Title: The Biggest Unknowns in Functional MRI
1The Biggest Unknowns in Functional MRI
- Peter A. Bandettini, Ph.D
- bandettini_at_nih.gov
- Unit on Functional Imaging Methods
-
- Functional MRI Facility
- Laboratory of Brain and Cognition
- National Institute of Mental Health
2Functional Neuroimaging Techniques
Non-invasive
Log Size (mm)
Invasive
Log Time (sec)
3(No Transcript)
4Uses
Understanding normal brain organization and
changes -networks involved with specific tasks
(low to high level processing) -changes over
time (seconds to years) -correlates of behavior
(response accuracy, performance
changes) Clinical research -correlates of
specifically activated networks to clinical
populations -presurgical mapping
Future Uses
Complementary use for clinical diagnosis -utiliza
tion of clinical research results -prediction of
pathology Clinical treatment and
assessment -drug, therapy, rehabilitation,
biofeedback -epileptic foci mapping -drug
effects Non clinical uses -complementary use
with behavioral, anatomical, other modality
results -lie detection -prediction of behavior
tendencies -brain/computer interface
5(No Transcript)
6Type of fMRI research performed
Motor (black) Primary Sensory (red) Integrative
Sensory (violet) Basic Cognition
(green) High-Order Cognition (yellow) Emotion
(blue)
J. Illes, M. P. Kirschen, J. D. E. Gabrielli,
Nature Neuroscience, 6 (3) p.205, 2001
7Methodology
Technology
Neuroscience
Interpretation
Applications
8The Biggest Unknowns in Functional MRI
- Relationship between neuronal activity and fMRI.
- Sources of fMRI dynamic characteristics.
- Sources of spatial and temporal variability.
- Whats really in the noise?
- Whats resting state?
- Other sources of functional contrast?
- Ultimate temporal resolution?
- Ultimate spatial resolution?
- Ultimate clinical utility?
- Best processing and display methods?
- Optimal field strength?
9The Biggest Unknowns in Functional MRI
- Relationship between neuronal activity and fMRI.
- Sources of fMRI dynamic characteristics.
- Sources of spatial and temporal variability.
- Whats really in the noise?
- Whats resting state?
- Other sources of functional contrast?
- Ultimate temporal resolution?
- Ultimate spatial resolution?
- Ultimate clinical utility?
- Best processing and display methods?
- Optimal field strength?
10Motor Cortex
Auditory Cortex
S. M. Rao et al, (1996) Relationship between
finger movement rate and functional magnetic
resonance signal change in human primary motor
cortex. J. Cereb. Blood Flow and Met. 16,
1250-1254.
J. R. Binder, et al, (1994). Effects of stimulus
rate on signal response during functional
magnetic resonance imaging of auditory cortex.
Cogn. Brain Res. 2, 31-38
11Flow modulation is not necessarily the same as
BOLD modulation
12(No Transcript)
13fMRI responses in human V1 are proportional to
average firing rates in monkey V1?
Heeger, D. J., Huk, A. C., Geisler, W. S., and
Albrecht, D. G. 2000.Spikes versus BOLD What
does neuroimaging tell us about neuronal
activity? Nat. Neurosci. 3 631633.
0.4 spikes/sec -gt 1 BOLD
Rees, G., Friston, K., and Koch, C. 2000. A
direct quantitative relationship between the
functional properties of human and macaque V5.
Nat. Neurosci. 3 716723.
9 spikes/sec -gt 1 BOLD
14Magnitude
Jesmanowicz, P. A. Bandettini, J. S. Hyde, (1998)
Single shot half k-space high resolution EPI
for fMRI at 3T. Magn. Reson. Med. 40, 754-762.
15 Relationship between neuronal activity and BOLD.
Magnitude
16 Relationship between neuronal activity and BOLD.
Location
17T1 - weighted
Flow weighted
T2 weighted
BOLD weighted
P. A. Bandettini, E. C. Wong, Echo - planar
magnetic resonance imaging of human brain
activation, in "Echo Planar Imaging Theory,
Technique, and Application" (F. Schmitt, M.
Stehling, R. Turner, Eds.), p.493-530, Springer -
Verlag, Berlin, 1997
T1 and T2 weighted
Flow and BOLD weighted
18Anatomy
BOLD
Perfusion
P. A. Bandettini, E. C. Wong, Magnetic resonance
imaging of human brain function principles,
practicalities, and possibilities, in
"Neurosurgery Clinics of North America
Functional Imaging" (M. Haglund, Ed.), p.345-371,
W. B. Saunders Co., 1997.
19NeuroImage, 19 132-144, (2003).
