Automatic luminal border detection in IVUS images: A textural approach

Segmentation of arterial wall boundaries from intravascular images is very important problem for many applications in the study of plaque characteristics mechanical properties of the arterial wall its border reconstruction and its measurements such as internal and external lumen radius intimae size and media adventitia radius . we present a texture feature extraction approach for segmentation of the arterial wall from intravascular ultrasound IVUS images in polar domain. in a properly built intimaee space using luminal measuring we applied the lumen border to a gray-level cooccurrence matrices GLCM operator and used fuzzy c-means FCM clustering to classify then utilize a curve fitting functio. furthermore the result compared with physician labelling.

کلیدواژه ها:

intravascular ultrasound IVUS images ، texture segmentaion ، feature extraction ، feature calssification