A new method for automatic lumen detection in intracoronary OCT images

Intracoronary optical coherence tomography (OCT) is a catheter based medical imaging technique that provide high resolution imaging of coronary lumen structures. However these images affected by catheter and stents shadows during pullback procedure. In order to overcome this problem, we present a new approach to detect the lumen boundary automatically using fuzzy system. At the first, the OCT images are mapped into the normalized polar OCT (NPOCT) space, and then primal lumen boundary is estimated by image processing methods. Afterwards, lumen boundary is detected by a fuzzy system precisely. Finally the results are remapped into the OCT image space. The proposed approach is compared with manual lumen detection (MLD), and HD and AD distance results are obtained.