Accelerated fMRI reconstruction using Matrix Completion with Sparse Recovery via Split Bregman

详细信息    查看全文

• 作者：Priya Aggarwal ; ^{priyaa@iiitd.ac.in" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Anubha Gupta ^{anubha@iiitd.ac.in" class="auth_mail" title="E-mail the corresponding author}Author Vitae
• 关键词：Accelerated functional MRI ; Nuclear norm ; l1 minimization ; Sparse recovery ; Compressed sensing
• 刊名：Neurocomputing
• 出版年：2016
• 出版时间：5 December 2016
• 年：2016
• 卷：216
• 期：Complete
• 页码：319-330
• 全文大小：6069 K

文摘

In this work, we propose a new method of accelerated functional MRI reconstruction, namely, Matrix Completion with Sparse Recovery (MCwSR). The proposed method combines low rank condition with transform domain sparsity for fMRI reconstruction and is solved using state-of-the-art Split Bregman algorithm. We compare results with state-of-the-art fMRI reconstruction algorithms. Experimental results demonstrate better performance of MCwSR method compared to the existing methods with reference to normalized mean squared error (NMSE) and other reconstruction quality metrics. In addition, the proposed method is able to preserve voxel activation maps on brain volume. None of the other existing methods is able to demonstrate this property. This shows that the proposed method is accurate and faster, and preserves the voxel activation maps that is the key to study fMRI data.