Recruiting & HR

B0 homogeneity throughout the monkey brain is strongly improved in the sphinx position as compared to the supine position

Description
B0 homogeneity throughout the monkey brain is strongly improved in the sphinx position as compared to the supine position
Published
of 5
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Technical Note B 0  Homogeneity Throughout the Monkey Brain IsStrongly Improved in the Sphinx Position asCompared to the Supine Position Julien Valette, MSc, 1 Martine Guillermier, MSc, 2 Fawzi Boumezbeur, MSc, 1 Cyril Poupon, PhD, 1 Alexis Amadon, PhD, 1 Philippe Hantraye, PhD, 1,2 andVincent Lebon, PhD 1 Purpose:  To map B 0  distortions throughout the monkey  brain in the two positions commonly used for NMR studies(thepronesphinxpositionandthesupineposition)inorder to test the hypothesis that B 0  homogeneity in the sphinx position is significantly improved as compared to the su-pine position. Materials and Methods:  Three macaque monkeys wereinstalledinthetwopositionsina3Twhole-bodyMRsystem withoutshimcorrection.B 0 mapswereacquiredusinga3Dgradient double-echo sequence, and field dispersionthroughout the brain was quantified. In addition, fieldmaps and localized  1 H spectra were acquired after first-order shimming was performed. Results:  The field maps collected in the three animals werehighly reproducible. B 0  dispersion throughout the brain was typically two to three times greater in the supine posi-tion than in the sphinx position. Although first-order shim-ming proved relatively more efficient in the supine position,B 0  dispersion still remained greater in the supine than inthe sphinx position. These findings can be explained by thethickness of outer brain tissues. Conclusion:  This work demonstrates that the sphinx posi-tion is highly favorable in terms of B 0  homogeneity. It should prove useful for NMR exploration of the monkey  brain, particularly at high fields where B 0  inhomogeneity associated with susceptibility artifacts is increased. Key Words:  monkey brain; sphinx position; supine posi-tion; B 0  homogeneity   J. Magn. Reson. Imaging 2006;23:408–412.© 2006 Wiley-Liss, Inc. HOMOGENEITY OF THE STATIC MAGNETIC FIELD iscrucial for many NMR applications, including func-tional MRI (fMRI) and MR spectroscopy (MRS). Poor shimming results in geometric distortions on echo-pla-nar images and increased linewidth on MR spectra.NMRexploration of the monkey brain has developedconsiderably over the last decade, including fMRIstudies of cognitive processes (1–10) and MRS studiesof animal models of brain disease (11–13). Reviewingthese studies reveals variability in the orientation of the monkey head relative to B 0 . Contrary to human or rodent studies, in which the head always has thesame orientation in the magnet, monkey studies areperformed in either a prone sphinx position (with themonkey face being oriented to the end of the magnet)or a conventional supine position. In experimentalsetups in which the monkeys sit in a vertical bore(3,4,7,8), the brain has the same orientation relativeto B 0  as it does in the supine position. It is possible tochoose these different head orientations because of the morphological properties of monkeys, and be-cause most primate studies are performed on humanNMR systems that are large enough to hold animalsin both positions; however, a literature review revealsthat the chosen orientation is rarely justified. Thesupine and sphinx positions are unlikely to be equiv-alent in terms of B 0  homogeneity throughout the brain. Indeed, air cavities (sinuses, nasal fossae, andauditory canal) are not regularly distributed aroundthe head. Additionally, the thickness of the outer tissue interface (skin, fat, muscle, and bone) aroundthe primate brain varies significantly. Therefore, B 0 homogeneity within the brain is expected to vary sig-nificantly with the orientation of the head, as previ-ously shown in humans (14).In this context our purpose was to investigate B 0 homogeneity throughout the monkey brain in thesphinx and supine positions, and to test the hypothesisthat B 0  homogeneity is significantly improved in thesphinx position compared to the supine position,mostly due to the distribution of outer tissues aroundthe brain. To illustrate the practical effect of head ori-entation, we also acquired B 0  maps and localized  1 Hspectra after first-order shimming for both head orien-tations. 