High-resolution 3D MRI can provide detailed anatomical information and is favorable for accurate quantitative analysis. However, due to the limited data acquisition time and other physical constraints such as breath-holding, multi-slice 2D images are often acquired. The 2D images usually have a larger slice thickness than the in-plane resolution. To reconstruct the high- resolution 3D MRI, we propose to use a super-resolution network with three orthogonal multi-slice 2D images as the input. We validated the proposed method on brain MRIs and achieved good results in terms of mean absolute difference, mean squared difference and image details with visual inspection.
This abstract and the presentation materials are available to members only; a login is required.