ADAPTIVE RECONSTRUCTION OF MAGNETIC RESONANCE IMAGES

    公开(公告)号:EP3828579A1

    公开(公告)日:2021-06-02

    申请号:EP19212205.9

    申请日:2019-11-28

    摘要: The present disclosure relates to a method comprising: providing a trained machine learning model. The trained machine learning model is configured for reconstructing images from input data. The method comprises: receiving (201) a multidimensional array comprising M dimensional acquired data; determining (205) a subset of values of at least one K selected dimension of the array; for each value of the subset determining (207) a M-K dimensional array comprising the acquired data corresponding to the value, resulting in a set of M-1 dimensional arrays; inputting (209) the set of M-K dimensional arrays to the trained machine learning model, and receive a reconstructed image from the trained machine learning model.

    SELECTION OF ROTATION ANGLE FOR REDUCED FIELD OF VIEW IMAGING

    公开(公告)号:EP3805775A1

    公开(公告)日:2021-04-14

    申请号:EP19202159.0

    申请日:2019-10-09

    发明人: Benkert, Thomas

    摘要: The invention relates to a method for determining imaging parameters for an MR image generated by an MR imaging system of an object under examination with an imaging sequence in which a part of the object under examination is imaged with a reduced field of view obtained with 2 dimensional spatially selective RF excitation pulses and magnetic field gradients applied to 2 different directions, wherein the 2 dimensional spatially selective RF excitation pulses are tilted by a rotation angle alpha larger 0 with regard to an imaging plane where the MR image is generated, the method comprising:
    - determining a set of image sequence parameters of the imaging sequence,
    - determining a frequency offset of off-resonant tissue potentially present in the object under examination,
    - determining an allowed maximum position shift of the off-resonant tissue along a slice selection direction,
    - determining the rotation angle which leads to the allowed maximum shift for the off-resonant tissue based on the determined set of image sequence parameters,
    - providing the determined rotation angle to the MR imaging system in order to allow the MR imaging system to generate the MR image using the determined rotation angle in the imaging sequence.

    VERFAHREN ZUR AUFNAHME VON MAGNETRESONANZDATEN UMFASSEND DIE AUFNAHME VON FETTSELEKTIVEN NAVIGATORDATEN MITTELS SIMULTANER MEHRSCHICHTBILDGEBUNG

    公开(公告)号:EP3588122A1

    公开(公告)日:2020-01-01

    申请号:EP18180414.7

    申请日:2018-06-28

    发明人: Zeller, Mario

    摘要: Verfahren zur Aufnahme von Magnetresonanzdaten eines Aufnahmebereichs eines Patienten, insbesondere wenigstens eines Teils eines Kopfes des Patienten, mit einer Magnetresonanzeinrichtung (35), wobei zur Bewegungskorrektur zwischen Aufnahmezeitabschnitten (18) zur Aufnahme von Magnetresonanzdaten in Navigatorzeitabschnitten (17) mittels einer Fettnavigatorsequenz, umfassend ein fettselektives Anregungsmodul mit wenigstens einem Hochfrequenzpuls und ein in der aufgenommenen Schicht unterabtastendes Auslesemodul, Navigatordaten (20) aufgenommen werden, aus denen zur Bewegungskorrektur der Magnetresonanzdaten zu verwendende Bewegungsdaten ermittelt werden, wobei die Navigatordaten (20) mittels der simultane Mehrschichtbildgebung nutzenden Fettnavigatorsequenz nach dem auf mehrere aufzunehmende Fettnavigatorschichten wirkenden Anregungsmodul im Auslesemodul simultan aus den angeregten Fettnavigatorschichten aufgenommen werden.

    AUTOMATED NONINVASIVE DETERMINING THE FERTILITY OF A BIRD'S EGG

    公开(公告)号:EP3523649A1

    公开(公告)日:2019-08-14

    申请号:EP18796987.8

    申请日:2018-11-13

    摘要: Shown herein is a method of automated noninvasive determining the fertility of a bird's egg (14), comprising the following steps: conveying a plurality of bird eggs (14) sequentially or in parallel into an NMR apparatus (18), subjecting the bird eggs (14) to an NMR measurement, such as to generate a 3-D NMR image of at least a part of each of said eggs (14), said 3-D NMR image having a spatial resolution in at least one dimension of 1.0 mm or less, preferably of 0.50 mm or less, wherein said part of the egg (14) includes the germinal disc of the respective egg (14), determining a prediction of the fertility according to at least one of the following two procedures: (i) deriving at least one feature from each of said 3-D NMR images, and employing said at least one feature in a feature-based classifier for determining a prediction of the fertility, and (ii) using a deep learning algorithm, and in particular a deep learning algorithm based on convolutional neural networks, generative adversarial networks, recurrent neural networks or long short-term memory networks.

    MAGNETIC RESONANCE IMAGING GUIDED THERAPY SYSTEM

    公开(公告)号:EP3484582A1

    公开(公告)日:2019-05-22

    申请号:EP17739567.0

    申请日:2017-07-13

    摘要: It is an object of the invention to improve the information about the position of a structure of interest (e.g. one or more organs and possibly their surroundings, a therapeutic target and/or organs at risk). This object is achieved by a magnetic resonance imaging system configured for determining whether a structure of interest is within a predefined region of interest, wherein the magnetic resonance imaging system comprises—a memory for storing computer program code means for causing the magnetic resoance system to carry out the following steps: —applying a multiband magnetic resonance imaging sequence in order to simultaneously acquire a first slice of magnetic resonance data a first slice location and a second slice of magnetic resonance data from a second and different slice location, wherein the first and second slice location define the region of interest and; —a determination step for determining based on the first and slice of magnetic resonance data and prior knowledge about the structure of interest and/or its surroundings whether the structure of interest is within the region of interest; wherein the magnetic resonance imaging system further comprises a processor for executing the computer program code means.