Quantitative Analysis of Cerebrospinal Fluid Flow in Complex Regions by Using Phase Contrast Magnetic Resonance Imaging Natalia Flo ´ rez, 1 Roger Bouzerar, 2 David Moratal, 3,4 Marc-Etienne Meyer, 2 LuisMartı´-Bonmatı´, 1 Olivier Bale ´ dent 2 1 Department of Radiology, Hospital Quiro ´ n, Valencia, Spain 2 Department of Imaging and Biophysics, University Hospital, Amiens, France 3 Center for Biomaterials and Tissue Engineering, Universitat Polite ` cnica de Vale ` ncia, Vale ` ncia, Spain 4 Department of Electronic Engineering, Universitat Polite ` cnica de Vale ` ncia, Vale ` ncia, Spain Received 28 May 2010; accepted 27 April 2011 ABSTRACT: To develop a method for segmenting cerebrospinal fluid (CSF) regions with complex, inhomogeneous pulsatile patterns in phase contrast magnetic resonance imaging (PC-MRI) sequences. Our approach used various temporal features of flow behavior as input attributes in an unsupervised k-means classification algorithm. CSF flow parameters for the cervical subarachnoid spaces and the pontine cistern were calculated in 26 healthy volunteers. Background and aliasing corrections were applied automatically. The algorithm’s reproducibility was determined by calculating two parameters (area and stroke volume) while varying the initially selected seed point. The influence of background correction on these parameters was also assessed. The method was highly reproducible, with coefficients of variation of 3 and 4% for the cervical stroke volume and area, respec- tively. In an analysis of variance, background correction did not have a statistically significant effect on either the stroke volume (p 5 0.32) or the CSF net mean flow (p 5 0.69) at the C2C3 level. The method presented here enables rapid, reproducible, quantitative analysis of CSF flow in complex regions such as the C2C3 subarachnoid spaces and the pontine cistern. V V C 2011 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 21, 290–297, 2011; Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/ima.20294 Key words: CSF flow; PC-MRI; k-means; segmentation; pontine cistern I. INTRODUCTION The brain is probably the most complex structure in the human body. With a view to understanding neurological disorders such as hydrocephalus, the brain’s motion and the distribution of cerebrospinal fluid (CSF) pulsations within the craniospinal axis in response to rhythmic cerebral blood volume variations during the cardiac cycle have been extensively investigated (Enzmann and Pelc, 1991; Enzmann and Pelc, 1992; Bradley, 1992; Alperin et al., 1996; Henry-Feugeas et al., 2000; Egnor et al; 2002). The combination of phase contrast magnetic resonance imaging (PC-MRI) and peripheral cardiac gating has enabled the precise, noninvasive, quantitative analysis of CSF and blood flow variations during the cardiac cycle (Thomsen et al., 1990; Nitz et al., 1992; Greitz et al., 1993; Henry-Feugeas et al., 1993; Greitz, 1993; Enz- mann and Pelc, 1993; Bhadelia et al., 1995; Kim et al., 1999). This technique has facilitated the study of the central nervous system’s pulsation at several intracranial and extracranial locations (Alperin et al., 1996; Bhadelia et al., 1997; Miyati et al., 2003; Baledent et al., 2004; Wagshul et al., 2006). The analysis of blood and CSF flow dynamics within the brain enables the calculation of periodi- cal, intracranial volume changes, and the simulation and assessment of intracranial compliance and pressure (Alperin, 2004; Baledent et al., 2006; Miyati et al., 2007). Several authors have proposed CSF flow parameters for use in the diagnosis of hydrocephalus and the prediction of shunt responses (Bradley et al., 1996; Luetmer et al., 2002). Nevertheless, the reproducibility and reliability of flow quanti- zation using PC-MRI depend greatly on the postprocessing method used to accurately delineate the region of interest (ROI) (Wolf et al., 1993; Pelc, 1995). The first studies of automatic or semi-auto- matic segmentation methods using PC-MRI images were applied to blood vessels (Kozerke et al., 1999; Yuan et al., 1999; Box et al., 2003). To study neurological and cerebrovascular diseases, the Grant sponsor: Generalitat Valenciana; Grant number: CTBPRB/2004/342 FPI. Correspondence to: Olivier Bale ´dent; e-mail: olivier.baledent@chu-amiens.fr ' 2011 Wiley Periodicals, Inc.