Vessel Delineation in Retinal Images using Leung-Malik filters and Two Levels Hierarchical Learning

Ehsan S. Varnousfaderani, Siamak Yousefi, Christopher Bowd, Akram Belghith, Michael H. Goldbaum

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Blood vessel segmentation is important for the analysis of ocular fundus images for diseases affecting vessel caliber, occlusion, leakage, inflammation, and proliferation. We introduce a novel supervised method to evaluate performance of Leung-Malik filters in delineating vessels. First, feature vectors are extracted for every pixel with respect to the response of Leung-Malik filters on green channel retinal images in different orientations and scales. A two level hierarchical learning framework is proposed to segment vessels in retinal images with confounding disease abnormalities. In the first level, three expert classifiers are trained to delineate 1) vessels, 2) background, and 3) retinal pathologies including abnormal pathologies such as lesions and anatomical structures such as optic disc. In the second level, a new classifier is trained to detect vessels and non-vessel pixels based on results of the expert classifiers. Qualitative evaluation shows the effectiveness of the proposed expert classifiers in modeling retinal pathologies. Quantitative results on two standard datasets STARE (AUC = 0.971, Acc=0.927) and DRIVE (AUC = 0.955, Acc =0.903) are comparable with other state-of-the-art vessel segmentation methods.

Original languageEnglish (US)
Pages (from-to)1140-1147
Number of pages8
JournalAMIA ... Annual Symposium proceedings. AMIA Symposium
Volume2015
StatePublished - Jan 1 2015
Externally publishedYes

Fingerprint

Learning
Pathology
Area Under Curve
Retinal Vessels
Optic Disk
Blood Vessels
Inflammation
Datasets

All Science Journal Classification (ASJC) codes

  • Medicine(all)

Cite this

Vessel Delineation in Retinal Images using Leung-Malik filters and Two Levels Hierarchical Learning. / Varnousfaderani, Ehsan S.; Yousefi, Siamak; Bowd, Christopher; Belghith, Akram; Goldbaum, Michael H.

In: AMIA ... Annual Symposium proceedings. AMIA Symposium, Vol. 2015, 01.01.2015, p. 1140-1147.

Research output: Contribution to journalArticle

Varnousfaderani, Ehsan S. ; Yousefi, Siamak ; Bowd, Christopher ; Belghith, Akram ; Goldbaum, Michael H. / Vessel Delineation in Retinal Images using Leung-Malik filters and Two Levels Hierarchical Learning. In: AMIA ... Annual Symposium proceedings. AMIA Symposium. 2015 ; Vol. 2015. pp. 1140-1147.
@article{61464774a48445bc8661e08724186ce5,
title = "Vessel Delineation in Retinal Images using Leung-Malik filters and Two Levels Hierarchical Learning",
abstract = "Blood vessel segmentation is important for the analysis of ocular fundus images for diseases affecting vessel caliber, occlusion, leakage, inflammation, and proliferation. We introduce a novel supervised method to evaluate performance of Leung-Malik filters in delineating vessels. First, feature vectors are extracted for every pixel with respect to the response of Leung-Malik filters on green channel retinal images in different orientations and scales. A two level hierarchical learning framework is proposed to segment vessels in retinal images with confounding disease abnormalities. In the first level, three expert classifiers are trained to delineate 1) vessels, 2) background, and 3) retinal pathologies including abnormal pathologies such as lesions and anatomical structures such as optic disc. In the second level, a new classifier is trained to detect vessels and non-vessel pixels based on results of the expert classifiers. Qualitative evaluation shows the effectiveness of the proposed expert classifiers in modeling retinal pathologies. Quantitative results on two standard datasets STARE (AUC = 0.971, Acc=0.927) and DRIVE (AUC = 0.955, Acc =0.903) are comparable with other state-of-the-art vessel segmentation methods.",
author = "Varnousfaderani, {Ehsan S.} and Siamak Yousefi and Christopher Bowd and Akram Belghith and Goldbaum, {Michael H.}",
year = "2015",
month = "1",
day = "1",
language = "English (US)",
volume = "2015",
pages = "1140--1147",
journal = "AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium",
issn = "1559-4076",
publisher = "American Medical Informatics Association",

}

TY - JOUR

T1 - Vessel Delineation in Retinal Images using Leung-Malik filters and Two Levels Hierarchical Learning

AU - Varnousfaderani, Ehsan S.

AU - Yousefi, Siamak

AU - Bowd, Christopher

AU - Belghith, Akram

AU - Goldbaum, Michael H.

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Blood vessel segmentation is important for the analysis of ocular fundus images for diseases affecting vessel caliber, occlusion, leakage, inflammation, and proliferation. We introduce a novel supervised method to evaluate performance of Leung-Malik filters in delineating vessels. First, feature vectors are extracted for every pixel with respect to the response of Leung-Malik filters on green channel retinal images in different orientations and scales. A two level hierarchical learning framework is proposed to segment vessels in retinal images with confounding disease abnormalities. In the first level, three expert classifiers are trained to delineate 1) vessels, 2) background, and 3) retinal pathologies including abnormal pathologies such as lesions and anatomical structures such as optic disc. In the second level, a new classifier is trained to detect vessels and non-vessel pixels based on results of the expert classifiers. Qualitative evaluation shows the effectiveness of the proposed expert classifiers in modeling retinal pathologies. Quantitative results on two standard datasets STARE (AUC = 0.971, Acc=0.927) and DRIVE (AUC = 0.955, Acc =0.903) are comparable with other state-of-the-art vessel segmentation methods.

AB - Blood vessel segmentation is important for the analysis of ocular fundus images for diseases affecting vessel caliber, occlusion, leakage, inflammation, and proliferation. We introduce a novel supervised method to evaluate performance of Leung-Malik filters in delineating vessels. First, feature vectors are extracted for every pixel with respect to the response of Leung-Malik filters on green channel retinal images in different orientations and scales. A two level hierarchical learning framework is proposed to segment vessels in retinal images with confounding disease abnormalities. In the first level, three expert classifiers are trained to delineate 1) vessels, 2) background, and 3) retinal pathologies including abnormal pathologies such as lesions and anatomical structures such as optic disc. In the second level, a new classifier is trained to detect vessels and non-vessel pixels based on results of the expert classifiers. Qualitative evaluation shows the effectiveness of the proposed expert classifiers in modeling retinal pathologies. Quantitative results on two standard datasets STARE (AUC = 0.971, Acc=0.927) and DRIVE (AUC = 0.955, Acc =0.903) are comparable with other state-of-the-art vessel segmentation methods.

UR - http://www.scopus.com/inward/record.url?scp=84981720579&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84981720579&partnerID=8YFLogxK

M3 - Article

VL - 2015

SP - 1140

EP - 1147

JO - AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium

JF - AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium

SN - 1559-4076

ER -