applications of real time fmri: pain treatment and

APPLICATIONS OF REAL TIME FMRI:PAIN TREATMENT AND SUBSTANCE ABUSE TREATMENT

MARCH 2007

THIS WORK SUPPORTED BY NIH:

MH067290-01 NS050642-03 DA-4-7748 DA-4-7748 NS049673-01 DA021877-01A1

Applications of Real Time fMRI

Applications of Real Time fMRI - Phase II

Virtual Reality and Real Time fMRI

Virtual Reality and Real Time fMRI - Phase II

Novel Methods for Functional Brain Imaging

Measurement and Control of Patterned Brain Activation U i R l Ti fMRI

SPECIAL THANKS FOR HELP AND GUIDANCENora Volkow, NIDARo Nemeth, NIDADave Thomas, NIDALarry Stanford, NIDALinda Porter, NINDS

1

COLLABORATIVE TEAM

SEAN MACKEY–Axel Lucca–Deepak Soneji

JOHN GABRIELI

–Fumiko Maeda

–Alison Adcock

GARY GLOVER

JOHN PAULY

CHRISTOPHER DECHARMS–Chloe Hutton–Susan Landau–Brett Mensh–Kristen Lutomski–Saxon MacLeod–Debbie Scacco–Dave Hagewood

AFFILIATION

2

TALK OUTLINE

OVERVIEW OF REAL TIME FMRI

LEARNED CONTROL OVER BRAIN ACTIVATION AND PAIN

RTFMRI IN CHRONIC PAIN

SUBSTANCE ABUSE - PRELIMINARY EXPERIENCES

3

DESCARTES VIEW OF BRAIN, AND PAIN

4

IS IT POSSIBLE TO VISUALIZE THE MECHANISMS UNDERLYING PERCEPTION IN REAL TIME?

fMRI

5

CAN MRI BECOME A THERAPEUTIC MODALITY?

Today3TMRI

Diagnostic Radiology

DIAGNOSTIC

• MRI provides answer

• Very broad application

DIAGNOSTIC



6

CAN MRI BECOME A THERAPEUTIC MODALITY?

Today3TMRI

DIAGNOSTIC



DIAGNOSTIC



THERAPEUTIC

• MRI provides patient improvement

• Application in areas of severe need

THERAPEUTIC

• MRI provides patient improvement

• Application in areas of severe need

Tomorrow?

Diagnostic Radiology Neuroimaging Therapy

7

RTFMRI AS A POTENTIAL NEW INTERFACE TO THE NERVOUS SYSTEM

+•Control of screen cursor demonstrated in monkeys•Potential to control prosthetics

Multi-electrode recording

•Used in epilepsy and elsewhere•EEG Neurofeedback•Control of screen cursor demonstrated in people

EEG-based measurement

•Drive centers in the brain that control global functioning in order to remediate disease•Currently applied in Parkinson’s disease, efforts underway in others

Deep brain stimulation

•Restore hearing through direct stimulation of the nervous system in the profoundly deaf•Potential to move on to vision as well

Cochlear implant

+

Wires

Wires

Wires

Wires

+

+

8

RTFMRI AS A POTENTIAL NEW INTERFACE TO THE NERVOUS SYSTEM

+•Control of screen cursor demonstrated in monkeys•Potential to control prosthetics

Multi-electrode recording

•Used in epilepsy and elsewhere•Used in anesthesia monitoring•Control of screen cursor demonstrated in people

EEG-based measurement

•Non-Invasive•No tissue damage•Reasonable localization

Neuroimaging/Cognitive

•Drive centers in the brain that control global functioning in order to remediate disease•Currently applied in Parkinson’s disease, efforts underway in others

Deep brain stimulation

•Restore hearing through direct stimulation of the nervous system in the profoundly deaf•Potential to move on to vision as well

Cochlear implant

+

Wires

Wires

Wires

Wires

+

+

+

Photons

9

Slice Thicknesse.g., 6 mm

Number of Slicese.g., 10

OBLIQUE SLICES

MRI: IMAGES OF ANATOMY –PHYSICAL STRUCTURE

ConsecutiveSlicesThroughSPACE

10

~2s

Same SpatialSlice FollowedThrough TIME ...

fMRI: IMAGES OF PHYSIOLOGY –FUNCTION

10 min ofdata collection

fMRI FUNCTIONAL IMAGES ARE DERRIVED FROM CHANGES IN T2*-SENSITIVE LOW-RESOLUTION IMAGES OVER TIME

11

Statistical Mapsuperimposed on

anatomical MRI image

~2s

Same SpatialSlice FollowedThrough TIME

Condition 1

Condition 2 ...

fMRI FUNCTIONAL IMAGES ARE DERRIVED FROM DEVIATIONS IN LOW-RESOLUTION ANATOMICAL IMAGES


Hours/daysof analysis

12



~2s


Condition 1

Condition 2 ...


