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

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

To evaluate the efficacy of a corneal tomography Imaging model in predicting postoperative vault based on preoperative corneal topography in Implantable Collamer Lens (ICL) surgery.

Accurate vault prediction is crucial for Implantable Collamer Lens (ICL) surgery safety and efficacy. Current methods using preoperative biometrics and regression formulas show limited accuracy due to parameter variability and incomplete utilization of corneal topography data. To address this, we developed a deep learning model that predicts postoperative vault while generating anterior chamber morphology images from preoperative data, enabling personalized surgical planning.

Inclusion Criteria:(1) stable myopia (≤0.50D/year change for 2 years), (2) ACD ≥2.80mm, (3) intact corneal endothelium (≥2000 cells/mm²), and (4) no confounding ocular/systemic conditions.

Exclusion Criteria:(1) glaucoma-spectrum disorders or retinal vasculopathies, (2) prior corneal/intraocular surgery, (3) compromised corneal endothelium, (4) uncontrolled systemic diseases, and (5) pregnancy/lactation.