A Narrow-Band Level-Set Method with Dynamic Velocity for Neural Stem Cell Cluster Segmentation Nezamoddin N. Kachouie and Paul Fieguth Department of Systems Design Engineering, University of Waterloo, Waterloo, Canada nnezamod@engmail.uwaterloo.ca pfieguth@uwaterloo.ca http://www.ocho.uwaterloo.ca Abstract. Neural Stem Cells (NSCs) have a remarkable capacity to proliferate and differentiate to other cell types. This ability to differen- tiate to desirable phenotypes has motivated clinical interests, hence the interest here to segment Neural Stem Cell (NSC) clusters to locate the NSC clusters over time in a sequence of frames, and in turn to perform NSC cluster motion analysis. However the manual segmentation of such data is a tedious task. Thus, due to the increasing amount of cell data being collected, automated cell segmentation methods are highly desired. In this paper a novel level set based segmentation method is proposed to accomplish this segmentation. The method is initialization insensitive, making it an appropriate solution for automated segmentation systems. The proposed segmentation method has been successfully applied to NSC cluster segmentation. 1 Introduction Neural Stem Cells (NSCs) as building blocks of the brain can proliferate and dif- ferentiate into all neural phenotypes. Progress in the analysis of NSC functional properties is required for development of clinically applicable procedures for stem cell transplantation and for the treatment of various incurable diseases. NSC can be used to repair damaged neuro-degenerative processes such as Alzheimer and to repair brain injuries such as stroke. Due to the universal attributes of NSCs, there has been great interest to develop a practical automated approach to measure and extract NSCs properties from microscopic cell images and track individual cells over time. To accomplish this task the NSC clusters must first be segmented. In practice, due to the presence of clutter, corrupted and blurred images, manual cell segmentation is a tedious task. An automated cell segmentation system may eliminate the onerous process of manual cell segmentation, extracting cell features from microscopic images. Several methods have been developed for region segmentation such as region growing, watershed and thresholding methods [2, 5]. Recently researchers have M. Kamel and A. Campilho (Eds.): ICIAR 2005, LNCS 3656, pp. 1006–1013, 2005. c  Springer-Verlag Berlin Heidelberg 2005