Bashir Bagheri Nakhjavanlo - دانشکده کامپیوتر و ریاضیات، دانشگاه آزاد اسلامی واحد فیروزکوه، فیروزکوه، ایران
Monireh Ayari - دانشکده کامپیوتر، دانشگاه آزاد اسلامی واحد کرج، کرج، ایران
Nima Aberomand - دانشکده کامپیوتر، دانشگاه آزاد اسلامی واحد شهرقدس، تهران، ایران - دانشکده کامپیوتر - دانشگاه تگزاس در آرلینگتون - تگزاس - امریکا

The presence of disturbances in the cornea of the eye leads to structural changes in this area which changes the eye from a spherical cone-shaped form and it indicating the Keratoconus. This condition leads to corneal cavity in some areas of the eye, causing problems such as vision loss, blurred vision, irregular astigmatism, sensitivity to light and etc. in the person's vision system. People with a thin cornea should not have laser surgery on their eyes, as this will cause a sharp increase in the cornea. Various methods have been used to treat corneal keratinization, one of them is implantation of rings. Certainly, any treatment can cause a series of complications in individuals which is not excluded from the implantation of the ring for the treatment of cornea. By recognizing complications such as changing people's vision, one can choose the type of rings and make a bi-directional pattern. Considering a number of features in rigging, including corneal depth, rim thickness, and rim angle, can help with cure and non-complications in people with corneal inclination. This research attempts to present a simulation method based on the implanting characteristics of a ring in a set of eye images that have cornea ventricles that can predict complications after planting the rings. The use of image processing principles as well as artificial neural network structures is considered in this research. According to the scientific viewpoint of an ophthalmologist, information can be provided to the training system and then received after processing results. Simulation results indicate that the proposed approach is similar to the previous one and more optimized.

Keratoconus, Planting Rings, Image Processing, Neural Networks