Exploratory dijkstra forest based automatic vessel segmentation: Applications in video indirect ophthalmoscopy (VIO)

Rolando Estrada, Carlo Tomasi, Michelle T. Cabrera, David K. Wallace, Sharon F. Freedman, Sina Farsiu

Research output: Contribution to journalArticle

26 Citations (Scopus)

Abstract

We present a methodology for extracting the vascular network in the human retina using Dijkstra's shortest-path algorithm. Our method preserves vessel thickness, requires no manual intervention, and follows vessel branching naturally and efficiently. To test our method, we constructed a retinal video indirect ophthalmoscopy (VIO) image database from pediatric patients and compared the segmentations achieved by ourmethod and state-of-the-art approaches to a human-drawn gold standard. Our experimental results show that our algorithm outperforms prior stateof-the-art methods, for both single VIO frames and automatically generated, large field-of-view enhanced mosaics. We have made the corresponding dataset and source code freely available online.

Original languageEnglish (US)
Pages (from-to)327-339
Number of pages13
JournalBiomedical Optics Express
Volume3
Issue number2
DOIs
StatePublished - Feb 1 2012
Externally publishedYes

Fingerprint

Ophthalmoscopy
vessels
retina
arts
field of view
methodology
Art
Blood Vessels
Retina
Databases
Pediatrics

ASJC Scopus subject areas

  • Biotechnology
  • Atomic and Molecular Physics, and Optics

Cite this

Exploratory dijkstra forest based automatic vessel segmentation : Applications in video indirect ophthalmoscopy (VIO). / Estrada, Rolando; Tomasi, Carlo; Cabrera, Michelle T.; Wallace, David K.; Freedman, Sharon F.; Farsiu, Sina.

In: Biomedical Optics Express, Vol. 3, No. 2, 01.02.2012, p. 327-339.

Research output: Contribution to journalArticle

Estrada, Rolando ; Tomasi, Carlo ; Cabrera, Michelle T. ; Wallace, David K. ; Freedman, Sharon F. ; Farsiu, Sina. / Exploratory dijkstra forest based automatic vessel segmentation : Applications in video indirect ophthalmoscopy (VIO). In: Biomedical Optics Express. 2012 ; Vol. 3, No. 2. pp. 327-339.
@article{3075f7dc2f3f451a9f7dfa4bc77bd956,
title = "Exploratory dijkstra forest based automatic vessel segmentation: Applications in video indirect ophthalmoscopy (VIO)",
abstract = "We present a methodology for extracting the vascular network in the human retina using Dijkstra's shortest-path algorithm. Our method preserves vessel thickness, requires no manual intervention, and follows vessel branching naturally and efficiently. To test our method, we constructed a retinal video indirect ophthalmoscopy (VIO) image database from pediatric patients and compared the segmentations achieved by ourmethod and state-of-the-art approaches to a human-drawn gold standard. Our experimental results show that our algorithm outperforms prior stateof-the-art methods, for both single VIO frames and automatically generated, large field-of-view enhanced mosaics. We have made the corresponding dataset and source code freely available online.",
author = "Rolando Estrada and Carlo Tomasi and Cabrera, {Michelle T.} and Wallace, {David K.} and Freedman, {Sharon F.} and Sina Farsiu",
year = "2012",
month = "2",
day = "1",
doi = "10.1364/BOE.3.000327",
language = "English (US)",
volume = "3",
pages = "327--339",
journal = "Biomedical Optics Express",
issn = "2156-7085",
publisher = "The Optical Society",
number = "2",

}

TY - JOUR

T1 - Exploratory dijkstra forest based automatic vessel segmentation

T2 - Applications in video indirect ophthalmoscopy (VIO)

AU - Estrada, Rolando

AU - Tomasi, Carlo

AU - Cabrera, Michelle T.

AU - Wallace, David K.

AU - Freedman, Sharon F.

AU - Farsiu, Sina

PY - 2012/2/1

Y1 - 2012/2/1

N2 - We present a methodology for extracting the vascular network in the human retina using Dijkstra's shortest-path algorithm. Our method preserves vessel thickness, requires no manual intervention, and follows vessel branching naturally and efficiently. To test our method, we constructed a retinal video indirect ophthalmoscopy (VIO) image database from pediatric patients and compared the segmentations achieved by ourmethod and state-of-the-art approaches to a human-drawn gold standard. Our experimental results show that our algorithm outperforms prior stateof-the-art methods, for both single VIO frames and automatically generated, large field-of-view enhanced mosaics. We have made the corresponding dataset and source code freely available online.

AB - We present a methodology for extracting the vascular network in the human retina using Dijkstra's shortest-path algorithm. Our method preserves vessel thickness, requires no manual intervention, and follows vessel branching naturally and efficiently. To test our method, we constructed a retinal video indirect ophthalmoscopy (VIO) image database from pediatric patients and compared the segmentations achieved by ourmethod and state-of-the-art approaches to a human-drawn gold standard. Our experimental results show that our algorithm outperforms prior stateof-the-art methods, for both single VIO frames and automatically generated, large field-of-view enhanced mosaics. We have made the corresponding dataset and source code freely available online.

UR - http://www.scopus.com/inward/record.url?scp=84863964004&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84863964004&partnerID=8YFLogxK

U2 - 10.1364/BOE.3.000327

DO - 10.1364/BOE.3.000327

M3 - Article

C2 - 22312585

AN - SCOPUS:84863964004

VL - 3

SP - 327

EP - 339

JO - Biomedical Optics Express

JF - Biomedical Optics Express

SN - 2156-7085

IS - 2

ER -