- 专利标题: Methods for scan-specific k-space interpolation reconstruction in magnetic resonance imaging using machine learning
-
申请号: US16603336申请日: 2018-03-14
-
公开(公告)号: US11694373B2公开(公告)日: 2023-07-04
- 发明人: Mehmet Akcakaya , Steen Moeller
- 申请人: REGENTS OF THE UNIVERSITY OF MINNESOTA
- 申请人地址: US MN Minneapolis
- 专利权人: REGENTS OF THE UNIVERSITY OF MINNESOTA
- 当前专利权人: REGENTS OF THE UNIVERSITY OF MINNESOTA
- 当前专利权人地址: US MN Minneapolis
- 代理机构: Quarles & Brady LLP
- 国际申请: PCT/US2018/022348 2018.03.14
- 国际公布: WO2018/187005A 2018.10.11
- 进入国家日期: 2019-10-07
- 主分类号: G06K9/00
- IPC分类号: G06K9/00 ; G06T11/00 ; G01R33/48 ; G01R33/483 ; G01R33/56 ; G01R33/561 ; G06N3/04 ; G06N3/08
摘要:
Methods for reconstructing images from undersampled k-space data using a machine learning approach to learn non-linear mapping functions from acquired k-space lines to generate unacquired target points across multiple coils are described.
公开/授权文献
信息查询