1 Comissariat a` l’energie atomique, Service hospitalier Fre´de´ric Joliot,Orsay, France. 2 Unite´ de recherche associe´e comissariat a` l’energie atomique–centrenational de la recherche scientifique 2210, Orsay, France.*Address reprint requests to: V.L., CEA-SHFJ, 4 Place du Ge´ne´ralLeclerc, 91401 Orsay, France. E-mail: lebon@shfj.cea.fr Received February 22, 2005; Accepted November 30, 2005.DOI 10.1002/jmri.20511Published online 2 February 2006 in Wiley InterScience (www.interscience.wiley.com).  JOURNAL OF MAGNETIC RESONANCE IMAGING 23:408–412 (2006)© 2006 Wiley-Liss, Inc.  408  MATERIALS AND METHODS  Animal Preparation and Positioning  MRstudieswereconductedonthreemacaquemonkeys( Macaca fascicularis  , body weight   7 kg). The exper-iments were performed in accordance with the recom-mendations of the European Community (86/609) andthe French National Committee (87/848). The animals were anesthetized with an intramuscular ketamine-xy-lazineinjection.TheyweremonitoredwiththeuseofanMR-compatible Maglife system (Schiller Me´dical SA, Wissembourg, France). First the animal was positionedin the sphinx position, with its head being firmly main-tained between the side wedges of a human head coil. A  bite-barwasusedtokeeptheheadhorizontal.Localizer scans were acquired to carefully position the “center” of the brain at the magnet isocenter. Localizer scans wereobtained using a short TR gradient-echo sequence that acquires three orthogonal slices during a single acqui-sition (the k-spaces of the three slices are acquired inan interleaved manner). As a result, the signal of twoorthogonal slices (which are partly saturated) appearsas dark lines on the third slice, and the intersection of thedarklinesisthemagnetisocenter.Thefirstdataset  was acquired in the sphinx position (detailed below). Then the monkey was repositioned in the supine posi-tion. The orientation of the head and position of the bed were recursively adjusted until the brain position rela-tive to the isocenter in the second experiment matchedthe position in the first experiment. A second data set  was then acquired. NMR Acquisitions  Experiments were conducted on a 3 Tesla whole-body Medspec NMR system (Bruker Biospin, Ettlingen, Ger-many) equipped with a gradient coil reaching 45 mT/min 310  s. A human volume head coil was used, whichmade it possible to place the monkey head in bothpositions (sphinx and supine) without moving the coilpositioninthemagnet,andthereforelimitpossiblebiasassociated with coil sensitivity. B  0   Mapping Without Shimming   All shim currents were set to zero before the field maps were acquired. The B 0  field map was obtained using a 3D gradient double-echo sequence in which the twoechoes were separated by   t     3 msec, yielding twophase maps. The  B 0  map was then determined by thedifference between the two 3D maps divided by   t  , fol-lowed by a 3D unwrapping algorithm to resolve thephase cycling issue. The unwrapping algorithm was based on the 2D algorithm “weighted full multigridleast-square phase unwrapping” (15), and unwrappingalong the third dimension was obtained by minimizingthe phase difference between two subsequent slices. The double-echo sequence parameters were TR     50msec, TE1    6 msec, TE2    9 msec, and FOV     128mm with 1-mm resolution in all three directions. Effect of First-Order Shimming   To illustrate the practical effect of head orientation onthe MR images, we also acquired phase maps after performing automatic first-order shimming in both thesupine and sphinx positions in each monkey. The practical effect of head orientation on MR spectra  was assessed by measuring the water