An improved PSF mapping method for EPI distortion correction in human brain at ultra high field (7T)
Object To further improve the quality and robustness of the point-spread function (PSF) mapping method for fully automatic and accurate correction of geometric distortions in EPI at ultra high field such as 7 Tesla with high fidelity. Materials and methods Conventional gradient-echo EPI and correspo...
Saved in:
Published in | Magma (New York, N.Y.) Vol. 24; no. 3; pp. 179 - 190 |
---|---|
Main Authors | , , , , , |
Format | Journal Article |
Language | English |
Published |
Berlin/Heidelberg
Springer-Verlag
01.06.2011
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | Object
To further improve the quality and robustness of the point-spread function (PSF) mapping method for fully automatic and accurate correction of geometric distortions in EPI at ultra high field such as 7 Tesla with high fidelity.
Materials and methods
Conventional gradient-echo EPI and corresponding PSF reference data in phantoms and in in vivo measurements of the human brain were acquired at 7 Tesla. To accurately determine and correct geometric distortions, i.e., in peripheral areas, the method calculates the local shift in the distorted phase-encoding dimension instead of the non-distorted spin-wrap encoding dimension. The results of the proposed method are compared with those originally demonstrated (Zaitsev et al. Magn Reson Med (
2004
) 52:1156–1166).
Results
The results show that the proposed method allows measurement and correction of the geometric distortions in EPI with high accuracy, and reduction of residual blurring. In addition, this method prevents errors in the shift map induced by PSF-ghost artifacts.
Conclusion
More precise mapping and correction of pixel shifts and blurring is accomplished with the proposed improvements. Errors in the shift map that are caused by PSF-ghost artifacts in the multi-shot PSF acquisition, e.g., from small motion during the reference scan, do not affect the improved shift estimation method. |
---|---|
Bibliography: | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
ISSN: | 0968-5243 1352-8661 |
DOI: | 10.1007/s10334-011-0251-1 |