Submitted on May 30, 2008
Revised on August 6, 2008
Accepted on August 7, 2008
Live Neuron Morphology Automatically Reconstructed from Multiphoton and Confocal Imaging Data
Bradley E Losavio1, Yong Liang2, Alberto Santamaria-Pang2, Ioannis A. Kakadiaris2, Costa M. Colbert2, and Peter Saggau1*
1 Baylor College of Medicine
2 University of Houston
* To whom correspondence should be addressed. E-mail: psaggau{at}bcm.edu.
We have developed a fully automated procedure for extracting dendritic morphology from multiple 3D image stacks produced by laser scanning microscopy. By eliminating human intervention, we ensure that the results are objective, quickly generated, and accurate. The software suite accounts for typical experimental conditions by reducing background noise, removing pipette artifacts, and aligning multiple overlapping image stacks. The output morphology is appropriate for simulation in compartmental simulation environments. In this paper, we validate the utility of this procedure by comparing its performance on live neurons and test specimen with other fully- and semi-automated reconstruction tools.
