brain imaging: reality/hype
Post on 02-Jun-2015
329 Views
Preview:
TRANSCRIPT
Introduction to neuroimaging: neurons, images, inference
Geoffrey K Aguirre, MD, PhDcfn.upenn.edu/aguirre
Center for Neuroscience & Society
Monday, March 8, 2010
GK Aguirre
neural activity functional MRI image
Monday, March 8, 2010
GK Aguirre
raw data over time processed result
Monday, March 8, 2010
GK Aguirre
processed result
lying
inference
Monday, March 8, 2010
GK Aguirre
neural activity
Monday, March 8, 2010
GK AguirreMonday, March 8, 2010
GK Aguirre
x-ray human head
Monday, March 8, 2010
Griff Wason
CT scanner
Monday, March 8, 2010
GK Aguirre
axial CT scan of the brain
Monday, March 8, 2010
Griff Wason
Allan M. Cormack Godfrey Hounsfield
1979 Nobel Prize in Physiology or Medicine
Monday, March 8, 2010
GK Aguirre
inject radio-labeled sugar or water
Monday, March 8, 2010
GK Aguirre
co-incidence detection for positron emission tomography
Monday, March 8, 2010
GK Aguirre
PET scan of the brain
Monday, March 8, 2010
GK AguirreMonday, March 8, 2010
GK AguirreMonday, March 8, 2010
B0
M0B0
H2O
Monday, March 8, 2010
B0
Y
X
Z90° radio-frequency
pulse
high energy state
Monday, March 8, 2010
T1 signal
B0
Y
X
Z
high energy state
Y
X
Z
low energy state
Y
X
Z
protons relax
RF energy released
time
RF
ener
gy T1 time constant of decay
Monday, March 8, 2010
GK Aguirre
T1 weighted T2 weighted
Monday, March 8, 2010
GK AguirreFLAIR pulse sequence
Monday, March 8, 2010
Griff Wason
Paul Lauterbur Sir Peter Mansfield
2003 Nobel Prize in Physiology or Medicine
Monday, March 8, 2010
GK Aguirre
diffusion weighted (DWI) gradient echo
FLAIR MRA
Monday, March 8, 2010
GK Aguirre
susceptibility artifact from iron oxide particlessuspended in beeswax in hair
Monday, March 8, 2010
GK Aguirre
hemoglobin contains iron atoms
Monday, March 8, 2010
paramagnetic diamagnetic
GK Aguirre
Linus Pauling
Monday, March 8, 2010
Deoxyhemoglobin creates magnetic field inhomo-geneities, causing spin dephasing
Y
X
Z
protons de-phase
Monday, March 8, 2010
GK Aguirre
gradient echoplanar image
Monday, March 8, 2010
Neuro-vascular coupling
↑ blood flow
↑ blood volume
↓ [deoxy-Hgb]
physiologic delay
Monday, March 8, 2010
GK Aguirre
neuro-imagingcoupling
T2*
time(secs)
imag
ing
sign
alMonday, March 8, 2010
GK Aguirre
raw data over time processed result
Monday, March 8, 2010
GK Aguirre
voxels
3x3x3 mm
~ 10 million neurons
Monday, March 8, 2010
time
1 2 3 4 5 6 7 8 9neur
al a
ctiv
ity
fMR
I sig
nal
time (seconds)
Monday, March 8, 2010
GK Aguirre
A hemodynamic proxy of neural activity
neuro-imagingcoupling
T2*
time(secs)
imag
ing
sign
al
Monday, March 8, 2010
GK Aguirre
A nearly linear system
T2*
neural activity BOLD fMRI signal
low-pass~ linearsystem
Monday, March 8, 2010
The BOLD fMRI system in spaceSpatial resolution of about 0.025 cm3 In a typical 3 x 3 x 3 mm voxel, there are ~ 10 million neurons
Monday, March 8, 2010
-0.5
0
0.5
1
0 2 4 6 8 10 12 14time(secs)
Spatial resolution of ~0.025 cm3 (3x3x3 mm, ~10 million neurons)
Temporal resolution of 3-4 s
imag
ing
sign
al
Monday, March 8, 2010
GK Aguirre
Compare conditions over timeBO
LD fM
RI s
igna
l
+ + +
Monday, March 8, 2010
0.0 -7.0
Color-coded statistical result:
t-statistic0.0 7.0
BOLD
fMR
I sig
nal
t-value = 2.1
GK AguirreMonday, March 8, 2010
surfacereconstruction
hi-resanatomical
digitalsmoothing
thresholding
Neuroaesthetics
GK AguirreMonday, March 8, 2010
The BOLD fMRI signal has no direct, absolute interpretation.
Must be compared between states studied close together in time.
A relative signal
Monday, March 8, 2010
Neural activity?
A hemodynamic proxy of neural activity
• fMRI measures blood flow; differences in vascular response can confound differences in neural activity
time(secs)
younger
older
imag
ing
sign
aldrugs, hormonal states, age, gender
GK AguirreMonday, March 8, 2010
A generalizable result?
Studied population
healthy collegestudents
middle-agedlawyers
children
? ?
GK AguirreMonday, March 8, 2010
An integrated measure
The BOLD fMRI signal integrates neural activity over seconds and millimeters.
