Retinal vasculature segmentation using principal spanning forests

Erhan Bas, Esra Ataer-Cansizoglu, Deniz Erdogmus, Jayashree Kalpathy-Cramer

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

We propose an automated morphology reconstruction method for curvilinear network analysis. The proposed approach first projects samples to the ridge of the intensity image of the curvilinear system. Then, a manifold deviation measure is utilized to approximate the ridge with piecewise linear segments between the projected samples. A nonparametric system workflow based on the kernel interpolation and density estimation is provided for the derivations without any user defined meta-parameter, i.e. hard threshold for segmentation. Lastly, a rigorous sampling strategy using the manifold deviation measure that can be used for robust sparse tree reconstruction is provided. The proposed approaches have been tested on a small set of representative retinal scans. Preliminary qualitative results indicate the effectiveness of the method.

Original languageEnglish (US)
Title of host publicationProceedings - International Symposium on Biomedical Imaging
Pages1792-1795
Number of pages4
DOIs
StatePublished - 2012
Externally publishedYes
Event2012 9th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2012 - Barcelona, Spain
Duration: May 2 2012May 5 2012

Other

Other2012 9th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2012
CountrySpain
CityBarcelona
Period5/2/125/5/12

Fingerprint

Electric network analysis
Interpolation
Sampling
Spatial Analysis
Workflow
Forests

Keywords

  • Principal graphs
  • resampling on manifolds

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

Bas, E., Ataer-Cansizoglu, E., Erdogmus, D., & Kalpathy-Cramer, J. (2012). Retinal vasculature segmentation using principal spanning forests. In Proceedings - International Symposium on Biomedical Imaging (pp. 1792-1795). [6235930] https://doi.org/10.1109/ISBI.2012.6235930

Retinal vasculature segmentation using principal spanning forests. / Bas, Erhan; Ataer-Cansizoglu, Esra; Erdogmus, Deniz; Kalpathy-Cramer, Jayashree.

Proceedings - International Symposium on Biomedical Imaging. 2012. p. 1792-1795 6235930.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Bas, E, Ataer-Cansizoglu, E, Erdogmus, D & Kalpathy-Cramer, J 2012, Retinal vasculature segmentation using principal spanning forests. in Proceedings - International Symposium on Biomedical Imaging., 6235930, pp. 1792-1795, 2012 9th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2012, Barcelona, Spain, 5/2/12. https://doi.org/10.1109/ISBI.2012.6235930
Bas E, Ataer-Cansizoglu E, Erdogmus D, Kalpathy-Cramer J. Retinal vasculature segmentation using principal spanning forests. In Proceedings - International Symposium on Biomedical Imaging. 2012. p. 1792-1795. 6235930 https://doi.org/10.1109/ISBI.2012.6235930
Bas, Erhan ; Ataer-Cansizoglu, Esra ; Erdogmus, Deniz ; Kalpathy-Cramer, Jayashree. / Retinal vasculature segmentation using principal spanning forests. Proceedings - International Symposium on Biomedical Imaging. 2012. pp. 1792-1795
@inproceedings{350bcf1b26ca45a19074c43241756c1e,
title = "Retinal vasculature segmentation using principal spanning forests",
abstract = "We propose an automated morphology reconstruction method for curvilinear network analysis. The proposed approach first projects samples to the ridge of the intensity image of the curvilinear system. Then, a manifold deviation measure is utilized to approximate the ridge with piecewise linear segments between the projected samples. A nonparametric system workflow based on the kernel interpolation and density estimation is provided for the derivations without any user defined meta-parameter, i.e. hard threshold for segmentation. Lastly, a rigorous sampling strategy using the manifold deviation measure that can be used for robust sparse tree reconstruction is provided. The proposed approaches have been tested on a small set of representative retinal scans. Preliminary qualitative results indicate the effectiveness of the method.",
keywords = "Principal graphs, resampling on manifolds",
author = "Erhan Bas and Esra Ataer-Cansizoglu and Deniz Erdogmus and Jayashree Kalpathy-Cramer",
year = "2012",
doi = "10.1109/ISBI.2012.6235930",
language = "English (US)",
isbn = "9781457718588",
pages = "1792--1795",
booktitle = "Proceedings - International Symposium on Biomedical Imaging",

}

TY - GEN

T1 - Retinal vasculature segmentation using principal spanning forests

AU - Bas, Erhan

AU - Ataer-Cansizoglu, Esra

AU - Erdogmus, Deniz

AU - Kalpathy-Cramer, Jayashree

PY - 2012

Y1 - 2012

N2 - We propose an automated morphology reconstruction method for curvilinear network analysis. The proposed approach first projects samples to the ridge of the intensity image of the curvilinear system. Then, a manifold deviation measure is utilized to approximate the ridge with piecewise linear segments between the projected samples. A nonparametric system workflow based on the kernel interpolation and density estimation is provided for the derivations without any user defined meta-parameter, i.e. hard threshold for segmentation. Lastly, a rigorous sampling strategy using the manifold deviation measure that can be used for robust sparse tree reconstruction is provided. The proposed approaches have been tested on a small set of representative retinal scans. Preliminary qualitative results indicate the effectiveness of the method.

AB - We propose an automated morphology reconstruction method for curvilinear network analysis. The proposed approach first projects samples to the ridge of the intensity image of the curvilinear system. Then, a manifold deviation measure is utilized to approximate the ridge with piecewise linear segments between the projected samples. A nonparametric system workflow based on the kernel interpolation and density estimation is provided for the derivations without any user defined meta-parameter, i.e. hard threshold for segmentation. Lastly, a rigorous sampling strategy using the manifold deviation measure that can be used for robust sparse tree reconstruction is provided. The proposed approaches have been tested on a small set of representative retinal scans. Preliminary qualitative results indicate the effectiveness of the method.

KW - Principal graphs

KW - resampling on manifolds

UR - http://www.scopus.com/inward/record.url?scp=84864851796&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84864851796&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2012.6235930

DO - 10.1109/ISBI.2012.6235930

M3 - Conference contribution

AN - SCOPUS:84864851796

SN - 9781457718588

SP - 1792

EP - 1795

BT - Proceedings - International Symposium on Biomedical Imaging

ER -