DoDTI has been designed to provide a tool for quantification of diffusion tensor imaging including visualization and analyzing techniques of DT-MRI. Diffusion tensor imaginge techniques in physics and computer science is a rapidly growing field. The application to clinical research is active but has been hampered by the lack of good tools for analyzing DT-MRI data. For this reason, we introduce an easy to use diffusion tensor analysis tool designed to meet these needs in both clinical research and technical research. We will continue to incorporate new features into DoDTI. Currently this website is still being constructed. We will provide more detailed information on DoDTI as soon as possible.
The first version of DoDTI was originally developed by Dr. Hae-Jeong Park in collaboration with Dr. Carl-Fredrik Westin (Dept. Radiology, Harvard Medical School) in the technical aspect and Dr. Martha E. Shenton (Dept. Psychiatry, Harvard Medical School) in the clincal and neuroscientific aspect. Now it is a continuing project between Laboratory of Molecular Neuroimaging, Yonsei Univerisity College of Medicine (Director, Dr. Park) and Laboratory of Mathmatics in Imaging, Brigham and Women's Hospital, Harvard Medical School (Director Dr. Westin).
Among many collaborators involved in this project, we would like to especially appreicate the following collabortors.
Professor Robert W. McCarley (Head of Dept. Psychiatry, Harvard Medical School, Deputy Chief of Staff, Brockton VAMC)
Professor Jong-Doo Lee (Head of Dept. Nuclear Medicine, Yonsei University, College of Medicine)
Professor Ron Kikinis (Director of Surgical Planning Lab. Brigham and Women's Hospital, Harvard Medical School)
Dr. Marek Kubicki (Dept. Psychiatry, Harvard Medical School)
Anders Brun (Dept. Biomedical engineering, Linko:ping Univ. Sweden)
Dr. Seung-Koo Lee (Dept. Radiology, Yonsei University, College of Medicine)
DoDTI is a platform independent tool implemented in Matlab (Mathworks Inc.), and is modular to facilitate maintenance and updates.
The basic features of the tool are:
1) 2D DT-MRI visualization where several types of glyphs are available to display with different coloring schemes with a background of the fractional anisotropy and the apparent diffusion coefficient.
2) A fully automated method for DT-MRI tractography where seed-points and stopping criteria can be assigned either from automated method or in an user-configurable manner. For the tractography, a 4th order Runge-kutta integration solver is used with a choice of several regularization schemes including TENS to better handle crossing fiber bundles.
3) Selection of fiber traces according to a user-defined criteria such as the length of a fiber, maximal angle difference, minimum distance between end points of a fiber. In addition, manually drawn regions of interest (ROIs) can be used for the selection of fibers of interest intersecting those ROIs.
4) Information about the properties of fibers of interest is available, such as the mean anisotropy along the fiber. Integration of functional MRI activation map in SPM99 (Wellcome Department of Cognitive Neurology, UK) format can be used as ROIs to enable the exploration of functional integration in relation with anatomical connectivity. Measuring fractional anisotropy and cross-sectional area of fiber bundles along the pathway is also possible.
5) The 3D visualization techniques include both streamlines and streamtubes with or without background grayscale images and renderings of ROIs for better understanding of fiber connecitivity and efficiency. In addition, individual fibers can be displayed with color-coded fractional anisotropy.
Current version of DoDTI is v. 1.0.
Hae-Jeong Park, Ph.D.
email: parkhj@yumc.yonsei.ac.kr
Laboratory
of Molecular Neuroimaging Technology
Dept. Diagnostic Radiology, Yonsei University, College of
Medicine, KOREA