Region of interest (ROI)



13



~2s


Condition 1

Condition 2 ...


Region of interest (ROI)



Time

fMRISignal

(% change)

ROI Time Course

Condition

14

OVERVIEW OF METHOD

MRI acquires real time fMRI data (spiral or EPI)

MRI acquires real time fMRI data (spiral or EPI)

Real time fMRI analysis•Motion correction•Temporal filtering•Spatial filtering•Event-related averages•Pattern comparison

Real time fMRI analysis•Motion correction•Temporal filtering•Spatial filtering•Event-related averages•Pattern comparison

Subjects (or patients or clinicians) watch their cognitive processes unfold ‘live’, depicted as simulated displays

Subjects (or patients or clinicians) watch their cognitive processes unfold ‘live’, depicted as simulated displays

Image from story courtesy

15

RTFMRI SETUP

16

RTFMRI-BASED TRAINING – A MORE PRECISE, ANATAMICALLY TARGETED MEASURE THAN TRADITIONAL AUTONOMIC ‘BIOFEEDBACK’

60’s Predominantly measures of global arousal.

Most useful if you want to teach relaxation

AUTONOMIC FUNCTIONHeart RateBreath RateSkin ConductanceSkin Temperature

CONSEQUENCESMEASURES

17

RTFMRI-BASED TRAINING – A MORE PRECISE, ANATAMICALLY TARGETED MEASURE THAN TRADITIONAL AUTONOMIC ‘BIOFEEDBACK’

TODAY

60’s

Can measure the very specific neurophysiological functions associated with the >100 individual brain areas.

Can measure patterns of activation evolving across multiple brain areas.

Potential to train subjects to produce very specific neurophysiological effects.

BRAIN FUNCTION

Predominantly measures of global arousal.

Most useful if you want to teach relaxation

AUTONOMIC FUNCTIONHeart RateBreath RateSkin ConductanceSkin Temperature

CONSEQUENCESMEASURES

18

CHALLENGES WITH fMRI AS A MEASURE OF BRAIN FUNCTION

Patient care costs can easily run into $100k/yr/patient for many CNS diseases. Invasive CNS procedures can easily cost this much for a single procedure.

Isn’t MRI way too expensive to really be practical?

$$

SPATIALSCALE

TIMESCALE

It may not be necessary to control individual neurons to achieve important applications: eg drugs, deep brain stimulation.

There may be ~107 neurons in the areas that fMRI is measuring – that’s no way to measure the code.

Cognitive processing is closer to the seconds timescale. We’ll use better temporal methods as soon as they come along.

Neural activation is on a msec timescale, diseases lead to long-term changes in brain function. fMRIsignals evolve over a few seconds.

POTENTIAL RESOLUTIONPROBLEM

“Neurophysiologist reaction”

19

CHALLENGES WITH fMRI AS A MEASURE OF BRAIN FUNCTION

Patient care costs can easily run into $100k/yr/patient for many CNS diseases. Invasive CNS procedures can easily cost this much for a single procedure.

Isn’t MRI way too expensive to really be practical?

$$

OVER TIME, IMAGING

TECHNOLOGY WILL EVOLVE TO

NEW USES:

SPATIALSCALE

TIMESCALE

It may not be necessary to control individual neurons to achieve important applications: eg drugs, deep brain stimulation.

There may be ~107 neurons in the areas that fMRI is measuring – that’s no way to measure the code.

Cognitive processing is closer to the seconds timescale. We’ll use better temporal methods as soon as they come along.

Neural activation is on a msec timescale, diseases lead to long-term changes in brain function. fMRIsignals evolve over a few seconds.

