3D cardiac cine reconstruction from free-breathing 2D real-time image acquisitions using iterative motion correction

MR imaging is well suited for diagnosis, treatment and study of various cardiac diseases affecting the functionality and morphology of the heart. MR imaging provides good tissue contrast and can achieve high spatial and temporal resolution. Most current MR acquisition methods require breath-holds during the acquisition or employ respiratory gating to avoid image corruption caused by respiratory motion. Also cardiac gating is applied to achieve time resolved sampling for functional analysis. Breath-holds can be difficult for patients and gating methods can be undermined by irregular motion patterns. Real-time imaging offers a potential solution to both these issues, but poses its own challenges. We present initial results for a reconstruction pipeline that takes multiple stacks of 2D real-time, short-axis images acquired during free-breathing and computes the respiratory deformations to reconstruct a coherent 3D+t volume of the beating heart.