The pymidas package is an integrated, open source, open development port of the MIDAS (Metabolite Imaging and Data Analysis System) software package, which was initially programmed in IDL. It contains multiple applications that can be combined into a pipeline, including:

1. importer - beta - code to import Siemens and GE data (MRIs and EPSI) into MIDAS projects.
2. volumizer - beta - code to volumize Siemens and GE data (MRIs and EPSI) into MIDAS format.
3. common - code used by two or more applications in PyMidas. Includes the libxml.py module used to access MIDAS XML files.
4. midas_browser - beta - GUI based selection tool for picking MIDAS Subject.xml nodes on which to act.
5. compare - beta - simple GUI based tool for picking/comparing data in two binary files.
6. epsi2 - alpha - porting the IDL epsi2 code to Python.
7. fdft - alpha - code to perfrom FFT and other processing steps
8. lite - alpha - code to remove lipid contamination via Papoulis-Gerschberg algorithm
9. util - alpha - useful code and utilities for running/testing code

All (some) modules have built in examples for running/using the code. Though the test data may not exist in your repository ... sorry. Typically just type:

The MIDAS project provides an integrated set of MRI and MRSI processing functions. The PyMIDAS software supports a "whole-brain" MRSI acquisition method that has been implemented on MRI systems from three major manufacturers.

This effort is based on code initially funded by NIH grant 1R01EB008387-01A1 that became the Vespa project. Other funding sources include NIH grants: U01CA264039, and many others

We've had a lot of people support this effort, but I'd like to mention Philip Semanchuk (http://pyspoken.com), Dr. Jeffrey Alger and Dr. Alexander Giuffrida for their early contributions.

This project is currently licensed BSD 3-Clause License, see LICENSE.md file for details

• Brian J. Soher - Principal Investigator - Duke Radiology

Many. To be included later.

• Lots of online ('google') searches have inspired this work. We have tried to list these in each module as relevant.