POTENTIAL RESOLUTIONPROBLEM

20

RTFMRI AND COGNITIVE TRAINING TAKE-HOME EXERCISE

Next WedProduce 1% modulation in rACC activation

DUE DATEASSIGNMENTPOSSIBLE CREDIT

21


??Take control over your own reward and endorphin systems…



22


FutureDecrease pain and suffering

??Take control over your own reward and endorphin systems…



23

REAL TIME FMRI TRAINING OF BRAIN FUNCTION

Pre Post

R L R LActivation target

Learned regulation of spatially localized brain activation using real-time fMRI. NeuroImage (2004) 21, 436-443deCharms, R. C., Christoff, K., Glover, G. H., Pauly, J. M., Whitfield, S., and Gabrieli, J. D.

24

IMPACT OF RTFMRI TRAINING ON BRAIN ACTIVATION

A) Pre-Training BOLD Individual

B) Post-Training BOLD Individual

0 60s

60 120 180 240s

C) Average

D) Average

E) Concurrent EMG

0Time (s)

% s

igna

l ∆%

sig

nal ∆

mV

2

1

-1

-2

2

1

-1

-2

-101

.6

0

.2

.4

.6

.2

.4

0

25

TIME COURSE OF TRAINING EFFECT AND CONTROLS

B)A) C) D) E)

Training,ROI

Whole brain control

ShamTraining,ROI

MotorTask,ROI

Post-TrainingTest,ROI

Sess

ion

I

0

.5

Sess

ion

IISe

ssio

n III

Sess

ion

ISe

ssio

n II

Sess

ion

III

Sess

ion

ISe

ssio

n II

Sess

ion

III

Expe

rimen

tal

Sham

+ Fe

edba

ck

-Fee

dbac

k

% s

igna

l ∆

26

CAN THIS APPROACH BE USED IN CLINICALLY IMPORTANT AREAS?

27

TRANSLATING BASIC RESEARCH IN PAIN INTO A NEW POTENTIAL THERAPEUTIC APPLICATION AREA: NEUROIMAGING

APPLIED QUESTIONBASIC RESEARCH

…and subjects with different pain sensitivities show differences in a similar group of brain regions

CAN NEURAL PLASTICITY BE ANATOMICALLY TARGETED?

Pain, and brain, can be changed substantially by mechanisms of plasticity

CAN SUBJECTS SHIFT THEIR PAIN TOLERANCE OR PERCEPTION?

There are large individual differences in pain perception…

CAN SUBJECTS BE TRAINED TO MORE EFFECTIVELY COGNITIVELY CONTROL PAIN?

Pain can be powerfully modulated by cognitive processes including attention, placebo effect, hypnosis, and many others involving a matrix of brain regions

012345678910

0 2 4 6 8 10stimulus

pain

Rainville…Bushnell, Science 1997

28

POTENTIAL TARGETS IN THE PAIN CONTROL SYSTEM

DCN

ThalamusVPL/VPM

SI/SII

MI/SMA

Insula

LateralOrbito-Frontal

AnteriorCingulate

Rostral ACC

Amygdala

PeriaqueductalGray

Pons/Parabrachial

nucleus

RostralVentromedial

Medulla

Spinal CordDorsal Horn

Ascending Pain

Perception System

Descending Pain Control

System

Invasive Electrical NeurostimulationProvides Pain Relief

Internal capsule

Potential Target

29

RTFMRI TRAINING PROTOCOL IN HEALTHY SUBJECTS

Decrease60s

Rest30s

Increase60s

CYCLE, (3 blocks, 150s total)

BLOCK DESIGNPain Pain

Control over brain activation and pain learned by using real-time functional MRI.Proceedings of the National Academy of Sciences (2005)deCharms, R. C., Maeda, F., Glover, G. H., Ludlow, D., Pauly, J. M., Soneji, D., Gabrieli, J. D., and Mackey, S. C.

30


Decrease60s

Rest30s

Increase60s


RUN, (5 cycles + ratings, 13min). 1-5 RUNS per TRAINING DAY)

BriefingPre-TestsAnatomicals

Cycle 1150s

Cycle 2150s

Cycle 3150s

Cycle 4150s

Cycle 5150s

After ScanRatings Debrief

BLOCK DESIGNPain Pain

Control over brain activation and pain learned by using real-time functional MRI.Proceedings of the National Academy of Sciences (2005)deCharms, R. C., Maeda, F., Glover, G. H., Ludlow, D., Pauly, J. M., Soneji, D., Gabrieli, J. D., and Mackey, S. C.

