Automated reconstruction of neuronal morphology based on local geometrical and global structural models.
Bottom Line:
Digital reconstruction of neurons from microscope images is an important and challenging problem in neuroscience.We first formulate a model structure, then develop an algorithm for computing it by carefully taking into account morphological characteristics of neurons, as well as the image properties under typical imaging protocols.The method has been tested on the data sets used in the DIADEM competition and produced promising results for four out of the five data sets.
View Article:
PubMed Central - PubMed
Affiliation: Qiushi Academy for Advanced Studies, Zhejiang University, 38 ZheDa Road, Hangzhou 310027, China. tingzhao@gmail.com
ABSTRACT
Show MeSH
Digital reconstruction of neurons from microscope images is an important and challenging problem in neuroscience. In this paper, we propose a model-based method to tackle this problem. We first formulate a model structure, then develop an algorithm for computing it by carefully taking into account morphological characteristics of neurons, as well as the image properties under typical imaging protocols. The method has been tested on the data sets used in the DIADEM competition and produced promising results for four out of the five data sets. |
Related In:
Results -
Collection
getmorefigures.php?uid=PMC3104133&req=5
Mentions: Preprocessing for the CF Dataset In the CF images neurons are brown and nuclei are blue. In some areas they overlap due to the nature of the bright-field point spread function which does not optically section in z. This overlap can not be resolved by extracting a certain color component as in some regions light from both objects is present. We solved the problem based on a color mixing model. In this model, there is an average (bright) background B and there are two different materials, m1 and m2, in the imaging field and each material absorbs light of a certain color, c1 and c2, respectively. At any voxel x, the total observed intensity I(x) is:5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ I(x)= B - [c_1, c_2] \cdot \left[ \begin{array}{c} p_1(x)\\ p_2(x) \\ \end{array} \right] \label{eq:color_mix} $$\end{document}where pi(x) denotes the absorption contribution from material mi. We estimate B as the average intensity of each frame which is a good estimate as the background occupies most of each frame. c1 and c2 are calculated reliably by taking the average values of the two separable peaks in the color histogram of B − I. Once we have these three values, we compute the amount of each material at each voxel by solving Eq. 5 by linear regression. The result is shown in Fig. 14.Fig. 14 |
View Article: PubMed Central - PubMed
Affiliation: Qiushi Academy for Advanced Studies, Zhejiang University, 38 ZheDa Road, Hangzhou 310027, China. tingzhao@gmail.com