Limits...
Automated detection of soma location and morphology in neuronal network cultures.

Ozcan B, Negi P, Laezza F, Papadakis M, Labate D - PLoS ONE (2015)

Bottom Line: In addition to the usual difficulties associated with the processing of fluorescent images, these types of stacks contain a small number of images so that only a small number of pixels are available along the z-direction and it is challenging to apply conventional 3D filters.Extensive numerical experiments show that our algorithms are computationally efficient and highly accurate in segmenting the somas and separating contiguous ones.The algorithms presented in this paper will facilitate the development of a high-throughput quantitative platform for the study of neuronal networks for HCS applications.

View Article: PubMed Central - PubMed

Affiliation: Dept. of Mathematics, University of Houston, Houston, Texas, United States of America.

ABSTRACT
Automated identification of the primary components of a neuron and extraction of its sub-cellular features are essential steps in many quantitative studies of neuronal networks. The focus of this paper is the development of an algorithm for the automated detection of the location and morphology of somas in confocal images of neuronal network cultures. This problem is motivated by applications in high-content screenings (HCS), where the extraction of multiple morphological features of neurons on large data sets is required. Existing algorithms are not very efficient when applied to the analysis of confocal image stacks of neuronal cultures. In addition to the usual difficulties associated with the processing of fluorescent images, these types of stacks contain a small number of images so that only a small number of pixels are available along the z-direction and it is challenging to apply conventional 3D filters. The algorithm we present in this paper applies a number of innovative ideas from the theory of directional multiscale representations and involves the following steps: (i) image segmentation based on support vector machines with specially designed multiscale filters; (ii) soma extraction and separation of contiguous somas, using a combination of level set method and directional multiscale filters. We also present an approach to extract the soma's surface morphology using the 3D shearlet transform. Extensive numerical experiments show that our algorithms are computationally efficient and highly accurate in segmenting the somas and separating contiguous ones. The algorithms presented in this paper will facilitate the development of a high-throughput quantitative platform for the study of neuronal networks for HCS applications.

No MeSH data available.


Related in: MedlinePlus

Extraction of soma support.The figure illustrates the detection of the cross section of the soma from a representative image taken from the bottom of a confocal image stack. (A) Denoised image. (B) Region obtained by applying the thresholding strategy within the region identified by the 2D mask as described in the text. (C) Detected support region obtained after applying a combination of filling and erosion operators to the image from Panel B.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4390318&req=5

pone.0121886.g008: Extraction of soma support.The figure illustrates the detection of the cross section of the soma from a representative image taken from the bottom of a confocal image stack. (A) Denoised image. (B) Region obtained by applying the thresholding strategy within the region identified by the 2D mask as described in the text. (C) Detected support region obtained after applying a combination of filling and erosion operators to the image from Panel B.

Mentions: Unfortunately, the application of a conventional image enhancement approach may also fail to improve the low signal-to-noise ratio of the images. On the other hand, the inspection of the images suggests that the soma’s support region does not change significantly from an optical section to the next one in this region of the stack (by contrast, there is often significant variability in the top optical sections). Therefore, we applied the following simple, yet effective strategy to detect a soma’s 3D support in each image stack. We computed the average of three successive images in a substack and use this value as an intensity threshold for the corresponding image. We applied this thresholding approach within the region identified by the mask we obtained from the 2D soma detection algorithm. We applied this approach only for the images located in the bottom half of the substack. Note that the extracted region may still contain some gaps. Hence, we applied a combination of filling and erosion operators [36, 37] to improve soma detection. An illustration of the application of this approach is given in Fig. 8 where we show a representative image of a neuron from the bottom of a confocal stack, the binary mask of the region obtained from the application of this localized thresholding strategy, and the final detected soma obtained after the application of the morphological operators. We remark that this simple thresholding procedure is successful because it is applied inside the box obtained by the masks Mi which tightly encases the soma.


Automated detection of soma location and morphology in neuronal network cultures.

Ozcan B, Negi P, Laezza F, Papadakis M, Labate D - PLoS ONE (2015)

Extraction of soma support.The figure illustrates the detection of the cross section of the soma from a representative image taken from the bottom of a confocal image stack. (A) Denoised image. (B) Region obtained by applying the thresholding strategy within the region identified by the 2D mask as described in the text. (C) Detected support region obtained after applying a combination of filling and erosion operators to the image from Panel B.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4390318&req=5

pone.0121886.g008: Extraction of soma support.The figure illustrates the detection of the cross section of the soma from a representative image taken from the bottom of a confocal image stack. (A) Denoised image. (B) Region obtained by applying the thresholding strategy within the region identified by the 2D mask as described in the text. (C) Detected support region obtained after applying a combination of filling and erosion operators to the image from Panel B.
Mentions: Unfortunately, the application of a conventional image enhancement approach may also fail to improve the low signal-to-noise ratio of the images. On the other hand, the inspection of the images suggests that the soma’s support region does not change significantly from an optical section to the next one in this region of the stack (by contrast, there is often significant variability in the top optical sections). Therefore, we applied the following simple, yet effective strategy to detect a soma’s 3D support in each image stack. We computed the average of three successive images in a substack and use this value as an intensity threshold for the corresponding image. We applied this thresholding approach within the region identified by the mask we obtained from the 2D soma detection algorithm. We applied this approach only for the images located in the bottom half of the substack. Note that the extracted region may still contain some gaps. Hence, we applied a combination of filling and erosion operators [36, 37] to improve soma detection. An illustration of the application of this approach is given in Fig. 8 where we show a representative image of a neuron from the bottom of a confocal stack, the binary mask of the region obtained from the application of this localized thresholding strategy, and the final detected soma obtained after the application of the morphological operators. We remark that this simple thresholding procedure is successful because it is applied inside the box obtained by the masks Mi which tightly encases the soma.

Bottom Line: In addition to the usual difficulties associated with the processing of fluorescent images, these types of stacks contain a small number of images so that only a small number of pixels are available along the z-direction and it is challenging to apply conventional 3D filters.Extensive numerical experiments show that our algorithms are computationally efficient and highly accurate in segmenting the somas and separating contiguous ones.The algorithms presented in this paper will facilitate the development of a high-throughput quantitative platform for the study of neuronal networks for HCS applications.

View Article: PubMed Central - PubMed

Affiliation: Dept. of Mathematics, University of Houston, Houston, Texas, United States of America.

ABSTRACT
Automated identification of the primary components of a neuron and extraction of its sub-cellular features are essential steps in many quantitative studies of neuronal networks. The focus of this paper is the development of an algorithm for the automated detection of the location and morphology of somas in confocal images of neuronal network cultures. This problem is motivated by applications in high-content screenings (HCS), where the extraction of multiple morphological features of neurons on large data sets is required. Existing algorithms are not very efficient when applied to the analysis of confocal image stacks of neuronal cultures. In addition to the usual difficulties associated with the processing of fluorescent images, these types of stacks contain a small number of images so that only a small number of pixels are available along the z-direction and it is challenging to apply conventional 3D filters. The algorithm we present in this paper applies a number of innovative ideas from the theory of directional multiscale representations and involves the following steps: (i) image segmentation based on support vector machines with specially designed multiscale filters; (ii) soma extraction and separation of contiguous somas, using a combination of level set method and directional multiscale filters. We also present an approach to extract the soma's surface morphology using the 3D shearlet transform. Extensive numerical experiments show that our algorithms are computationally efficient and highly accurate in segmenting the somas and separating contiguous ones. The algorithms presented in this paper will facilitate the development of a high-throughput quantitative platform for the study of neuronal networks for HCS applications.

No MeSH data available.


Related in: MedlinePlus