31


Decrease60s

Rest30s

Increase60s




Cycle 1150s

Cycle 2150s

Cycle 3150s

Cycle 4150s

Cycle 5150s


BLOCK DESIGN

ROI TARGET: rostral Anterior Cingulate Cortex

Pain Pain

32


Decrease60s

Rest30s

Increase60s




Cycle 1150s

Cycle 2150s

Cycle 3150s

Cycle 4150s

Cycle 5150s


BLOCK DESIGN


Pain Pain

SUBJECT INSTRUCTIONS: Written text describing cognitive modulation of pain

•Attend to pain vs. attend away•Perceive the pain as more intense vs. less intense•Perceive the pain as harmful vs. only a tactile sensation

33


Decrease60s

Rest30s

Increase60s




Cycle 1150s

Cycle 2150s

Cycle 3150s

Cycle 4150s

Cycle 5150s

After ScanRatings

0 50 100s

-101234

fMR

I BO

LD d

iffer

ence

Debrief

BLOCK DESIGN


Pain Pain



SUBJECT DISPLAYS

34

rtfMRI-BASED TRAINING LEADS TO SPATIALLY-SPECIFIC CHANGES IN BRAIN ACTIVATION

MEASURE: Thresholded T-statistic, (INCREASE – DECREASE) last run VS. (INCREASE – DECREASE) first run

35

HEALTHY SUBJECTS LEARN INCREASED CONTROL OVER BRAIN ACTIVATION THROUGH THE COURSE OF TRAINING

Training run 1

Trainingrun 2

Training run 3

Final testrun 4

0

0.1

0.2

0.3

0.4

0.5

0.6

rAC

Cac

tivat

ion

(BO

LD) * †

MEASURE: Brain Activation, BOLD % Signal Change,(Increase Period – Decrease Period) from each pair of blocks,

Averaged over N=8 Subjects

36

HEALTHY SUBJECTS LEARN INCREASED CONTROL OVER PAINTHROUGH THE COURSE OF TRAINING

Training run 1

Training run 2

Training run 3

Final testrun 4

-10

0

10

20

30

40

50

Pain

inte

nsity

ratin

g (%

diff

eren

ce)

** †

MEASURE: Pain Intensity Rating % Difference,(Increase Period Rating – Decrease Period Rating)/Average from each pair of blocks,

Averaged over N=8 Subjects

37

THE TIMECOURSE OF LEARNING OF CONTROL OVER BRAIN ACTIVATION MIRRORS THE TIME COURSE FOR CONTROL OVER PAIN

Training run 1

Training run 2

Training run 3

Final testrun 4

-10

0

10

20

30

40

50

Pain

inte

nsity

ratin

g (%

diff

eren

ce)

** †

Training run 1

Trainingrun 2

Training run 3

Final testrun 4

0

0.1

0.2

0.3

0.4

0.5

0.6

rAC

Cac

tivat

ion

(BO

LD) * †

38

LEARNED CONTROL OVER BRAIN ACTIVATION IN RACC LEADS TO CORRESPONDING CHANGES IN PAIN INTENSITY RATINGS FOR A CONCURRENT THERMAL STIMULUS

-0.5 0 0.5 1 1.5-40

-20

0

20

40

60

80

100

rACC activation (BOLD)

Pain

inte

nsity

(% d

iffer

ence

)

y=28.167x + 7.013 R=0.368 p<0.00076

39

FOUR CONTROL GROUPS WERE TRAINED USING SIMILAR OR IDENTICAL PROCEDURES BUT IN THE ABSENCE OF RACC RTFMRI INFORMATION

BLIND CONTROLControl for cognitive effects.

Received identical training to the experimental group, but unknown to them the rtfMRI displays that they saw corresponded to activation from a previously-tested experimental subject’s rACC, rather than their own rACC.

GROUP IV

BLIND CONTROLControl for spatial and physiological specificity

Received identical training to the experimental group, but using rtfMRI information derived from a posterior cingulate cortex region not involved in pain processing, to examine spatial and physiological specificity.

GROUP III

Control for identical training without rtfMRI

Received identical instructions to the experimental group, and the same period of training, but with no rtfMRI information, to test the effect of identical practice alone.

GROUP II

Control for effects of extended attention training

Received purely behavioral training for twice as long as the experimental group, but they had no rtfMRI feedback. They were additionally instructed to focus attention on the thermal stimuli during “increase” periods.

