An accurate determination of the arterial input function (AIF) is necessary for quantification of cerebral blood flow (CBF) using dynamic susceptibility contrast-enhanced magnetic resonance imaging. In this study, we developed a method for obtaining the AIF automatically using fuzzy c-means (FCM) clustering. The validity of this approach was investigated with computer simulations. We found that this method can automatically extract the AIF, even under very noisy conditions, e.g., when the signal-to-noise ratio is 2. The simulation results also indicated that when using a manual drawing of a region of interest (ROI) (manual ROI method), the contamination of surrounding pixels (background) into ROI caused considerable overestimation of CBF. We applied this method to six subjects and compared it with the manual ROI method. The CBF values, calculated using the AIF obtained using the manual ROI method [CBF ( Index terms: fuzzy c-means clustering; dynamic susceptibility contrast-enhanced MR imaging; arterial input function; cerebral blood flow; quantification MAGNETIC RESONANCE IMAGING (MRI) techniques that measure cerebral perfusion have become increasingly important. Compared with methods using single photon emission computed tomography (SPECT) or positron emission tomography (PET), quantification of cerebral perfusion using dynamic MRI has advantages such as improved spatial resolution, no patient exposure to ionizing radiation, and the opportunity to combine morphological and functional information during a single imaging session (1).The use of an intravascular contrast agent in combination with dynamic susceptibility-contrast (DSC) MRI for measurement of cerebral perfusion is an attractive concept, although not completely straightforward (1). For quantification of perfusion parameters such as cerebral blood flow (CBF) and cerebral blood volume (CBV), in terms of the absolute values using DSC-MRI, the arterial input function (AIF) of the contrast agent entering the tissue has to be determined (2,3,4). When an artery (e.g., internal carotid artery) runs through imaged slices of the cortex, the AIF can be obtained non-invasively from the arterial pixels. Traditional methods for the extraction of arterial pixels require the manual drawing of a region of interest (ROI) (5). Vonken et al (5) obtained the AIF by segmenting the arterial pixels manually from a slice through the neck with the internal carotid arteries. However, due to finite spatial resolution and large statistical noise, manual ROI analysis may lead to serious inconsistencies in the definition of arterial pixels. In addition, it may be subjective.Rempp et al developed an interactive computer program to determine AIF automatically (2). In their program, AIF is determined by following two steps. In the first step, certain parameters describing the concentration-time curves, such as the full width at half maximum (FWHM), the maximum concentration (MC), and the moment of maximum concentration (MMC), are calculated pixel by pixel for the whole brain. The mean ...