AUTHOR=Hong Dan , Huang Chenxi , Yang Chenhui , Li Jianpeng , Qian Yunhan , Cai Chunting 

TITLE=FFA-DMRI: A Network Based on Feature Fusion and Attention Mechanism for Brain MRI Denoising

JOURNAL=Frontiers in Neuroscience

VOLUME=Volume 14 - 2020

YEAR=2020

URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2020.577937

DOI=10.3389/fnins.2020.577937

ISSN=1662-453X

ABSTRACT=Magnetic Resonance Imaging (MRI) is an indispensable tool in the diagnosis of brain diseases due to its painlessness and safety. Nevertheless, Rician noise is inevitably injected in the process of image acquisition, which results in poor observation and interferes with the treatment. Owing to the complexity of Rician noise, using the elimination method of Gaussian to remove noise does not perform well. Therefore, the feature fusion and attention network (FFA-DMRI) is proposed to separate noise from observed MRI. Inspired by the Attention-guided CNN network (ADNet) and Convolutional Block Attention Module (CBAM), a spatial attention mechanism is specially designed to obtain the area-of-interest in MRI. Additionally, the feature fusion block concatenates local with global information, which makes full use of the multilevel information and boosts the expressive ability of network. The comprehensive experiments on Alzheimer’s Disease Neuroimaging Initiative dataset (ADNI) have exhibited high effectiveness of FFA-DMRI. Meanwhile, it preserves the crucial brain details. Moreover, in terms of visual inspection, the denoising results are also consistent with human perception.