Velocardiofacial Syndrome as a Genetic Model for Schizophrenia

NA-MICNational Alliance for Medical Image Computing http://na-mic.org

Velocardiofacial Syndrome as a Genetic Model for Schizophrenia

Marek KubickiDBP2, Brigham and Women’s Hospital, Harvard Medical School

National Alliance for Medical Image Computing http://na-mic.org

Team

HARVARD

• Marek Kubicki, Sylvain Bouix, • Katharina Quintus, Tri Ngo, Julien Siebenthall,• Doug Markant, Usman Khan, Doug Terry and Andrew Raush

NAMIC

• Polina Gollard MIT• Brad Davis Kitware• Mahnaz Maddah, Jim Miller GE• Steve Piper, CF Westin, John Melanakos, others

National Alliance for Medical Image Computing http://na-mic.org

NAMIC Roadmap Project: Stochastic Tractography

Goals:

1. To develop the software for stochastic tractography2. To apply the software to clinical population3. To present results of the study at international forum,

popularize it and invite collaborations4. Use the feedback to improve and develop additional tools5. To apply for NIH funding

National Alliance for Medical Image Computing http://na-mic.org

To Develop Slicer Module for Stochastic Tractography

1. VTK slicer version released in 2008. 2. Improved Python version released in January 2009. 3. Updated version released in October of 2009.

National Alliance for Medical Image Computing http://na-mic.org

To Apply Software to Clinical Population

1. Few abstracts have been presented at international symposiums (HBM, Biological Psychiatry, ACNP, WPA). 2. One original paper written up, and has been submitted to Biological Psychiatry (Dorsal and Ventral Semantic

Processing Stream and their Abnormalities in Schizophrenia)3. Another three projects that use stochastic tractography are in progress (WM connectivity within the networks defined

by resting state fMRI in schizophrenia; Connectivity within the emotional network in schizophrenia; and fMRI/DTI study of semantic associates).

National Alliance for Medical Image Computing http://na-mic.org

To Popularize Software and to Invite Collaborations

1. Software has been applied to clinical studies at several institutions in US (MGH, MIT, University of Utah)

2. We have collaborative project with Max Plank Institute, where we analyze connectivity within the default network.

Installation/Training/Dissemination:1. Adriana Sampaio - University of Minho - Portugal2. Aristotle Voineskos - University of Toronto - Canada3. Kang Uk Lee - Kangwon National University Hospital - Korea4. Gudrun Rosenberger - Innsbruck Medical University - Austria5. Motoaki Nakamura - Yokohama City University of Medicine -

Japan

National Alliance for Medical Image Computing http://na-mic.org

To Improve the Software, and to Develop Additional Needed Tools

1. Label Statistics  (Quintus)2. ResampleDTI (Budin)3. ResampleScalar/Vector/DWI Volume (Budin)4. Rician LMMSE image filter (Niethammer)5. Python Convert Fiducials to Labelmap (Siebenthal)6. Python Convert Volume to NUMPY (Siebenthal)7. Python Load Volume from NUMPY File (Siebenthal)In progress:1. filtered two tensor tractography (Malcolm/Rathi)2. dispersion/curving of tensor fields (Savadjiev)3. CUDA enabled affine registration (Malcolm/Rathi)4. Initialization of EMSegmentation using k-means for brain parcellation (Srinivasan/Bouix)5. Software designed with/for the PNL:6. ACPC transform (Aucoin/Bouix)7. EMSegment  (Davis/Pohl/Bouix)8. Diffusion Editor in module tab (Kirsten?/Bouix)Software extensively benchmarked/tested/bugtracked by PNL:1. Every registration module (Rausch/Bouix)2. Most diffusion and tractography modules3. DICOM to nrrd converter (Bouix)4. Mahnaz's clustering tractography slicer module.

National Alliance for Medical Image Computing http://na-mic.org

To Apply for NIH Funding

Collaboration between NAMIC (Jim Miller , Mahnaz Maddah at GE, CF Westin at BHW) and PNL (Marek Kubicki, Peter Savadjiev, Martha Shenton)- Collaborative R01 “White Matter Imaging and Analysis in Schizophrenia” (to be submitted February 5th).

Specific Aims:To map the anatomy of four distinctive fiber bundles that are part of the semantic processing stream:

Uncinate Fasciculus (UF), Inferior Occipito-Frontal Fasciculus (IOFF), Inferior Longitudinal Fasciculus (ILF), and Arcuate Fasciculus (AF).

To map and to delineate further the localization, extent, nature, and temporal dynamics of WM abnormalities related to myelin abnormalities, tract architecture, and inflammation in patients with SZ. We will use Cross Relaxation Imaging (CRI), sensitive to myelin content (Underhill et al., 2009), DTI measures of curvature and dispersion indices (CI&DI- Savadjijev et al., 2009), both sensitive to tract architecture, and free water (FW- Pasternak et al., 2009), sensitive to Wm swelling. WM measures will be combined with GM volumes, cortical thickness and cortical complexity measures of fronto-temporal regions, and clinical/medication data will be collected and analyzed.

To develop, test, and disseminate robust tool for WM Analysis. We will develop one freely available software module (for 3D Slicer; http://3dslicer.org) that will enable WM tract extraction and will include tools for statistical group comparison where data from any imaging modality (i.e., DTI, CRI, CI, DI, FW, etc.) can be combined and analyzed along the WM bundles.

National Alliance for Medical Image Computing http://na-mic.org

White Matter Imaging and Analysis in Schizophrenia