Abstract

q-Ball imaging (QBI) is a high-angular-resolution diffusion imaging (HARDI) method that is capable of resolving complex, subvoxel white matter (WM) architecture. QBI requires time-intensive sampling of the diffusion signal and large diffusion wavevectors. Here we describe a reconstruction scheme for QBI, termed multiple wavevector fusion (MWF), that substantially boosts the sampling efficiency and signal-to-noise ratio (SNR) of QBI. The MWF reconstruction operates by nonlinearly fusing the diffusion signal from separate low and high wavevector acquisitions. The combination of wavevectors provides the benefits of the high SNR of the low wavevector signal and the high angular contrast-to-noise ratio (CNR) and peak separation of the high wavevector signal. The MWF procedure provides a framework for combining diffusion tensor imaging (DTI) and QBI. Numerical simulations show that MWF of DTI and QBI provides a more accurate estimate of the diffusion orientation distribution function (ODF) than QBI alone. The accuracy improvement can be translated into an efficiency gain of 274-377%. An intravoxel peak connectivity metric (IPCM) is presented that calculates the peak connectivity between an ODF and its neighboring voxels. In human WM, MWF reveals more detailed WM architecture than QBI as measured by the IPCM for all sampling schemes presented.