. The radiology workstation 50 also includes one or more user input devices, such as an illustrative keyboard 56, a spoken microphone 58 via which a radiologist may dictate radiology reports, and so forth. The radiology workstation 50 also typically includes an electronic processor and may be implemented in whole or in part as a computer with appropriate peripherals (e.g., displays 52, 54, user input devices 56, 58). Although a single radiology workstation 50 is illustrated, more generally, one, two, three, or more radiology workstations may be provided, for example, to serve radiologists' staff, and/or the MRI IT infrastructure may include other similar clinician workstations (not shown), such as physician office computer(s).

With continued reference to FIG. 1, an illustrative brain MRI examination is diagrammatically depicted. The MRI controller 30 controls the MRI scanner 10 to acquire k-space data 60 for a field of view (FOV) prescribed by a clinician for a brain MRI examination. The acquired k-space data 60 comprises oversampled k-space data sufficient to allow reconstruction of the prescribed FOV without aliasing or warping. K-space data 60, including oversampled k-space data, is reconstructed in an image reconstruction operation 62 (e.g., using fourier transform image reconstruction, iterative reconstruction, or another image reconstruction algorithm suitable for the spatial encoding employed in the data acquisition). The output of the image reconstruction operation 62 is an extended field of view (EFOV) image 64 that encompasses and is larger than the prescribed FOV. However, portions of the image 64 that lie outside the prescribed FOV can have some aliasing or warping. In a cropping operation 66, the image 64 of the EFOV is cropped to the prescribed FOV to generate an image 68 of the prescribed FOV. Cropping of the image removes portions that may have aliasing or warping, leaving the image 68 without aliasing or warping.

When tissue or anatomical structures are present outside the FOV, MRI artifacts known as aliasing or wrap-around occur. If the MRI signal is sufficiently sampled, aliasing from tissue outside the FOV can be eliminated by using oversampling. For two-dimensional imaging, aliasing occurs in the phase and frequency directions. Frequency oversampling can be applied to eliminate aliasing along the frequency coding direction without any time penalty. However, when the tissue is outside the FOV in the phase encoding direction, then k-space oversampling in the phase encoding direction is performed at the expense of longer scan time. The phase and frequency oversampled data are then removed by cropping so that the resulting image contains only the specified FOV without any aliasing or warping in the phase encoding direction. Similarly, if there is an anatomical structure outside the prescribed FOV, oversampling and cropping may be performed in the slice selection direction, providing an image with the prescribed FOV without any aliasing or warping in the slice selection direction.

Typically, the FOV defining image 68 is the final image stored in the PACS 42, 44 along with relevant metadata (such as FOV 70 tags) and acquisition parameters (such as slice thickness, resolution, etc.). The images 68 may then be later retrieved to the radiology workstation 50 for review by the radiologist. However, a problem may arise if the radiologist determines that the prescribed FOV of the image 68 is insufficient. For example, the radiologist may determine that the desired anatomical region is not within the prescribed FOV, or the radiologist may observe suspicious features (e.g., possible brain lesions) at the periphery of the prescribed FOV in the image 68. Typically, in such a case, the radiologist will need to determine the clinical findings based on the image 68 despite its potentially insufficient field of view, or will have to request a callback brain MRI exam with an updated prescribed FOV to obtain an image with an updated FOV.

This problem is solved by retaining and reusing oversampled k-space data as disclosed herein. As shown diagrammatically in fig. 1, k-space data 60 is also stored in PACS 40, 42, preferably labeled with relevant metadata such as a label of the extended field of view (EFOV)72 and a label of the angle 74 of the k-space data 60. (it should be noted that the various metadata 70, 72, 74 may be stored in any suitable format, and in particular may be stored in a non-explicit format, e.g., the EFOV 72 may not be stored as an explicit spatial dimension, but rather may be stored as data acquisition parameters for an MRI examination, from which the EFOV 72 may be derived by known calculations).

With continuing reference to figure 1 and with further reference to figure 2, when an image 68 specifying a FOV is retrieved at the radiology workstation 50, it is preferably tagged with EFOV 72 metadata. Fig. 2 shows the display of the image 68 on the display 52 (or display 54) of the radiology workstation 50. In addition to displaying the image 68, the display also includes an indication 80 of the EFOV 72. In the illustrative example of fig. 2, the image 68 is an axial slice of the brain, and the Phase Encoding (PE) direction is along the anteroposterior anatomical direction. Thus, the EFOV 72 includes additional image content that extends beyond the prescribed FOV along the PE direction, and the spatial region of this additional image content is indicated in the illustrative example by the dashed rectangular outline 80 of the available extension in the illustrative display presentation of fig. 2. Other graphical representations of the indication 80 of the EFOV 72 are contemplated, such as using hatching or cross-hatching, shading, color, etc. to refer to the additional image content available. The illustrative display also includes a graphically indicated text message 82 (represented by a wavy symbol (i.e., "-" in FIG. 2)) asking whether to increase the field of view to the field of view indicated by indication 80. Preferably, although not necessarily, the text message 82 informs the user that the additional image content that may be generated for the region indicated by the indication 82 may have aliasing or warping. If the user selects to increase the FOV, for example by selecting the "RR" (re-reconstruction) selection button 84 using the mouse pointer 86 in the illustrative example, but other user inputs for requesting an increase in the field of view are anticipated, the oversampled k-space data in the phase encoding direction is reused for this purpose. As shown in fig. 1, a re-reconstruction request 87 generated by selection of the button 84 triggers retrieval of k-space data 60 comprising oversampled k-space data from the PACS 40, 42 and re-reconstruction of this data 60 is performed by the reconstruction operation 62. However, instead of then transferring the re-reconstructed image 64 to the cropping operation 66, it is instead transmitted as a re-reconstructed image 64rr of the EFOV 72 to the radiology workstation 50 for display on the displays 52, 54. In this way, the radiologist is provided with additional image content in the region of the indication 80 (although there may be some aliasing or warping).

Referring briefly to fig. 3, it will be appreciated that a similar process may be performed to provide the user with the option of extending the image into an extended field of view (EFOV) along the Slice Selection (SS) direction, in the illustrative example of fig. 3, the axial brain slice is along the anatomical lateral medial direction. Again, a text message 82 is presented asking whether to increase the field of view to that indicated by indication 90. If the user selects to increase the FOV, for example, by selecting the "RR" (re-reconstruction ") selection button 84 using the mouse pointer 86 in the illustrative example, the oversampled k-space data in the slice selection direction is reused for this purpose. (in this case, the oversampled k-space data are additional peripheral slices of k-space data acquired to fill in the region indicated by the indication 90). K-space data comprising oversampled k-space data in the slice selection direction is retrieved from the PACS 40, 42 and a reconstruction of this data is performed by a reconstruction operation 62. However, instead of then cropping to a specified FOV, the uncut image with EFOV is instead transmitted as a re-reconstructed image of the EFOV to the radiology workstation 50 for display on the displays 52, 54. In this way, the radiologist is provided with additional image content (again, possibly with some aliasing or warping) in the region of the indication 90.

In the illustrative example of fig. 1, all image reconstructions or re-reconstructions are via a reconstruction operation 62 performed by the MRI controller 30. However, other distributions of computational load are contemplated. For example, the PACS 42, 44 may include programming to perform a re-reconstruction operation, or the radiology workstation 50 may be programmed to perform a re-reconstruction operation. Further, while the image display operation and the request to expand the field of view are performed by the radiology workstation 50 in the illustrative example, this may be done at another workstation (such as the clinician's workstation).

Although not illustrated, when the re-reconstructed image 64rr with EFOV is displayed at the radiology workstation 50 (or other clinician's workstation), it is contemplated to include the indication 80 to indicate to the radiologist or other clinician which regions of the image of the EFOV may have some aliasing or wrap around.

Referring now to fig. 4, in another application, k-space data acquired for one MRI examination is reused to provide oversampled k-space data for another MRI examination of adjacent or overlapping fields of view. The example of fig. 4 applies this to spinal MRI examinations. The leftmost image 100 of fig. 4 depicts an image of a thoracic MRI study 102. In order to reconstruct the image 100 without aliasing or warping, it is necessary to reconstruct k-space data for the (prescribed) FOV of the image 100 together with oversampled k-space data from a region extending in the superior anatomic direction "above" the FOV of the image 100, and also together with oversampled k-space data from a region extending in the inferior anatomic direction "below" the FOV of the image 100. Typically, acquiring this oversampled k-space data in the upper and lower extension regions will be done as part of a thoracic MRI study 102.

However, it is recognized herein that in many MRI spine examination scenarios (such as spine MRI examinations that are required in conjunction with pathologies such as bone metastases, multiple sclerosis, spinal cord compression, etc.), a thoracic spine MRI examination 102 in conjunction with a cervical spine MRI examination 104 and/or in conjunction with a lumbar spine MRI examination 106 is typically required. As diagrammatically shown in the right side of fig. 4, at least a portion of the k-space data of the cervical MRI exam 104 (if performed with the thoracic MRI exam 102) may be reused to supply oversampled k-space data in an upper anatomical region 110 located "above" the thoracic FOV of the image 100. Likewise, at least a portion of the k-space data of the lumbar MRI exam 106 (if performed with the thoracic MRI exam 102) may be reused to supply oversampled k-space data located in the lower anatomical region 112 "below" the thoracic FOV of the image 100. To reuse the data in this manner, the three spinal MRI examinations 102, 104, 106 should have the same angle. Preferably they should also have other common acquisition parameters (e.g. slice thickness, resolution, etc.), but this can be compensated by performing a re-sampling of the re-used oversampled k-space data.

Although not indicated in fig. 4, in a similar manner, reconstruction of k-space data of the cervical MRI exam 104 may suitably utilize at least a portion of the k-space data of the thoracic MRI exam 102 to supply oversampled k-space data in a lower anatomical region relative to the cervical FOV (i.e., in a region "below" the cervical FOV). Likewise, reconstruction of k-space data of the lumbar MRI exam 106 may suitably utilize at least a portion of the k-space data of the thoracic MRI exam 102 to supply oversampled k-space data in an upper anatomical region relative to the lumbar FOV (i.e., in a region "above" the lumbar FOV). In the lumbar example, the k-space data of the thoracic MRI 102 provides only half of the required oversampled k-space data, as the thoracic MRI 102 cannot provide oversampled k-space data in the lower anatomical region relative to the lumbar FOV (i.e., in the region "below" the lumbar FOV). However, with the above described re-use of k-space data, the k-space data acquisition for a series of cervical/thoracic/lumbar MRI examinations is significantly reduced. For example, if the average length of a set of cervical, thoracic and/or lumbar MRI examinations is 100 minutes, it is estimated that the disclosed method of reusing k-space data from adjacent spinal MRI examinations as oversampled k-space data as described above will save an average of 40 minutes per patient, thereby reducing the average patient scan time to 60 minutes.

Although a spinal MRI examination is used as an illustrative example in fig. 4, it will be appreciated that the disclosed methods are more generally applicable to MRI examination of any two (or more) adjacent body parts, for example, along the spine (as per fig. 4), along long bones, and along the abdomen/pelvis, as additional examples.

As another application, the method may be applied to adjacent stations of a multi-station imaging examination, where successive stations are adjacent or overlapping along the axial direction 24 (as indicated in fig. 1). In this case, the FOV is the station field of view of the multi-station MRI exam, and the k-space data includes k-space data of the station FOV and k-space data of adjacent or overlapping station fields of view of the multi-station MRI exam adjacent or overlapping the station FOV. The reconstruction operation 62 includes reconstructing k-space data including k-space data of adjacent or overlapping station fields of view as oversampled k-space data. The reconstruction 62 generates an extended image of an extended field of view (EFOV) that encompasses the station FOV and extends into adjacent or overlapping station fields of view. The cropping operation 66 operates to crop the extended image to the station FOV to generate an image of the station FOV without aliasing or warping. By this mechanism, the k-space data acquired for each station FOV is reduced, thereby reducing the total scan time for performing a multi-station MRI examination of a patient.

With reference to fig. 5, application of the disclosed method of reusing oversampled k-space data to provide an enlarged field of view (as described with reference to fig. 1-3) is described. In this application, potential patient callback MRI checks are avoided. Patient recall of anatomical structures or pathologies due to errors in properly prescribing a FOV or otherwise missing clinical interest can result in delays in patient diagnosis, inconvenience to the patient, reduced workflow efficiency, and loss of potential revenue. By saving the oversampled k-space data in the PACS 40, 42 as disclosed herein, when this occurs and the relevant or suspected anatomy is close to the FOV, cropped, or just outside the FOV (but within the EFOV), then the image can be re-reconstructed, eliminating the need for a callback. In the example of fig. 5, at a clinician workstation (e.g., radiology workstation 50), the clinician is presented with a display 110 (left hand side of fig. 5) of an image 112 having a prescribed field of view (FOV). The display also includes an indication 114 of additional image content that may be provided by utilizing the oversampled k-space data stored in the PACS 42, 44 as well as the user interface dialog text and controls 82, 84 previously described with reference to fig. 2 and 3. The clinician notes feature 116 at the lower perimeter of the FOV of image 112. Typically, a clinician will likely need a callback MRI exam requiring a larger or displaced field of view in order to capture the features 116. However, using the disclosed method, the clinician selects button 84 using a mouse (or alternatively some other user interface device) that controls a mouse pointer 86 in order to request a re-reconstruction of the extended field of view (EFOV). The re-reconstruction (without subsequent cropping) produces an image 120 of the EFOV within which image 120 the features 116 are well away from the periphery of the image. This method for reducing callback MRI examinations is more generally applicable to imaging any body part as long as cropped or peripheral or missing anatomical structures or pathologies fall within the region covered by the phase/frequency oversampling.

The invention has been described with reference to the preferred embodiments. Modifications and alterations will occur to others upon reading and understanding the preceding detailed description. It is intended that the exemplary embodiment be construed as including all such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.