Possibilities of Application of Neuro-Fuzzy Networks for Ophthalmologic Image Classification

The use of artificial intelligence technologies in image analysis tasks is rapidly developing every year, thanks to which the possibilities of using artificial neural networks in the diagnosis of various diseases have reached a completely new level. The results of thousands of studies using neural networks in ophthalmology show diagnostic accuracy results comparable and sometimes even superior to those achieved by a human performer. However, the transparency of the results obtained and the adaptability of the operation of such systems remain low. In the future, the use of convolutional neuro-fuzzy networks in image analysis tasks will probably improve the results. The work examines the existing experience in using neuro-fuzzy networks for problems of classification of medical images and demonstrates the possibilities of practical implementation of a neuro-fuzzy network for problems of classification of glaucoma ophthalmic images.