Width-wise vessel bifurcation for improved retinal vessel segmentation

Tariq M. Khan; Mohammad A.U. Khan; Naveed Ur Rehman; Khuram Naveed; Imran Uddin Afridi; Syed Saud Naqvi; Imran Raazak

doi:10.1016/j.bspc.2021.103169

Width-wise vessel bifurcation for improved retinal vessel segmentation

Tariq M. Khan, Mohammad A.U. Khan, Naveed Ur Rehman, Khuram Naveed^*, Imran Uddin Afridi, Syed Saud Naqvi, Imran Raazak

^*Corresponding author for this work

Department of Electrical and Computer Engineering - Signal Processing and Machine learning

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

3 Citations (Scopus)

Abstract

Vessel local characteristics such as noise, illumination, and direction vary significantly in a fundus image, making it difficult to segment the vessel tree structure as a whole. To facilitate vessel detection, an alternative procedure proposed here, whereby retinal vessels first classified into two categories, large and small. Then, for its unique characteristics, each group has been processed with its own enhancement and detection filter. The sensitivity of the proposed method is boosted by capturing tiny vessels through a directional filter bank followed by its associated triple-stick filtering. Additionally, the specificity of the proposed method is enhanced through noise suppression attributed largely to the proposed BM3D filtering and multi-scale line detection approach. As a result, the detection accuracy on the DRIVE, STARE, and CHASE DB1 datasets is significantly improved, with scores of 0.9610, 0.9586, and 0.9578, respectively.

Original language	English
Article number	103169
Journal	Biomedical Signal Processing and Control
Volume	71
Issue	Part A
Number of pages	11
ISSN	1746-8094
DOIs	https://doi.org/10.1016/j.bspc.2021.103169
Publication status	Published - Jan 2022

Keywords

Diabetic retinopathy
Multiscale line detector
Retinal vessels
Triple stick filter

Access to Document

10.1016/j.bspc.2021.103169

Cite this

@article{ebd5690f59874cfcac4ccaedd0d44f81,

title = "Width-wise vessel bifurcation for improved retinal vessel segmentation",

abstract = "Vessel local characteristics such as noise, illumination, and direction vary significantly in a fundus image, making it difficult to segment the vessel tree structure as a whole. To facilitate vessel detection, an alternative procedure proposed here, whereby retinal vessels first classified into two categories, large and small. Then, for its unique characteristics, each group has been processed with its own enhancement and detection filter. The sensitivity of the proposed method is boosted by capturing tiny vessels through a directional filter bank followed by its associated triple-stick filtering. Additionally, the specificity of the proposed method is enhanced through noise suppression attributed largely to the proposed BM3D filtering and multi-scale line detection approach. As a result, the detection accuracy on the DRIVE, STARE, and CHASE DB1 datasets is significantly improved, with scores of 0.9610, 0.9586, and 0.9578, respectively.",

keywords = "Diabetic retinopathy, Multiscale line detector, Retinal vessels, Triple stick filter",

author = "Khan, {Tariq M.} and Khan, {Mohammad A.U.} and Rehman, {Naveed Ur} and Khuram Naveed and Afridi, {Imran Uddin} and Naqvi, {Syed Saud} and Imran Raazak",

note = "Publisher Copyright: {\textcopyright} 2021 Elsevier Ltd",

year = "2022",

month = jan,

doi = "10.1016/j.bspc.2021.103169",

language = "English",

volume = "71",

journal = "Biomedical Signal Processing and Control",

issn = "1746-8094",

publisher = "Elsevier BV",

number = "Part A",

}

TY - JOUR

T1 - Width-wise vessel bifurcation for improved retinal vessel segmentation

AU - Khan, Tariq M.

AU - Khan, Mohammad A.U.

AU - Rehman, Naveed Ur

AU - Naveed, Khuram

AU - Afridi, Imran Uddin

AU - Naqvi, Syed Saud

AU - Raazak, Imran

PY - 2022/1

Y1 - 2022/1

N2 - Vessel local characteristics such as noise, illumination, and direction vary significantly in a fundus image, making it difficult to segment the vessel tree structure as a whole. To facilitate vessel detection, an alternative procedure proposed here, whereby retinal vessels first classified into two categories, large and small. Then, for its unique characteristics, each group has been processed with its own enhancement and detection filter. The sensitivity of the proposed method is boosted by capturing tiny vessels through a directional filter bank followed by its associated triple-stick filtering. Additionally, the specificity of the proposed method is enhanced through noise suppression attributed largely to the proposed BM3D filtering and multi-scale line detection approach. As a result, the detection accuracy on the DRIVE, STARE, and CHASE DB1 datasets is significantly improved, with scores of 0.9610, 0.9586, and 0.9578, respectively.

AB - Vessel local characteristics such as noise, illumination, and direction vary significantly in a fundus image, making it difficult to segment the vessel tree structure as a whole. To facilitate vessel detection, an alternative procedure proposed here, whereby retinal vessels first classified into two categories, large and small. Then, for its unique characteristics, each group has been processed with its own enhancement and detection filter. The sensitivity of the proposed method is boosted by capturing tiny vessels through a directional filter bank followed by its associated triple-stick filtering. Additionally, the specificity of the proposed method is enhanced through noise suppression attributed largely to the proposed BM3D filtering and multi-scale line detection approach. As a result, the detection accuracy on the DRIVE, STARE, and CHASE DB1 datasets is significantly improved, with scores of 0.9610, 0.9586, and 0.9578, respectively.

KW - Diabetic retinopathy

KW - Multiscale line detector

KW - Retinal vessels

KW - Triple stick filter

UR - http://www.scopus.com/inward/record.url?scp=85115230666&partnerID=8YFLogxK

U2 - 10.1016/j.bspc.2021.103169

DO - 10.1016/j.bspc.2021.103169

M3 - Journal article

AN - SCOPUS:85115230666

SN - 1746-8094

VL - 71

JO - Biomedical Signal Processing and Control

JF - Biomedical Signal Processing and Control

IS - Part A

M1 - 103169

ER -

Width-wise vessel bifurcation for improved retinal vessel segmentation

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this