20(No Transcript)
21(No Transcript)
22Relationship between neuronal activity and BOLD.
23The Biggest Unknowns in Functional MRI
- Relationship between neuronal activity and fMRI.
- Sources of fMRI dynamic characteristics.
- Sources of spatial and temporal variability.
- Whats really in the noise?
- Whats resting state?
- Other sources of functional contrast?
- Ultimate temporal resolution?
- Ultimate spatial resolution?
- Ultimate clinical utility?
- Best processing and display methods?
- Optimal field strength?
24Sources of BOLD dynamic characteristics.
Yacoub E, Le TH, Ugurbil K, Hu X (1999) Magn Res
Med 41(3)436-41
Courtesy of Arno Villringer
25Sources of BOLD dynamic characteristics.
Pre-undershoot
26Post-undershoot
27Sources of BOLD dynamic characteristics.
Lu, et al (2004), ISMRM 271
28Different stimulus ON periods
Dynamic Nonlinearity Assessment
measured
linear
BOLD Response
Signal
Stimulus timing
0.25 s
0.5 s
1 s
2 s
20 s
Brief stimuli produce larger responses than
expected
R. M. Birn, Z. Saad, P. A. Bandettini, (2001)
Spatial heterogeneity of the nonlinear dynamics
in the fMRI BOLD response. NeuroImage, 14
817-826.
29BOLD response is nonlinear
Observed response
Linear response
Short duration stimuli produce larger responses
than expected
30Spatial Heterogeneity of BOLD Nonlinearity
R. M. Birn, Z. Saad, P. A. Bandettini, (2001)
Spatial heterogeneity of the nonlinear dynamics
in the fMRI BOLD response. NeuroImage, 14
817-826.
31Results visual task
Nonlinearity
Magnitude
Latency
R. M. Birn, Z. Saad, P. A. Bandettini, (2001)
Spatial heterogeneity of the nonlinear dynamics
in the fMRI BOLD response. NeuroImage, 14
817-826.
32Results motor task
8
f (SD)
6
4
2
0
1
2
3
4
5
Stimulus Duration
33Results motor task
Nonlinearity
Magnitude
Latency
34Reproducibility
Visual task
Motor task
Nonlinearity1
Nonlinearity1
Nonlinearity2
Nonlinearity2
Experiment 1
Experiment 2
Experiment 1
Experiment 2
35Different stimulus OFF periods
measured
BOLD Response
Signal
linear
Stimulus timing
2 s
3 s
4 s
8 s
16 s
Brief stimulus OFF periods produce smaller
decreases than expected
36Sources of this Nonlinearity
37BOLD Correlation with Neuronal Activity
Logothetis et al. (2001) Neurophysiological
investigation of the basis of the fMRI signal
Nature, 412, 150-157.
P. A. Bandettini and L. G. Ungerleider, (2001)
From neuron to BOLD new connections. Nature
Neuroscience, 4 864-866.
38The Biggest Unknowns in Functional MRI
- Relationship between neuronal activity and fMRI.
- Sources of fMRI dynamic characteristics.
- Sources of spatial and temporal variability.
- Whats really in the noise?
- Whats resting state?
- Other sources of functional contrast?
- Ultimate temporal resolution?
- Ultimate spatial resolution?
- Ultimate clinical utility?
- Best processing and display methods?
- Optimal field strength?
39Sources of spatial and temporal variability.
Latency and Magnitude From Subject to Voxel.
40- Rapid event-related design with varying ISI
8 ON
25 ON
50 ON
75 ON
R. Birn, et al (2001), OHBM 971
41Estimated Impulse Response
Predicted Responses to 20 s stimulation
R. Birn, et al (2001), OHBM 971
42Sources of spatial and temporal variability.
Spatial Variation
43The Biggest Unknowns in Functional MRI
- Relationship between neuronal activity and fMRI.
- Sources of fMRI dynamic characteristics.
- Sources of spatial and temporal variability.
- Whats really in the noise?
- Whats resting state?
- Other sources of functional contrast?
- Ultimate temporal resolution?
- Ultimate spatial resolution?
- Ultimate clinical utility?
- Best processing and display methods?
- Optimal field strength?
