View Slides - Arlington Technology Association

Report
Using Near-Infrared Light to
Image how the Brain Rewires
after Injury and Rehabilitation
George Alexandrakis
Department of Bioengineering
University of Texas at Arlington
Functional Near-Infrared Spectroscopy
(fNIRS)
 Near infrared (NIR) light (750-900
nm) can penetrate the human
scalp/skull, reach the brain, and its
absorbed by the oxygenated
hemoglobin (HbO) and deoxygenated hemoglobin (HbR) in the
tissue.
 fNIRS can be thought of as the
optical analogue of fMRI for imaging
the cerebral cortex.
Physiological Basis of the fNIRS Signal:
Changes in Brain Oxygenation
• The neuronal firing occurring when the brain does work results into
an oversupply of oxygen in areas that doe the work – this is known
as neurovascular coupling
• The increase in cerebral blood oxygenation during functional
activation is mostly due to an increase in blood flow velocity, and
occurs without a significant swelling of the blood vessels.
Functional Near-Infrared Spectroscopy
(fNIRS) Imaging as an alternate
technology for imaging brain function
Main body of the CW-5 system
Geometry of CW-5 probe on the motor
cortex (circle: light source; square:
detector).
What does an fNIRS measurement look like?
Tapping
∆HbO
Full time-series
100
Average time-series
200
seconds
300
400
Source-detector
geometry on head
∆HbO
fNIRS Image
0
10
20
30
seconds
40
What do fNIRS images look like?
Time series (5sec/frame)
Patient with Right Hemiparesis
Right Finger Tapping
Block average (30 sec)
Don’t people usually lie inside
fMRIa tube to map their brains?
Functional Magnetic Resonance Imaging (fMRI)
An example of the information one gets from fMRI:
A map of change in deoxygenated blood during a specified activity
Brain activity is superimposed onto an anatomical map of the brain in 3D
Vandermeeren, et al 2003
Limitations of fMRI
• Extremely rigid stabilization of
the head
• High magnetic fields and rapid RF
gradients
• High intensity acoustic
environment
• Highly propensity for movement
artifacts
• 50% of normal 5 yr-old can have a
successful fMRI
Wilke, et al 2003
Validation of fNIRS Imaging with fMRI
(a)
HbO
0 o1
o9
o8
o7
o6
o5
o4
o3
o2
HbR
o10 o11 o12 o13 o14
-2
0 o1
o9
o8
o7
o6
o5
o4
o3
o2
HbT
o10 o11 o12 o13 o14
-2
x1
x2
x3
x4
x5
0 o1
o9
o8
o7
o6
o5
o4
o3
o2
o10 o11 o12 o13 o14
-2
x1
x2
x3
x4
x5
x1
x2
x3
x4
x5
-4
-4
-4
-6 o15 o16 o17 o18 o19 o20
0
5
10
-6 o15 o16 o17 o18 o19 o20
0
5
10
-6 o15 o16 o17 o18 o19 o20
0
5
10
X3
X2
(b)
X4
X2
O16
O16
O17
O18
-1
0
1
2O19
-5
-1
0
1
2
-1
-5
0
1
2
-5
x 10
x 10
x 10
(a) fNIR and (b) fMRI images, showing good spatial correlation during a 15-second finger tapping task. The data were averaged over 10 blocks.
Specifically, (a) shows HbO, HbR, and HbT activation maps that are reconstructed from fNIR measurement: x represents sources and o represents
detectors; (b) shows axial (left panel), coronal (middle), and sagittal (right) views of fMRI images during finger tapping.
Limitations:
• Can only see the cortical surface of the brain but not deeper
• Lower spatial resolution (~1 cm)
In what clinical applications could using
fNIRS have an advantage over fMRI?
“…So far, fNIRS has been promoted in a number of fields in which fMRI is
limited due to the constraints induced by the scanning environment and the
experimental measurements take place in a more comfortable and natural
environment… Only these devices can be utilized for example in infant and
children developmental studies, in neuro-rehabilitation assessment, and in
simultaneous brain activation studies on multiple subjects.”