GROUP I

40

THE LEARNED CONTROL OVER PAIN REQUIRES SPATIALLY-SPECIFIC RTFMRI INFORMATION

rACCexper.group

-40

-30

-20

-10

0

10

20

30

40

50C

hang

e in

pai

n ra

ting

(% d

iffer

ence

)Pain Intensity

41


rACCexper.group

Attention controlgroup I

– rtfMRIcontrolgroup II

PCC control

group III

Yoked control

group IV

-40

-30

-20

-10

0

10

20

30

40

50C

hang

e in

pai

n ra

ting

(% d

iffer

ence

)

† † † *** * ** ***† †

Pain Intensity

p<.01† †p<.001† † †p<.05 vs. experimental*p<.01 vs. experimental **p<.001 vs. experimental***

CONTROL GROUPS,NO EFFECT

EFFECT

42


rACCexper.group

Attention controlgroup I

– rtfMRIcontrolgroup II

PCC control

group III

Yoked control

group IV

-40

-30

-20

-10

0

10

20

30

40

50C

hang

e in

pai

n ra

ting

(% d

iffer

ence

)

† † † *** * ** ***† † *** * *** **

UnpleasantnessPain Intensity

p<.01† †p<.001† † †p<.05 vs. experimental*p<.01 vs. experimental **p<.001 vs. experimental***

CONTROL GROUPS,NO EFFECT

EFFECT

43

CAN THE PICTURES OF YOUR HEAD PROVIDE RELIEF?

44

RTFMRI TRAINING PROTOCOL IN PAIN PATIENTS

Decrease60s

Rest30s

Increase60s




Cycle 1150s

Cycle 2150s

Cycle 3150s

Cycle 4150s

Cycle 5150s

After ScanRatings

0 50 100s

-101234

fMR

I BO

LD d

iffer

ence

Debrief

BLOCK DESIGN




SUBJECT DISPLAYS

NO PAINFULEXTERNAL STIMULI

45

PATIENT REPORT OF PAIN MEASURES

PRIOR TO SCANNING

Pain Rating Index

46

PATIENT REPORT OF PAIN MEASURES

PRIOR TO SCANNING

Pain Rating Index

AFTER SCANNING

Pain Rating Index

47

CHANGE IN PAIN RATINGS FOLLOWING RTFMRI TRAINING IN CHRONIC PAIN PATIENTS

MPQVAS

rACC rtfMRIexperimental

group

Cha

nge

in p

ain

ratin

g (%

)

10

20

30

40

50

60

70

80

p<.001

† † †

† † †

48

PATIENTS WHO LEARNED TO CONTROL RACC ACTIVATION SHOWED A CHANGE IN PAIN, OTHERS DID NOT

--0.20.2 00 0.20.2 0.40.4 0.60.6 0.80.8 11 1.21.2 1.41.4 1.61.6 1.81.8rACCrACC change in activation (BOLD)change in activation (BOLD)

y=0.422x + 0.085. R=0.9170. y=0.422x + 0.085. R=0.9170. p<0.01p<0.01

--101000

10102020303040405050606070708080

Cha

nge

in V

AS

(%)

Cha

nge

in V

AS

(%)

00 0.20.2 0.40.4 0.60.6 0.80.8 11 1.21.2 1.41.4 1.61.6 1.81.8

y=0.376x +0.392. R=0.9173. y=0.376x +0.392. R=0.9173. p<0.01p<0.01

3030404050506060707080809090

100100

Cha

nge

in M

PQ (%

)C

hang

e in

MPQ

(%)

2020

rACCrACC change in activation (BOLD)change in activation (BOLD)

49

A CONTROL GROUP, TRAINED USING AUTONOMIC BIOFEEDBACK, DID NOT SHOW THE SAME CHANGES IN PAIN

p<.02p<.02

MPQVAS

rACC rtfMRIexperimental

group

Autonomic feedbackcontrol group

Cha

nge

in p

ain

ratin

g (%

)

10

20

30

40

50

60

70

80

p<.001† † †

† † †

† † †

EFFECTCONTROL GROUP,

NO EFFECT

50

MOTIVATIONS FOR NEUROIMAGING THERAPY IN CHRONIC PAIN TREATMENT

No surgery.No physical intervention required.

NON-INVASIVE

Technology grows to meet need:Less expensivePHYSICIAN’S OFFICE IMAGING MAY BECOME FEASIBLE

REVERSIBLE

NON-PHARMACOLOGIC

Potentially low risk.Can be terminated if unsuccessful.

