Segmentation of carotid artery in ultrasound images: Method development and evaluation technique

Segmentation of carotid artery lumen in two-dimensional and three-dimension al ultrasonography is an important step in computerized evaluation of arter ial disease severity and in finding vulnerable atherosclerotic plaques susc eptible to rupture causing stroke. Because of the complexity of anatomical structures, noise as well as the requirement of accurate segmentation, inte ractions are necessary between observers and the computer segmentation proc ess. In this paper a segmentation process is described based on the deforma ble model method with only one seed point to guide the initialization of th e deformable model for each lumen cross section. With one seed, the initial contour of the deformable model is generated using the entropy map of the original image and mathematical morphology operations. The deformable model is driven to fit the lumen contour by an internal force and an external fo rce that are calculated, respectively, with geometrical properties of defor med contour and with the image gray level features. The evaluation methodol ogy using distance-based and area-based metrics is introduced in this paper . A contour probability distribution (CPD) method for calculating distance- based metrics is introduced. The CPD is obtained by generating contours of the lumen using a set of possible seed locations. The mean contour can be c ompared to a manual outlined contour to provide accuracy metrics. The varia nce computed from the CPD can provide metrics of local and global variabili ty. These metrics provide a complete performance evaluation of an interacti ve segmentation algorithm and a means for comparing different algorithm set tings. (C) 2000 American Association of Physicists in Medicine. [S0094-2405 (00)02208-2].