M. Ferrari, V. Quaresima / NeuroImage 63 (2012) 921–935
FNIRS applications that I have pursued so far:
• Monitoring brain plasticity during rehabilitation in children with cerebral palsy (CP)
• Monitoring brain activation while playing Wii games
• Continuous bedside monitoring of brain health in traumatic brain injury (TBI)
• Real-time guidance for treatment optimization (electrical stimulation) in stroke
Cerebral Palsy (CP)
• What is Cerebral Palsy?
– Lesion or stroke in the cortex or in the
subcortex caused before or during birth.
– Results in damage to the motor control
areas of the brain which leads to
restricted movements.
– Affects 1 in 500 children
• Currently Used Functional
Assessments
– Physical Assessments
• Measures refined motion
• Measures ability to perform daily activities
– EEG measures neuronal activity
– fMRI measures hemodynamic activity
fNIRS for Cerebral Palsy
(b)
(a)
Left Hemiparesis
(c)
Cortical
Subject 1
R
L
(d)
(e)
Right Hemiparesis
(f)
Subcortical
R
Subject 2
L
(g)
(h)
Right Hemiparesis
(i)
Subcortical
R
Healthy, Control
L
Subject 3
Subject with CP (note lesion) recruits secondary
cortical regions for finger tapping
Texas Scottish Rite Hospital for Children
Pirate Camps 2012-2013
Constraint Induced Movement Therapy (CIMT)
Group Activity, Outside Activities, Craft, Fine Motor Stations, Gross Motor Stations,
Large Group Game, Personal Goals (tying laces, holding a fork, buttoning a shirt)
2 weeks, 6 hours per day
Correlation between improved hand
use and cortical activation laterality
R2 = 0.94
P < 0.01
- Increased contralateral cortical activity after CIMT (Laterality index improves)
- Improvement in uni-manual hand use after CIMT (Melbourne test benefit)
- Increase laterality and hand use significantly correlate (p < 0.01)
fNIRS detected a relapse at six moths post-therapy
Healthy Children
n=6
Before
Rehabilitation
Immediately after
Rehabilitation
Children with CP
n=6
Six months after
Rehabilitation
• Less impaired subjects did not benefit as much from CIMT, as proven by the non-significant changes in their clinical
scores post-therapy.
• There were no children with CP in this study that significantly improved both their unimanual and bimanual abilities after
CIMT while also normalizing their sensorimotor activation patterns, as measured by fNIRS.
• The laterality index and the resting-state functional connectivity appeared to be normalized, compared to the
corresponding patterns seen in the controls group, immediately post-CIMT but then relapsed at the six-month time
point. On the other hand, the blood flow metric remained at near-normal values even at the six-month time point.
• ‘Booster’ treatments would be needed to maintain long term benefits.
Measuring Brain Activity While Playing
Nintendo Wii and I-Pad
Probe Configuration
L
Right Handed
Movements
R
PMC
PMC
M1
M1
S1
SMA
S1
PMC
M1
PPC
PPC
PPC
SMA
S1
M1
S1
PPC
Activation Pattern Changes
Time Signals
Baseball
(Wii)
Boxing
(Wii)
Guitar Hero
(I-pad)
Activation Maps
PMC
M1
S1
SMA
PPC
PMC
M1
S1
PPC
SMA
PPC
PMC
M1
S1
PPC
PMC
M1
S1
PMC
M1
S1
PPC
SMA
PMC
M1
S1
PPC
Transcranial Direct Current Stimulation
(tDCS)
• tDCS is a form of neurostimulation which delivers
a constant, low current to the brain by two 5 x 5
cm2 electrodes (anode and cathode)
• tDCS is potentially helpful in treating a wide
range of physiological disorders
–
–
–
–
–
–
Depression
Stroke
Alzheimer’s
Chronic Pain
Epilepsy
Parkinson’s Disease
How Does tDCS Work?
• tDCS stimulation modulates brain function by
changing the neuron membrane polarization.
• A change in the membrane polarization leads
to changes in neuron, synaptic, and network
activity.
• The membrane polarization will either
increase or decrease depending on the
current direction.
Two Types of tDCS Stimulation
• Anodal Stimulation
Cathode
Anode
– The anode is placed over
the area being excited.
• Cathodal Stimulation
– The cathode is placed
over the area being
suppressed.
