A ring selection platform for treatment of keratoconus

Keratoconus is an eye disease in which the cornea thins and, due to intraocular pressure, loses its symmetrical shape, tending to be conical. These changes in the shape of the cornea cause disturbances in the refraction of light and its entry into the eye, leading to decreased visual clarity and vision problems. As one of the best ways to treat this disease in its advanced stages, ophthalmologists use intracorneal ring segment implantation. Indeed, they implant a suitable ring segment in the patient's cornea to increase the strength of the cornea and help it return to its original symmetrical state. To help ophthalmologists select a suitable ring segment for a specific patient with unique characteristics of keratoconus, creating a platform based on data derived from mechanical simulations of intrastromal corneal ring segment implantation and a machine learning approach trained on this data would be highly beneficial. In this research, at first, the cornea model of a patient was prepared using ophthalmological data, and after assigning its properties according to previous research, simulations of intracorneal ring segment implantation were performed with various values of basic parameters such as ring segment radius, ring segment angle, ring segment implantation depth, and ring segment position. Then, with the help of a genetic algorithm and the extraction of Zernike coefficients, parameters related to the quality of vision after surgery were obtained for each simulation. Two algorithms, random forest and deep learning, were trained on the simulation data, and a platform was created that predicts the surgical outcome for the same patient by varying parameters. The accuracy of predictions was evaluated using metrics such as mean squared error, root mean squared error, and coefficient of determination. By combining mechanical simulations and machine learning, this approach offers a practical, data-driven solution to improve treatment planning for keratoconus patients. It enables ophthalmologists to personalize treatments, leading to better surgical outcomes and enhanced quality of life for patients.