Determination of arterial input function for quantification of cerebral blood flow with dynamic susceptibility contrast-enhanced MR imaging using fuzzy clustering

Abstract

When quantifying the perfusion parameters such as cerebral blood flow (CBF) using dynamic susceptibility contrast-enhanced magnetic resonance imaging (DSC-MRI), the arterial input function (AIF) of contrast agent has to be determined. In this study, we developed a method for obtaining the AIF automatically using fuzzy c-means (FCM) clustering. First, a mask region of interest (ROI) was drawn around the internal carotid artery. Second, FCM clustering was applied to the data in this ROI and the cluster centroids were calculated. The cluster centroid with the highest maximum concentration, earliest maximum concentration and smallest FWHM of the time-concentration curve (TCC) was determined as the arterial pixels and the AIF was obtained from the mean TCC in these pixels. We applied this method to six subjects and compared it with a manual ROI method. The difference between the CBF values calculated using the AIF obtained by FCM clustering [CBF(fuzzy)] and that obtained by the manual ROI method [CBF(manual)] ranged from 0.92% to 122% [38.6/spl plusmn/37.7% (mean/spl plusmn/SD)]. The CBF(manual) values were generally overestimated compared with the CBF(fuzzy) values, while the CBF(fuzzy) values became closer to the CBF values found in the literature. In conclusion, FCM clustering appears to be promising for determination of AIF, because it allows automatic, rapid and accurate extraction of arterial pixels.