Tissue microstructure estimation using a deep network inspired by a dictionary-based framework

• Published in: Medical image analysis Vol. 42; pp. 288 - 299
• Main Author: Ye, Chuyang
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.12.2017; Elsevier BV
• Subjects: Algorithms; Anisotropy; Brain; Compartments; Connectome - methods; Deep network; Diffusion; Diffusion Magnetic Resonance Imaging - methods; Diffusion MRI; Humans; Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Iterative methods; Machine Learning; Magnetic resonance imaging; Mathematical models; Maximum likelihood estimates; Microstructure; Neuroimaging; Neuroimaging - methods; NMR; NODDI; Noise reduction; Nuclear magnetic resonance; Parameter estimation; Sparse reconstruction; Studies; Tissue microstructure; NODDI; Deep network; Diffusion MRI; Sparse reconstruction; Tissue microstructure

•We have proposed a deep network for NODDI microstructure estimation.•The network structure is motivated by a dictionary-based estimation framework.•The network comprises two cascaded stages, where weights are jointly learned.•We have also extended the network to allow incorporation of neighbor info...