440.25 Hz Breathing at 3T
3ms
Image
26ms
Power Spectra
Respiration map
49ms
0 0.25
0.5 Hz
450.68 Hz Cardiac rate at 3T
3ms
Image
26ms
Power Spectra
Cardiac map
49ms
0 0.68 (aliased) 0.5 Hz
46Temporal S/N vs. Image S/N
PHANTOMS
SUBJECTS
1400 1200 1000 800 600 400 200
Temporal S/N
Temporal S/N
0 200 400 600 800
1000 1200 1400
Image S/N
Image S/N
N. Petridou
47Signal / Thermal Noise
Signal / Physiologic Noise
Signal to Noise Ratio
Optimal for fMRI
Resolution, Speed, Surface Coils, Field Strength,
etc..
48(No Transcript)
49(No Transcript)
50Whats really in the noise?
Spontaneous Fluctuation Correlation
0.3 Hz
51Whats really in the noise?
Correlation with External Measures
52The Biggest Unknowns in Functional MRI
- Relationship between neuronal activity and fMRI.
- Sources of fMRI dynamic characteristics.
- Sources of spatial and temporal variability.
- Whats really in the noise?
- Whats resting state?
- Other sources of functional contrast?
- Ultimate temporal resolution?
- Ultimate spatial resolution?
- Ultimate clinical utility?
- Best processing and display methods?
- Optimal field strength?
53What is resting state?
Decreases during activation
54Are decreases related to resting correlations?
What is resting state?
55What is resting state?
Clinical applications?
56The Biggest Unknowns in Functional MRI
- Relationship between neuronal activity and fMRI.
- Sources of fMRI dynamic characteristics.
- Sources of spatial and temporal variability.
- Whats really in the noise?
- Whats resting state?
- Other sources of functional contrast?
- Ultimate temporal resolution?
- Ultimate spatial resolution?
- Ultimate clinical utility?
- Best processing and display methods?
- Optimal field strength?
57Other sources of functional contrast?
Blood Volume
58Other sources of functional contrast?
59CBF
BOLD
Simultaneous Perfusion and BOLD imaging during
graded visual activation and hypercapnia
N12
60Computed CMRO2 Changes
40
30
20
10
0
-10
-20
-30
-40
Subject 2
Subject 1
R. Hoge et al.
61Other sources of functional contrast?
Direct Neuronal Current Imaging
62Other sources of functional contrast?
In Vitro Results
Active condition black line Inactive condition
red line
A 0.15 Hz activity, on/off frequency B
activity C scanner noise (cooling-pump)
Petridou, et al (2003), HBM
63Other sources of functional contrast?
Temperature
Yablonskiy, D. A., J. J. H. Ackerman, et al.
(2000). "Coupling between changes in human brain
temperature and oxidative metabolism during
prolonged visual stimulation." Proceedings of the
National Academy of Sciences of the United States
of America 97(13) 7603-7608.
64The Biggest Unknowns in Functional MRI
- Relationship between neuronal activity and fMRI.
- Sources of fMRI dynamic characteristics.
- Sources of spatial and temporal variability.
- Whats really in the noise?
- Whats resting state?
- Other sources of functional contrast?
- Ultimate temporal resolution?
- Ultimate spatial resolution?
- Ultimate clinical utility?
- Best processing and display methods?
- Optimal field strength?
65First Event-related fMRI Results
Blamire, A. M., et al. (1992). Dynamic mapping
of the human visual cortex by high-speed magnetic
resonance imaging. Proc. Natl. Acad. Sci. USA
89 11069-11073.
66R. L. Savoy, et al., Pushing the temporal
resolution of fMRI studies of very brief visual
stimuli, onset variability and asynchrony, and
stimulus-correlated changes in noise oral, 3'rd
Proc. Soc. Magn. Reson., Nice, p. 450. (1995).
67P. A. Bandettini, Functional MRI temporal
resolution in "Functional MRI" (C. Moonen, and
P. Bandettini., Eds.), p. 205-220, Springer -
Verlag,. 1999.
68Ultimate temporal resolution?
Voxel-wise hemodynamic variation
Magnitude
69Relative dynamics obtained by precise activation
timing modulation
Preliminary results (with Savoy et al. 1995)
Hemi-Field Experiment
Right Hemisphere
Left Hemisphere
70500 ms
500 ms
Right Hemifield
Left Hemifield
2.5 s
-
0 s
- 2.5 s
71Ultimate temporal resolution?
Task Timing Modulation
72Ultimate temporal resolution?
In an ideal worldno hemodynamic variation over
space.
73Ultimate temporal resolution?
Neuronal Communication Timing
74The Biggest Unknowns in Functional MRI
- Relationship between neuronal activity and fMRI.
