Computer-vision-based extraction of neural dendrograms

This work introduces a new approach to the characterization of neural cells by means of semi-automated generation of dendrograms; data structures whic h describe the inherently hierarchical nature of neuronal arborizations. De ndrograms describe the branched structure of neurons in terms of the length , average thickness and bending energy of each of the dendritic segments an d allow in a straightforward manner, the inclusion of additional measures. The bending energy quantifies the complexity of the shape and can be used t o characterize the spatial coverage of the arborizations (the bending energ y is an alternative for other complexity measures such as the fractal dimen sion). The new approach is based on the partitioning of the cell's outer co ntour as a function of the high curvature points followed by a syntactical analysis of the segmented contours. The semi-automated method is robust and is an improvement on the time consuming manual generation of the dendrogra ms. Several experimental results are included in this paper which illustrat e and corroborate the effectiveness of the approach. The technique presente d in this paper is limited to planar neurons but could be extended to a 3D approach. (C) 1999 Published by Elsevier Science B.V. All rights reserved.