Automated stent strut coverage and apposition analysis of in-vivo intra coronary Optical Coherence Tomography images

Several studies have proven that intra-vascular OCT is an appropriate imaging modality able to evaluate stent strut apposition and coverage in coronary arteries. Currently image processing is performed manually resulting in a very time consuming and labor intensive procedure. We propose an algorithm for fully automatic individual stent strut apposition and coverage analysis in coronary arteries. The vessel lumen and stent strut are automatically detected and segmented through analysis of the intensity profiles of the A-scan lines. From these data, apposition and coverage can then be estimated automatically. The algorithm was validated using manual measurement (performed by two trained cardiologists) as a reference. 108 images were taken at random from in-vivo pullbacks from 9 different patient presenting 'real-life' situations (i.e. blood residual, small luminal objects and artifacts). High Pearson's correlation coefficients were found (R = 0.96 - 0.95) between the automated and manual measurements while Bland-Altman statistics showed no significant bias with good limits of agreement. As such, it was shown that the presented algorithm provides a robust and a fast tool to automatically estimate apposition and coverage of stent struts in in-vivo pullbacks. This will be important for the integration of this technology in clinical routine and large clinical trials.