Automatic Identification of Diabetic Retinopathy in Retinal Images Using Ensemble Learning

  • Flávio Araújo UFPI
  • Romuere Silva UFPI
  • André Macedo UFPI
  • Kelson Aires UFPI
  • Rodrigo Veras UFPI


Diabetic Retinopathy (DR) is one of the major complications of diabetes mellitus and can cause blindness. The diagnosis of DR is performed by visual analysis of retinal images being exudates (fat deposits) the mains patterns traced by a specialist doctor. This paper presents a new method for DR detection in color retinal images. The proposed algorithm combines two images classification methodologies using an ensemble learning. The first Methodology extracts the image attributes using Speeded Up Robust Features (SURF) algorithm and determines the presence of DR using a Support Vector Machine (SVM). The second Methodology consists in the classification of an image based on the segmentation of exudates regions. The experimental validation was performed on a public image database, DIARETDB1.


Agurto, C., Murray, V., Barriga, E., Murillo, S., Pattichis, M., Davis, H., Russel, S., Abràmoff, M., and Soliz, P. (2010). Multiscale am-fm methods for diabetic retinopathy lesion detection. IEEE Trans. on Medical Imaging, pages 502–512.

Bay, H., Tuytelaars, T., and Gool, L. V. (2006). Surf: Speeded up robust features. In In ECCV, pages 404–417.

Bernardes, R., Serranho, P., and Lobo, C. (2011). Digital ocular fundus imaging: A review. Ophthalmologica, 4(226):161–181.

Giancardo, L., Meriaudeau, F., Karnowski, T. P., Li, Y., Garg, S., Tobin, K. W., and Chaum, E. (2012). Exudate-based diabetic macular edema detection in fundus images using publicly available datasets. Medical Image Analysis, 16(1):216–226.

Gutterres, D. M., Maestri, M., Feller, N., Nedel, L., and Duncan, B. (2011). Análise de imagens de retinografia digital com imagej customizado. Workshop de Informática Médica.

Harangi, B., Antal, B., and Hajdu, A. (2012). Automatic exudate detection with improved naïve-bayes classifier. In IEEE International Symposium on Computer-Based Medical Systems, pages 1–6.

Haykin, S. (2001). Neural Networks: A Comprehensive Foundation. Prentice Hall, 2 edition.

Kauppi, T., Kalesnykiene, V., Kamarainen, J. K., Lensu, L., Sorri, I., Kalviainen, H., and Pietila, J. (2007). Diaretdb1 diabetic retinopathy database and evaluation protocol. In Conference on Medical Image Understanding and Analysis, pages 1–5.

Kavitha, D. and Shenbaga, S. (2005). A cellular neurofuzzy network for contrast enhancement of fundus images with retinopathies. In Proceeding ICS’06 Proceedings of the 10th WSEAS international conference on Systems.

Köse, C., Sevik, U., Ikibas, C., and Erdöl, H. (2012). Simple methods for segmentation and measurement of diabetic retinopathy lesions in retinal fundus images. In Computer Methods and Programs in Biomedicine, pages 274–293.

María, G., Clara, I. S., María, I. L., Daniel, A., and Hornero, R. (2009). Neural network based detection of hard exudates in retinal images. Computer Methods and Programs in Biomedicine, 93:9–19.

Osareh, A., Mirmehdi, M., Thomas, B., and Markham, R. (2002). Comparative exudate classification using support vector machines and neural networks. In 5th International Conference on Medical Image Computing and Computer-Assisted Intervention.

Papa, J. and Rocha, A. (2011). Image categorization through optimum path forest and visual words. In Intl. Conference on Image Processing (ICIP), pages 3586–3589.

Ponti, M. (2011). Combining classifiers: From the creation of ensembles to the decision fusion. In (SIBGRAPI-T), pages 1–10.

Rafael C. Gonzalez, R. E. W. (2008). Digital Image Process. Prentice Hall, 3 edition.

Ram, K. and Sivaswamy, J. (2009). Multi-space clustering for segmetation of exudates in retinal color photographs. In Annual Int. Conf. of the IEEE Enginnersing in Medicine and Biology, pages 1437–1440.

Rocha, A., Carvalho, T., Jelinek, H., Goldenstein, S., and Wainer, J. (2012). Points of interest and visual dictionaries for automatic retinal lesion detection. IEEE Transactions on Biomedical Engineering, 59(8):2244–2253.

Sopharak, A., Dailey, N. M., Uyyanonvara, B., Barman, S., Williamson, T., New, T. N., and Moe, A. Y. (2010). Machine learning approach to automatic exudate detection in retinal images from diabetic patients. Journal of Modern Optics, (57):124–135.

Sopharak, A., Uyyanonvara, B., and Barman, S. (2009). Automatic exudates detection from non-dilated diabetic retinopathy retinal images using c-means clustering. Sensors, pages 2148–2161.

Walter, T., Klevin, J., Massin, P., and Erginay, A. (2002). A contribution of image processing to the diagnosis of diabetic retinopathy-detection of exudates in color fundus images of the human retina. IEEE Transactions on Medical Imaging, 21(10):1236 – 1243.

Zana, F. and Klein., J. C. (2001). Segmentation of vessel-like patterns using mathematical morphology and curvature evaluation. IEEE Transactions on Image Processing, 10(7):1010 – 1019.
ARAÚJO, Flávio; SILVA, Romuere; MACEDO, André; AIRES, Kelson; VERAS, Rodrigo. Automatic Identification of Diabetic Retinopathy in Retinal Images Using Ensemble Learning. In: SIMPÓSIO BRASILEIRO DE COMPUTAÇÃO APLICADA À SAÚDE (SBCAS), 13. , 2013, Maceió/AL. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2013 . p. 1238-1247. ISSN 2763-8952.