Predictive Performance of a Generative Model for Corneal Tomography After Implantable Collamer Lens Implantation

What this Trial Is About

This research aims to assess the effectiveness of a corneal tomography imaging model in predicting postoperative vault after Implantable Collamer Lens (ICL) surgery. Accurate prediction of vault is important for the safety and success of ICL surgery. Current prediction methods are limited due to variable parameters and incomplete use of corneal topography data. The study evaluates a deep learning model that predicts vault and creates images of the anterior chamber from preoperative data, helping to personalize surgical planning. The study observes eyes that have undergone ICL surgery performed by experienced surgeons. The corneal tomography generation model analyzes the surgical data to assess its performance in predicting vault, including measures of accuracy, area under the curve (AUC), sensitivity, and specificity. This study is observational and focuses on evaluating this diagnostic model rather than testing a treatment. Participants will have their preoperative corneal topography data collected and analyzed by the model. Researchers will measure the primary outcome of the model's ability to predict vault seven days after surgery, using AUROC (Area Under the Receiver Operating Characteristic curve). Secondary outcomes include sensitivity and specificity of the model's vault prediction. The study involves participants aged 18 to 45 years with stable myopia and healthy corneal endothelium, with no confounding ocular or systemic conditions. Participation duration and follow-up details are based on the postoperative evaluation at day 7.

Predictive Performance of a Generative Model for Corneal Tomography After ICL Implantation