Outline Basics Of BOLD FMRI

Basics of BOLD fMRIOutline MR Basic Principles Spin Hardware Sequences Basics of BOLD fMRISusceptibility and BOLD fMRIA few tradetrade-offsSpring 2007 fMRI Analysis Course1Spring 2007 fMRI Analysis CourseThe MR roomSpring 2007 fMRI Analysis Course2Scanner Internals3Macroscopic: Brain SystemsSpring 2007 fMRI Analysis Course4Microscopic: Neuronal FunctionAction Potentials & Neurotransmitter TraffickingSpring 2007 fMRI Analysis Course5Spring 2007 fMRI Analysis Course61

Hemodynamic Measure of Brain Function (1881)NeuronsArmBrainVeinArteryArteriolesCapillary BedVenulesAngelo Mosso1 - 3 cmSpring 2007 fMRI Analysis CoursePressure Traces“Bertino”Bertino”7Spring 2007 fMRI Analysis Course8Hemodynamic Response PropertiesBlood Oxygen Level Dependent (BOLD) Magnitude of signal changes is quite smallarteriolescapillarybedBaseline xyDeoxy-HbNeural activity increasesBlood flow increases (“reactive hyperemia”hyperemia”)Deoxyhemoglobin concentration decreasesMagnetic field homogeneity increasesGradient echo EPI signal increases 0.5 to 3% at 1.5 T (or smaller) Too small to see in individual images Always considering differences or timetime-course changesin image intensityvenules Response is delayed and quite slow ( 10 seconds)“Activated”Activated”MxySignal Extracting temporal information is tricky, but possible Even short events have a rather long responseMosinθT2* taskT2* controlStaskScontrolΔSTEoptimumSpring 2007 fMRI Analysis Coursetime9Blood Oxygen Level Dependent(BOLD) Contrast ActivationSpring 2007 fMRI Analysis Course10TimeTime-Course Response in fMRI Brief neuronal events canelicit a (positive) bloodflow and oxygenationresponse.SlowerStart of Rise andEvent Fall in 10 s NegativeResponse Reponses to events asbrief as 50 ms have beenrecorded.Functional MRI response to avisual stimulus of duration 2sSpring 2007 fMRI Analysis Course11Spring 2007 fMRI Analysis Course122

Response to periodic flashes of lightProcessed ImageTypical Functional Image VolumeAnatomic ImageSpring 2007 fMRI Analysis Course13Spring 2007 fMRI Analysis Course14Slice TerminologyfMRI Experiment Stages: Prep1) Prepare subjectConsent form Safety screeningInstructions2) Shimmingputting body in magnetic field makes it non-uniform VOXEL(Volumetric Pixel)Slice Thicknesse.g., 6 mmIn-plane resolutione.g., 192 mm / 64 3 mmadjust 3 orthogonal weak magnets to make magnetic field as homogenous aspossible3 mm3) Sagittals6 mmSAGITTAL SLICETake images along the midline to use to plan slicesIN-PLANE SLICE3 mmNumber of Slicese.g., 10Matrix Sizee.g., 64 x 64Field of View (FOV)e.g., 19.2 cmSpring 2007 fMRI Analysis Course15Spring 2007 fMRI Analysis Course16fMRI Experiment Stages: FunctionalsMRI vs. fMRI5) Take functional (T2*) imagesimages are indirectly related to neural activity high resolution(1 mm)usually low resolution images (3x3x5 mm)all slices at one time a volume (sometimes also called an image)sample many volumes (time points) (e.g., 1 volume every 2 seconds for 150volumes 300 sec 5 minutes)4D data: 3 spatial, 1 temporalMRIfMRIlow resolution( 3 mm but can be better)one image fMRIBlood Oxygenation Level Dependent (BOLD) signalindirect measure of neural activityfirst volume(2 sec to acquire) neural activitySpring 2007 fMRI Analysis Course17Spring 2007 fMRI Analysis Course many images(e.g., every 2 sec for 5mins)Î blood oxygen Î fMRI signal183