- Sources of fMRI dynamic characteristics.
- Sources of spatial and temporal variability.
- Whats really in the noise?
- Whats resting state?
- Other sources of functional contrast?
- Ultimate temporal resolution?
- Ultimate spatial resolution?
- Ultimate clinical utility?
- Best processing and display methods?
- Optimal field strength?
75Magnitude
Jesmanowicz, P. A. Bandettini, J. S. Hyde, (1998)
Single shot half k-space high resolution EPI
for fMRI at 3T. Magn. Reson. Med. 40, 754-762.
76Ocular Dominance Column Mapping using fMRI
Menon, R. S., S. Ogawa, et al. (1997). Ocular
dominance in human V1 demonstrated by functional
magnetic resonance imaging. J Neurophysiol
77(5) 2780-7.
Optical Imaging
R. D. Frostig et. al, PNAS 87 6082-6086, (1990).
77(No Transcript)
78Parallel acquisition (16 radio frequency
channels)
Custom-built Radio-frequency (RF) coil
Nova Medical, Inc.
79Parallel acquisition (16 radio frequency
channels)
Receiver Hardware
80Individual coil images
Single combined image
81Parallel acquisition (16 radio frequency
channels)
Large improvement in signal-to-noise ratio (SNR)
- Increased resolution
- Increased imaging speed
- Increased sensitivity
82Single Shot EPI
T2 decay
EPI Readout Window
20 to 40 ms
83Multishot Imaging
T2 decay
EPI Window 2
84Partial k-space imaging
85SENSE Imaging
5 to 30 ms
Pruessmann, et al.
86Ultimate spatial resolution?
Resolving columns with single shot EPI is a goal..
using SENSE, 32 channels, 7T, and perhaps
partial k-space we might get to 0.5 mm3
87The Biggest Unknowns in Functional MRI
- Relationship between neuronal activity and fMRI.
- Sources of fMRI dynamic characteristics.
- Sources of spatial and temporal variability.
- Whats really in the noise?
- Whats resting state?
- Other sources of functional contrast?
- Ultimate temporal resolution?
- Ultimate spatial resolution?
- Ultimate clinical utility?
- Best processing and display methods?
- Optimal field strength?
88Ultimate clinical utility?
Needs Real time feedback Characterization of
confounding effects Robust yet incisive set of
probe tasks Baseline information?
89The Biggest Unknowns in Functional MRI
- Relationship between neuronal activity and fMRI.
- Sources of fMRI dynamic characteristics.
- Sources of spatial and temporal variability.
- Whats really in the noise?
- Whats resting state?
- Other sources of functional contrast?
- Ultimate temporal resolution?
- Ultimate spatial resolution?
- Ultimate clinical utility?
- Best processing and display methods?
- Optimal field strength?
90Best processing and display methods?
Processing
fMRI data, and noise is time and space varying in
predictable and unpredictable ways over several
temporal and spatial scales
Signal and noise models Model free, open ended,
methods?
Classification methods? Multivariate
methods? Connectivity (across time and space
scales?)
91Best processing and display methods?
Display
To convey -collapsed multidimensional
data -sense of data quality
Surface Glass brain ROI Time courses Example
slices Connectivity maps? Quality index?
92The Biggest Unknowns in Functional MRI
- Relationship between neuronal activity and fMRI.
- Sources of fMRI dynamic characteristics.
- Sources of spatial and temporal variability.
- Whats really in the noise?
- Whats resting state?
- Other sources of functional contrast?
- Ultimate temporal resolution?
- Ultimate spatial resolution?
- Ultimate clinical utility?
- Best processing and display methods?
- Optimal field strength?
93Optimal Field Strength?
Utility vs. Difficulty
Difficulty Shimming (generally lower T2 and
T2) RF penetration effects Stability Utility
Higher SNR Better susceptibility contrast Better
ASL perfusion contrast (longer T1)
94Functional Imaging Methods Unit Functional MRI
Facility
Post Docs Hauke Heekeren David Knight Anthony
Boemio Niko Kriegeskorte
Staff Scientists Sean Marrett Jerzy
Bodurka Frank Ye Wen-Ming Luh Rasmus Birn
Computer Specialist Adam Thomas
Scanning Technologists Karen Bove-Bettis Paula
Rowser Alda Ottley Ellen Condon
Graduate Student Natalia Petridou
Program Assistant Kay Kuhns
95Unit on Functional Imaging FMRI Core Facility
http//sodium.nimh.nih.gov/upload T165.ppt