No drug-related side effects.Uses endogenous physiological, neurotransmitter systems

RATIONALEIMPLICATION

51CAN THIS APPROACH COOL THE FIRES OF CHRONIC PAIN?

52

INTERFACE FOR CHRONIC PAIN PATIENTS

A Better View of Brain Disorders Science 313, 1377-1379 (8 Sept, 2006)Miller, G

53

CAN REAL TIME FMRI LEAD TO A NEW, MECHANISTICALLY-BASED, COMPUTER GUIDED FORM OF COGNITIVE INTERVENTION?

54

ONGOING STUDY: LONG-TERM RTFMRI TRAINING PROTOCOL IN PAIN PATIENTS

Decrease60s

Rest30s

Increase60s




Cycle 1150s

Cycle 2150s

Cycle 3150s

Cycle 4150s

Cycle 5150s


BLOCK DESIGN

ROI TARGETS: (Two groups) 1. rostral Anterior Cingulate Cortex2. Training using rACC and bilateral insula

PROTOCOL: Training over 6 consecutive sessions, approximately 6 weeks

SUBJECT DISPLAY

NO PAINFULEXTERNAL STIMULI

55

WILL NEUROIMAGING THERAPY PRODUCE LONG-TERM DECREASES IN CHRONIC PAIN?

3

3.5

4

4.5

5

5.5

6

6.5

7

1 6

Training Week

VAS

Pain

Rat

ing

Training Group (N=21)21% decreaseDecrease seen in 16/21

NOTE:Preliminary, Unpublished Data!No Control Group to DatePlacebo Effects Are Likely

56

WILL NEUROIMAGING THERAPY PRODUCE LONG-TERM DECREASES IN CHRONIC PAIN?

Training Group (N=21)21% decreaseDecrease seen in 16/21

3

3.5

4

4.5

5

5.5

6

6.5

7

1 6Training Week

VAS

Pain

Rat

ing

Waiting List Control (N=24)5% decrease

3

3.5

4

4.5

5

5.5

6

6.5

7

1 6Training Week

VAS

Pain

Rat

ing

EFFECT CONTROL, NO EFFECT


57

COMPARISON OF EFFECT ACROSS TWO TRAINING SITES/SCANNERS

Comparison Across Training Sites (N=10/11)

3

3.5

4

4.5

5

5.5

6

6.5

7

1 6

Training Week

VAS

Pain

Rat

ing

Together Stanford Omneuron


58

CAN RTFMRI-BASED TRAINING BE USED IN SUBSTANCE ABUSE?PROTOCOL DETAIL

SELF-INDUCED CRAVINGTASK

REST 30s

RATE 20s

DECREASE 30s

REST 30s

RATE 20s

INCREASE 30s

RTFMRI TRAININGTASK

STIMULUS-INDUCEDCRAVING TASK

OVERVIEW OF DISPLAY TO SUBJECTS

INCREASE CRAVING

INCREASE CRAVING

DECREASE CRAVING

DECREASE CRAVING

GRAPH UP GRAPH DOWN

0 50 100s-101234

fMR

I BO

LD

0 50 100s-101234

fMR

I BO

LD

0 50 100s-101234

fMR

I BO

LD

0 50 100s-101234

fMR

I BO

LD

REST

REST

REST

REST

REST REST

59

SOME REFERENCES

Control over brain activation and pain learned by using real-time functional MRI.Proceedings of the National Academy of Sciences (2005) deCharms, R. C., Maeda, F., Glover, G. H., Ludlow, D., Pauly, J. M., Soneji, D., Gabrieli, J. D., and Mackey, S. C.

Learned regulation of spatially localized brain activation using real-time fMRI. NeuroImage (2004) 21, 436-443deCharms, R. C., Christoff, K., Glover, G. H., Pauly, J. M., Whitfield, S., and Gabrieli, J. D.

Functional brain imaging using a blood oxygenation sensitive steady state. Magn Reson Med (2003) 50, 675-683Miller, K. L., Hargreaves, B. A., Lee, J., Ress, D., deCharms, R. C., and Pauly, J. M.

WE ARE ACTIVELY ENROLLING CHRONIC PAIN PATIENTS FOR OUR CURRENT TRIAL

WE ARE INITIATING NEW COLLABORATIVE STUDY SITES

WE ARE ADDING RESEARCHERS TO OUR TEAM

THANK YOU…

applications of real time fmri: pain treatment and

Documents