A Polynomial Fitting Improved Bayesian Reconstruction Method for Whole Brain Volumetric MRSI Metabolite Images

Yufang Bao1, *, Andrew Maudsley2
1 Department of Mathematics and Computer Science/Center of Defense and Homeland Security (CDHS), UNC Fayetteville State University, NC, USA
2 Department of Radiology, Miller School of Medicine, University of Miami, FL, USA

© 2013 Bao and Maudsley

open-access license: This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 International Public License (CC-BY 4.0), a copy of which is available at: This license permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

* Address correspondence to this author at the Department of Mathematics and Computer Science, UNC Fayetteville State University, Fayetteville, NC 28301; Tel: 910-672-2437(o); Fax: 910-672-1070; E-mail:


In this paper, a polynomial fitting improved Bayesian approach is proposed for the reconstruction of volumetric metabolite images from long echo time (TE) whole brain proton magnetic resonance spectroscopic imaging (MRSI) data. The proposed algorithm uses a modified EM (expectation maximization) algorithm that takes into account the partial volume effects contained inside a thick slice MRSI. It incorporates high resolution volumetric magnetic resonance imaging (MRI) as a priori information. It further integrates the polynomial fitting method to smooth out artificial edges before the high resolution metabolite images are reconstructed. Our proposed reconstruction method has successfully extended our existing reconstruction of two dimensional (2D) metabolite images to 3D cases. The experimental results show that resolution enhanced volumetric metabolite images are reconstructed.

Keywords: MRSI k-space data, volumetric metabolite images, Bayesian image reconstruction, polynomial fitting.