|
|
Hard Exudates Detection Method Based on Background-Estimation and SVM Classifier |
1 School of Electronics and Information Engineering, Tianjin Polytechnic University, Tianjin 300387, China
2 Tianjin Medical University Eye Hospital, Tianjin 300384, China
3 Tianjin Medical University Metabolic Disease Hospital, Tianjin 300070, China |
|
|
Abstract Hard exudates (HE) are early symptoms of diabetic retinopathy (DR) and main symptom of macular edema. Hence, HE detection is very important for clinical diagnosis. In this paper, a new method based on backgroundestimation and SVM classifier for hard exudates detection is presented. Firstly, foreground map containing all bright objects is obtained by background-estimation. The HE candidates are gotten using the edge information based on Kirsch operator, and then the optic disc is removed. Finally, the shape features, histogram statistic features and phase features of the HE candidates are extracted before using the SVM classifier so that the accurate extraction of HE is obtained. Our method has been tested on the public databases of DIARETDB1 and HEI-MED. The experiment results show that the method’s sensitivity is 97.3% and the specificity is 90% at the image level, the mean sensitivity is 84.6% and the mean predictive value is 94.4% at the lesion level. The performance of the proposed method shows considerable efficiency for hard exudates detection.
|
|
|
|
|
[1]Walter T, Klein JC, Massin P, et al. A contribution of image processing to the diagnosis of diabetic retinopathy detection of exudates in color fundus images of the human retina [J]. IEEE Transactions on Medical Imaging, 2002, 21(10): 1236-1243.
[2]Welfer D, Scharcanski J, Marinho DR. A coarsetofine strategy for automatically detecting exudates in color eye fundus images [J]. Computerized Medical Imaging Graphics, 2010, 34(3): 228-235.
[3]Gandhi M, Dhanasekaran R. Diagnosis of diabetic retinopathy using morphological process and SVM classifier[C] //International Conference on Communication and Signal Processing. Washington: IEEE, 2013: 873-877.
[4]Ravishankar S, Jain A, Mittal A. Automated feature extraction for early detection of diabetic retinopathy in fundus images[C] //Computer Society Conference on Computer Vision and Pattern Recognition Workshops. Piscataway: IEEE, 2009: 210-217.
[5]Sopharak A, Uyyanonvara B, Barman S. Automatic exudate detection for diabetic retinopathy screening [J]. Science Asia, 2009, (35): 80-88.
[6]Zhang Xiaohui, Chutatape O. Topdown and bottomup strategies in lesion detection of background diabetic retinopathy[C] //Proceedings of 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Diego: IEEE, 2005 (2): 422-428.
[7]Jayakumari C, Maruthi R. Detection of hard exudates in color fundus images of the human retina [J]. Procedia Engineering, 2012,(30): 297-302.
[8]Wang Huan, Hsu W, Goh KG. An effective approach to detect lesions in color retinal images[C] //IEEE Conference on Computer Vision and Pattern Recognition. Hilton Head Island: IEEE, 2000: 181-186.
[9]Osareh A. Automated identification of diabetic retinal exudates and the optic disc [D]. Bristol: University of Bristol, 2004.
[10]Sánchez C, Hornero R, López M. A novel automatic image processing algorithm for detection of hard exudates based on retinal image analysis [J]. Medical Engineering and Physics, 2008, 30(3): 350-357.
[11]García M, Sánchez C, López M. Neural network based detection of hard exudates in retinal images [J]. Computer Methods and Programs in Biomedicine, 2009, 93(1): 9-19.
[12]Niemeijer M, Ginneken B, Russell SR, et al. Automated detection and differentiation of drusen, exudates and cottonwool spots in digital color fundus photographs for diabetic retinopathy diagnosis [J]. Investigative Ophthalmology & Visual Science, 2007, 48(5): 2260-2267.
[13]Sae\|Tang W, Chiracharit W, Kumwilaisak W. Exudates detection in fundus image using nonuniform illumination background subtraction[C] //TENCON 2010 IEEE Region 10 Conference. Piscataway: Institute of Electrical and Electronics Engineers, 2010: 204-209.
[14]Sagar AV, Balasubramaniam S, Chandrasekaran V. A Novel Integrated Approach Using Dynamic Thresholding and Edge Detection (IDTED) for Automatic Detection of Exudates in Digital Fundus Retinal Images[C] //Proceedings of the International Conference on Computing: Theory and Applications. Kolkata: IEEE, 2007: 705-710.
[15]Fleming AD, Philip S, Goatman KA, et al. Automated microaneurysm detection using local contrast normalization and local vessel detection[J]. IEEE Transactions on Medical Imaging, 2006, 25: 1223-1232.
[16]Vincent L. Morphological grayscale reconstruction in image analysis: applications and efficient algorithms[J]. IEsEE Transactions on Image Processing, 1993, 2(2): 176-201.
[17]Kirsch RA. Computer determination of the constituent structure of biological images[J]. Computers and Biomedical Research, 1971, 4(3): 315-328.
[18]Sánchez CI, García M, Mayo A. Retinal image analysis based on mixture models to detect hard exudates[J]. Medical Image Analysis, 2009, 13(4): 650-658.
[19]Morrone MC, Owens RA. Feature detection from local energy [J]. Pattern Recognition Letters, 1987, 6(5): 303-313.
[20]Venkatesh S, Owens RA. An energy feature detection scheme [C] //Proceedings of International Conference on Image Processing. Singapore City: IEEE, 1989: 553-557.
[21]Kovesi P. Image features from phase congruency [J]. Journal of Computer Vision Research, 1999, 1(3): 1-26.
[22]Kauppi T, Kalesnykiene V, Kamarainen J. The DIARETDB1 diabetic retinopathy database and evaluation protocol [C] //Proceedings of the British Machine Vision Conference. Warwick: British Machine Vision Association, 2007: 61-65.
[23]Giancardo L, Meriaudeau F. Exudatebased diabetic macular edema detection in fundus images using publicly available datasets[J]. Medical Image Analysis, 2012, 16(1): 216-226.
[24]Akobeng A. Understanding diagnostic tests 2: likelihood ratios, pre and posttest probabilities and their use in clinical practice [J]. Acta Paediatrica, 2007, 96(4):487-491.
|
|
|
|