High resolution segmentation of neuronal tissues from low depth-resolution em imagery

Daniel Glasner; Tao Hu; Juan Nunez-Iglesias; Lou Scheffer; Shan Xu; Harald Hess; Richard Fetter; Dmitri Chklovskii; Ronen Basri

doi:10.1007/978-3-642-23094-3_19

High resolution segmentation of neuronal tissues from low depth-resolution em imagery

Daniel Glasner, Tao Hu, Juan Nunez-Iglesias, Lou Scheffer, Shan Xu, Harald Hess, Richard Fetter, Dmitri Chklovskii, Ronen Basri

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

2 Citations (Scopus)

Abstract

The challenge of recovering the topology of massive neuronal circuits can potentially be met by high throughput Electron Microscopy (EM) imagery. Segmenting a 3-dimensional stack of EM images into the individual neurons is difficult, due to the low depth-resolution in existing high-throughput EM technology, such as serial section Transmission EM (ssTEM). In this paper we propose methods for detecting the high resolution locations of membranes from low depth-resolution images. We approach this problem using both a method that learns a discriminative, over-complete dictionary and a kernel SVM. We test this approach on tomographic sections produced in simulations from high resolution Focused Ion Beam (FIB) images and on low depth-resolution images acquired with ssTEM and evaluate our results by comparing it to manual labeling of this data.

Original language	English
Title of host publication	Energy Minimazation Methods in Computer Vision and Pattern Recognition - 8th International Conference, EMMCVPR 2011, Proceedings
Pages	261-272
Number of pages	12
DOIs	https://doi.org/10.1007/978-3-642-23094-3_19
Publication status	Published - 22 Aug 2011
Externally published	Yes
Event	International Conference on Energy Minimization Methods in Computer Vision and Pattern Recognition (EMMCVPR) 2011 - St. Petersburg, Russian Federation Duration: 25 Jul 2011 → 27 Jul 2011 Conference number: 8th https://link.springer.com/book/10.1007/978-3-642-23094-3 (Proceedings)

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	6819 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	International Conference on Energy Minimization Methods in Computer Vision and Pattern Recognition (EMMCVPR) 2011
Abbreviated title	EMMCVPR 2011
Country/Territory	Russian Federation
City	St. Petersburg
Period	25/07/11 → 27/07/11
Internet address	https://link.springer.com/book/10.1007/978-3-642-23094-3 (Proceedings)

Keywords

Connectomics
Segmentation of neuronal tissues
Sparse over-complete representation
Task-driven dictionary learning

Access to Document

10.1007/978-3-642-23094-3_19

Cite this

Glasner, D., Hu, T., Nunez-Iglesias, J., Scheffer, L., Xu, S., Hess, H., Fetter, R., Chklovskii, D., & Basri, R. (2011). High resolution segmentation of neuronal tissues from low depth-resolution em imagery. In Energy Minimazation Methods in Computer Vision and Pattern Recognition - 8th International Conference, EMMCVPR 2011, Proceedings (pp. 261-272). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6819 LNCS). https://doi.org/10.1007/978-3-642-23094-3_19

Glasner, Daniel ; Hu, Tao ; Nunez-Iglesias, Juan et al. / High resolution segmentation of neuronal tissues from low depth-resolution em imagery. Energy Minimazation Methods in Computer Vision and Pattern Recognition - 8th International Conference, EMMCVPR 2011, Proceedings. 2011. pp. 261-272 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{0f095eeb40c247d59242337dbcc0d98a,

title = "High resolution segmentation of neuronal tissues from low depth-resolution em imagery",

abstract = "The challenge of recovering the topology of massive neuronal circuits can potentially be met by high throughput Electron Microscopy (EM) imagery. Segmenting a 3-dimensional stack of EM images into the individual neurons is difficult, due to the low depth-resolution in existing high-throughput EM technology, such as serial section Transmission EM (ssTEM). In this paper we propose methods for detecting the high resolution locations of membranes from low depth-resolution images. We approach this problem using both a method that learns a discriminative, over-complete dictionary and a kernel SVM. We test this approach on tomographic sections produced in simulations from high resolution Focused Ion Beam (FIB) images and on low depth-resolution images acquired with ssTEM and evaluate our results by comparing it to manual labeling of this data.",

keywords = "Connectomics, Segmentation of neuronal tissues, Sparse over-complete representation, Task-driven dictionary learning",

author = "Daniel Glasner and Tao Hu and Juan Nunez-Iglesias and Lou Scheffer and Shan Xu and Harald Hess and Richard Fetter and Dmitri Chklovskii and Ronen Basri",