linewidth in a large voxel (  15 mL) positioned in the center of the brain. The voxel geometry was chosen to cover a typical brain chemical shift imaging (CSI) box. Using a stimu-lated-echo acquisition mode (STEAM) sequence(TE/TM  25/21 msec), first-order shimming was per-formed locally and a water spectrum was acquired. For the two positions (sphinx and supine), the geometry of the CSI-like voxel was defined relative to the magnet isocenter, with the two-voxel geometry (3.5  4  1 cmin the supine position and 3.5  1  4 cm in the sphinx position)differingbya90°Xrotation.Aslightdifferencecouldresultfromthefactthatthemonkeyheadwasnot perfectly orthogonal between both positions. Therefore,amanualcorrection(  1mm)waseventuallyperformedto restore the symmetric positioning in the coronal andaxial planes. Voxel positioning, shimming, and spec-trum acquisition were performed for both head orienta-tions in two monkeys. RESULTS  As demonstrated in Fig. 1, very similar positioning of the monkey head was achieved during the two phasesof the experiments (the sphinx and the supine mea-surements). Given the spatial resolution of the localizer scans, the position differences between the supine andsphinx positions are on the order of 1 mm.Images acquired in both positions were transformedfor display in the same reference frame. The relativeorientations of the brain in the magnet were almost perpendicular,allowingforamanualregisteringofdata sets with 90° and 180° rotations. A comparison of sagittal, coronal, and axial slices in both positions is presented in Fig. 2 for one monkey. Thethree monkeys presented very similar field maps and ex-hibitedsystematicallygreaterB 0 distortionsinthesupineposition. In particular, B 0  dispersion was greater in thefrontal and dorsal brain when the subject was positionedsupine. In the supine-positioned frontal brain, B 0  disper-sionalongthedorsoventralaxiswastwotothreetimesaslarge as in the sphinx position (2.3    0.3 vs. 0.8    0.2ppm,  N   3 , mean  SD,  P   0.005). The same holds truealong the axis from the extreme frontal lobe to the brainstem(3.6  0.3vs.1.7  0.1ppm,mean  SD, P   0.006). The axes along which the B 0  dispersions were measuredare shown on Fig. 2.First-order shimming proved to be relatively moreefficient in the supine position than in the sphinx posi-tion. In the supine position, B 0  dispersion was reduced by    45% after shimming, whereas the reduction was  20% in the sphinx position. However, this was insuf-ficient to bring B 0  dispersion down to the level mea-sured in the sphinx position. Compared to the sphinx position,B 0 dispersioninthesupinepositionwasabout three times higher without shimming, and about twotimes higher after first-order shimming. This phenom-enon was observed when B 0  dispersion was measuredalong both the dorsoventral axis and the axis from theextreme frontal lobe to the brain stem. B  0   Homogeneity in the Monkey Brain   409  For both experiments in which manual shimming was performed in a CSI-like voxel (Fig. 3), B 0  homoge-neity was improved in the sphinx position (full width at half maximum (FWHM) on water peak   11.6  0.8 vs.14.9  2.2 Hz). DISCUSSION  As shown on the B 0  maps, the head orientation relativeto the magnet had a significant effect on B 0  homogene-ity throughout the monkey brain when no shimming was performed. This effect was almost identical in thethree monkeys. The possible contribution of magnet imperfection to B 0  inhomogeneities measured in mon-keys can be considered negligible, since the brain(  5–6 cm diameter) was at the isocenter of the 100-cm whole-body magnet. The potential effect of the support-ing frame (i.e., the bite-bar and the side wedges) on B 0 homogeneity was tested on a spherical water phantom.Phase maps were acquired with and without bite-bar and side wedges, and no detectable effect on B 0  homo-geneity was demonstrated, as expected with MR-com-patible certified materials. As previously demonstratedin humans (16), anatomical considerations are morelikely to explain the