Research

My primary research interests are Medical Image Analysis and Computer Vision. My current areas of focus are image registration, time series growth modeling, and longitudinal atlas building.

Fluid Registration and Atlas Toolkit (FRAT)

NITRC Home Page
Doxygen Documentation

FRAT provides c++ libraries and applications for performing fluid registration based operations on 2D and 3D images. The registration method is based on the large displacement diffeomorphic metric mapping (LDDMM) registration method and implements discretized fluid registration. This registration method is then applied to time series analysis, cross-sectional atlas building, and longitudinal atlas building. The individual tool components are:

Publications

An Optimal Control Approach for Deformable Registration
Gabriel L. Hart, Christopher Zach, Marc Niethammer
IEEE Computer Vision and Pattern Recognition, Workshop on Mathematical Methods in Biomedical Image Analysis, 2009.

An Optimal Control Approach for the Registration of Time Series
Marc Niethammer, Gabriel Hart, Christopher Zach
IEEE Conference on Decision and Control, 2009.

Other Projects

Conservative Registration

Growth Modeling for Cross-Sectional Atlas Building