Age-Related Macular Degeneration Screening Using Data Mining Approaches

This paper investigates the use of tabular approach to classifying retinal images according to whether they feature Age-related Macular Degeneration (AMD), a retina condition that causes blindness in old age. The novelty of the proposed approach is that it is not founded on feature segmentation, instead entire image encodings are used. Features in the form of statistical parameters extracted directly and indirectly from the images are considered. For the evaluation two publically available, retinal fundus image data sets were used. The evaluation was conducted in the context of AMD screening. Excellent results were produced: Sensitivity and AUC of 90% and over were recorded for binary-class classification problem.