Use mean shift to track neuronal axons in 3D

Hongmin Cai; Xiaoyin Xu; Ju Lu; Jeff W. Lichtman; S. P. Yung; Stephen T. Wong

doi:10.1109/LSSA.2006.250405

Use mean shift to track neuronal axons in 3D

Hongmin Cai, Xiaoyin Xu, Ju Lu, Jeff W. Lichtman, S. P. Yung, Stephen T. Wong

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Morphology is very important in help neuroscientists understand neuronal functions and connectivity of neurons. Using confocal microscopy researchers can acquire 3D images of neuronal axons in high resolution and study how axons innervate muscular fibers. To test different innervation models, researchers need to track every single axons and its branches in 3D. A robust segmentation and tracking method is needed to follow each axon in 3D. Challenges are that axons may appear touching each other in the image and make it difficult to segment. In addition, split and merge of axons require judicious image processing to correctly track axons in these cases. We present a 3-step segmentation and tracking algorithm to address these problems. Our proposed method includes nonlinear anisotropic diffusion for noise removal and edge enhancement, morphological operation for edge detection, and mean shift for tracking in three dimensions. The method can segment contacting objects and track the axons when they merge or split.

Original language	English (US)
Title of host publication	2006 IEEE/NLM Life Science Systems and Applications Workshop, LiSA 2006
DOIs	https://doi.org/10.1109/LSSA.2006.250405
State	Published - Dec 1 2006
Event	2006 IEEE/NLM Life Science Systems and Applications Workshop, LiSA 2006 - Bethesda, MD, United States Duration: Jul 13 2006 → Jul 14 2006

Other

Other	2006 IEEE/NLM Life Science Systems and Applications Workshop, LiSA 2006
Country/Territory	United States
City	Bethesda, MD
Period	7/13/06 → 7/14/06

ASJC Scopus subject areas

Health(social science)
Assessment and Diagnosis
Medicine(all)
Health Information Management
Electrical and Electronic Engineering
Human-Computer Interaction
Computer Science Applications
Signal Processing

Access to Document

10.1109/LSSA.2006.250405

Cite this

@inproceedings{dc879db76c8d46c2a27add3d7c32f73a,

title = "Use mean shift to track neuronal axons in 3D",

abstract = "Morphology is very important in help neuroscientists understand neuronal functions and connectivity of neurons. Using confocal microscopy researchers can acquire 3D images of neuronal axons in high resolution and study how axons innervate muscular fibers. To test different innervation models, researchers need to track every single axons and its branches in 3D. A robust segmentation and tracking method is needed to follow each axon in 3D. Challenges are that axons may appear touching each other in the image and make it difficult to segment. In addition, split and merge of axons require judicious image processing to correctly track axons in these cases. We present a 3-step segmentation and tracking algorithm to address these problems. Our proposed method includes nonlinear anisotropic diffusion for noise removal and edge enhancement, morphological operation for edge detection, and mean shift for tracking in three dimensions. The method can segment contacting objects and track the axons when they merge or split.",

author = "Hongmin Cai and Xiaoyin Xu and Ju Lu and Lichtman, {Jeff W.} and Yung, {S. P.} and Wong, {Stephen T.}",

year = "2006",

month = dec,

day = "1",

doi = "10.1109/LSSA.2006.250405",

language = "English (US)",

isbn = "1424402786",

booktitle = "2006 IEEE/NLM Life Science Systems and Applications Workshop, LiSA 2006",

note = "2006 IEEE/NLM Life Science Systems and Applications Workshop, LiSA 2006 ; Conference date: 13-07-2006 Through 14-07-2006",

}

TY - GEN

T1 - Use mean shift to track neuronal axons in 3D

AU - Cai, Hongmin

AU - Xu, Xiaoyin

AU - Lu, Ju

AU - Lichtman, Jeff W.

AU - Yung, S. P.

AU - Wong, Stephen T.

PY - 2006/12/1

Y1 - 2006/12/1

N2 - Morphology is very important in help neuroscientists understand neuronal functions and connectivity of neurons. Using confocal microscopy researchers can acquire 3D images of neuronal axons in high resolution and study how axons innervate muscular fibers. To test different innervation models, researchers need to track every single axons and its branches in 3D. A robust segmentation and tracking method is needed to follow each axon in 3D. Challenges are that axons may appear touching each other in the image and make it difficult to segment. In addition, split and merge of axons require judicious image processing to correctly track axons in these cases. We present a 3-step segmentation and tracking algorithm to address these problems. Our proposed method includes nonlinear anisotropic diffusion for noise removal and edge enhancement, morphological operation for edge detection, and mean shift for tracking in three dimensions. The method can segment contacting objects and track the axons when they merge or split.

AB - Morphology is very important in help neuroscientists understand neuronal functions and connectivity of neurons. Using confocal microscopy researchers can acquire 3D images of neuronal axons in high resolution and study how axons innervate muscular fibers. To test different innervation models, researchers need to track every single axons and its branches in 3D. A robust segmentation and tracking method is needed to follow each axon in 3D. Challenges are that axons may appear touching each other in the image and make it difficult to segment. In addition, split and merge of axons require judicious image processing to correctly track axons in these cases. We present a 3-step segmentation and tracking algorithm to address these problems. Our proposed method includes nonlinear anisotropic diffusion for noise removal and edge enhancement, morphological operation for edge detection, and mean shift for tracking in three dimensions. The method can segment contacting objects and track the axons when they merge or split.

UR - http://www.scopus.com/inward/record.url?scp=34548864942&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34548864942&partnerID=8YFLogxK

U2 - 10.1109/LSSA.2006.250405

DO - 10.1109/LSSA.2006.250405

M3 - Conference contribution

AN - SCOPUS:34548864942

SN - 1424402786

SN - 9781424402786

BT - 2006 IEEE/NLM Life Science Systems and Applications Workshop, LiSA 2006

T2 - 2006 IEEE/NLM Life Science Systems and Applications Workshop, LiSA 2006

Y2 - 13 July 2006 through 14 July 2006

ER -

Use mean shift to track neuronal axons in 3D

Abstract

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this