Rights statement: The final publication is available at Springer via http://dx.doi.org/10.1007/s11760-019-01515-3
Accepted author manuscript, 1.14 MB, PDF document
Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License
Final published version
Research output: Contribution to Journal/Magazine › Journal article › peer-review
Research output: Contribution to Journal/Magazine › Journal article › peer-review
}
TY - JOUR
T1 - GGM classifier with multi-scale line detectors for retinal vessel segmentation
AU - Khan, M.A.U.
AU - Khan, T.M.
AU - Naqvi, S.S.
AU - Aurangzeb Khan, M.
N1 - The final publication is available at Springer via http://dx.doi.org/10.1007/s11760-019-01515-3
PY - 2019/11/1
Y1 - 2019/11/1
N2 - Persistent changes in the diameter of retinal blood vessels may indicate some chronic eye diseases. Computer-assisted change observation attempts may become challenging due to the emergence of interfering pathologies around blood vessels in retinal fundus images. The end result is lower sensitivity to thin vessels for certain computerized detection methods. Quite recently, multi-scale line detection method proved to be worthy for improved sensitivity toward lower-caliber vessels detection. This happens largely due to its adaptive property that responds more to the longevity patterns than width of a given vessel. However, the method suffers from the lack of a better aggregation process for individual line detectors. This paper investigates a scenario that introduces a supervised generalized Gaussian mixture classifier as a robust solution for the aggregate process. The classifier is built with class-conditional probability density functions as a logistic function of linear mixtures. To boost the classifier’s performance, the weighted scale images are modeled as Gaussian mixtures. The classifier is trained with weighted images modeled on a Gaussian mixture. The net effect is increased sensitivity for small vessels. The classifier’s performance has been tested with three commonly available data sets: DRIVE, SATRE, and CHASE_DB1. The results of the proposed method (with an accuracy of 96%, 96.1% and 95% on DRIVE, STARE, and CHASE_DB1, respectively) demonstrate its competitiveness against the state-of-the-art methods and its reliability for vessel segmentation.
AB - Persistent changes in the diameter of retinal blood vessels may indicate some chronic eye diseases. Computer-assisted change observation attempts may become challenging due to the emergence of interfering pathologies around blood vessels in retinal fundus images. The end result is lower sensitivity to thin vessels for certain computerized detection methods. Quite recently, multi-scale line detection method proved to be worthy for improved sensitivity toward lower-caliber vessels detection. This happens largely due to its adaptive property that responds more to the longevity patterns than width of a given vessel. However, the method suffers from the lack of a better aggregation process for individual line detectors. This paper investigates a scenario that introduces a supervised generalized Gaussian mixture classifier as a robust solution for the aggregate process. The classifier is built with class-conditional probability density functions as a logistic function of linear mixtures. To boost the classifier’s performance, the weighted scale images are modeled as Gaussian mixtures. The classifier is trained with weighted images modeled on a Gaussian mixture. The net effect is increased sensitivity for small vessels. The classifier’s performance has been tested with three commonly available data sets: DRIVE, SATRE, and CHASE_DB1. The results of the proposed method (with an accuracy of 96%, 96.1% and 95% on DRIVE, STARE, and CHASE_DB1, respectively) demonstrate its competitiveness against the state-of-the-art methods and its reliability for vessel segmentation.
KW - Diabetic retinopathy
KW - Image segmentation
KW - Retinal images
KW - Vessel segmentation
KW - Classification (of information)
KW - Digital storage
KW - Eye protection
KW - Gaussian distribution
KW - Large scale systems
KW - Ophthalmology
KW - Probability density function
KW - Scales (weighing instruments)
KW - Conditional probability density
KW - Retinal blood vessels
KW - Retinal fundus images
KW - Retinal image
KW - Retinal vessel segmentations
KW - State-of-the-art methods
KW - Blood vessels
U2 - 10.1007/s11760-019-01515-3
DO - 10.1007/s11760-019-01515-3
M3 - Journal article
VL - 13
SP - 1667
EP - 1675
JO - Signal, Image and Video Processing
JF - Signal, Image and Video Processing
SN - 1863-1703
IS - 8
ER -