With the acceptance of the manuscript "Using pulse oximetry to account for high and low frequency physiological artifacts in the BOLD signal" by myself and Vibhas Deshpande in NeuroImage, I have released an updated version of the code as used in the analysis described in the paper.  I will post the full citation when it it officially comes out.  Please cite if you use this code.

1) Modified use of the PhLEM_setup command to take care of most users demands on model generation

2) Ability to generate the RV+RRF (Birn et al. 2008) and HR+CRF (Chang et al. 2009) models.

3) Inclusion of raw, unfiltered version of pulse oximetry data.

4) Generation of simple down-sampled version of the pulse oximetry data

5) New optional loader functions, including those that work with Exvolt data

I have just posted a new version of the PhLEM Toolbox (version 1.02) which has new versions of the siemen's loader functions that should run significantly faster than the previous loops.  Most of the code changes were due to code borrowed from V. Deshpande and J. Grinstead over at Siemens.  All other code is the same as before.

There was a bug in the seimens_PPO_loader.m routine which clipped the time-series incorrectly.  This has hopefully been fixed, but until we know precisely how the Siemens WIP sequence starts and ends the streaming of this signal you should be careful when using the pulse-oximetry signal.