effect of head orientation relative tothe magnet. When a macaque is positioned supine theB 0 fieldlinesgothroughthenasalfossaeontheirwaytotheventralsideofthefrontallobe(Fig.4).Moreover,theouter air–tissue interface is made of thin frontal andparietal tissue layers. This short air–tissue transition(6.2  1.3mmforthethreemonkeysasmeasuredonT  1 images) results in large B 0  inhomogeneities in the dor-salregionsofthefrontallobe.InthesphinxpositiontheB 0  field is parallel to the anteroposterior axis of the brain. This orientation is more favorable to B 0  homoge-neitybecausethethickarchoftheeyebrows(12.5  2.5mm) separates the brain from the air. In the posterior zone of the brain, the interface thickness is comparable Figure 1.  Localizer scans acquired in thesphinx and supine positions during the sameexperiment. Dark lines intersect at the mag-net isocenter, demonstrating very similar po-sitioning of the brain relative to the isocenter in the sphinx and supine positions. 410 Valette et al.  inbothpositions(10.3  1.5mminthesphinxposition,and 11.3    1.2 mm in the supine position). This ex-plains why B 0  distortion in the supine position is lesspronouncedintheoccipitalregionthaninthefrontalor parietal regions. The hypothesis that outer tissue inter-faces (skin, fat, muscle, and bone) heavily dominate B 0 inhomogeneity within the brain is in agreement withthefactthatanadaptedmoldcan“reject”theair–tissueinterface far from the brain (17,18). When the macaqueis held in the sphinx position, this shielding effect isnaturally achieved by facial tissues, although air-filledsinuses are likely to decrease the air–tissue interface inthe frontal zone for the sphinx position. The field maps presented here reflect the “natural”homogeneity of B 0  (i.e., when no shim currents areapplied). Although first-order shimming was rela-tively more efficient in the supine position, it did not enable us to bring B 0  dispersion down to the levelmeasured in the sphinx position. After first-order shimming was performed, B 0  dispersion through theentire brain was about twofold higher in the supineposition compared to the sphinx position. This obser- vation is consistent with our measurements of water linewidth within a “CSI-like” voxel centered in the brain. After first-order shimming, the water linewidth was smaller in the sphinx position (11.6 Hz) com- Figure 2.  Comparison of sag-ittal, coronal, and axial slicesin the sphinx and supine posi-tions for one monkey (all shimcurrents were set to zero). Thefield map is superimposed onthe magnitude image. Thesmall black dot is the isocenter position on the current slice.Eachcolorstepofthefieldmapstands for 0.3 ppm. B 0  disper-sion was measured along thedorsoventral axis (dotted line)and the axis from the extremefrontal lobe to the brain stem(dashed line). Figure 3.  Position of the CSI-like voxel in the monkey head. B  0   Homogeneity in the Monkey Brain   411  pared to the supine position (14.9 Hz), reflecting thefact that B 0  dispersion in the 15 mL voxel was   1.3times higher in the supine position. The 1.3-fold dif-ference in water linewidth is smaller than the twofoldincrease in B 0  dispersion measured along two axescovering the entire brain. This can be explained inpart by the fact that shimming is more efficient inspatially limited areas of the brain (e.g., a 15-mL  voxel) than in the entire monkey brain (  200 mL).For localized spectroscopy performed on MR researchsystems, second-order shimming might correct for B 0 dispersion within a given voxel of interest, with the useof field map-guided shimming procedures. For NMR studies focusing on a limited brain area, such as theoccipital lobe, second-order shimming might partly compensate for natural B 0  distortions. However, high-order shim coils are unable to homogenize B 0  through-out the entire brain, since B 0  distribution throughout the brain can hardly be described by second-order spherical harmonics. The inability to correct for whole- brain B 0  inhomogeneities results in severe geometricdistortions