Intermodal registration of CTA and IVUS-VH, and its application on CTA-based plaque composition analysis

Plaque composition analysis is a critical tool in identifying vulnerable atherosclerotic plaques. Intravascular ultrasound with spectral analysis of the backscattered radio frequency (RF) signals (IVUS-VH) is currently considered as the gold standard for the evaluation of coronary plaque composition, while CT coronary angiography (CTA) has been proposed as a potential non-invasive counterpart. In this paper, we have developed a novel intermodal registration framework for IVUS-VH and CTA, which is based on a novel composition-to-appearance transformation technique and a mutual-information-based registration algorithm. A detailed and complete intermodal registration system was presented. In addition, based on the registered images, we have developed a principal component analysis-based plaque composition classification tool, which has been tested and validated on 6 sets of IVUS-VH and CTA images. Experimental results have shown that our approach has the potential of improving the classification accuracy of plaque compositions on CTA.