This repository contains all necessary functions to fit the free water elimination DTI (fwdti).

The fit procedures were based on the work proposed by Hoy et al. (2014). For more details on the implementation are reported in article.md. For an example of how to run this procedure in a real data please give a look to the ipython notebook run_realdata.ipynb. For a quantitative evaluation of the technique using monte carlo simulation please see the notebook run_sim1.ipynb (this evaluation reproduces the results of Hoy et al. (2014)). Finally in the notebook supplementary_info.ipynb, some of the details of the free water elimination model implementations are explored (for instance this notebook will show why the scipy.optimize.leastsq is used for the non-linear fwdti fit instead of using the must recent sicpy's optimeze module scipy.optimize.least_squares).

When publishing or disclose any result obtained through the use of the this procedures please include the following reference:

Neto Henriques, R. (...)

1. Neto Henriques, R. (...)

2. Hoy et al. 2014 Optimization of a free water elimination two-compartment model for diffusion tensor imaging. Neuroimage 103:323-33. doi: 10.1016/j.neuroimage.2014.09.053