Artery/vein classification of retinal blood vessels using feature selection

Vishal Vijayakumar, Dara Koozekanani, Robert White, James Kohler, Sohini Roychowdhury, Keshab K Parhi

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

5 Citations (Scopus)

Abstract

Automated classification of retinal vessels in fundus images is the first step towards measurement of retinal characteristics that can be used to screen and diagnose vessel abnormalities for cardiovascular and retinal disorders. This paper presents a novel approach to vessel classification to compute the artery/vein ratio (AVR) for all blood vessel segments in the fundus image. The features extracted are then subjected to a selection procedure using Random Forests (RF) where the features that contribute most to classification accuracy are chosen as input to a polynomial kernel Support Vector Machine (SVM) classifier. The most dominant feature was found to be the vessel information obtained from the Light plane of the LAB color space. The SVM is then subjected to one time training using 10-fold cross validation on images randomly selected from the VICAVR dataset before testing on an independent test dataset, derived from the same database. An Area Under the ROC Curve (AUC) of 97.2% was obtained on an average of 100 runs of the algorithm. The proposed algorithm is robust due to the feature selection procedure, and it is possible to get similar accuracies across many datasets.

Original languageEnglish (US)
Title of host publication2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1320-1323
Number of pages4
Volume2016-October
ISBN (Electronic)9781457702204
DOIs
StatePublished - Oct 13 2016
Event38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016 - Orlando, United States
Duration: Aug 16 2016Aug 20 2016

Other

Other38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016
CountryUnited States
CityOrlando
Period8/16/168/20/16

Fingerprint

Retinal Vessels
Blood vessels
Feature extraction
Veins
Arteries
Support vector machines
Cardiovascular Abnormalities
ROC Curve
Area Under Curve
Blood Vessels
Classifiers
Color
Polynomials
Databases
Light
Testing
Datasets
Support Vector Machine

Cite this

Vijayakumar, V., Koozekanani, D., White, R., Kohler, J., Roychowdhury, S., & Parhi, K. K. (2016). Artery/vein classification of retinal blood vessels using feature selection. In 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016 (Vol. 2016-October, pp. 1320-1323). [7590950] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/EMBC.2016.7590950

Artery/vein classification of retinal blood vessels using feature selection. / Vijayakumar, Vishal; Koozekanani, Dara; White, Robert; Kohler, James; Roychowdhury, Sohini; Parhi, Keshab K.

2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016. Vol. 2016-October Institute of Electrical and Electronics Engineers Inc., 2016. p. 1320-1323 7590950.

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

Vijayakumar, V, Koozekanani, D, White, R, Kohler, J, Roychowdhury, S & Parhi, KK 2016, Artery/vein classification of retinal blood vessels using feature selection. in 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016. vol. 2016-October, 7590950, Institute of Electrical and Electronics Engineers Inc., pp. 1320-1323, 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016, Orlando, United States, 8/16/16. https://doi.org/10.1109/EMBC.2016.7590950
Vijayakumar V, Koozekanani D, White R, Kohler J, Roychowdhury S, Parhi KK. Artery/vein classification of retinal blood vessels using feature selection. In 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016. Vol. 2016-October. Institute of Electrical and Electronics Engineers Inc. 2016. p. 1320-1323. 7590950 https://doi.org/10.1109/EMBC.2016.7590950
Vijayakumar, Vishal ; Koozekanani, Dara ; White, Robert ; Kohler, James ; Roychowdhury, Sohini ; Parhi, Keshab K. / Artery/vein classification of retinal blood vessels using feature selection. 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016. Vol. 2016-October Institute of Electrical and Electronics Engineers Inc., 2016. pp. 1320-1323
@inproceedings{77222b7d81fd44229a214ef17042e10c,
title = "Artery/vein classification of retinal blood vessels using feature selection",
abstract = "Automated classification of retinal vessels in fundus images is the first step towards measurement of retinal characteristics that can be used to screen and diagnose vessel abnormalities for cardiovascular and retinal disorders. This paper presents a novel approach to vessel classification to compute the artery/vein ratio (AVR) for all blood vessel segments in the fundus image. The features extracted are then subjected to a selection procedure using Random Forests (RF) where the features that contribute most to classification accuracy are chosen as input to a polynomial kernel Support Vector Machine (SVM) classifier. The most dominant feature was found to be the vessel information obtained from the Light plane of the LAB color space. The SVM is then subjected to one time training using 10-fold cross validation on images randomly selected from the VICAVR dataset before testing on an independent test dataset, derived from the same database. An Area Under the ROC Curve (AUC) of 97.2{\%} was obtained on an average of 100 runs of the algorithm. The proposed algorithm is robust due to the feature selection procedure, and it is possible to get similar accuracies across many datasets.",
author = "Vishal Vijayakumar and Dara Koozekanani and Robert White and James Kohler and Sohini Roychowdhury and Parhi, {Keshab K}",
year = "2016",
month = "10",
day = "13",
doi = "10.1109/EMBC.2016.7590950",
language = "English (US)",
volume = "2016-October",
pages = "1320--1323",
booktitle = "2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Artery/vein classification of retinal blood vessels using feature selection

AU - Vijayakumar, Vishal

AU - Koozekanani, Dara

AU - White, Robert

AU - Kohler, James

AU - Roychowdhury, Sohini

AU - Parhi, Keshab K

PY - 2016/10/13

Y1 - 2016/10/13

N2 - Automated classification of retinal vessels in fundus images is the first step towards measurement of retinal characteristics that can be used to screen and diagnose vessel abnormalities for cardiovascular and retinal disorders. This paper presents a novel approach to vessel classification to compute the artery/vein ratio (AVR) for all blood vessel segments in the fundus image. The features extracted are then subjected to a selection procedure using Random Forests (RF) where the features that contribute most to classification accuracy are chosen as input to a polynomial kernel Support Vector Machine (SVM) classifier. The most dominant feature was found to be the vessel information obtained from the Light plane of the LAB color space. The SVM is then subjected to one time training using 10-fold cross validation on images randomly selected from the VICAVR dataset before testing on an independent test dataset, derived from the same database. An Area Under the ROC Curve (AUC) of 97.2% was obtained on an average of 100 runs of the algorithm. The proposed algorithm is robust due to the feature selection procedure, and it is possible to get similar accuracies across many datasets.

AB - Automated classification of retinal vessels in fundus images is the first step towards measurement of retinal characteristics that can be used to screen and diagnose vessel abnormalities for cardiovascular and retinal disorders. This paper presents a novel approach to vessel classification to compute the artery/vein ratio (AVR) for all blood vessel segments in the fundus image. The features extracted are then subjected to a selection procedure using Random Forests (RF) where the features that contribute most to classification accuracy are chosen as input to a polynomial kernel Support Vector Machine (SVM) classifier. The most dominant feature was found to be the vessel information obtained from the Light plane of the LAB color space. The SVM is then subjected to one time training using 10-fold cross validation on images randomly selected from the VICAVR dataset before testing on an independent test dataset, derived from the same database. An Area Under the ROC Curve (AUC) of 97.2% was obtained on an average of 100 runs of the algorithm. The proposed algorithm is robust due to the feature selection procedure, and it is possible to get similar accuracies across many datasets.

UR - http://www.scopus.com/inward/record.url?scp=85009072543&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85009072543&partnerID=8YFLogxK

U2 - 10.1109/EMBC.2016.7590950

DO - 10.1109/EMBC.2016.7590950

M3 - Conference contribution

VL - 2016-October

SP - 1320

EP - 1323

BT - 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -