(19)
(11)EP 2 728 371 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
27.07.2022 Bulletin 2022/30

(21)Application number: 12191104.4

(22)Date of filing:  02.11.2012
(51)International Patent Classification (IPC): 
G01R 33/561(2006.01)
G01R 33/48(2006.01)
G01R 33/565(2006.01)
(52)Cooperative Patent Classification (CPC):
G01R 33/4822; G01R 33/561; G01R 33/56509

(54)

Segmented 3D Cartesian MR data acquisition using a randomized sampling pattern for compressed sensing image reconstruction

Segmentierte dreidimensionale kartesische MR Datenaufnahme unter Verwendung eines stochastischen Abtastmusters für die Bildrekonstruktion mittels komprimierter Abtastung

Acquisition Cartésienne tridimensionnelle segmentée de données de résonance magnétique utilisant une grille d'échantillonnage aléatoire pour la reconstruction d'image par moyen de détection comprimée


(84)Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

(43)Date of publication of application:
07.05.2014 Bulletin 2014/19

(73)Proprietor: Universitätsklinikum Freiburg
79106 Freiburg (DE)

(72)Inventors:
  • Li, Guobin
    79115 Freiburg (DE)
  • Hennig, Jürgen
    79100 Freiburg (DE)
  • Zaitsev, Maxim
    79117 Freiburg (DE)

(74)Representative: Kohler Schmid Möbus Patentanwälte 
Partnerschaftsgesellschaft mbB Gropiusplatz 10
70563 Stuttgart
70563 Stuttgart (DE)


(56)References cited: : 
JP-A- 3 004 837
US-A1- 2009 082 656
JP-A- 7 194 575
  
  • NADINE GDANIEC ET AL: "Novel Sampling Strategy for Abdominal Imaging with Incomplete Breathholds", PROC.INTL.SOC.MAG.RESON.MED. 20, 5 May 2012 (2012-05-05), page 600, XP55054347,
  • TOLGA CUKUR ET AL: "Signal Compensation and Compressed Sensing for Magnetization-Prepared MR Angiography", IEEE TRANSACTIONS ON MEDICAL IMAGING, IEEE SERVICE CENTER, PISCATAWAY, NJ, US, vol. 30, no. 5, 1 May 2011 (2011-05-01), pages 1017-1027, XP011321148, ISSN: 0278-0062, DOI: 10.1109/TMI.2011.2116123
  • T.C. BASSE-LUESEBRINK ET AL: "Application of Compressed Sensing to 19F turbo spin echo chemical shift imaging", PROC.INTL.SOC.MAG.RESON.MED. 19, 1 January 2011 (2011-01-01), page 1502, XP55054350, ISSN: 1090-7807, DOI: 10.1016/j.jmr.2010.09.006
  • WILMAN A H ET AL: "PERFORMANCE OF AN ELLIPTICAL CENTRIC VIEW ORDER FOR SIGNAL ENHANCEMENT AND MOTION ARTIFACT SUPPRESSION IN BREATH-HOLD THREE-DIMENSIONAL GRADIENT ECHO IMAGING", MAGNETIC RESONANCE IN MEDICINE, ACADEMIC PRESS, DULUTH, MN, US, vol. 38, no. 5, 1 November 1997 (1997-11-01), pages 793-802, XP000741761, ISSN: 0740-3194
  • Gdaniec N. ET AL: "Towards Clinical Robustness in Abdominal Water-Fat Imaging", ESMRMB 2012 Congress, 5 October 2012 (2012-10-05), XP055794870, Retrieved from the Internet: URL:http://link.springer.com/content/pdf/1 0.1007%2Fs10334-012-0322-y.pdf>
  
Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


Description


[0001] The present invention relates to a method of magnetic resonance imaging (=MRI) technology for reduction of motion related artifacts in 3D MR data acquisition with multiple segments. In particular, the invention pertains to a method of magnetic resonance (=MR) imaging for the acquisition of a three-dimensional (=3D) dataset of an object of investigation, where spatial encoding by three mutually orthogonal magnetic field gradients is performed such, that signal is readout under a read-gradient in one spatial direction and spatial encoding in the other two spatial directions is performed by applying phase encoding gradients in the other two spatial directions prior to signal acquisition and data acquisition is performed in a sequential manner such, that at each acquisition step signal is acquired under said readout gradient, but with different combinations of the two phase encoding gradients.

Field of Technology



[0002] Such method is known from reference [3] (Busse et al.).

Background Art



[0003] The scan time of 3D MRI is long and is prone to artifacts caused by patient motion.

[0004] Recently developed k-space random sampling techniques (also known as Compressed Sensing, see reference [2]) support the image reconstruction using randomly sampled k-space datasets.

[0005] Recently developed view reordering (see reference [3]) supports the flexible filling of 3D gridded k-space with multiple segments.

[0006] In a recently developed method for dynamic MR angiography progressive filling of k-space is performed to allow for a trade-off of temporal resolution and image sharpness (see reference [4])

[0007] Navigator echo (see reference [5]) is an easy and reliable way for motion detection, which was already used to monitor the respiration motion, swallowing cased motion etc.

[0008] Nadine Gdaniec et al.: "Novel Sampling Strategy for Abdominal Imaging with Incomplete Breathholds", PROC. INTL. SOC. MAG. RESON. MED. 20, 5 May 2012, discloses a magnetic resonance imaging sampling strategy for abdominal imaging with incomplete breathholds.

[0009] Nadine Gdaniec et al.: "Towards Clinical Robustness in Abdominal Water-Fat Imaging", ESMRMB 2012 Congress, 5 October 2012, discloses a magnetic resonance imaging method comprising navigator-based motion detection for automatic scan termination.

[0010] To the present moment no method for 3D structural imaging is known which is capable of being interrupted due to the patient's or subject's motion, where it would be possible to reconstruct high quality anatomic images with only partial data and where the quality of the complete data set would not be compromised to allow for the flexibility of interrupting the scan.

Contents of the Invention



[0011] An object of the present invention is to propose a method for reduction of motion related artifacts in 3D MR data acquisition, where data are acquired under a readout gradient for spatial encoding in the first spatial dimension (referred to as K1). Data in the 2nd and 3rd spatial dimension are spatially encoded according using two mutually orthogonal phase encoding gradients (referred to as K2 and K3). The invention applies to any MR sequence suitable for this spatial encoding strategy. This includes 3D-acquisition with steady state techniques like gradient echoes, contrast prepared gradient echo techniques like MDEFT and MPRAGE as well as 3D-multiecho-techniques (3D-TSE, SPACE).

[0012] According to the present invention, these objects are achieved by modifying the initially defined generic method in that the temporal sequence of spatial encoding is applied such, that first a basic dataset containing N3D(b) phase encoding steps is acquired using a randomized sampling pattern in the two phase encoding directions, where the sampling density of the randomized sampling pattern decreases monotonously from the center of k-space to generate a first dataset suitable for reconstruction of strongly undersampled data by use of the compressed sensing method to generate a first 3D image volume of the object of investigation with a predefined sufficient minimum quality, that data acquisition then proceeds by acquiring a complimentary dataset in which N3D(c) additional phase encoding steps are acquired supplementing the already acquired data such, that the density of sampling points is increased allowing the reconstruction of 3D-image volumes of increasing image quality until data are acquired with a predefined sampling density to reconstruct a final 3D-volume of optimum quality, that spatial encoding in the acquisition of the basic as well as the complimentary dataset is performed in a segmented fashion in L segments with Nseg phase encoding steps in each segment, in which the total k-space sampling pattern of the basic or complimentary dataset is subdivided into a predefined number of Nseg concentric or rectilinear sectors containing L phase encoding steps each and for the acquisition of each phase encoding step in each segment one phase encoding step is selected from each sector such that each segment contains one phase encoding step from each sector, and that suitable monitoring of motion of the object of investigation is applied such that data acquisition is terminated whenever involuntary motion exceeds a predefined threshold, where motion monitoring is performed either using navigators built into the measurement sequence or by extrinsic motion detection devices.

[0013] In order to achieve the above mentioned objective, the present invention proposes a method comprising the following steps: the complete acquisition is divided into two parts: basic acquisition and complementary acquisition. Basic acquisition is performed at the beginning. Complementary acquisition is performed immediately after the basic acquisition is finished. The selection and sequence of phase encoding sampling points in the K2-K3-plane ('view ordering') is prepared for basic acquisition and complementary acquisition separately. Motion monitoring is performed regularly during the data acquisition. When motional displacement above a predefined threshold is detected, MR data acquisition stops. Image reconstruction is performed when motion occurs in the phase of complementary acquisition.

[0014] K-space data acquired in said complete acquisition is randomly sampled in the K2-K3-plane, and compressed Sensing reconstruction is used. The view ordering is optimized using the theory in Compressed Sensing. The amount of acquired data is determined empirically by the optimally desired image quality.

[0015] K-space data acquired in said basic acquisition is randomly sampled. Data acquired in the basic acquisition is a subset of the data acquired in the complete acquisition. The sampling pattern is optimized using the theory in Compressed Sensing. The amount of acquired data is determined empirically by the lowest acceptable image quality.

[0016] Said complementary acquisition is the acquisition of data acquired in complete acquisition, but not acquired in basic acquisition.

[0017] Said view reordering comprises:
  • First random sampling patterns with N3D(b) phase encoding points for the basic acquisition and N3D(c) phase encoding points for the complimentary acquisition are created. Sampling patterns for basic and complimentary acquisition are defined independently with the boundary condition that each phase encoding step is acquired only once.
  • The thus defined sampled k-space for each type of acquisition is subdivided into Nseg concentric or rectilinear sectors containing L phase encoding steps each such that L≈N3D/Nseg. L, Nseg and N3D are different between basic and complimentary acquisition. Sectors can be either concentric (see Fig.2) or rectilinear (see Fig.3) to generate favorable contrast with different parameter settings, for instance, concentric sectors are favorable for proton-density weighted imaging, and rectilinear sectors are favorable for T2-weighted imaging in acquisition using 3D TSE sequences.
  • Said subdividing sampled k-space into Nseg concentric or rectilinear sectors is performed by sorting all phase encoding points by their position in a defined direction, and equally dividing the sorted list into Nseg sectors.
  • Data acquisition is performed in a segmented manner in L segments with Nseg phase encoding steps acquired in each segment.
  • For each segment each phase encoding step is selected from a different sector such that each segment contains one phase encoding step from each sector.
  • For acquisition schemes based on steady state sequences like gradient echoes the order of assignment can be arbitrary. For non-steady state acquisition schemes like TSE and contrast prepared gradient echo sequences the order of assignment follows any of the strategies described in the literature (ref.3 and references therein).
  • For the selection of phase encoding points within sector different strategies can be applied: In a fully random approach points are selected randomly within each sector. In a limited random approach sectors are further subdivided into regions (see Fig.2~4) and points are selected from corresponding regions in the different sectors. This approach can be used to minimize the length of the k-space trajectory within each segment, which may be beneficial in terms of i.e. eddy current behavior, and limitations of total gradient power.


[0018] Said View ordering of basic acquisition and complementary acquisition is prepared separately.

[0019] One option to implement the said motion monitoring is: the first echo of each train is acquired with only readout gradient; motion is detected by comparing the acquired echo to a reference echo acquired at the beginning of the basic acquisition. Significant difference of two echoes indicates the occurrence of object motion; the direction of readout gradient could be adjusted in each examination.

[0020] Another option to implement the said motion monitoring is: image reconstruction is performed regularly when new data is acquired. Motion is detected by comparing the new reconstructed image with the previous reconstructed image. Significant inconsistence of two reconstructed images indicates the occurrence of object motion.

[0021] Another option to implement the said motion monitoring is to use external devices, for example optical sensor, or a camera, breathing belt, etc.

[0022] Said image reconstruction is performed by algorithms developed in the area of Compressed Sensing technique.

[0023] Said MR data acquisition with multiple echo trains includes, but is not limited to these sequences: 3D Turbo Spin Echo sequences; 3D Gradient Echo sequences, which acquire multiple echoes after each magnetization preparation pulse, for example 3D MPRAGE.

Description of Accompanying Drawings



[0024] In order to make the aforesaid and other features and advantages of the present invention more apparent to those skilled in the art, preferred embodiments of the present invention will be described in detail below by referring to the accompanying drawings, wherein identical numerals represent the same parts, and in the accompanied drawings:

Fig. 1 is an illustration of the complete acquisition, basic acquisition, and complementary acquisition, where gridded Poisson-disk sampling is used in complete acquisition and basic acquisition;

Fig. 2 is an illustration of the concentric view ordering;

Fig. 3 is an illustration of the rectilinear view ordering;

Fig. 4 is an illustration of the interleaved random trajectories used in basic acquisition and complementary acquisition;

Fig. 5 illustrates the detection of motion by comparing navigator echoes acquired at the beginning of each segment; and

Fig. 6 shows a comparison of the reconstruction with all data, including motion corrupted data, and the reconstruction without motion corrupted data.


Embodiments of the Invention



[0025] In order to make the object, technical solution and advantage of the present invention more apparent, the present invention will be further described in detail in conjunction with accompanying drawings and embodiments. It should be understood that the embodiments explained here are merely to explain the present invention; they are not to define the present invention.

[0026] Fig. 1 shows the complete acquisition (left) consisting of two parts: basic acquisition (middle) and complementary acquisition (right).

[0027] Fig. 2 depicts the concentric view ordering, where k-space is divided into concentric sectors (see solid circles). Each sector is subdivided into multiple regions (see dashed lines).

[0028] Fig. 3 shows the rectilinear view ordering, where k-space points are divided into multiple sectors in the direction of the 2nd phase encoding K2 (see solid lines). Each sector is subdivided into multiple regions in the direction of the 3rd phase encoding K3 (see dashed lines).

[0029] Fig. 4 is an illustration of the concentric view ordering with limited random trajectories, where there are 5 sectors and each sector is further subdivided into 8 regions.

[0030] Fig. 5 demonstrates a significant change of acquired navigators shows the time when the motion happened in the knee examination.

[0031] In Fig. 6

a) is the k-space sampling pattern of the complete acquisition;

b) ~d) are three perpendicular slices reconstructed with all acquired data;

e) is a mask showing the actually acquired k-space data before motion happened;

f) ~h) are the same slices as in b)~d), but reconstructed without motion corrupted data.



[0032] According to the invention, the numbers Nseg and L are different between the basic and the complimentary acquisition and chosen such that NsegL≈N3D(b) and N3D(c) respectively.

[0033] In another embodiment of the invention, the sectors subdividing the k-space sampling pattern for the basic or complimentary dataset are arranged concentrically in the k-space spanned by the two phase encoding gradients.

[0034] In another embodiment of the invention, the the sectors subdividing the k-space sampling pattern for the basic or complimentary dataset are rectangular.

[0035] In a class of further embodiments of the invention, the selection of phase encoding points within each sector is performed randomly within each sector.

[0036] In an alternative class of embodiments, each sector is further subdivided into regions and phase encoding points within each segment are chosen from corresponding regions.

[0037] In another class of embodiments of the invention, for data acquisition a 3D-multiecho-technique like 3D-TSE or SPACE is used such, that the echo train length is equal to Nseg.

[0038] In an alternative class of embodiments, for data acquisition a 3D-gradient echo like 3D-spoiled FLASH, 3D-FLASH, 3D-trueFISP or 3D-VIBE is used.

[0039] Still another class of embodiments of the invention is characterized in that for data acquisition a contrast prepared 3D-gradient echo technique like MPRAGE or MDEFT is used.

[0040] The present invention discloses a method for reduction of motion artifacts in 3D MR data acquisition with multiple segments, comprising: the complete acquisition is divided into two parts: basic acquisition and complementary acquisition. Basic acquisition is performed at the beginning. Complementary acquisition is performed after the basic acquisition is finished. View Reordering is prepared for basic acquisition and complementary acquisition separately. Motion monitoring is performed regularly during the data acquisition. Whenever motion is detected, data acquisition stops. Image reconstruction is performed when motion occurs in the phase of complementary acquisition. The final reconstructed image is free of motion artifacts.

References



[0041] 
  1. [1] View-sharing in keyhole imaging: Partially compressed central k-space acquisition in time-resolved MRA at 3.0 T Hadizadeh, Dariusch R.; Gieseke, Juergen; Beck, Gabriele; et al. Source: EUROPEAN JOURNAL OF RADIOLOGY Volume: 80 Issue: 2 Pages: 400-406 DOI: 10.1016/j.ejrad.2010.04.020 Published: NOV 2011
  2. [2] Lustig, M.; Donoho, D.L.; Santos, J.M.; Pauly, J.M.; , "Compressed Sensing MRI," Signal Processing Magazine, IEEE , vol.25, no.2, pp.72-82, March 2008 doi: 10.1109/MSP.2007.914728
  3. [3] Effects of refocusing flip angle modulation and view ordering in 3D fast spin echo. Busse, R. F., Brau, A. C.S., Vu, A., Michelich, C. R., Bayram, E., Kijowski, R., Reeder, S. B. and Rowley, H. A. (2008), Magnetic Resonance Medicine, 60: 640-649. doi: 10.1002/mrm.21680
  4. [4] Buildup of Image Quality in View-Shared Time-Resolved 3D CE-MRA. Casey P. Johnson, Thomas W. Polley, James F. Glockner, Phillip M. Young, and Stephen J. Riederer. Magnet.Reson.Med. 2012 doi: 10.1002/mrm.24466
  5. [5] A Motion Monitored SPACE Sequence for Isotropic Cardotid Wall Imaging. Zhang Q et al, ISMRM 2012, p3831



Claims

1. A method of magnetic resonance imaging for the acquisition of a three-dimensional dataset of an object of investigation, where spatial encoding by three mutually orthogonal magnetic field gradients is performed such, that signal is readout under a read-gradient in one spatial direction k1 and spatial encoding in the other two spatial directions k2, k3 is performed by applying phase encoding gradients in the other two spatial directions k2, k3 prior to signal acquisition and data acquisition is performed in a sequential manner such, that at each acquisition step signal is acquired under said readout gradient, but with different combinations of the two phase encoding gradients,

wherein a complete dataset is acquired using a randomized sampling pattern in the k2-k3 plane, which comprises two datasets: a basic dataset and a complementary dataset,

wherein the temporal sequence of spatial encoding is applied such, that first a basic dataset containing N3D(b) phase encoding steps is acquired using a randomized sampling pattern in the two phase encoding directions, where the sampling density of the randomized sampling pattern decreases monotonously from the center of k-space to generate a first dataset suitable for reconstruction of strongly undersampled data by use of the compressed sensing method to generate a first 3D image volume of the object of investigation with a predefined sufficient minimum quality,

wherein data acquisition then proceeds by acquiring a complimentary dataset in which N3D(c) additional phase encoding steps are acquired supplementing the already acquired data such, that the sampling density allows the reconstruction of 3D-image volumes of increasing image quality until data are acquired with a predefined sampling density to reconstruct a final 3D-volume of optimum quality, and wherein monitoring of motion of the object of investigation is applied such that data acquisition is terminated whenever involuntary motion exceeds a predefined threshold, where motion monitoring is performed either using navigators built into the measurement sequence or by extrinsic motion detection devices,

characterized in

that spatial encoding in the acquisition of the basic as well as the complimentary dataset is performed according to a view ordering scheme in a segmented fashion in L segments with Nseg phase encoding steps in each segment, wherein according to the view ordering scheme, the total k-space sampling pattern of the basic or complimentary dataset is subdivided into a predefined number of Nseg concentric or rectilinear sectors containing L phase encoding steps each and for the acquisition of each phase encoding step in each segment one phase encoding step is selected from each sector such that each segment contains one phase encoding step from each sector, and

that the numbers Nseg and L are different between the basic and the complimentary acquisition and chosen such that Nseg and L are numbers and that NsegL≈N3D(b) and N3D(c) respectively.