Activation StatisticsfMRI process chainFunctional imagesPhase FixFunctional ImagesROI TimeCoursefMRISignal(% change) 2sRegistrationConditionTimeTime(secs)123 75021 3.33s0s n2Statistical Map1NormalizationTimey Xβ e1ConditionRegion of interest (ROI)2Spring 2007 fMRI Analysis Course192. 5 minSpring 2007 fMRI Analysis Course20Statistical Maps & Time CoursesStimulation protocols in fMRItime courseof activationbaseline restsuperimposed onanatomical MRI imageUse stat maps to pick regions% signal changeThen extract the time courseimagesstimulationhaemodynamicresponse functionSpring 2007 fMRI Analysis Course21Spring 2007 fMRI Analysis Course2D Î 3D22Design Jargon: Runssession: all of the scans collected from one subject in one dayrun (or scan): one continuous period of fMRI scanning ( 5-7 min)experiment: a set of conditions you want to compare to each othercondition: one set of stimuli or one task4 stimulus conditions 1 baseline condition (fixation)A session consists of one or more experiments.Each experiment consists of several (e.g., 1-8) runsMore runs/expt are needed when SNR is low or the effect is weak.Thus each session consists of numerous (e.g., 5-20) runs (e.g., 0.5 – 3hours)Spring 2007 fMRI Analysis Course23Spring 2007 fMRI Analysis Course244

Susceptibility in MRDesign Jargon: Paradigm or ProtocolAll susceptibilityeffects increasewith Bo fieldThe good.paradigm (or protocol): the set of conditions and their order used in aparticular runrunepoch: one instance of aconditionThe bad.first “objects right” epochsecond “objects right” epochvolume #1(time 0)epoch8 vol x 2 sec/vol 16 secvolume #105(time 105 vol x 2 sec/vol 210 sec 3:30)The ugly.TimeSpring 2007 fMRI Analysis Course25Susceptibility in Temporal LobesSpring 2007 fMRI Analysis Course26What is the source of susceptibility?BoThe magnet has a spatially uniformfield but your head is magnetic 1) deoxyHeme is paramagneticPattern of B field outsidemagnetic object in a uniformfield 2) Water is diamagnetic (χ -10-5)3) Air is paramagnetic (χ 4x10-6)Spring 2007 fMRI Analysis Course27Susceptibility effects occur near magnetically disdis-similar materialsSpring 2007 fMRI Analysis Course28Bo map in head: it’s the air tissueinterface Bo1.5T3TPing-pong ball in H20:Field maps (ΔTE 5ms), black lines spaced by0.024G (0.8ppm at 3T)Spring 2007 fMRI Analysis CourseSagittal Bo field maps at 3T29Spring 2007 fMRI Analysis Course305

Other Sources of Susceptibility You ShouldBe Aware of of Local susceptibility gradients:2 effectsSagittal Bo field map at 3T Local dephasing of the signal(signal loss) within a voxel,mainly from thru-plane gradients Local geometric distortions,(voxel location improperlyreconstructed) mainly from localin-plane gradients (in PEdirection).Those fillings might be a problem Spring 2007 fMRI Analysis Course31Spring 2007 fMRI Analysis Course32Susceptibility in EPI can give either acompression or expansionBandwidth is asymmetric in EPI(Distortion is 100x more in phase direction)kyThe phase error (and thusdistortions) are in the phaseencode direction.υ1 υ2ϕ Δυ τAltering the direction kspace istraversed causes either localcompression or expansion.choose your poison kxδt 0.5ms3T whole body gradientsδt 0.005msSpring 2007 fMRI Analysis Course33Susceptibility Causes Image DistortionSpring 2007 fMRI Analysis Course34With fast gradients, add parallel imagingΔk 2πFOV{Use shortest possible encodingEchoplanar Image,Δθ α encode time α 1/BWzAcquisition:SENSESMASHReduced k-spacesamplingFolded images ineach receiver channelReconstruction:Folded datasets Coil sensitivitymaps3T head gradientsField near sinusSpring 2007 fMRI Analysis CourseEncode time 34, 26, 22, 17ms35Spring 2007 fMRI Analysis Course366

