SLIDE 1 NBML, Webinars, Second Meeting
Resting State fMRI: I: From Basics to Advance Applications in in
Hamed Ekhtiari, MD, PhD, National Brain Mapping Laboratory (NBML)
SLIDE 2
SLIDE 3
Main Functional Neuroimaging Softwares
SLIDE 4
SLIDE 5
SLIDE 6
SLIDE 7
sMRI tfMRI dMRI rfMRI
SLIDE 8
SLIDE 9
Clinically Meaningful Image-Derived Phenotypes (IDPs)
SLIDE 10
Automated Clinically Meaningful Image-Derived Phenotypes (IDPs)
SLIDE 11
sMRI tfMRI dMRI rfMRI
Automated Clinically Meaningful Image-Derived Phenotypes (IDPs)
SLIDE 12
rfMRI
Automated Clinically Meaningful Image-Derived Phenotypes (IDPs)
SLIDE 13
rfMRI
1771 Automated Image-Derived Phenotypes (IDPs)
SLIDE 14
rfMRI
? Clinically Meaningful Automated Image-Derived Phenotypes (IDPs)
SLIDE 15
sMRI 39 tfMRI 16 dMRI 675 rfMRI 1771
Automated Clinically Meaningful Image-Derived Phenotypes (IDPs)
SLIDE 16
1100 Health Related Factors (11 Cat) 2501 IDPs (6 Cat)
SLIDE 17
What is Resting Functional MRI?
rfMRI 1771
SLIDE 18
SLIDE 20 Rest Task Rest Task Rest Task Rest Task Rest
Task-based fMRI (Task>Rest) What happens during REST? (Rest>Task)
SLIDE 21
SLIDE 22
Default Mode Network (DMN)
SLIDE 23
SLIDE 24 Lets do some calculations!
24
SLIDE 25 13 * 12 = ?
25
156
SLIDE 26 15 * 16 = ?
26
240
SLIDE 27 23 * 21 = ?
27
483
SLIDE 28 24 * 15 = ?
28
360
SLIDE 29
SLIDE 30
SLIDE 31 Environmental Cues
Executive Control
SLIDE 32 Environmental Cues
Executive Control
SLIDE 33
SLIDE 34
SLIDE 35
SLIDE 36 Spontaneous Low Frequency Fluctuations during Rest Have Meaningful Signals
SLIDE 37
How to Analyze These Time Series of Data (rfMRI)?
SLIDE 38
SLIDE 39
SLIDE 40
SLIDE 41 (Cole, et al., 2010)
Default Mode Network with Seeds in PCC
SLIDE 42 (Cole, et al., 2010)
Default Mode Network with Seeds in PCC
SLIDE 43 (Cole, et al., 2010)
SLIDE 44
SLIDE 45
SLIDE 46
SLIDE 47
SLIDE 48
SLIDE 49
Temporal Concatenation
SLIDE 50
Temporal Concatenation
SLIDE 51
From Group ICA to Individual Maps
SLIDE 52 Beckmann, et al., 2005
SLIDE 53
SLIDE 54 Amplitude of Low Frequency Fluctuation (ALFF)
(Zang et al., 2007, 2008)
SLIDE 55 Regional Homogeneity (ReHo)
- Similarity of the time courses within usually 27 neighboring voxels
- Measured with Kendall Coefficient of Concordance (KCC) (0-1)
- Recorded as a value for the central voxel
- Results in a voxel wise KCC values in the individualized maps
(Zang et al., 2004, Liu, et al., 2010)
3 3
SLIDE 56
SLIDE 57
rfMRI
Now, What is this?
SLIDE 58
IDPs: 55 (out of 100) Nodes (Regional ALFF) and 1695 Edges (Connectivity)!
SLIDE 59
Number Matters?
SLIDE 60
1100 Health Related Factors & 2501 Image-Derived Phenotypes (IDPs)
sMRI 39 tfMRI 16 dMRI 675 rfMRI 1771
What do you do with this database?
SLIDE 61 Levels of Data Analysis
- 1. Univariate Correlations (one by one)
- 2. Covariates of No Interest or Confounders (Age, Gender, and etc)
- 3. Data Driven Multivariate Analysis
- 4. Hypothesis Driven Analysis
1100 Health Related Factors & 2501 Image-Derived Phenotypes (IDPs) What do you do with this database?
SLIDE 62 2.8 million testing
FDR (P= 3.8 . 10−5) and Bonf (P=1.8 . 10−8)
- 1. Univariate Correlations (one by one)
SLIDE 63
- 1. Univariate Correlations
SLIDE 64
SLIDE 65