Invention Grant
- Patent Title: Method and computer for local correction of gradient non-linearities
-
Application No.: US15994309Application Date: 2018-05-31
-
Publication No.: US10663550B2Publication Date: 2020-05-26
- Inventor: Thorsten Feiweier , Marcel Dominink Nickel
- Applicant: Siemens Healthcare GmbH
- Applicant Address: DE Erlangen
- Assignee: Siemens Healthcare GmbH
- Current Assignee: Siemens Healthcare GmbH
- Current Assignee Address: DE Erlangen
- Agency: Schiff Hardin LLP
- Priority: com.zzzhc.datahub.patent.etl.us.BibliographicData$PriorityClaim@214f409b
- Main IPC: G01R33/56
- IPC: G01R33/56 ; G01R33/565 ; G01R33/563

Abstract:
In a method and magnetic resonance (MR) apparatus for generating an approximated trace-weighted MR image of a subject, at least three diffusion-weighted MR images of the subject are created, based on different diffusion encoding directions and diffusion gradient fields with essentially equal diffusion encoding strengths. A weighting factor is determined by an optimization method based on spatial inhomogeneities of the diffusion gradient fields for each image point of the at least three diffusion-weighted MR images. Based on the weighting factors, the at least three diffusion-weighted MR images are combined to generate the approximated trace-weighted MR image. Furthermore, an ADC map is determined based on approximated trace-weighted images and the effective diffusion encoding strength. Furthermore, a trace-weighted MR image is synthesized with nominal diffusion encoding strength based on a trace-weighted image with effective diffusion encoding strength.
Public/Granted literature
- US20180348325A1 METHOD AND COMPUTER FOR LOCAL CORRECTION OF GRADIENT NON-LINEARITIES Public/Granted day:2018-12-06
Information query