Bright Retinal Lesions Detection using Color Fundus Images Containing Reflective Features

Recently, the research community has developed many techniques to detect and diagnose diabetic retinopathy with retinal fundus images. This is a necessary step for the implementation of a large scale screening effort in rural areas where ophthalmologists are not available. In the United States of America, the incidence of diabetes is increasing among the young population. Retina fundus images of patients younger than 20 years old present a high amount of reflectance due to the Nerve Fibre Layer (NFL). Generally, the younger the patient the more the reflectance is visible. We are not aware of algorithms able to explicitly deal with this type of artifact. This paper presents a technique to detect bright lesions in patients with a high degree of reflective NFL. First, the candidate bright lesions are detected using image equalization and histogram analysis. Then, a classifier is trained using texture descriptors (Multi-scale Local Binary Patterns) and other statistical features in order to remove the false positives in the lesion detection. Finally, the area of the lesions is used to diagnose diabetic retinopathy. Our database consists of 33 images from a telemedicine network currently under active development. When determining moderate to severe diabetic retinopathy using the bright lesions detected, the algorithm achieves a sensitivity of 100% at a specificity of 100% with a leave-one-out test.