Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Transcript of multimodality
imaging modalities: TBI*
Why multiple modalities?
Medical imaging analytics @ PICSL
#1 in 3 independently
evaluated anatomical mapping challenges
Murphy 2011 CT/Lung
Multi-atlas challenge 2012
ANTs is optimized
for integrating modalities
by spatial alignment
Traumatic Brain Injury:
processing speed deficits
Multivariate Characterization of Thalamus in TBI
volume of each sub-region of the thalamus ( T1 )
fractional anisotropy in each region ( DTI )
Cerebral blood flow (ASL)
Processing speed (psychometrics)
The 3429 sessions are from
32 different international sites
13 years of data collection, for both 1.5T and 3.0T data, 3 different scanner manufacturers, and
2/3 of these scans are HD subjects, 1/3 are normal controls.
Hongzhi Wang, Paul Yushkevich
Thalamus Related White Matter
Integrity & Cortical thickness in TBI
i.e. processing speed
Identify relationships between thalamus & other modalities (via SCCAN) to reduce dimensionality &
identify a network predictive of processing speed.
Prof Hans I Johnson
University of Iowa
related to TBI
Within TBI sub-group, what
predicts variation in processing speed?
Employ a TBI-specific hypothesis to filter a large multimodality dataset down to a manageable and interpretable scale (
<50 tests total
Quantified thalamus variation with 3 modalities.
Show thalamic projections and thalamo-cortical regions affected by TBI.
Within TBI subjects, we further identify a cortico-connective network that is related to processing speed.
Framework is extensible to numerous related analysis problems.
N=40, 22 TBI
matched age, gender
mean age 30+/-10
educ = 14y +/- 2
* joint work with
Junghoon Kim of
Moss Rehab & Ben Kandel,
Jeff Duda of PICSL
Penn Image Computing
& Science Laboratory
UPENN Dept of Radiology
BrainMeasure ~ 1 + Age + Gender + Education + TBI
all 40 subjects
ProcessingSpeed ~ Age + Gender + Education + Duration + Brain
Supported by: 5R01EB006266-04, P01-AG032953-02,
White matter integrity
TBI ~ cortical thickness
GM & WM Eigenanatomy
SCCAN uses thalamus variation to extract white matter regions.
Also use thalamus variation to extract cortical regions.
SCCAN uses thalamus variation to extract white matter regions related to TBI.
q < 0.05