MGA is an MRI preprocessing pipeline built with HOF (Heterogeneous Optimization Framework) methodology. MGA prepares neuro-oncology clinical imaging studies for scientific analysis such as multispectral ROI analysis, in both longitudinal and cross-sectional studies. MGA works on DICOM images from a single MRI study. Includes perfusion (DSC sequence based) analysis and DTI analysis, and spatially co-registers all study images to an atlas template and to a template image within the study. MGA is designed to reasonably minimize user configuration steps and to accept a broad range of MRI submodalities, including support for clinical image quality. The distribution is available as a stand-alone docker.io virtual appliance for 64-bit Linux environments (see detailed instructions in the installation package):
https://bitbucket.org/mmilch01/mga_docke...