Repository logo
 
Publication

FMRI 3D registration based on Fourier space subsets using neural networks

dc.contributor.authorFreire, Luís
dc.contributor.authorGouveia, A. R.
dc.contributor.authorGodinho, F. M.
dc.date.accessioned2013-12-23T18:37:03Z
dc.date.available2013-12-23T18:37:03Z
dc.date.issued2010
dc.description.abstractIn this work, we present a neural network (NN) based method designed for 3D rigid-body registration of FMRI time series, which relies on a limited number of Fourier coefficients of the images to be aligned. These coefficients, which are comprised in a small cubic neighborhood located at the first octant of a 3D Fourier space (including the DC component), are then fed into six NN during the learning stage. Each NN yields the estimates of a registration parameter. The proposed method was assessed for 3D rigid-body transformations, using DC neighborhoods of different sizes. The mean absolute registration errors are of approximately 0.030 mm in translations and 0.030 deg in rotations, for the typical motion amplitudes encountered in FMRI studies. The construction of the training set and the learning stage are fast requiring, respectively, 90 s and 1 to 12 s, depending on the number of input and hidden units of the NN. We believe that NN-based approaches to the problem of FMRI registration can be of great interest in the future. For instance, NN relying on limited K-space data (possibly in navigation echoes) can be a valid solution to the problem of prospective (in frame) FMRI registration.por
dc.identifier.citationFreire L, Gouveia AR, Godinho FM. FMRI 3D registration based on Fourier space subsets using neural networks. In 2010 Annual International Conference of the IEEE - Engineering in Medicine and Biology Society (EMBC). IEEE; 2010. p. 5624-7.por
dc.identifier.isbn978-1-4244-4123-5
dc.identifier.urihttp://hdl.handle.net/10400.21/3028
dc.language.isoengpor
dc.peerreviewedyespor
dc.publisherIEEEpor
dc.relation.publisherversionhttp://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=5628038por
dc.subjectArtificial neural networkspor
dc.subjectBiomedical imagingpor
dc.subjectImage registrationpor
dc.subjectSolid modelingpor
dc.subjectThree dimensional displayspor
dc.subjectTime series analysispor
dc.subjectComputer simulationpor
dc.subjectFourier analysispor
dc.subjectImage interpretation, Computer-assistedpor
dc.subjectImaging, Three-dimensionalpor
dc.subjectMagnetic resonance imagingpor
dc.subjectNeural networkspor
dc.subjectTime factorspor
dc.titleFMRI 3D registration based on Fourier space subsets using neural networkspor
dc.typebook part
dspace.entity.typePublication
oaire.citation.endPage5627por
oaire.citation.startPage5624por
rcaap.rightsrestrictedAccesspor
rcaap.typebookPartpor

Files

Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
FMRI 3D registration based on Fourier space subsets using neural networks.pdf
Size:
162.7 KB
Format:
Adobe Portable Document Format
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description: