BrainLLE performs image feature extraction based on local linear embedding as descibed in publication: Locally linear embedding (LLE) for MRI based Alzheimer's disease classification. Liu, X., Tosun, D., Weiner, M.W., and Schuff, N. NeuroImage 83, 148-157, 2013.
Brainography is a free, open source MATLAB software package for brain network graph and anatomic surface visualization based on atlas files in NIftI/ANALYZE format. Brainography is used in a variety of applications, including functional and structural connectivity analysis, and lesion- and cortical structure illustration. Brainography is written for Windows, Linux and Mac Os X with valid MATLAB licenses. Brainography has been developed in the Imaging Data Evaluation and Analytics Lab (IDEAL) of the Weill Cornell Department of Radiology.
The NMR Animations Powerpoint presentation illustrates the vector model for NMR under the action of first, the main magnetic field, and second, an applied off-resonance RF pulse from the viewpoint of the rotating frame. The animations were created by Matson with Mathematica.
Pseudo-Continuous Arterial Spin Labeling
The Powerpoint animations show the magnetizations of flowing spins under the action of three different pseudo-continuous arterial spin labeling pulses: First, the originally proposed pseudo-continuous inversion pulse. Next, the originally proposed null pulse. And finally, the improved null pulse. The pulses are described in: Improved Pseudo-Continuous Arterial Spin Labeling for Mapping Brain Perfusion. Nezamzadeh, M., Matson, G.B., Young, K., Weiner, M.W., and Schuff, N. JMagn Reson Imaging31, 1419-1427, 2010.
It has been argued that a Poisson Disk distribution is superior to a random distribution for use as a sparse sampling pattern for MRI. This Matlab program p_disk_dist.m calculates a Poisson disk distribution on a Cartesian frame.
This Matlab program spiral_vdsn.m generates a constant density spiral starting from the center of k-space out to a selected portion of k-space, and then switches over to a variable density spiral. In this scheme, only the outer regions of k-space are under-sampled.
The FourPhasePerf program is designed to obtain quantitative measures of cerebral blood flow from arterial spin labeling (ASL) MRI data. The program was written in Matlab. It calculates hemodynamic parameters based on a four phase ASL model [1-2]. In this dual-compartment distributed perfusion model, the bolus arrival time are decomposed into transit delays through large arteries and delays through smaller arteries and arterioles, expressed as arterial-arteriole transit time (aaTT), before the spin labels reach the capillary bed and perfusion into brain tissue.
- Li, K.L., et al., Four-phase single-capillary stepwise model for kinetics in arterial spin labeling MRI. Magn Reson Med, 2005. 53(3): p. 511-8.
- Liu, Y., et al., Arterial spin labeling MRI study of age and gender effects on brain perfusion hemodynamics. Magn Reson Med, 2012. 68(3): p. 912-22.
CGoct3D is a program to perform K-Bayes optimization of multi-slice ASL-perfusion datasets [1,2]. The optimization performed in this program differs from that of the papers in that it uses a faster conjugate gradients approach rather than Expectation-Maximation.
- J. Kornak, K. Young, N. Schuff, A. Du, A. A. Maudsley, and M. W. Weiner, "K-Bayes Reconstruction for Perfusion MRI I: Concepts and Application," Journal of Digital Imaging, 2008.
- J. Kornak and K. Young, "K-Bayes Reconstruction for Perfusion MRI II: Modeling and Technical Development," Journal of Digital Imaging 2008.
ValMap is a command line voxel-wise statistical analysis program for images. Images can be gray matter density, jacobian images, etc. The linear model is implemented, i.e. designs that can be modeled as Y=AB, where Y is a vector or matrix of dependent variables, B is a vector or matrix of parameters to be estimated, and A is a design matrix.
Note: This program is no longer supported by the CIND.
scalar_voxMediation is a command line tool that will test for the mediation effect, when the independent variables are scalar and the dependent variables are maps (perfusion, deformation, FA, etc.). It assumes that you have already done a voxel-wise analysis of the simple models, then computes the Sobel test for the mediation effects.
Note: This program is no longer supported by the CIND.
MATPULSE is a GUI-based program written in Matlab that enables calculation of Shinnar-LeRioux (SLR) frequency pulses, and also provides for design of RF pulses by an Optimal Control method.
MATCOIL is a GUI-based program written in Matlab that enables both single surface coils and surface coil configurations to be designed, and the resulting B1 fields and sensitivity fields to be calculated and displayed.
The MATMRI program program is written in Matlab and simulates 3D MPRAGE experiments with inclusion of magnetization transfer (MT) effects using the formulations for MT provided by Helms and Hagberg (Helms G, Hagberg GE. Pulsed saturation of the standard two-pool model for magnetization transfer. Part I: The steady state. Concepts Magn Reson. 2004 Mar;21A(1):37–49).
This program transforms a classified MRI image down to the resolution of SI data, taking account of pulse and slice profiles, spatial response functions and metabolite shift. Values are reported in terms of tissue contribution per voxel. HDF format supported.
NOTE: The following software are no longer in active use.
This program will identify the largest contiguous solid within an image, by applying erosion, threshholding, dilation and logical AND with the original threshold image. The program supports the HDF version 4.x format.
This program converts Dicom data to HDF version 4.2 format. The program calls the windows command-line version of the dicom2 program for translation of the Dicom header. All dicom files growing from the parent directory are cataloged, and combined into a single HDF volume where possible. The program uses position vectors within each Dicom slice to verify slice ordering and data completeness.
This program uses K-means cluster analysis to produce a 3 tissue classification on 3 channel MRI data. Expected inputs are a T1-Weighted image, and T2-weighted and a Pd-Weighted image. Initial cluster members are generated using histogram analysis of T1 data, and the clustering performed is identical to the SAS "Fastclus" procedure (circa 1992). HDF format supported.
This program supports a broad range of processing on image data in the HDF format, including logical operations, rigid body transformations, format translations, and many other functions. The program is intended for batch processing of large data sets, primarily with non-interactive processing.
Converts Siemens Vision system MR images (ACR-NEMA format) to HDF format.