Atlas of Human Brain_人类大脑图谱

1.1MacroscopicAtlas1.1.1AnatomicalPreparationsSeventeenheadsusedforthisstudywerefrombodiesdonatedtotheDepartmentofAnatomy,H.-HeineUniversityofDuesseldorf.Thestudieswereperformedinaccordancewithestablishedethicalstandards.Thecadaverswereperfusedviatheradialveinsfirstwithphysiologicalsaline,thenwithfixative.Thecomposi-tionofthefixativechangedasthestudyprogressed.Theheadswereremovedfromtheperfusedcadaverswithin36hoursafterdeathandwereplacedinawoodenboxwithearbarsandheadhold-erstoallowexactpositioningandfixingforsubsequentsectioning.Inthiswaythecoordinatesofthepositionboxandthedesiredplaneofanatomicsectioningwereexactlyaligned.Thethreeheadspresent-edinthisatlaswereperfusedwithafixa-tivecontaining10%formalin,glycerol,andIncidin(Henkel,Duesseldorf).Theheadswerecoveredwithlinenandfixa-tivetopreventthepreparationfromdry-ingoutbetweenperformanceoftheMRIandthefreezingprocess.1.1.2MagneticResonanceImaging(MRI)PilotMRIScans:PilotMRIscanswereperformedwiththeheadsinthewoodenboxestodeterminetheeffectoffixationandtodetectpathologicalorartifactualchanges.Outof17headsusedforthispartofthestudy,4showedsevereneu-ropathologicalalterations,suchashemor-rhage,focallesionsortumor,whichwerenotknownbeforetheMRIwasper-formed.Fiveheadswereexcludedfortechnicalreasons,namely:artifactsduetoincompletefixation,brainswellingorinsufficientimageresolution(MRimag-ingforthisatlasstartedat1984).Additionallytwoheadswerelaterdis-cardedfromthisstudybecauseofprob-lemswithcutting.MRScanOrientations:Allheadswerescannedinthehorizontal,coronal,sagit-tal,andobliquecoronalplanesaccordingtoastandardizedprotocolThecorre-spondingsectionplanesweredefinedby(a)theintercommissurallineandverticalsintersectingthislineinthecenteroftheanteriorandposteriorcommissures,and(b)theverticaltothebrainstemaxis(Meynert’splane).Fordefiningaddition-alplanes,weusedthecantho-meatallineandtheorbitalaxis.MRSystemandScanModes:Forinvitroimagingandinvivo/invitrocorre-lation,scansandmeasurementswereper-formedona0.15Teslasuperconductivemagnet(Vista2035,PickerInternational).Forinvitroimagingweusedamultislicedouble-echosequence[spin-echo-sequence(SE)5000/40/160]withfourexcitationstorenderproton(N(H))-andT2-weightedimages.Theslicethicknesswas5mmwithamatrixof256x256andafieldofviewof250mm,resultinginapixelsizeof0.96mm2andavoxelsizeof4.8mm3.Thesescansarepresentedinthepagesprecedingtheatlasfiguresofthehorizontal,coronal,andsagittalsections(Sections,2.1,2.2and2.3,respectively).Forinvivo/invitrocorrelation,weusedasingle-slicetechniquewithSEandinver-sionrecovery(IR)sequenceswiththefol-lowingparameters:SE(TR:200,800,1600,3200,6400mscombinedwithTE:30,40,60,80,120,160ms)andIR(TR:620,1240,2480mscombinedwithTI:32,100,150,200,300,400,500,600,700ms).Theslicethicknesswas5mmwithamatrixof256x256andafieldofviewof250mm.Forinvivoimagingweuseda1.5Teslaimager(Vista2055HPQ,PickerInternational)withstandardprotocols:SE3000/100,resultinginproton-andT2-weightedimages;andIR5000/600,resultinginT1-weightedimages,andmultislicepartialsaturation(PS)sequence,PS500/10resultinginT1-weightedimageswithpartialfatsuppres-sion.Theslicethicknesswas5mmwithamatrixof256x256andafieldofviewof250mm.TheseimagesarepresentedinSections2.1to2.3facingeachmacro-scopicanatomicalsection.1.1.3MRImages:Invivo/invitroCorrelationInautopsybrains,relaxationtimes,sig-nal-to-noiseratio,andcontrastarealteredduetoabsenceofthefloweffect,cessa-tionofmetabolism,andfixationproce-MaterialandMethods10102030405060708090100110120130140150VBWMCPONGMRNHIOFTMN(H)in-vitro/N(H)in-vivo(%)SE-Seq.IR-Seq.in-vivoin-vitro01002003004005006007008009001000CSFVBWMCPONGMRNHIOFTMT2inms01002003004005006007008009001000VBWMCPONGMRNHIOFTMT1inms1548in-vivoin-vitroFigure1.ComparisonofT1(spin-latticerelax-ation)times,T2(spin-spinrelaxation)times,andN(H)valuesfordifferenttissuesbetweeninvivoandinvitroconditions.TheN(H)valuesarecalcu-latedfromSE-sequences(redsquares)andIR-sequences(blacksquares).Theyarepresentedasquotientofinvitrovaluestoinvivovaluesin%.(DatafromLongerich,1989).Abbreviations:CSF:cerebrospinalfluid;VB:vitreousbody;OF:orbitalfat;ON:opticnerve;HI:hippocampus;WM:whitematter;GM:graymatter,CP:cerebralpeduncle;RN:rednucleus;TM:temporalismuscle.1_front_pdf.qxd02.09.200313:24Seite5dures.Toperformcomparativeanalysisbetweenlivingandpostmortemtissuetheheadsoftwovolunteersandoftwocadav-erheadswerescannedwiththesamescanningparameters.T1(spin-lattice)relaxationtimes,T2(spin-spin)relaxationtimes,andN(H)(protondensity)val-uesoftheinvivoandinvitrotissuesarerepresentedinFig.1.Examplesofregres-sioncurvesofinvivo/invitromeasure-mentsofsignalintensitiesareshowninFig.2.Comparedwiththeinvivoresults,theinvitrotissuesshowlarge,nonlinearreductioninT1relaxationtimesandN(H)valuesandsmall(negligible)changesinT2relaxationtimes.Astablecorrelationexistsbetweentheinvitroandinvivorelaxationcurves.Thedifferentsignalintensitiesweremeasuredfromthesametissueunderinvitroandinvivoconditionsandcanbeexplainedbythebiophysicalalterationsinthedifferentfractionsofwater(“three-fractionmodel”)(GröschandNoack,1976)bytemper