公开(公告)号：US10859659B2

公开(公告)日：2020-12-08

申请号：US16174366

申请日：2018-10-30

Applicant: GE Precision Healthcare LLC

Inventor： Weinan Tang ,  Longqing Wang ,  Qingyu Dai ,  Dan Wu ,  Haoyang Xing

IPC: G01R33/565 ,  G01R33/56 ,  G01R33/567

Abstract: Embodiments of the present invention provide a motion monitoring method during MR imaging, comprising: acquiring a noise of a receiving coil before or after each imaging repetition time of an imaging scanning sequence; determining main coil channels associated with a motion of a scanned object in the receiving coil; determining a sum of squares of amplitudes of noises of the respective main coil channels; and filtering the sum of squares of amplitudes of noises of the main coil channels to obtain a motion track of the scanned object.

公开(公告)号：US11107211B2

公开(公告)日：2021-08-31

申请号：US15930884

申请日：2020-05-13

Applicant: GE Precision Healthcare LLC

Inventor： Dan Wu ,  Kun Wang ,  Weinan Tang ,  Longqing Wang

IPC: G06T7/00 ,  G06T7/80

Abstract: The present application provides a medical imaging method and system and a non-transitory computer-readable storage medium. The medical imaging method comprises identifying an image quality type of a medical image based on a trained learning network, and generating, based on the identified image quality type, a corresponding control signal for controlling a medical imaging system.

公开(公告)号：US11540788B2

公开(公告)日：2023-01-03

申请号：US17074849

申请日：2020-10-20

Applicant: GE Precision Healthcare LLC

Inventor： Qingyu Dai ,  Qilin Lu ,  Yaan Ge ,  Kun Wang ,  Longqing Wang

IPC: G06K9/00 ,  A61B6/04 ,  G06N3/02 ,  A61B6/00 ,  G06T7/70 ,  G06T7/00

Abstract: Embodiments of the present invention provide a method for identifying a body position of a detection object in medical imaging, a medical imaging system, and a computer-readable storage medium. The method comprises: receiving an image group by a trained deep learning network, the image group comprising a plurality of pre-scan images in a plurality of directions obtained by pre-scanning a detection object; and outputting body position information of the detection object by the deep learning network.

公开(公告)号：US20210123993A1

公开(公告)日：2021-04-29

申请号：US17065058

申请日：2020-10-07

Applicant: GE Precision Healthcare LLC

Inventor： Longqing Wang ,  Weinan Tang ,  Kun Wang ,  Hongbin Wang ,  Yongchuan Lai ,  Jiabin Yao

IPC: G01R33/24 ,  G01R33/54 ,  G01R33/483

Abstract: Embodiments of the present invention provide a magnetic resonance imaging method and system and a computer-readable storage medium. The method comprises: performing pre-scanning, wherein a first sequence and a second sequence are separately performed on a plurality of slices; in the first sequence, two echoes are continuously obtained to respectively obtain first image data and second image data having a first phase offset; in the second sequence, two echoes are continuously obtained to respectively obtain third image data and fourth image data having a second phase offset, the first phase offset and the second phase offset having opposite directions but the same angle; obtaining a plurality of radio-frequency field maps respectively corresponding to the plurality of slices based on the first image data and the second image data, and obtaining a plurality of static magnetic field maps respectively corresponding to the plurality of slices based on at least one of the following two groups: a first image and a third image, and a second image and a fourth image; and calculating formal scanning parameters suitable for a corresponding slice based on at least one of the following two components: each radio-frequency field map, and a corresponding static magnetic field map.

公开(公告)号：US11287493B2

公开(公告)日：2022-03-29

申请号：US17065058

申请日：2020-10-07

Applicant: GE Precision Healthcare LLC

Inventor： Longqing Wang ,  Weinan Tang ,  Kun Wang ,  Hongbin Wang ,  Yongchuan Lai ,  Jiabin Yao

IPC: G01V3/00 ,  G01R33/24 ,  G01R33/483 ,  G01R33/54

Abstract: Embodiments of the present invention provide a magnetic resonance imaging method and system and a computer-readable storage medium. The method comprises: performing pre-scanning, wherein a first sequence and a second sequence are separately performed on a plurality of slices; in the first sequence, two echoes are continuously obtained to respectively obtain first image data and second image data having a first phase offset; in the second sequence, two echoes are continuously obtained to respectively obtain third image data and fourth image data having a second phase offset, the first phase offset and the second phase offset having opposite directions but the same angle; obtaining a plurality of radio-frequency field maps respectively corresponding to the plurality of slices based on the first image data and the second image data, and obtaining a plurality of static magnetic field maps respectively corresponding to the plurality of slices based on at least one of the following two groups: a first image and a third image, and a second image and a fourth image; and calculating formal scanning parameters suitable for a corresponding slice based on at least one of the following two components: each radio-frequency field map, and a corresponding static magnetic field map.

公开(公告)号：US20210128084A1

公开(公告)日：2021-05-06

申请号：US17074849

申请日：2020-10-20

Applicant: GE Precision Healthcare LLC

Inventor： Qingyu Dai ,  Qilin Lu ,  Yaan Ge ,  Kun Wang ,  Longqing Wang

IPC: A61B6/04 ,  A61B6/00 ,  G06N3/02

Abstract: Embodiments of the present invention provide a method for identifying a body position of a detection object in medical imaging, a medical imaging system, and a computer-readable storage medium. The method comprises: receiving an image group by a trained deep learning network, the image group comprising a plurality of pre-scan images in a plurality of directions obtained by pre-scanning a detection object; and outputting body position information of the detection object by the deep learning network.