Fuzzy image fusion application in detecting coronary layers in IVUS pictures

Intra Vascular UltraSound (IVUS) is a medical imaging technique, which is based on inserting an ultrasound catheter inside a vessel and producing real-time high-resolution images from the inner side of the vessel. In this paper, processing of fused IVUS frames in order to automatically detecting the coronary inner layers is provided. To reach this goal, preprocessing actions including substitution of the catheter region with the average brightness of the whole image, wavelet transform and edge-preserving smoothing are performed first. Then detecting the borders applying deformable models (distance-potential snake is used according to these images topology) has been displayed. Finally, using fuzzy integral operators for fusion in three levels including data-level, feature- level and decision-level, the mentioned process is done again and the good effect of image fusion on the IVUS frames of NIOC hospital data bank has been displayed. This method has the advantage of being fully automated, without needing the initial contour to be manually determined. The merit of this method comparing with similar methods is also its acceptable executing time, which is very important for curing patients.