3T MAGNETOM AllegraWhat can you do?ss EPI PAT Single shotTE 30 msConventional64x64 with PAT x264x64 with PAT x2128x128 with PAT x2192x128 Good shimming (first & second order)Thinner slices (Drawback: Takes more to cover the brain)Shorter TE (Drawback: BOLD contrast is optimized for TE T2*local)T2*local)“Z-shimming”shimming” Repeat measurement several times with an applied zgradients that rewind the dephasing,dephasing, Pick the right gradient afterward on apixel by pixel basis. (Drawback: multi shot or longer encode). Yang et al.MRM 39 p402, 1998.Use special RF pulse with builtbuilt-in prephasing in just the right places.(Drawback: long RF pulse, prepre-phasing differs from person to person)Glover et al. Proceed. ISMRM p298, 1998.The “mouth shim”shim” paramagnetic material in roof of mouth. Wilson,Jenkinson,Jenkinson, Jezzard,Jezzard, Proceed. ISMRM p205, 2002.Distortion correction based on a measured field map (drawback: cannotcannotrecover signal dropout or fully correct “overlapping”overlapping” intensities)MultiMulti-shot imaging methods (drawback: more motion sensitive)Fancy pulse sequences (best to have local physicist): 180 degreerefocusing pulses to reverse distortion (GRASE)/Multiple refocusingrefocusingpulses pulses singlesingle-shot FSE, UU-Flare4 channel tx/rx array coilMAGNETOM Allegra. Courtesy Bruker Medical and USA Instruments.Spring 2007 fMRI Analysis Course37Spring 2007 fMRI Analysis CourseSingleSingle-shot Gradient Echo EPITR (repetition time) Determines how much magnetization is allowed torecover before it is knocked over again by the next rfpulse From a pure signal strength perspective, waiting forvery long TR’TR’s (5 seconds ) allows for maximalsignalsignal-toto-noise (SNR) Noise is MR dominated by physiologic noise (notthermal noise) Requires many images in both conditions to reliablydistinguish activation (which requires shorter TR’TR’s) fMRI can be performed as fast as TR 100ms Bottom line: use as short a TR as you can Parameters you can choose TRSlice thickness/gapNumber of slices/slice acquisition orderTEBandwidthMatrix sizeField of viewFlip angle All of these parameters can be appropriatelyapplied over a wide range of valuesSpring 2007 fMRI Analysis Course3839Spring 2007 fMRI Analysis CourseFlip Angle40Number of slices A given flip angle will maximize the SNR (ErnstAngle) Angle) at long TR’TR’s ( 3s) this is 90 degrees This angle is dependent upon the TR Incorrect angles may sensitize your BOLD scans to ininflow artifacts (bad) [Lu et al, NeuroImage 17, 943–943–955 (2002)] Bottom line: For TR of 11-2s, a flip angle of around 606070 degrees is optimal Separate slices in EPI are typically squeezed into a TRinterval Many factors influence # of slices that fit in a TR Length of TRTE (determines center of blue box)Matrix size (determines length of blue box)Bandwidth (determines length of blue box) Bottom line: collect as many slices as you canθ cos 1 ( exp (TR / T1 ) )Spring 2007 fMRI Analysis Course41Spring 2007 fMRI Analysis Course427

So far Slice ThicknessLong TR maximized SNRShort TR maximizes fMRI statsLong TR provides many slicesShort TR provides few slices SNR in MRI is proportional to voxel volume (thinnerslices - less SNR) Thinner slices reduces partial volume effects Thinner slices reduces throughthrough-plan dephasing What is the size of the structure of interest? Isotropic voxel size is preferred The above suggests imaging only brain regions ofinterest (to minimize slices) But processing decisions also play a role Whole brain data is much easier to spatially normalizeMotion correction works best with thin slicesIn general TR’TR’s between 1s and 2s are not too badSpring 2007 fMRI Analysis Course43Spring 2007 fMRI Analysis Course44TE (echo time)Bandwidth Optimum TE is shorter at high field (say 30ms at 3Tversus 50ms at 1.5T) Shorter TE reduces signal loss due to fieldinhomogeneities, but also reduces BOLD effect Rate at which points are sampled (the echoes aredigitized) High bandwidth implies a high sampling rate Sampling of the order of 128 kHz 128kHz/64matrix 2000Hz/pixel Noise is proportional to sampling rate High bandwidth means faster data acquisition (andmore slices can be acquired, with less T2 blurring)Spring 2007 fMRI Analysis Course45Spring 2007 fMRI Analysis Course46Matrix SizeField of View (FOV) Matrix size impact everything Voxel size determined by field of view and matrix size Increasing matrix size decreases voxel size and thus SNR Increasing matrix and FOV maintains constant voxel size, butincreases N and therefore increases SNR Intravoxel dephasing reduced somewhat with smaller voxels(bigger matrix)Δx FOVxNxΔy FOVyNy FOV 200mm/64 matrix 3.125mm voxel dimension Recall SNR proportional to voxel volumeSpring 2007 fMRI Analysis Course47Spring 2007 fMRI Analysis Course488

Spring 2007 fMRI Analysis Course 1 Outline MR Basic Principles Spin Hardware Sequences Basics of BOLD fMRI Susceptibility and BOLD fMRI A few trade-offs Spring 2007 fMRI Analysis Course 2 Basics of BOLD fMRI Spring 2007 fMRI Analysis Course 3 The MR room Spring 2007 fMR