Comparative analysis of dendritic architecture of identified neurons using the Hausdorff distance metric

Adi Mizrahi, Eyal Ben-Ner, Matthew J. Katz, Klara Kedem, J. Gustavo Glusman, Frederic Libersat*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

24 Scopus citations


Dendritic trees often are complex, three-dimensional structures. Comparative morphologic studies have not yet provided a reliable measure to analyze and compare the geometry of different dendritic trees. Therefore, it is important to develop quantitative methods for analyzing the three- dimensional geometry of these complex trees. The authors developed a comparison measure based on the Hausdorff distance for comparing quantitatively the three-dimensional structure of different neurons. This algorithm was implemented and incorporated into a new software package that the authors developed called NeuroComp. The authors tested this algorithm to study the variability in the three-dimensional structure of identified central neurons as well as measuring the structural differences between homologue neurons. They took advantage of the uniform dendritic morphology of identified interneurons of an insect, the giant interneurons of the cockroach. More specifically, after establishing a morphometric data base of these neurons, the authors found that the algorithm is a reliable tool for distinguishing between dendritic trees of different neurons, whereas conventional metric analysis often is inadequate. The authors propose to use this method as a quantitative tool for the investigation of the effects of various experimental paradigms on three-dimensional dendritic architecture. (C) 2000 Wiley-Liss, Inc.

Original languageAmerican English
Pages (from-to)415-428
Number of pages14
JournalJournal of Comparative Neurology
Issue number3
StatePublished - 3 Jul 2000
Externally publishedYes


  • Dendritic trees
  • Geometry
  • Morphometry
  • Three-dimensional reconstruction


Dive into the research topics of 'Comparative analysis of dendritic architecture of identified neurons using the Hausdorff distance metric'. Together they form a unique fingerprint.

Cite this