Abstract:In vivo diffusion weighted MR imaging (DWI) can probe not only microstructural but also functional information of the tissue, thus it has tremendous potentials in clinical applications and neuroscience research. Due to technical limitations, traditional DWI uses single shot acquisition method to suppress motion artifacts, which inevitably has low spatial resolution, low signal to noise ratio (SNR) and severe image distortion artifacts. In this study, a high resolution multi-shot DWI technique based on an optimized variable-density spiral (VDS) sampling trajectory and a new reconstruction method with common bi-dimensional information (multiple-direction and multiple-channel) enhancement, called CBIR, to preserve SNR and reduce scan time. The proposed CBIR method consists of two steps. First, under-sampled and complementary k-space data is collected using interleaved VDS. The complementary trajectories can be combined to form a full or over sampled k-space. Second, common information among different diffusion encoding directions is used to calculate the final images. The image without diffusion encoding (b=0) is reconstructed using nonlinear joint SENSE (JSENSE) instead of linear parallel imaging. Results show that compared to conventional parallel imaging techniques, CBIR can significantly improves image quality in terms of SNR. Temporal and spatial resolution can also be increased. The proposed CBIR method might break through the barriers of separate signal acquisition and independent image reconstruction for different direction diffusion weighted images, which can definitely improve the image speed and accuracy.

CBIR : Common Bi-dimensional Information Enhanced Reconstruction for DWI