700 likes | 897 Views
Columbia fMRI. Hirsch, J., et al. Neuroscience 2005 Functional Brain Imaging. Joy Hirsch, Ph.D., Professor Director, fMRI Research Center Columbia University Medical Center NI Basement. www.fmri.org. A Brief Outline. Columbia fMRI. Hirsch, J., et al.
E N D
Columbia fMRI Hirsch, J., et al Neuroscience 2005 Functional Brain Imaging Joy Hirsch, Ph.D., Professor Director, fMRI Research Center Columbia University Medical Center NI Basement www.fmri.org
A Brief Outline Columbia fMRI Hirsch, J., et al I. The Principle of functional specificity A. Single Areas B. Multiple Areas II. Brain Mapping Techniques A. Lesion- Based Methods Visual field loss Aphasia Personality Changes B. Cardiovascular Based Methods Positron Emission Tomography, PET Functional Magnetic Resonance Imaging, fMRI C. Electromagnetic-Based Methods SSEP Somatosensory Potentials Cortical Stimulation Magnetoencephalography, MEG Electroencephalography, EEG III. Future Directions of Brain Mapping
Columbia fMRI Hirsch, J., et al I. The principle of functional specificity A. Specializations of single brain areas
Columbia fMRI Hirsch, J., et al Homunculus: Map of Sensory/Motor Function
7 7 6 6 5 5 Columbia fMRI 4 4 Hirsch, J., et al Calcarine Sulcus Primary Visual Cortex Flashing LED Display
Columbia fMRI Hirsch, J., et al FUNCTIONAL SPECIFCITY BASED ON RETINOTOPY
Rotation Eccentricity V1, V2 Boundary Columbia fMRI Hirsch, J., et al
Columbia fMRI Hirsch, J., et al I. Principle of functional specificity A. Specializations of single brain areas B. Specializations of multiple brain areas
Columbia fMRI Hirsch, J., et al Functional Organization of Visual Cortex
Columbia fMRI Hirsch, J., et al BRAIN MAP OF OBJECT NAMING (MANY SUBJECTS) Centers of mass Anatomical Region Area x y z Hemis Wernicke Broca Broca Sup. Motor Superior Temporal Gyrus Inferior Frontal Gyrus Inferior Frontal Gyrus Medial Frontal Gyrus 57 49 409 -26 10 25-6 9 25 8 53 LLLL Connectivity Principle: Neural circuits in the brain connect working parts to execute complex tasks. Hirsch, R-Moreno, Kim, Interconnected large-scale systems for three fundamental cognitive tasks revealed by functional MRI. Journal of Cognitive Neuroscience, 13(3), 389-405, 2001.
Columbia fMRI Hirsch, J., et al II. Brain Mapping Techniques A. Lesion-Based Methods
Columbia fMRI Hirsch, J., et al 1. Visual Field Loss Harrington, 1964v R lesion Previous Surgical lesion Visual Field BINOCULAR FLASHING LIGHTS Left Eye Right Eye
THE BRAIN IS ORGANIZED BY DEDICATING SPECIFIC AREAS TO SPECIFIC FUNCTIONS 2. Aphasia Neuroscience and Medicine Year BROCA Aphasia and lesions in GFi 1841 HARLOW Phineas Gage 1861 WERNICKE Aphasia and lesions in GTs 1874 Columbia fMRI Hirsch, J., et al
3. Personality Changes Columbia fMRI Hirsch, J., et al Phineas Gage Damasio, H., et al; Science 264: 1102-1105, 20 May 1994
Columbia fMRI Hirsch, J., et al II. Brain Mapping Techniques B. Cardiovascular Based Methods Positron Emission Tomography, PET
TECHNICAL MILESTONES IN IMAGING DAMADIAN Discovery that biological tissues have different relaxation rates 1971 HOUNSFIELD CORMACK Invention of Computed Tomography LAUTERBUR First MR image 1972 MANSFIELD First MRI of a body part invention of EPI (scans whole brain in secs.) 1976 TER-POGOSSOAN SOKOLOFF First PET studies of brain metabolism 1977 Columbia fMRI Hirsch, J., et al
Source of Signal Columbia fMRI Hirsch, J., et al Positron Emission Tomography Radionuclides that emit positrons such as 15O and 18F are introduced into the brain. H215O behaves like H216O and indicates blood flow (rCBF) (half life = 123 seconds) integration time ≈ 60 seconds. 18F – deoxyglucose behaves like deoxyglucose and indicates metabolic activity (half-life = 110 minutes) integration time ≈ 20 minutes PET SCANNER From: www.epub.org.br/cm/n011pet/pet.htm
Columbia fMRI Hirsch, J., et al Principle of PET A2 Positron and electron annihilation and emission of gamma rays PET is based on the radioactive decay of positrons from the nucleus of the unstable atoms (15O has 8 protons and 7 neutrons) Gamma ray Site of positron annihilation (imaged point) A1 Positron emission in the brain Electron Gamma ray photon 0-9mm Unstable radionuclide Positron resolution limit From: Principles of Neural Science (4th. Ed.) Kandel, Schwartz, & Jessell, p. 377.
