Limits...
Automated reconstruction of neuronal morphology based on local geometrical and global structural models.

Zhao T, Xie J, Amat F, Clack N, Ahammad P, Peng H, Long F, Myers E - Neuroinformatics (2011)

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
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.

Show MeSH
An OP image stack was contaminated by different levels of Gaussian noise: aσ = 20, bσ = 40, cσ = 60, dσ = 80, eσ = 100
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3104133&req=5

Fig12: An OP image stack was contaminated by different levels of Gaussian noise: aσ = 20, bσ = 40, cσ = 60, dσ = 80, eσ = 100

Mentions: Testing for Robustness to Noise We also tested our method on different noise levels to show its strength. The testing images were created by adding Gaussian noise to the OP image stack (Fig. 11). We used five noise levels, measured by the standard deviation of the noise distribution: σ = 20, σ = 40, σ = 60, σ = 80, σ = 100 where values in the image range from 0 to 255. The neuron becomes less visible when σ is larger (Fig. 12), making tracing progressively more difficult. To quantitatively evaluate how robust our method is to noise, we calculated the scores of tracing quality using the DIADEM metric (http://www.diademchallenge.org/metric.html). The results were also compared with those obtained from the free software NeuroStudio (Wearne et al. 2005), which was developed based on the Rayburst Sampling tracing algorithm. As shown in Fig. 13, our method produced reasonable result (score = 0.866) even when the noise level is at 100. In contrast, NeuroStudio’s performance degrades rapidly reaching a score of 0 when the noise is at 80. Moreover, our method outperformed NeuroStudio significantly for all the noise levels, including the case where there is no noise.Fig. 12


Automated reconstruction of neuronal morphology based on local geometrical and global structural models.

Zhao T, Xie J, Amat F, Clack N, Ahammad P, Peng H, Long F, Myers E - Neuroinformatics (2011)

An OP image stack was contaminated by different levels of Gaussian noise: aσ = 20, bσ = 40, cσ = 60, dσ = 80, eσ = 100
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3104133&req=5

Fig12: An OP image stack was contaminated by different levels of Gaussian noise: aσ = 20, bσ = 40, cσ = 60, dσ = 80, eσ = 100
Mentions: Testing for Robustness to Noise We also tested our method on different noise levels to show its strength. The testing images were created by adding Gaussian noise to the OP image stack (Fig. 11). We used five noise levels, measured by the standard deviation of the noise distribution: σ = 20, σ = 40, σ = 60, σ = 80, σ = 100 where values in the image range from 0 to 255. The neuron becomes less visible when σ is larger (Fig. 12), making tracing progressively more difficult. To quantitatively evaluate how robust our method is to noise, we calculated the scores of tracing quality using the DIADEM metric (http://www.diademchallenge.org/metric.html). The results were also compared with those obtained from the free software NeuroStudio (Wearne et al. 2005), which was developed based on the Rayburst Sampling tracing algorithm. As shown in Fig. 13, our method produced reasonable result (score = 0.866) even when the noise level is at 100. In contrast, NeuroStudio’s performance degrades rapidly reaching a score of 0 when the noise is at 80. Moreover, our method outperformed NeuroStudio significantly for all the noise levels, including the case where there is no noise.Fig. 12

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
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.

Show MeSH