Transcript Velo-Cardio-Facial Syndrome - National Alliance for Medical Image

Velo-Cardio-Facial Syndrome
PRINCIPAL INVESTIGATOR:
Marek Kubicki, MD, PhD
INVESTIGATORS:
Zora Kikinis, PhD
Sylvain Bouix, PhD
Marc Niethammer, PhD
Martha Shenton, PhD
Christine Finn, MD
Raju Kucherlapati, MD
RESEARCH ASSISTANT:
Kate Smith, BA
National Alliance for Medical Image Computing
http://na-mic.org
May, 2007
Velo-Cardio-Facial Syndrome (VCFS)
• Deletion of the fragment of short arm of the 22nd
chromosome (single copies of 30 to 45 genes missing)
• Prevalence 1 in every 4000 newborns
• Clinical symptoms:
–
"velum" latin meaning soft palate
–
“kardia" greek meaning heart
–
"facial" latin having to do with the face
• Cognitive symptoms:
–
Psychomotor deficits
–
Learning and memory disabilities
–
Emotional abnormalities (flat affect and poor social interaction).
–
High incidence for schizophrenia and/or bipolar disorder in adult (30% VCFS patients
develop schizophrenia), and 4-5 genes have been suggested to be related to
schizophrenia (COMT- attention, memory, prefrontal function; RTN4R- axonal regeneration
and plasticity)
National Alliance for Medical Image Computing
http://na-mic.org
May, 2007
Aims
• Etiology of schizophrenia and
related diseases
• Prognosis of mental health diseases in
VCFS
• Early intervention
National Alliance for Medical Image Computing
http://na-mic.org
May, 2007
Project
• Subject recruitment
• Psychological interview
• DNA analysis, genotyping of the
22q11.2 region
• Brain imaging (MRI, DTI and fMRI)
• Analysis of imaging data, and genetic
correlations
National Alliance for Medical Image Computing
http://na-mic.org
May, 2007
Imaging Data Available
Two stages:
• Available now:
7 VCFS and 7 matched control cases scanned on 1.5 T magnet
(DTI and structural scans)
15 schizophrenics and 15 control cases scanned on 3T magnet
(DTI and fMRI) plus 1.5T chronic schizophrenia data (DTI and
structural scans)
• Available by the end of the year:
15 VCFS and 15 matched control cases scanned on 3T magnet
(DTI, structural scans and fMRI)
National Alliance for Medical Image Computing
http://na-mic.org
May, 2007
Hypotheses
Regions that we want to study with MRI
-
DLPC (executive function, memory)
-
Orbital Frontal Gyrus (emotion)
-
Cingulate Gyrus (attention, emotion)
-
Hippocampus (memory, learning)
Tracts that we want to study with DTI (frontal-temporal connections):
-
Fornix (memory)
-
Arcuate Fasciculus (language)
-
Cingulum Bundle (attention)
-
Uncinate Fasciculus (emotion, affective flattening)
Networks that want to study with fMRI :
-
Memory, attention, emotion, language (semantic processing)
National Alliance for Medical Image Computing
http://na-mic.org
May, 2007
Specific Requirements
- Automatic DLPC segmentation.
- Automatic segmentation of other structures (orbital
frontal cortex, cingulate gyrus, hippocampus).
- Tools for measuring anatomical connectivity between
these regions (optimal path analysis, stochastic
tractography).
- “Default network” and “rest” fMRI analysis.
- Anatomical (DTI), and functional (fMRI) connectivity
analysis.
National Alliance for Medical Image Computing
http://na-mic.org
May, 2007
Progress So Far
-
Automatic DLPC segmentation.
- John and Brad finished the slicer2 module, its being tested and used for first
episode schizophrenia study now.
-
Automatic segmentation of other structures (orbital frontal cortex, cingulate gyrus, hippocampus).
- We have limited experience with freesurfer and cingulate, have not tried
Killian’s segmentation for anything other than STG.
-
Tools for measuring anatomical connectivity between these regions (optimal path analysis, stochastic
tractography).
- Optimal Path Analysis- we used it, but its not yet in the slicer, there might be
some problems with it. Same with Stochastic tractography.
-
“Default network” fMRI analysis, functional connectivity analysis.
- Polina and Sandy might have some tools, but we have not tried them yet.
National Alliance for Medical Image Computing
http://na-mic.org
May, 2007
Potential Challenges for
Segmentation/Registration
•
Different brain shape
•
Thicker skull
•
Brain atrophy
•
Congenital abnormalities
•
Brain asymmetry
•
White matter lesions
National Alliance for Medical Image Computing
http://na-mic.org
May, 2007
Segmentation/Registration
Beneficial algorithms/software:
• Automatic segmentation procedures tailored for
VCFS (atlas?).
• Automatic parameterization of segmentation algorithm
(e.g., optimal parameter selection for EM segmenter
based on training data).
• Robust and easy to use registration.
Capabilities for batch processing.
National Alliance for Medical Image Computing
http://na-mic.org
May, 2007
VCFS Schizophrenia Candidate Genes
• COMT (controls dopamine degradation in prefrontal
cortex, related to attention and memory)
• RTN4R (also known as Nogo 66, related to axonal
regeneration and plasticity)
• PRODH
• ZDH8
• SNAP29
• TBX1
National Alliance for Medical Image Computing
http://na-mic.org
May, 2007