NA-MIC Brains Collaboration

Back to NA-MIC_External_Collaborations

Abstract

This project is not an NCBC collaboration grant, but instead a Continued Development and Maintenance of Software grant. The intent of this application is to update the BRAINS image analysis software developed at the University of Iowa.

Grant #

R01NS050568

Key Personnel

University of Iowa: Vincent Magnotta, PI, Hans Johnson, Jeremy Bockholt, and Nancy Andreasen. NA-MIC: Steve Pieper

Funding Duration

05/15/2007-03/31/2010

Projects

There are three main thrusts of this application as outlined below.

Implement an Automated Brain Analysis Pipeline

• BRAINS Automated Pipeline
• Support for Batch Processing
• Visual Review and Verification of Results
• Build GUI using Slicer3 Interface

Validation of Pipeline

• Evaluate MIND MCIC Human Phantom Sample Using Automated Pipeline
• Evaluate of DTI Reliability Across Sites and Platforms

Tools Developed

Here is a list of tools that are being developed as part of this project. A number of these tools have been contributed to the NITRC website. All of the tools are built upon the Slicer3 execution model and therefore will readily plug-in to Slicer3.

• BRAINSFit - Linear fitting of two 3D image sets using a mutual information similarity metric. NITRC Link
• BRAINSDemonsWarp - Non-linear registration of two image sets. This includes the diffeomorphic demons algorithm along with a number of useful enhancements. NITRC Link
• BRAINSMush - Tool to combine T1 and T2 datasets into a composite image. The resulting image is then used to perform brain extraction. NITRC Link
• BRAINSCut - Automated segmentation of subcortical and cerebellar regions of interest using an artificial neural network. NITRC Link
• BRAINSTracer - Tracing tool that uses thge vtkContourWidget for defining a region of interest manually. NITRC Link
• GTRACT - Diffusion tensor analysis and Fiber tracking tool. NITRC Link
• BRAINS Automated Pipeline - Fully automated image analysis pipeline for generation of volumetric measurements. NITRC Link

BRAINS Features

• BRAINS Gallery
• 

  BRAINS Surface showing cortical depth

• 

  BRAINS Surface showing curvature (ie sulcal and gyral labels)

• 

  BRAINS Continuous Tissue Classification

• 

  Automated segmentation of Subcortical nuclei using Neural Networks

• 

  BRAINS Image Registration using ITK Mutual Information

• 

  Lesion identification with FLAIR, T1 and T2 weighted images

• 

  Talairach lobar labels

• 

  GTRACT DTI fiber tracking

Publications

1. Powell S, Magnotta VA, Johnson H, Jammalamadaka VK, Pierson R, Andreasen NC. Registration and machine learning-based automated segmentation of subcortical and cerebellar brain structures. Neuroimage. 2008 Jan 1;39(1):238-47.