Year 1881 1890 Cerebral Blood Flow is Coupled to Neural Activity MOSSO Blood flow and cognitive events Blood Flow ROY & SHERRINGTON Relationship between neural activity and vascular changes Neural Activity Columbia fMRI Hirsch, J., et al
NEUROIMAGING: PET STRUCTURAL IMAGING: MRI Year HILAL First clinical MRI scanner 1981 PETERSON/FOX POSNER/RAICHLE PET study of human language Radiolabeled blood flow and neural events 1984 Columbia fMRI Hirsch, J., et al
Columbia fMRI Hirsch, J., et al II. Brain Mapping Techniques B. Cardiovascular Based Methods Positron Emission Tomography, PET • Source of signal and principles • Measurement techniques
Columbia fMRI Hirsch, J., et al Gamma Ray Detections to Location of Function From: Principles of Neural Science (4th. Ed.) Kandel,Schwartz, & Jessell, p. 377.
Columbia fMRI Hirsch, J., et al Injection of radioactive-labeled water
Columbia fMRI Hirsch, J., et al II. Brain Mapping Techniques B. Cardiovascular Based Methods Positron Emission Tomography, PET • Source of signal and principles • Measurement techniques • Computation for analysis
Analysis of PET Results Stimulation Fixation Difference Flashing Checkerboard Fixation Individual difference images Mean difference image Columbia fMRI Hirsch, J., et al From: Images of Mind by Posner, M. and Raichle, M. Scientific American Library, 1994, p. 24
Columbia fMRI Hirsch, J., et al II. Brain Mapping Techniques B. Cardiovascular Based Methods Positron Emission Tomography, PET 2. Functional Magnetic Resonance Imaging, fMRI
FUNCTIONAL MRI: fMRI OGAWA Blood Oxygen dependent signal EPI/MRI and neural events 1990 BELLIVEAU Cortical map of the human visual system: fMRI 1992 Columbia fMRI Hirsch, J., et al
Columbia fMRI Hirsch, J., et al II. Brain Mapping Techniques Functional Magnetic Resonance Imaging, fMRI • Source of signal and principles
Columbia fMRI Hirsch, J., et al MAGNETIC FIELD 1: Scanner Environment [1.5] T Protons align along an axis Protons Protons Outside Field Inside Field (scattered) (aligned)
RFi (precess) (wobble) Columbia fMRI Hirsch, J., et al MAGNETIC FIELD 2: Created when a radio frequency pulse(63.3 mgHz) is applied to aligned protons Protons precess around the axis and create a small current (MRI signal)
uniform field gradientfield Columbia fMRI Hirsch, J., et al MAGNETIC FIELD 3: A detectable radio frequency is emitted by the protons as they relax into their aligned state RFo The Radio frequency (RFo) is dependent upon field strength and therefore indicates location of origin Application of magnetic field gradient (mT) Location of signals are recorded
MRI Signal Intensity MAGNETIC FIELD 4: Local signal change of a single voxel over time is due to change in proportions of oxyhemoglobin/deoxyhemoglobin Change in MR Signal over time REST TASK REST REST TASK REST - 40 s - - 40 s - - 40 s - - 40 s - - 40 s - - 40 s - PHYSIOLOGY PHYSICS DEOXY HGB IS PARAMAGNETIC(Linus Pauling, 1936) NEURAL ACTIVATION IS ASSOCIATED WITH AN INCREASE IN BLOOD FLOW (Roy & Sherrington, 1890) AND DISTORTS THE LOCAL MAGNETIC FIELD CAUSING SIGNAL LOSS RESULT: REDUCTION IN THE PROPORTION OF DEOXY HGB IN THE LOCAL VASCULATURE RESULT: LESS DISTORTION OF THE MAGNETIC FIELD RESULTS IN LOCAL MR SIGNAL INCREASE Columbia fMRI Hirsch, J., et al
Peak Brief Stimulus Undershoot Initial Undershoot Columbia fMRI Hirsch, J., et al BOLD Impulse Response Model • Function of blood oxygenation, flow, volume (Buxton et al, 1998) • Peak (max. oxygenation) 4-6s poststimulus; baseline after 20-30s • Initial undershoot can be observed (Malonek & Grinvald, 1996) • Similar across V1, A1, S1… • … but differences across:other regions (Schacter et al 1997) individuals (Aguirre et al, 1998)