year = "2011",

month = aug,

day = "22",

doi = "10.1007/978-3-642-23094-3_19",

language = "English",

isbn = "9783642230936",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

pages = "261--272",

booktitle = "Energy Minimazation Methods in Computer Vision and Pattern Recognition - 8th International Conference, EMMCVPR 2011, Proceedings",

note = "International Conference on Energy Minimization Methods in Computer Vision and Pattern Recognition (EMMCVPR) 2011, EMMCVPR 2011 ; Conference date: 25-07-2011 Through 27-07-2011",

url = "https://link.springer.com/book/10.1007/978-3-642-23094-3",

}

Glasner, D, Hu, T, Nunez-Iglesias, J, Scheffer, L, Xu, S, Hess, H, Fetter, R, Chklovskii, D & Basri, R 2011, High resolution segmentation of neuronal tissues from low depth-resolution em imagery. in Energy Minimazation Methods in Computer Vision and Pattern Recognition - 8th International Conference, EMMCVPR 2011, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 6819 LNCS, pp. 261-272, International Conference on Energy Minimization Methods in Computer Vision and Pattern Recognition (EMMCVPR) 2011, St. Petersburg, Russian Federation, 25/07/11. https://doi.org/10.1007/978-3-642-23094-3_19

High resolution segmentation of neuronal tissues from low depth-resolution em imagery. / Glasner, Daniel; Hu, Tao; Nunez-Iglesias, Juan et al.
Energy Minimazation Methods in Computer Vision and Pattern Recognition - 8th International Conference, EMMCVPR 2011, Proceedings. 2011. p. 261-272 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6819 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - High resolution segmentation of neuronal tissues from low depth-resolution em imagery

AU - Glasner, Daniel

AU - Hu, Tao

AU - Nunez-Iglesias, Juan

AU - Scheffer, Lou

AU - Xu, Shan

AU - Hess, Harald

AU - Fetter, Richard

AU - Chklovskii, Dmitri

AU - Basri, Ronen

N1 - Conference code: 8th

PY - 2011/8/22

Y1 - 2011/8/22

N2 - The challenge of recovering the topology of massive neuronal circuits can potentially be met by high throughput Electron Microscopy (EM) imagery. Segmenting a 3-dimensional stack of EM images into the individual neurons is difficult, due to the low depth-resolution in existing high-throughput EM technology, such as serial section Transmission EM (ssTEM). In this paper we propose methods for detecting the high resolution locations of membranes from low depth-resolution images. We approach this problem using both a method that learns a discriminative, over-complete dictionary and a kernel SVM. We test this approach on tomographic sections produced in simulations from high resolution Focused Ion Beam (FIB) images and on low depth-resolution images acquired with ssTEM and evaluate our results by comparing it to manual labeling of this data.

AB - The challenge of recovering the topology of massive neuronal circuits can potentially be met by high throughput Electron Microscopy (EM) imagery. Segmenting a 3-dimensional stack of EM images into the individual neurons is difficult, due to the low depth-resolution in existing high-throughput EM technology, such as serial section Transmission EM (ssTEM). In this paper we propose methods for detecting the high resolution locations of membranes from low depth-resolution images. We approach this problem using both a method that learns a discriminative, over-complete dictionary and a kernel SVM. We test this approach on tomographic sections produced in simulations from high resolution Focused Ion Beam (FIB) images and on low depth-resolution images acquired with ssTEM and evaluate our results by comparing it to manual labeling of this data.

KW - Connectomics

KW - Segmentation of neuronal tissues

KW - Sparse over-complete representation

KW - Task-driven dictionary learning

UR - http://www.scopus.com/inward/record.url?scp=80051769353&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-23094-3_19

DO - 10.1007/978-3-642-23094-3_19

M3 - Conference Paper

AN - SCOPUS:80051769353

SN - 9783642230936

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 261

EP - 272

BT - Energy Minimazation Methods in Computer Vision and Pattern Recognition - 8th International Conference, EMMCVPR 2011, Proceedings

T2 - International Conference on Energy Minimization Methods in Computer Vision and Pattern Recognition (EMMCVPR) 2011

Y2 - 25 July 2011 through 27 July 2011

ER -

Glasner D, Hu T, Nunez-Iglesias J, Scheffer L, Xu S, Hess H et al. High resolution segmentation of neuronal tissues from low depth-resolution em imagery. In Energy Minimazation Methods in Computer Vision and Pattern Recognition - 8th International Conference, EMMCVPR 2011, Proceedings. 2011. p. 261-272. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-642-23094-3_19

High resolution segmentation of neuronal tissues from low depth-resolution em imagery

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Cite this