Research Article Open Access

Adaptive Neuro-Fuzzy Inference System Approach for the Automatic Screening of Diabetic Retinopathy in Fundus Images

S. Kavitha1 and K. Duraiswamy1
  • 1 ,
Journal of Computer Science
Volume 7 No. 7, 2011, 1020-1026

DOI: https://doi.org/10.3844/jcssp.2011.1020.1026

Submitted On: 8 May 2011 Published On: 28 June 2011

How to Cite: Kavitha, S. & Duraiswamy, K. (2011). Adaptive Neuro-Fuzzy Inference System Approach for the Automatic Screening of Diabetic Retinopathy in Fundus Images. Journal of Computer Science, 7(7), 1020-1026. https://doi.org/10.3844/jcssp.2011.1020.1026

Abstract

Problem statement: Diabetic retinopathy is one of the most significant factors contributing to blindness and so early diagnosis and timely treatment is particularly important to prevent visual loss. Approach: An integrated approach for extraction of blood vessels and exudates detection was proposed to screen diabetic retinopathy. An automated classifier was developed based on Adaptive Neuro-Fuzzy Inference System (ANFIS) to differentiate between normal and nonproliferative eyes from the quantitative assessment of monocular fundus images. Feature extraction was performed on the preprocessed fundus images. Structure of Blood vessels was extracted using Multiscale analysis. Hard Exudates were detected using CIE Color channel transformation, Entropy Thresholding and Improved Connected Component Analysis from the fundus images. Features like Wall to Lumen ratio in blood vessels, Texture, Homogeneity properties and area occupied by Hard Exudates, were given as input to ANFIS.ANFIS was trained with Back propagation in combination with the least squares method. Proposed method was evaluated on 200 real time images comprising 70 normal and 130 retinopathic eyes. Results and Conclusion: All of the results were validated with ground truths obtained from expert ophthalmologists. Quantitative performance of the method, detected exudates with an accuracy of 99.5%. Receiver operating characteristic curve evaluated for real time images produced better results compared to the other state of the art methods. ANFIS provides best classification and can be used as a screening tool in the analysis and diagnosis of retinal images.

  • 930 Views
  • 2,904 Downloads
  • 4 Citations

Download

Keywords

  • Monocular fundus
  • CIE color space
  • connected component analysis
  • Computer-Aided Diagnosis (CAD)
  • Gabor filter
  • automatic exudates
  • optic disk
  • lipids leaking
  • blood vessel
  • fundus images
  • True Negative (TN)