(spike rate vs. local field potential)
Detectable • a bulk change in neural activity
Undetectable • a change in population code
Monday, March 8, 2010
GK Aguirre
raw data over time processed result
Monday, March 8, 2010
GK Aguirre
processed result
love
inference
Monday, March 8, 2010
Three basic types of neuroimaging studies
The key question to ask for each
GK AguirreMonday, March 8, 2010
GK Aguirre
cognition / consciousness
T2*
. .
4system #1 system #2
workingmemory
physics / physiology
Monday, March 8, 2010
Three basic types of neuroimaging studies
The key question to ask for each
GK AguirreMonday, March 8, 2010
The brain area for “Love”?
Forward inference
?
which brain areas correspond to an isolated behavior?
Love
GK AguirreMonday, March 8, 2010
Forward inference
isolate behavior by subtracting conditions
The brain area for “Love”?
Spouse - Friend= Love
GK AguirreMonday, March 8, 2010
Spouse vs. Friend over time
The brain area for “Love”?
GK AguirreMonday, March 8, 2010
Forward inference
what if the “subtraction” includes other mental states?
= Familiarity?= Obligation?= Sexuality?
= Love
The brain area for “Love”?
GK AguirreMonday, March 8, 2010
Forward inference
relate variations in the behavior to variations in neural response
brai
n ac
tivity
amount of loveLOVE!Lovelove?
The brain area for “Love”?
GK AguirreMonday, March 8, 2010
Forward inference
relate individual differences in behavior to brain differences
brai
n ac
tivity
amount of love
The brain area for “Love”?
LOVE!Love
love?
GK AguirreMonday, March 8, 2010
Forward inferencehow was the behavior isolated?
GK AguirreMonday, March 8, 2010
Clinton induces “Conflict”?
Focal reverse inference
use local brain activity to identify mental states or emotions a situation evokes
conflict
Conflict
GK AguirreMonday, March 8, 2010
• Why we love• What your brain looks like on faith• What makes us moral• When worry hijacks the brain• How we get addicted• Marketing to your mind
• Inside the grieving brain• It feels good and everybody does it [scratching]• Mind reading is now possible• This is your brain on optimism• Hot flashes [fMRI of menopause]
Untrue statements make us feel disgust, just like seeing
rotten food
People with complicated grief experience paradoxical
pleasure during sadness
Scratching evokes a sense of pleasure because it
decreases memory of pain
GK AguirreMonday, March 8, 2010
Clinton induces “Conflict”?
Focal reverse inference
use local brain activity to identify mental states or emotions a situation evokes
conflict
Conflict
GK AguirreMonday, March 8, 2010
Clinton induces “Conflict”?
Focal reverse inferenceConflict
what if more than one emotion can activate a brain region?
?hope
decision making
unpleasant memories
GK AguirreMonday, March 8, 2010
Implementation
relate variations in stimulus properties or information processing to response
brai
n ac
tivity
speed of motion
“Boring neuroscience”
GK AguirreMonday, March 8, 2010
Focal reverse inferencehow strong is the association between local brain activity and the assumed evoked behavior?
GK AguirreMonday, March 8, 2010
Are you lying?
Distributed reverse inference
measure distributed patterns of response to classify mental states
GK Aguirre
I picked the ace of spades!
Monday, March 8, 2010
• Why we love• What your brain looks like on faith• What makes us moral• When worry hijacks the brain• How we get addicted• Marketing to your mind
• Inside the grieving brain• It feels good and everybody does it [scratching]• Mind reading is now possible• This is your brain on optimism• Hot flashes [fMRI of menopause]
fMRI data can determine what tool you are currently thinking
about (hammer or wrench)
Which of 10,000 different pictures you are viewing can
be read from your cortex
The scanner can determine if you are lying or telling the
truth
GK AguirreMonday, March 8, 2010
Are you lying?
Distributed reverse inference
1) train a computer to learn the pattern of activity seen with different mental states
GK Aguirre
Iʼll tell the truth
Monday, March 8, 2010
Are you lying?
Distributed reverse inference
1) train a computer to learn the pattern of activity seen with different mental states
GK Aguirre
Iʼll lie and say I have a jack
Monday, March 8, 2010
Are you lying?
Distributed reverse inference
2) classify an unknown brain pattern as belonging to a particular state
?
GK Aguirre
I picked the three of clubs!
Monday, March 8, 2010
Are you lying?
Distributed reverse inference
How stimulus or context dependent is the effect? Particularly relevant for lie detection...
?
GK Aguirre
I never killed a man in Reno just to watch
him die!
Monday, March 8, 2010
Multi-voxel classificationcan the classification be generalized beyond the training context?
GK AguirreMonday, March 8, 2010
Forward inferencedetermine which brain region is associated with an isolated behavior
Focal reverse inferenceuse localized brain activity to determine which mental states are evoked by a complex behavior
Multi-voxel classificationuse distributed patterns of brain activity to predict which mental state is being experienced
GK AguirreMonday, March 8, 2010
Forward inference
Focal reverse inference
Multi-voxel classification
how was the behavior isolated?
how strong is the association between local brain activity and the assumed evoked behavior?
can the classification be generalized beyond the training context?
GK AguirreMonday, March 8, 2010
Monday, March 8, 2010
top related