Detection of retinal microaneurysms using fractal analysis and feature extraction technique

In this paper a novel method for the improvement in the candidature detection of microaneurysm (MA) in fundus images has been proposed. In automated screening of diabetic retinopathy it becomes necessary to detect the MAs, which appear as small red dots on retinal fundus images to give the earliest possible sign of the disease. The proposed algorithm consists of two stages. The first stage comprises of image preprocessing and fractal analysis of retinal vascular structure. The effectiveness of the automated screening program gets increased as fractal analysis differentiates normal retina image form the abnormal one. Second stage aims at detection of a typical shape of MAs as the abnormal retinal image goes through canny edge detection and morphological reconstruction. True micro-aneurysms are discriminated from other retinal features based on the analysis of binary object parameters on each segmented candidate. The proposed algorithm has been applied on a set of 89 color fundus images from a published database. The implemented algorithm has achieved a best operating sensitivity of 89.5% and a specificity of 82.1% which makes it feasible for diabetic retinopathy screening.

[1]  Andrew K. C. Wong,et al.  A new method for gray-level picture thresholding using the entropy of the histogram , 1985, Comput. Vis. Graph. Image Process..

[2]  Peter F. Sharp,et al.  Automated microaneurysm detection using local contrast normalization and local vessel detection , 2006, IEEE Transactions on Medical Imaging.

[3]  P F Sharp,et al.  An image-processing strategy for the segmentation and quantification of microaneurysms in fluorescein angiograms of the ocular fundus. , 1996, Computers and biomedical research, an international journal.

[4]  Sheng Chen,et al.  A clustering technique for digital communications channel equalization using radial basis function networks , 1993, IEEE Trans. Neural Networks.

[5]  Lionel Marraffa,et al.  PARAMETRIC STUDY OF THERMAL AND CHEMICAL NONEQUILIBRIUM NOZZLE FLOW , 1991 .

[6]  Keith E. Muller,et al.  Contrast-limited adaptive histogram equalization: speed and effectiveness , 1990, [1990] Proceedings of the First Conference on Visualization in Biomedical Computing.

[7]  András Hajdu,et al.  Microaneurysm detection in retinal images using a rotating cross-section based model , 2011, 2011 IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[8]  G. Landini,et al.  Local Connected Fractal Dimensions and Lacunarity Analyses of 60 ° Fluorescein Angiograms , 2005 .

[9]  W. Doyle,et al.  Magnetization reversal in films with biaxial anisotropy , 1966 .

[10]  Gwénolé Quellec,et al.  Optimal Wavelet Transform for the Detection of Microaneurysms in Retina Photographs , 2008, IEEE Transactions on Medical Imaging.

[11]  Adrienne Csutak,et al.  A multi-level ensemble-based system for detecting microaneurysms in fundus images , 2010, 4th International Workshop on Soft Computing Applications.

[12]  Chikkannan Eswaran,et al.  An Automated Blood Vessel Segmentation Algorithm Using Histogram Equalization and Automatic Threshold Selection , 2011, Journal of Digital Imaging.

[13]  Richard E. Haskell,et al.  Transient signal propagation in lossless, isotropic plasmas , 1967 .

[14]  Peter Kulchyski and , 2015 .

[15]  J. Olson,et al.  Automated detection of microaneurysms in digital red‐free photographs: a diabetic retinopathy screening tool , 2000, Diabetic medicine : a journal of the British Diabetic Association.