• The direction of the
current goes from the
Anode to the Cathode.
Cathode
Anode
Detectors
Sources
Hand Device
CW-6
Data Box
Instrumentation Setup
EMG Box
tDCS Box
Protocol
Move dotted white line into
the white box, and keep it
there for 1 s. Then rest for
~12 s. Repeat this 20 times.
Hand device
FNIRS Probe Configuration
L
R
PMC
M1
S1
PPC
PMC
SMA
M1
S1
PPC
tDCS Electrode
fNIRS Source Location
fNIRS Detector Location
Suppression of critical motor control centers
creates transient ‘virtual’ brain lesions
Excited
Excited
Suppressed
Before
Suppressed
During
tDCS Anode Electrode
tDCS Cathode Electrode
fNIRS Source Location
fNIRS Detector Location
After
Indicated ROI
Exciting the cortical region
in use, increased the
activity in that hemisphere
Suppressing the cortical
region in use, resulted in
activity in both hemispheres.
Changes in arm muscle excitability and wrist flexion performance after tDCS
Suppressed
Excited
WE
Excited
Suppressed
WE
WF
WF
The perturbation tDCS (ptDCS) concept: A novel, fast method to
derive a personalized tDCS electrode arrangement that yields
maximal improvements in upper extremity performance
ptDCS Protocol
Typical tDCS protocol uses a current of 1 – 2mA for a
period of 20 minutes. Effects can last up to a week.
ptDCS applies current at 0.5 mA for 40 s. This reduced
the effects to ~3 minutes.
Schematic of the 20 different electrode montages
measured during a ptDCS protocol. Red squares
indicate anodal (excitation) and black squares
cathodal (suppression) stimulation. All were done
in a single ~2 hour session.
Shortening tDCS Effects
tDCS
tDCS +
Flexion
• tDCS applied with current of 0.5 mA for 40 s
• ΔHbO returned to pre-tDCS baseline levels within 3 minutes
• Allows 20 stimulations to be applied within a 2 hour period.
Arm Performance Metrics
• Reaction Time (RT): How quickly one responds to the stimulus.
• Error: How inaccurate a person puts the cursor into the target.
• Error x RT product: Measure of overall performance.
Important points found with ptDCS-guided treatment
so far:
• Current clinical scales are not specific enough in
reflecting targeted rehabilitation improvements as
they add apples to oranges
• General Systems Performance Theory (GSPT) is
proposed as a physiology-driven way to derive
quantitative performance metrics (Dr. G. Kondraske)
GSPT-based resource demand function
(RDF) provides quantitative insight into
the relationship between EMG and task
performance (reaction speed). The
RDF defines a minimum level of EMG
activation required to achieve a given
maximum reaction speed (green curve).
Inter-Subject Variability in Healthy Subjects
The typically studied tDCS montages (A and M) do
not help improve any healthy subjects.
Two tDCS montages (C and K) cause 3 healthy
subjects to worsen (p < 0.01).
A never before used dual PMC montage (O)
presented significant (p < 0.01) improvement in all
healthy subjects.
