Functional MRI: Physics & Data Acquisition - NBML

Functional MRI:Physics & Data AcquisitionVahid MalekianPost-doctoral Associate,School of Cognitive Sciences, IPM, Tehran.Email: vmalekian@ipm.ir1

2

fMRI is a non-invasive neuroimaging tool thatdetermines neural activity within the brain using MRIscanner.3

Advantages Non-invasive, no radiation Spatial resolution Easy for researchers to use Limitations Expensive! Metal free! Time resolution (many havestarted to combine with EEG) Need expert!4

MRI Killed the radiotracers, By Neuroskeptic , August 30, 2013www. discovermagazine.comK. Smith, et al., Functional magnetic resonance imaging is growing from showyadolescence into a workhorse of brain imaging. Nature, 2012: VOL 484.5

fMRI demonstrates brain activation by recording T2*signal changes due to increased blood oxygenationlevel, related to neural activity.6

fMRI Physics fMRI - Magnetic fields & spinsfMRI - Radio pulse & relaxation timesfMRI - Tissue contrastsfMRI - BOLD & functional contrastfMRI Data Acquisition Data Acquisition Experimental Design Areas for future research

8

B0 constant magneticfield Strong and homogenousfield Along z-axisAlways ON!Z1 Tesla 20,000x Earth’s magnetic field

In particle physics, spin is an intrinsic form of angular momentum carried byelementary particles and atomic nuclei. Hydrogen protons spinproducing a magnetic field spinningprotonN A magnetic field creates anelectrical charge when itrotates past a coil of wireSbarmagnet10

Outside scannerInside scanner (B0)The protons of the H2O molecules in our body align along B0

12

Frequency of precession of magnetic moments given byLarmor relationshipf g x B0f Larmor frequency (mHz)g Gyromagnetic ratio (mHz/Tesla)B0 Magnetic field strength (Tesla)g 43 mHz/TeslaLarmor frequencies of RICs MRIs3T 130 mHZ7T 300 mHz13

Spin Excitation (B1)XXXB0Magnetization vector does notprecess no induction in anycoilTransverse magnetizationvector precesses about the mainfield detected by a loopperpendicular to main field14

RF signal (B1)15

beforez axisafterRF Pulse (B1)x-y planez axisx-y plane

Dephasing in x-y plane spin-spin relaxationThe Second Law of ThermodynamicsRF pulseT2 relaxation

Different tissues have different T2 relaxation timesCSFGrey matterWhite matterFatTE (milliseconds)

Two reasons for dephasing in x-y plane Spin-spin interaction T2 Local magnetic field in-homogeneities T2*T2(*) time constantMagnetic field inhomogeneities

Too much to explain here Different gradients alongmagnetic field Lauterbur contribution

Blood Oxygen Level Dependent signalO2 is transported by haemoglobin (Hb)

22

23

Remember T2* and field inhomogeneities?DeoxyHb paramagneticstrong field inhomogeneitiesOxyHb diamagneticweak field inhomogeneitiesFast dephasingFast T2*Slower dephasingslower T2*

25

MRI studies brainanatomy. 3D with high spatialresolution Can distinguishdifferent types of tissue fMRI studies brainfunction. Functional (T2*) images 4D with lower spatialresolution but highertemporal resolution26

An fMRI experiment consists of a sequence of individualMR images, where one can study oxygenation changes inthe brain across timeDynamic Imaging27

1) MRI physics Optimal fMRI data acquisition is achieved through theconsideration and refinement of many aspects of MRI imagingtechniques.2) General Issues in fMRI experimental design The design of experiments reflects the temporal resolution offMRI. Task must be designed related to the experiment.3) Scanner hardware and environment The high magnetic field, restricted space and noise of scannermake special demands on the methods of stimulus presentationand subject accommodation within the system.28

High-field MRI (1.5T or greater) scannerBOLD effect (fMRI signal) increases with field strength Fast pulse sequenceEcho Planar Imaging (GRE-EPI) Stimulus presentation equipmentProjector to show visual stimuliResponse devices such as button box to record subject’sresponseHeadphones for auditory stimuli (and hearing protection)29

30

31

Controlling the timing and quality of cognitiveoperations to influence resulting brain processes What can we control?Experimental comparisons (what is to be measured?)Stimulus properties (what is presented?)Stimulus timing (when is it presented?)Subject instructions (what do subjects do with it?)32

What is to be measured? Motor ActivationWhat is presented? Hand PictureWhen is it presented? 30s-rest 30s-Act Total: 5 minWhat do subjects do with it? Right finger tappingWhich MRI Protocol?GE-EPI Sequence, Spatial & temporal resolutions (2.5*2.5*2.5mm3 and 3s), brain coverage (Whole), TE & FA (30 ms & 90)33

Result34

Small space (limited room for equipments)Claustrophobia (patient)Strong magnet (some equipments won’t work)Limited range of motion (signal destruction)Scanner noise (care about auditory experiments)Fatigue (experiment duration)35

It is important that the task induces the subject tothink or perform in the intended manner. Don’t make the study too predictable as this mayinfluence the subjects psychological state. Make sure to keep subjects on task by giving them justthe right amount of time to perform it. What you expect from subjects should fit with what theycan actually do. Keep in mind that subjects’ brains may be responding tothings you didn’t tell them to do.36

37

VERY IMPORTANT POINTYOU SHOULD KEEP IN MIND ABOUT MRI & fMRI38

39

AnalysisSpecificity vs. SensitivitySpatial-resolution vs. Sensitivity40

41

42

43

Data acquisition and reconstruction Developing fMRI techniques for higher spatial and temporal resolutions Contrast improvement Data Processing De-noising & enhancement Advanced functional analysis fMRI Clinical Applications Quantitative assessments for the treatments Pre-surgical planning for tumor patients Alzheimer, dementia, ADHD, epilepsy Physics & Engineering Simulations Hardware (coil design & peripherals) Multimodal imagingEEG-fMRI, MEG-fMRI and fNIR-fMRI44

Thank you 45

1) MRI physics Optimal fMRI data acquisition is achieved through the consideration and refinement of many aspects of MRI imaging techniques. 2) General Issues in fMRI experimental design The design of experiments reflects the temporal resolution of fMRI. Task must be designed related to the experiment. 3) Scanner hardware and environment