Low-rank plus sparse compressed sensing for accelerated pulmonary imaging

摘要

Introduction: Imaging of pulmonary function is of significant clinical value for diagnosis and follow-up investigations. To this end, Fourier Decomposition (FD) MRI was demonstrated to be a viable method for obtaining functional ventilation and perfusion maps during free-breathing [1]. However, this method critically depends on various acquisition parameters, including temporal resolution [2]. In a previous study, low-rank (L) plus sparse (S) matrix decomposition has been proven to be helpful to accelerate the dynamic cardiac MRI while maintaining image quality [3]. Inspired by this work, we propose a novel reconstruction approach on this idea for free-breathing pulmonary imaging to improve the scan efficiency.