In this chapter, we have seen different ways of computer assistance to the diagnosis of diabetic retinopathy, which is a very frequent and severe eye-disease: image enhancement, mass screening, and monitoring. Different algorithms within this framework have been presented and evaluated with encouraging results.
However, there are still improvements to be made. The first one is to use high-resolution images. We worked on images already used in centers of ophthalmology, but it is clear that acquisition techniques also improve and that in the coming years high-resolution images will become clinical standard. Future segmentation algorithm can make use of this high resolution (e.g. there will be more features for microaneurysm detection).
Another possible research axis is the inclusion of patient data into the algorithms. This a priori knowledge about the patient is used by physicians; it also could be used by automatic methods. For instance, we have observed, that the color of black people's eyes is quite different from the color of white people's, the color of a child's retina is different from the color of an adult's eye. This is precious information that could be used in order to enhance the performance of lesions detection algorithms.
Even if there is still progress to be made, the presented algorithms work well; a clinical trial is envisaged.
Was this article helpful?