Evaluation of blood vessel detection methods

R. Sadeghzadeh; M. Berks; S. M. Astley; C. J. Taylor

doi:10.1117/12.878723

Evaluation of blood vessel detection methods

R. Sadeghzadeh, M. Berks, S. M. Astley, C. J. Taylor

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

Abstract

We address the problem of evaluating the performance of algorithms for detecting curvilinear structures in medical images. As an exemplar we consider the detection of vessel trees which contain structures of variable width and contrast. Results for the conventional approach to evaluation, in which the detector output is compared directly with a groundtruth mask, tend to be dominated by the detection of large vessels and fail to capture adequately whether or not finer, lower contrast vessels have been detected successfully. We propose and investigate three alternative evaluation strategies. We demonstrate the use of the standard and new evaluation strategies to assess the performance of a novel method for detecting vessels in retinograms, using the publicly available DRIVE database. © 2011 Copyright Society of Photo-Optical Instrumentation Engineers (SPIE).

Original language	English
Title of host publication	Progress in Biomedical Optics and Imaging - Proceedings of SPIE\|Progr. Biomed. Opt. Imaging Proc. SPIE
Volume	7962
DOIs	https://doi.org/10.1117/12.878723
Publication status	Published - 2011
Event	Medical Imaging 2011: Image Processing - Lake Buena Vista, FL Duration: 1 Jul 2011 → …

Conference

Conference	Medical Imaging 2011: Image Processing
City	Lake Buena Vista, FL
Period	1/07/11 → …

Keywords

blood vessels
detection
Evaluation
retinograms
ROC analysis

Access to Document

10.1117/12.878723

Cite this

@inproceedings{445d7fc0885e48a386fcd9e7801216ad,

title = "Evaluation of blood vessel detection methods",

abstract = "We address the problem of evaluating the performance of algorithms for detecting curvilinear structures in medical images. As an exemplar we consider the detection of vessel trees which contain structures of variable width and contrast. Results for the conventional approach to evaluation, in which the detector output is compared directly with a groundtruth mask, tend to be dominated by the detection of large vessels and fail to capture adequately whether or not finer, lower contrast vessels have been detected successfully. We propose and investigate three alternative evaluation strategies. We demonstrate the use of the standard and new evaluation strategies to assess the performance of a novel method for detecting vessels in retinograms, using the publicly available DRIVE database. {\textcopyright} 2011 Copyright Society of Photo-Optical Instrumentation Engineers (SPIE).",

keywords = "blood vessels, detection, Evaluation, retinograms, ROC analysis",

author = "R. Sadeghzadeh and M. Berks and Astley, {S. M.} and Taylor, {C. J.}",

note = "Sadeghzadeh, R. Berks, M. Astley, S. M. Taylor, C. J. Conference on Medical Imaging 2011 - Image Processing Feb 14-16, 2011 Lake Buena Vista, FL Dynasil Corp/RMD Res, Amer Assoc Physicists Med, DQE Instruments, Inc, Ocean Thin Films, Inc, Univ Cent Florida, CREOL - Coll Opt & Photon, VIDA Diagnost, Inc, SPIE; Medical Imaging 2011: Image Processing ; Conference date: 01-07-2011",

year = "2011",

doi = "10.1117/12.878723",

language = "English",

isbn = "9780819485045",

volume = "7962",

booktitle = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE|Progr. Biomed. Opt. Imaging Proc. SPIE",

}

TY - GEN

T1 - Evaluation of blood vessel detection methods

AU - Sadeghzadeh, R.

AU - Berks, M.

AU - Astley, S. M.

AU - Taylor, C. J.

N1 - Sadeghzadeh, R. Berks, M. Astley, S. M. Taylor, C. J. Conference on Medical Imaging 2011 - Image Processing Feb 14-16, 2011 Lake Buena Vista, FL Dynasil Corp/RMD Res, Amer Assoc Physicists Med, DQE Instruments, Inc, Ocean Thin Films, Inc, Univ Cent Florida, CREOL - Coll Opt & Photon, VIDA Diagnost, Inc, SPIE

PY - 2011

Y1 - 2011

N2 - We address the problem of evaluating the performance of algorithms for detecting curvilinear structures in medical images. As an exemplar we consider the detection of vessel trees which contain structures of variable width and contrast. Results for the conventional approach to evaluation, in which the detector output is compared directly with a groundtruth mask, tend to be dominated by the detection of large vessels and fail to capture adequately whether or not finer, lower contrast vessels have been detected successfully. We propose and investigate three alternative evaluation strategies. We demonstrate the use of the standard and new evaluation strategies to assess the performance of a novel method for detecting vessels in retinograms, using the publicly available DRIVE database. © 2011 Copyright Society of Photo-Optical Instrumentation Engineers (SPIE).

AB - We address the problem of evaluating the performance of algorithms for detecting curvilinear structures in medical images. As an exemplar we consider the detection of vessel trees which contain structures of variable width and contrast. Results for the conventional approach to evaluation, in which the detector output is compared directly with a groundtruth mask, tend to be dominated by the detection of large vessels and fail to capture adequately whether or not finer, lower contrast vessels have been detected successfully. We propose and investigate three alternative evaluation strategies. We demonstrate the use of the standard and new evaluation strategies to assess the performance of a novel method for detecting vessels in retinograms, using the publicly available DRIVE database. © 2011 Copyright Society of Photo-Optical Instrumentation Engineers (SPIE).

KW - blood vessels

KW - detection

KW - Evaluation

KW - retinograms

KW - ROC analysis

U2 - 10.1117/12.878723

DO - 10.1117/12.878723

M3 - Conference contribution

SN - 9780819485045

VL - 7962

BT - Progress in Biomedical Optics and Imaging - Proceedings of SPIE|Progr. Biomed. Opt. Imaging Proc. SPIE

T2 - Medical Imaging 2011: Image Processing

Y2 - 1 July 2011

ER -

Evaluation of blood vessel detection methods

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this