Columbia fMRI Hirsch, J., et al BOLD ORIGIN BOLD Signal corresponds to local field potential (LFP) Logothetis, N.K., Pauls, , Augath, M, Torsten, T, Oeltermann, A, (2001) Neurophysiological investigation of the basis of the fMRI signal. Nature 412 150-157
Columbia fMRI Hirsch, J., et al II. Brain Mapping Techniques B. Cardiovascular Based Methods Positron Emission Tomography, PET 2. Functional Magnetic Resonance Imaging, fMRI • Source of signal and principles • Measurement techniques
Imaging While Naming Objects Columbia fMRI Hirsch, J., et al Scanner acquires the whole brain every [4] secs: [26] axial slices Resolution [1.5 x 1.5 x 4.5] mm Each voxel is analyzed seperately
COMPUTATIONS FOR fUNCTIONAL IMAGE PROCESSING RECONSTRUCTION ALIGNMENT VOXEL BY VOXEL ANALYSIS GRAPHICAL REPRESENTATION Columbia fMRI Acquisition Functional Brain Map from Nature388, 171-174 (1997) Kim, Relkin, Lee, & Hirsch
Columbia fMRI Hirsch, J., et al Brain Map of Object Naming(Single Subject)
Block Design Event-Related Design Columbia fMRI Hirsch, J., et al
One voxel = One test (t, F, ...) amplitude • General Linear Model • fitting • statistical analysis time Voxel Location Temporal series fMRI Columbia fMRI voxel time course Hirsch, J., et al
Columbia fMRI Hirsch, J., et al II. Brain Mapping Techniques B. Cardiovascular Based Methods Positron Emission Tomography, PET 2. Functional Magnetic Resonance Imaging, fMRI • Source of signal and principles • Measurement techniques • Computations for analysis
parametric one sample t-test two sample t-test paired t-test Anova AnCova correlation linear regression multiple regression F-tests etc… Columbia fMRI Hirsch, J., et al Voxel statistics… all cases of theGeneral Linear Model assume normality to account for serial correlations:
Columbia fMRI Hirsch, J., et al two-sample t-test t-statistic imageSPM{t} Image intensity compares size of effect to its error standard deviation • standard t-test assumes independenceignores temporal autocorrelation! voxel time series
90 100 110 -2 0 2 -10 0 10 90 100 110 + m + = a m = 100 a = 1 Fit the GLM Mean value voxel time series box-car reference function Regression Columbia fMRI
Columbia fMRI Hirsch, J., et al II. Brain Mapping Techniques B. Cardiovascular Based Methods Positron Emission Tomography, PET 2. Functional Magnetic Resonance Imaging, fMRI • Source of signal and principles • Measuremnet techniques • Computation for analysis • Individual brain maps
Standard Brain Mapping Tasks Columbia fMRI Hirsch, J., et al SENSORY Touch MOTOR Finger Thumb Tapping LANGUAGE VISION Reversing Checkerboard Picture Naming Listening to Words (passive) (active) (active) (passive) (passive) GPoC GPrC GOi GTT GFi GTs CaS From Hirsch, J., et al; Neurosurgery 47: 711-722, 2000
Conventional Imaging Functional Imaging Columbia fMRI Hirsch, J., et al Before Surgery After Surgery Tumor Tumor R Left Hand Movement Left Hand: Sensory/Motor CC 23 (AB)
Columbia fMRI Hirsch, J., et al Surgery Before After English Language Areas EnglishLanguage Areas R English Tumor LANGUAGE Italian Language Areas Italian Language Areas Tumor Italian a b
Columbia fMRI Hirsch, J., et al II. Brain Mapping Techniques C. Electromagnetic - Based Methods Somatosensory Evoked Potential, SSEP Direct Cortical Stimulation
Sensory Motor Mapping Columbia fMRI Hirsch, J., et al Direct Cortical Stimulation Craniotomy Localization fMRI SSEP “Twitching of hand, focal seizure involving arm ” Tag 3 Tag 3 Tag 5 “Twitching in 1st three digits” From Hirsch, J., et al; An Integrated Functional Magnetic Resonance Imaging Procedure for Preoperative Mapping of Cortical Areas Associated with Tactile, Motor, Language, and Visual Functions, Neurosurgery 47: 711-722, 2000. Tag 5