Thrombus Localization in Middle Cerebral Artery of Patient with Acute Ischemic Stroke on ncCT Image

A common cause of an Acute Ischemic Stroke (AIS) is a thrombus in the middle cerebral artery (MCA). Localizing the thrombus on a set of non-contrast computerized tomography (ncCT) image for some composite type is not easy. This work presents an algorithm for localizing thrombus in MCA region on ncCT using an image processing method. Given an ncCT image P, brain region is extracted by considering a homogeneous intensity area using k-mean cluster. A set of candidate regions is obtained by considering size and average intensity of each connected region. Also, location of the MCA is located to filter the candidate regions that are in the interested region. Experiment is perform on a number of images of the different composite of thrombus. The result shows that the proposed image-processing-based thrombus localization algorithm is able to automatically locate the thrombus on ncCT of both red blood cell-rich and fibrin-rich composite.