Activities and Findings

1. Efficient classification methods for individual brain activation maps
   • Development of novel classification techniques for three-dimensional region data.
   • Development of efficient and effective techniques for characterizing image data and facilitating similarity queries or content-based retrieval.
   • Initial attempts to construct of a framework for evaluating many aspects of association mining methods in medical image databases, and in particular methods that can be used to discover associations between brain activations and tasks performed.
   • Investigation of Fisher Linear Discriminant (FLD) analysis as a non-threshold based tool for comparing brain activation maps.
   • Integration of structural information, including structure volume as well as structure shape and lesion data. We are currently investigating the extension of our data mining approach to include structural brain information.

2. Efficient three-dimensional shape analysis using parametric surface modeling for regions of interest in structural brain maps
   • Development of a new framework for 3D surface object classification that combines a powerful shape description method with a set of effective pattern classification, feature selection, evaluation, and visualization techniques;
   • Development of a new multi-surface alignment algorithm that extends the above framework from the single object case to the multiple object case;
   • Development of a new spherical parameterizations algorithm that makes the above framework applicable to not only voxel surfaces but also general triangle meshes.

3. Ongoing development of a secure Heart Failure Database
   • Development of an extension to the secure clinical Computer Information System (CIS) at the Dartmouth-Hitchcock Medical Center in order to provide a more convenient data entry and retrieval platform

4. Development of new registration and 3D motion modeling tools
   • Creation of tools to cancel lung motion in images of pulmonary nodules
   • Creation of spherical harmonic-based models of heart motion and application to simulating and predicting heart dynamics

5. Biweekly educational seminars on biomedical topics