Phase unwrapping of MR images using ΦUN - A fast and robust region growing algorithm

We present a fully automated phase unwrapping algorithm (ΦUN) which is optimized for high-resolution magnetic resonance imaging data. The algorithm is a region growing method and uses separate quality maps for seed finding and unwrapping which are retrieved from the full complex information of the data. We compared our algorithm with an established method in various phantom and in vivo data and found a very good agreement between the results of both techniques. ΦUN, however, was significantly faster at low signal to noise ratio (SNR) and data with a more complex phase topography, making it particularly suitable for applications with low SNR and high spatial resolution. ΦUN is freely available to the scientific community.