Identified Optimal tDCS Electrode Arrangement in
Stroke Patients is Better than dtDCS
Anode
Cathode
TBI patient population
• 5 TBI patients were recruited for this study
– Glasgow coma score (GSC)
• GCS 6 (1 patient)
• GCS 13-14 (4 patients)
– Age: 53.8 ± 19.6 years
– Gender:
• 3 male
• 2 female
Fluctuation amplitude
Healthy Subject (GCS 15)
Patient 4 (GCS 6)
Amplitude
Patient 1 (GCS 14)
CT Images
L
Indicate lesions or
subdural/subarachnoid
hemorrhages
R
L
R
Resting-state connectivity of healthy subjects
Component 2
24 %
19 %
Component 4
Healthy 2 (GCS 15)
14 %
Component 3
Component 1
Healthy 1 (GCS 15)
17 %
Resting-state connectivity of TBI patients
GCS 15 (Healthy)
GCS 14
GCS 6
18 %
39 %
75 %
17 %
28 %
12 %
Indicate lesions or subdural/subarachnoid hemorrhages
Resting-state connectivity with respect to injury location
CT Images
Connectivity
Patient 1 (GCS 14)
Patient 3 (GCS 14)
Patient 2 (GCS 14)
Collaborators
UT Arlington
Dr. Hanli Liu (BME)
Dr. Fillia Makedon (CSE)
Dr. Fenghua Tian (BME)
UT Southwestern Medical
Center at Dallas
Dr. Timea Hodics (Neurology)
Dr. Anne Stowe (Neurology)
Current Graduate Students
Mr. Bilal Khan (UT Arlington)
Mr. Nathan Hervey (UT Arlington)
Mrs. Jianwei Cao (UT Arlington)
UT Dallas
Dr. Duncan MacFarlane (EE)
Texas Scottish Rite Hospital for
Children
Dr. Mauricio Delgado (Pediatric
Neurology)
Past Graduate Students
Ms. Ankita Chainani (UT Arlington)
Ms. Laura Shagman (UT Dallas)
Industry
Dr. Chester Wildey (MRRA Inc)
Mr. Robert Francis (Raytheon)
Grants
Current
NIH/NIBIB 1R01EB013313-01 (Alexandrakis), Near Infrared Brain Imaging for Guiding Treatment in Children with
Cerebral Palsy
UTA – UTD – THR Collaborative Research Program (Alexandrakis), A Near-Infrared Brain Imaging System for the
Continual Bedside Monitoring of Hemorrhagic Progression of Contusions after Traumatic Brain Injury
Recently Completed
UT Arlington – UT Dallas & Texas Health Resource Collaborative Research Program (Alexandrakis), A Near-Infrared Brain Imaging
System for the Continuous Bedside Monitoring of Intracranial Pressure Buildup in Patients with Traumatic Brain Injury
NSF/CPS 1125441 (Makedon), A Novel Human Centric CPS to Improve Motor/Cognitive Assessment and Enable Adaptive
Rehabilitation
UT Arlington – UT Dallas & Texas Health Resource Collaborative Research Program (Alexandrakis), A Breakthrough Probe Technology
for Translating Near-Infrared Brain Imaging into a Routine Clinical Tool for Assessing Motor Deficits in Children with Cerebral Palsy
United Cerebral Palsy Research & Educational Foundation (Delgado), Assessment of Neuroplasticity in Children with CP: A MultiSource-Detector Near Infrared Spectroscopy Imaging Study





Recent Publications
Hodics T*, Khan B*, Hervey N, Stowe A, Kondraske GV, Alexandrakis G. Functional near-infrared spectroscopy
maps cortical plasticity underlying altered motor performance induced by transcranial direct current
stimulation. J Biomed Opt. (Accepted October 2, 2013; *equal first authorship).
B. Khan, C. Wildey, R. Francis, F. Tian, M.I. Romero, M.R. Delgado, N.J. Clegg, L. Smith, H. Liu, D.L. MacFarlane,
G. Alexandrakis, “Improving optical contact for functional near infrared brain imaging with brush optodes,”
Biomed. Opt. Express, 3(5):878–898 (2012).
B. Khan, P. Chand, G. Alexandrakis, “Spatiotemporal relations of primary sensorimotor and secondary motor
activation patterns mapped by fNIR imaging,” Biomed. Opt. Express, 2(12):3367–3386 (2011).
F. Tian, M.R. Delgado, S.C. Dhamne, B. Khan, G. Alexandrakis, M.I. Romero, L. Smith, D. Reid, N.J. Clegg, H. Liu,
“Quantification of Functional Near Infrared Spectroscopy to Assess Cortical Reorganization in Children with
Cerebral Palsy,” Opt. Express 18(25):25973-25986, (2010).
B. Khan, F. Tian, K. Behbehani, M.I. Romero-Ortega, M.R. Delgado, N. Clegg, L. Smith, D. Reid, H. Liu, G.
Alexandrakis, “Identification of abnormal motor cortex activation patterns in children with cerebral palsy by
functional near infrared spectroscopy,” J. Biomed. Opt. 15(3):036008-1 – 14, (2010).
Publicity
http://www.sciencecodex.com/the_hair_brush_that_reads_your_mind
http://news.cnet.com/8301-27083_3-20020193-247.html (High-tech hair brush improves
optical brain scans)

similar documents