on whole-brain echo-planar imaging (EPI)even when high-order shim coils are available (19). Fi-nally, second-order shim coils are simply not availablefor the clinical MR systems used in the vast majority of primate studies.In conclusion, the orientation of the monkey headrelative to B 0  has a large effect on B 0  homogeneity, which is likely due to the uneven thickness of the outer tissue interface around the head. Global homogeneity throughout the brain is typically two to three times better in the sphinx position compared to the supineposition. This work should prove useful for MR explo-ration of the monkey brain, especially for whole-brainimaging and studies at high field strength, where the“natural” homogeneity of B 0  is strongly affected. REFERENCES 1. Denys K, Vanduffel W, Fize D, et al. The processing of visual shapein the cerebral cortex of human and nonhuman primates: a func-tional magnetic resonance imaging study. J Neurosci 2004:24:2551–2565.2. Fize D, Vanduffel W, Nelissen K, et al. The retinotopic organization of primate dorsal V4 and surrounding areas: a functional magnetic res-onance study in awake monkeys. J Neurosci 2003:23:7395–7406.3. Logothetis NK, Pauls J, Augath M, Trinath T, Oeltermann A. Neu-rophysiological investigation of the basis of the fMRI signal. Nature2001:412:150–157.4. Logothetis NK. MR imaging in the non-human primates: studies of function and of dynamic connectivity. Curr Opin Neurobiol 2003:13:630–642.5. Morita M, Nakahara H, Hayashi T. A rapid presentation event-related functional magnetic resonance imaging study of responseinhibition in macaque monkeys. Neurosci Lett 2004:356:203–206.6. Orban GA, Fize D, Peuskens H, et al. Similarities and differences inmotion processing between the human and macaque brain: evi-dence from fMRI. Neuropsychology 2003:41:1757–1768.7. Rainer G, Lee H, Logothetis NK. The effect of learning on the functionof monkey extrastriate visual cortex. PLoS Biol 2004:2:275–283.8. SerenoME,TrinathT,AugathM,LogothetisNK.Three-dimensionalshape representation in monkey cortex. Neuron 2002:33:635–652.9. Vanduffel W, Fize D, Mandeville JB, et al. Visual motion processinginvestigated using contrast agent-enhanced fMRI in awake behav-ing monkeys. Neuron 2001:32:565–577.10. Vanduffel W, Fize D, Peuskens H, et al. Extracting 3D from motion:differences in human and monkey intraparietal cortex. Science2002:298:413–415.11. Greco JB, Westmoreland SV, Ratai EM, et al. In vivo 1H MRS of  brain injury and repair during acute SIV infection in the macaquemodel of neuroAIDS. Magn Reson Med 2004:51:1108–1114.12. Mathew SJ, Shungu DC, Mao X, et al. A magnetic resonance spec-troscopic imaging study of adult nonhuman primates exposed toearly-life stressors. Biol Psychiatry 2003:54:727–735.13. Roitberg B, Khan N, Tuccar E, et al. Chronic ischemic stroke modelincynomolgusmonkeys:behavioral,neuroimagingandanatomicalstudy. Neurol Res 2003:25:68–78.14. Tyszka JM, Mamelak AN. Quantification of B 0  homogeneity varia-tionwithheadpitchbyregisteredthree-dimensionalfieldmapping. J Magn Reson 2002:159:213–218.15. Ghiglia DC, Pritt MD. Two-dimensional phase unwrapping theory,algorithms and software. New York: John Wiley and Sons, Inc.;1998. 512 p.16. Li S, Williams GD, Frisk TA, Arnold BW, Smith MB. A computer simulation of the static magnetic field distribution in the humanhead. Magn Reson Med 1995:34:268–275.17. Landuyt W, Sunaert S, Farina D, et al. In vivo animal functionalMRI: improved image quality with a body-adapted mold. J MagnReson Imaging 2002:16:224–227.18. Wilson JL, Jenkinson M, Jezzard P. Optimization of static fieldhomogeneity in human brain using diamagnetic passive shims.Magn Reson Med 2002:48:906–914.19. Jezzard P, Clare S. Sources of distortion in functional MRI data.Hum Brain Mapp 1999:8:80–85. Figure4.  OrientationofB 0 relativetothebrainin the sphinx and supine positions, and thick-ness of the air–brain interfaces for the anterior and posterior regions. 412 Valette et al.
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x