 
2. The method according to claim 1, characterized in that the sectors subdividing the k-space sampling pattern for the basic or complimentary dataset are arranged concentrically in the k-space spanned by the two phase encoding gradients.
 
3. The method according to claim 1, characterized in that the sectors subdividing the k-space sampling pattern for the basic or complimentary dataset are rectangular in the k-space spanned by the two phase encoding gradients.
 
4. The method according to claim 1, characterized in that the selection of phase encoding points within each sector is performed randomly within each sector.
 
5. The method according to anyone of claims 1 through 3, characterized in that each sector is further subdivided into regions and phase encoding points within each segment are chosen from corresponding regions.
 
6. The method according to anyone of claims 1 through 5, characterized in that for data acquisition a 3D-multiecho-technique is used such, that the echo train length is equal to Nseg.
 
7. The method according to anyone of claims 1 through 5, characterized in that for data acquisition a 3D-gradient echo is used.
 
8. The method according to anyone of claims 1 through 5, characterized in that for data acquisition a contrast prepared 3D-gradient echo technique is used.
 


Ansprüche

1. Bildgebungsverfahren der Magnetresonanz zur Erfassung eines dreidimensionalen Datensatzes von einem Untersuchungsobjekt, wobei die räumliche Kodierung durch drei zueinander orthogonale Magnetfeldgradienten so durchgeführt wird, dass das Signal unter einem Auslesegradienten in einer Raumrichtung k1 ausgelesen wird und räumliche Kodierung in den anderen beiden Raumrichtungen k2, k3 durch Anlegen von Phasenkodiergradienten in den anderen beiden Raumrichtungen k2, k3 vor der Signalerfassung durchgeführt wird und die Datenerfassung in sequentieller Weise so durchgeführt wird, dass bei jedem Erfassungsschritt ein Signal unter dem Auslesegradienten erfasst wird, jedoch mit unterschiedlichen Kombinationen der beiden Phasenkodiergradienten,

wobei ein vollständiger Datensatz unter Verwendung eines randomisierten Abtastmusters in der k2-k3-Ebene erfasst wird, der zwei Datensätze umfasst: einen Basisdatensatz und einen Komplementärdatensatz, wobei die zeitliche Abfolge der räumlichen Kodierung so angewendet wird, dass zuerst ein Basisdatensatz, der N3D(b) Phasenkodierschritte enthält, unter Verwendung eines randomisierten Abtastmusters in den zwei Phasenkodierrichtungen erfasst wird, wobei die Abtastdichte des randomisierten Abtastmusters monoton von der Mitte des k-Raums abnimmt, um einen ersten Datensatz zu erzeugen, der für die Rekonstruktion von stark unterabgetasteten Daten mit Hilfe des Compressed-Sensing-Verfahrens geeignet ist, um ein erstes 3D-Bildvolumen des Untersuchungsobjekts mit einer vordefinierten ausreichenden Mindestqualität zu erzeugen, wobei die Datenerfassung dann fortfährt, indem ein Komplementärdatensatz erfasst wird, in dem N3D(c) zusätzliche Phasenkodierschritte erfasst werden, welche die bereits erfassten Daten so komplementieren, dass die Abtastdichte die Rekonstruktion von 3D-Bildvolumen von zunehmender Bildqualität ermöglicht, bis Daten mit einer vordefinierten Abtastdichte erfasst werden, um ein endgültiges 3D-Volumen von optimaler Qualität zu rekonstruieren,

und wobei eine Überwachung der Bewegung des Untersuchungsobjekts so angewandt wird, dass die Datenerfassung beendet wird, immer wenn eine unwillkürliche Bewegung einen vordefinierten Schwellenwert überschreitet, wobei die Bewegungsüberwachung entweder unter Verwendung von in die Messsequenz eingebauten Navigatoren oder durch extrinsische Bewegungserfassungsvorrichtungen durchgeführt wird,

dadurch gekennzeichnet,

dass die räumliche Kodierung bei der Erfassung sowohl des Basis- als auch des Komplementärdatensatzes gemäß einem Ansicht-Ordnungs-Schema in segmentierter Weise in L Segmenten mit Nseg Phasenkodierschritten in jedem Segment durchgeführt wird,

wobei gemäß dem Ansicht-Ordnungs-Schema das gesamte k-Raum-Abtastmuster des Basis- oder Komplementärdatensatzes in eine vordefinierte Anzahl von Nseg konzentrischen oder geradlinigen Sektoren unterteilt wird, die jeweils L Phasenkodierschritte enthalten, und für die Erfassung jedes Phasenkodierschritts in jedem Segment ein Phasenkodierschritt aus jedem Sektor ausgewählt wird, so dass jedes Segment einen Phasenkodierschritt aus jedem Sektor enthält, und

dass die Zahlen Nseg und L zwischen der Basiserfassung und der Komplementärerfassung unterschiedlich sind und so gewählt werden, dass Nseg und L Zahlen sind, und dass Nseg*Lr≈N3D(b) beziehungsweise N3D(c).


 
2. Verfahren nach Anspruch 1, dadurch gekennzeichnet, dass die Sektoren, die das k-Raum-Abtastmuster für den Basis- oder Komplementärdatensatz unterteilen, konzentrisch in dem von den beiden Phasenkodiergradienten aufgespannten k-Raum angeordnet sind.
 
3. Verfahren nach Anspruch 1, dadurch gekennzeichnet, dass die Sektoren, die das k-Raum-Abtastmuster für den Basis- oder Komplementärdatensatz unterteilen, in dem von den beiden Phasenkodiergradienten aufgespannten k-Raum rechtwinklig sind.
 
4. Verfahren nach Anspruch 1, dadurch gekennzeichnet, dass die Auswahl der Phasenkodierpunkte innerhalb jedes Sektors nach dem Zufallsprinzip innerhalb jedes Sektors durchgeführt wird.
 
5. Verfahren nach einem der Ansprüche 1 bis 3, dadurch gekennzeichnet, dass jeder Sektor weiter in Bereiche unterteilt wird und Phasenkodierpunkte innerhalb jedes Segments aus entsprechenden Bereichen ausgewählt werden.
 
6. Verfahren nach einem der Ansprüche 1 bis 5, dadurch gekennzeichnet, dass für die Datenerfassung eine 3D-Multiecho-Technik verwendet wird, so dass die Länge der Echozüge gleich Nseg ist.
 
7. Verfahren nach einem der Ansprüche 1 bis 5, dadurch gekennzeichnet, dass für die Datenerfassung ein 3D-Gradientenecho verwendet wird.
 
8. Verfahren nach einem der Ansprüche 1 bis 5, dadurch gekennzeichnet, dass für die Datenerfassung eine Kontrast-aufbereitete 3D-Gradientenechotechnik verwendet wird.
 


Revendications

1. Procédé d'imagerie par résonance magnétique pour l'acquisition d'un jeu de données tridimensionnelles d'un objet d'investigation, procédé selon lequel un codage spatial au moyen de trois gradients de champ magnétique mutuellement orthogonaux est réalisé de telle sorte qu'un signal soit lu sous un gradient de lecture dans une direction spatiale k1 et selon lequel un codage spatial dans les deux autres directions spatiales k2, k3 est réalisé en appliquant des gradients de codage de phase dans les deux autres directions spatiales k2, k3 avant une acquisition de signal, et une acquisition de données est réalisée d'une manière séquentielle de telle sorte que, au niveau de chaque étape d'acquisition, le signal soit acquis sous ledit gradient de lecture mais avec des combinaisons différentes des deux gradients de codage de phase ;

dans lequel un jeu de données complet est acquis en utilisant une grille d'échantillonnage aléatoire dans le plan k2-k3, lequel jeu de données complet comprend deux jeux de données : un jeu de données de base et un jeu de données complémentaire ;

dans lequel la séquence temporelle de codage spatial est appliquée de telle sorte qu'en premier lieu, un jeu de données de base qui contient N3D(b) étapes de codage de phase soit acquis en utilisant une grille d'échantillonnage aléatoire dans les deux directions de codage de phase, où la densité d'échantillonnage de la grille d'échantillonnage aléatoire décroît de façon monotone depuis le centre de l'espace k, pour générer un premier jeu de données qui convient pour la reconstruction de données fortement sous-échantillonnées au moyen de l'utilisation du procédé par détection comprimée afin de générer un premier volume d'images 3D de l'objet d'investigation qui présente une qualité minimum suffisante prédéfinie ;

dans lequel l'acquisition de données passe ensuite à l'acquisition d'un jeu de données complémentaire selon lequel N3D(c) étapes de codage de phase additionnelles sont effectuées en complément des données déjà acquises, de telle sorte que la densité d'échantillonnage permette la reconstruction de volumes d'images 3D qui présentent une qualité d'image croissante jusqu'à ce que des données soient acquises selon une densité d'échantillonnage prédéfinie pour reconstruire un volume 3D final de qualité optimum ; et

dans lequel une surveillance du déplacement de l'objet d'investigation est effectuée de telle sorte que l'acquisition de données soit terminée chaque fois qu'un déplacement involontaire excède un seuil prédéfini, où la surveillance du déplacement est réalisée soit en utilisant des navigateurs qui sont intégrés à l'intérieur de la séquence de mesure, soit au moyen de dispositifs de détection de mouvement extrinsèques ;

caractérisé :

en ce que le codage spatial lors de l'acquisition du jeu de données de base de même que lors de l'acquisition du jeu de données complémentaire est réalisé conformément à un schéma de hiérarchisation par classement des vues d'une façon segmentée selon L segments moyennant Nseg étapes de codage de phase dans chaque segment ;

dans lequel, conformément au schéma de hiérarchisation par classement des vues, la grille d'échantillonnage de l'espace k total du jeu de données de base ou du jeu de données complémentaire est subdivisée selon un nombre prédéfini de Nseg secteurs concentriques ou rectilignes qui contiennent chacun L étapes de codage de phase, et pour la réalisation de chaque étape de codage de phase, dans chaque segment, une étape de codage de phase est sélectionnée à partir de chaque secteur de telle sorte que chaque segment contienne une étape de codage de phase en provenance de chaque secteur ; et

en ce que les nombres Nseg et L sont différents entre l'acquisition de base et l'acquisition complémentaire et ils sont choisis de telle sorte que Nseg et L soient respectivement des nombres et que Nseg * L ≈ N3D(c).


 
2. Procédé selon la revendication 1, caractérisé en ce que les secteurs qui subdivisent la grille d'échantillonnage de l'espace k pour le jeu de données de base ou pour le jeu de données complémentaire sont agencés de façon concentrique dans l'espace k qui est couvert par les deux gradients de codage de phase.
 
3. Procédé selon la revendication 1, caractérisé en ce que les secteurs qui subdivisent la grille d'échantillonnage de l'espace k pour le jeu de données de base ou pour le jeu de données complémentaire sont rectangulaires dans l'espace k qui est couvert par les deux gradients de codage de phase.
 
4. Procédé selon la revendication 1, caractérisé en ce que la sélection de points de codage de phase à l'intérieur de chaque secteur est réalisée de façon aléatoire à l'intérieur de chaque secteur.
 
5. Procédé selon l'une quelconque des revendications 1 à 3, caractérisé en ce que chaque secteur est en outre subdivisé selon des régions, et des points de codage de phase à l'intérieur de chaque segment sont choisis à partir de régions correspondantes.
 
6. Procédé selon l'une quelconque des revendications 1 à 5, caractérisé en ce que, pour l'acquisition de données, une technique à multiples échos 3D est utilisée de telle sorte que la longueur du train d'échos soit égale à Nseg.
 
7. Procédé selon l'une quelconque des revendications 1 à 5, caractérisé en ce que, pour l'acquisition de données, un écho de gradient 3D est utilisé.
 
8. Procédé selon l'une quelconque des revendications 1 à 5, caractérisé en ce que, pour l'acquisition de données, une technique par écho de gradient 3D préparé par contraste est utilisée.
 




Drawing














Cited references

REFERENCES CITED IN THE DESCRIPTION



This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

Non-patent literature cited in the description