Retrospective Respiratory Gating for Intravascular Ultrasound/Intravascular Optical Coherence Tomography Images

Clinically acquired free-breathing interventional cardiac images are usually polluted by cardiac and respiratory motion artifacts. ECG-gating is one of widely used strategies to suppress cardiac motion artifacts in clinic. In this study, retrospective respiratory gating is addressed to suppress respiratory motion artifacts in ECG-gated free-breathing intravascular ultrasound (IVUS) and intravascular optical coherence tomography (IV-OCT) image sequences. The dimensionality of the data matrix representing the image sequence is firstly reduced with Laplacian Eigenmaps algorithm. A signal reflecting the respiratory phases is then detected. Gating frames in the same respiratory phase are finally selected by detecting local extrema of the respiratory motion signal. The experimental results with clinically acquired image data show that the average inter-frame difference of the gated sequence decreases obviously compared with the original one. The visualization of longitudinal cuts is improved significantly. The method is easy to apply and has advantages of full-automation and low computation cost.