A Random Forest classifier-based approach in the detection of abnormalities in the retina
Classification of abnormalities from medical images using computer-based approaches is of growing interest in medical imaging. Timely detection of abnormalities due to diabetic retinopathy and age-related macular degeneration is required in order to prevent the prognosis of the disease. Computer-aided systems using machine learning are becoming interesting to ophthalmologists and researchers. We present here one such technique, the Random Forest classifier, to aid medical practitioners in accurate diagnosis of the diseases. A computer-aided diagnosis system is proposed for detecting retina abnormalities, which combines K means-based segmentation of the retina image, after due preprocessing, followed by machine learning techniques, using several low level and statistical features. Abnormalities in the retina that are classified are caused by age-related macular degeneration and diabetic retinopathy. Performance measures used in the analysis are accuracy, sensitivity, specificity, F-measure, and Mathew correlation coefficient. A comparison with another machine learning technique, the Naïve Bayes classifier shows that the classification achieved by Random Forest classifier is 93.58% and it outperforms Naïve Bayes classifier which yields an accuracy of 83.63%.
KeywordsRandom Forest classifier Naïve Bayes classifier Diabetic retinopathy images Age-related macular degeneration, K-means clustering
The authors wish to acknowledge medical experts for fruitful consultations during the course of this work.
The authors received support for this research from a grant from Department of Biotechnology, Government of India (No.BT/PR4256/BID/7/393/2012 dated 02.08.2012).
- 1.American Academy of Ophthalmology. Diabetic Retinopathy, Preferred Practice Pattern. San Francisco: American Academy of Ophthalmology, 2008. Available at: http://www.aao.org/ppp
- 3.Crabb JW, Miyagi M, Gu X, Shadrach K, West KA, Sakaguchi H, Kamei M, Hasan A, Yan L, Rayborn ME, Salomon RG, Hollyfield JG (2002) Drusen proteome analysis: an approach to the etiology of age-related macular degeneration. Proc Natl Acad Sci U S A 99(23):14682–14687CrossRefPubMedPubMedCentralGoogle Scholar
- 6.Abbadi NKE, Saadi EHA, Automatic detection of exudates in retinal images, IJCSI International Journal of Computer Science Issues 2013, Vol. 10, Issue 2, No 1, pp: 237–242Google Scholar
- 7.Akila T, Kavitha G, Detection and classification of hard exudates in human retinal fundus images using Clustering and Random Forest methods, International Journal of Emerging Technology and Advanced Engineering 2014, Vol. 4, Special Issue 2, pp. 24–29Google Scholar
- 8.Jayakumari C, Santhanam T (2007) Detection of hard exudates for diabetic retinopathy using contextual clustering and fuzzy ART neural network. Asian J Information Technol 6(8):842–846Google Scholar
- 11.R. Scherer, Multiple fuzzy classification schemes studies in fuzziness and soft computing (springer), 228 (2012)Google Scholar
- 12.Jayanthi D, Devi N, SwarnaParvathi S (2010) Automatic diagnosis of retinal diseases from color retinal images. Int J Computer Sci Information Security 7(1):234–238Google Scholar
- 23.Kavakiotis I, Tsave O, Salifoglou A, Maglaveras N, Vlahavas I, Chouvarda I, Machine learning and data mining methods in Diabetes Research, Computational and Structural Biotechnology Journal 2017, Elsevier, Vol 15, pp. 104–116Google Scholar
- 25.A. Lang, A. Carass, E. Sotirchos, P. Calabresi, J.L. Prince, Segmentation of retinal OCT images using a random forest classifier, Proceedings of SPIE-the International Society for Optical Engineering 2013, 8869: 1667494, available on: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3660978/
- 27.Saha R, RoyChowdhury A, Banerjee S. Diabetic retinopathy related lesions detection and classification using machine learning technology. Springer International Publishing Switzerland, Artificial Intelligence and Soft Computing (ICAISC), 2016, Vol. 9693, pp. 734–745Google Scholar
- 29.Decencière E, et al. TeleOphta: Machine learning and image processing methods for teleophthalmology. IRBM (2013), https://doi.org/10.1016/j.irbm.2013.01.010, TeleOphta: Machine learning and image processing methods for teleophthalmology
- 30.Kauppi T, Kalesnykiene V, et al. DIARETDB0-standard diabetic retinopathy database, calibration level 0. IMAGERET project 2007,http://www.it.lut.fi/project/imageret/diaretdb0
- 31.Kauppi T, Kalesnykiene V, et al. DIARETDB1-standard diabetic retinopathy database, calibration level 1. IMAGERET projects 2007, http://www.it.lut.fi/project/imageret/diaretdb1
- 33.RoyChowdhury A, Banerjee S, Segmentation of retina images to detect abnormalities arising from diabetic retinopathy. [preprint]Google Scholar
- 34.Liu D, Yu J, Otsu method and K-means, Ninth International Conference on Hybrid Intelligent Systems 2009, IEEE Computer Society, Vol. 1, pp. 344–349Google Scholar
- 35.Roychowdhury A, Banerjee S, Random Forests in the classification of diabetic retinopathy retinal images, Advanced Computational and Communication Paradigms, Proceedings of International Conference on Advanced Computational and Communication Paradigms (ICACCP-2017), Vol. 1, pp. 168–176Google Scholar
- 36.Breiman L (2001) Random Forests, machine learning. Springer 45(1):5–32Google Scholar
- 37.Witten I, Frank E, Hall M. Data mining: practical machine learning tools and techniques.3rd Edition, Morgan Kaufmann, 2011Google Scholar