Investigating the Prediction Accuracy of Recently Updated Intraocular Lens Power Formulas with Artificial Intelligence for High Myopia

The aim of this study was to investigate the prediction accuracy of intraocular lens (IOL) power formulas with artificial intelligence (AI) for high myopia. Cases of highly myopic patients (axial length [AL], >26.0 mm) undergoing uncomplicated cataract surgery with at least 1-month follow-up were included. Prediction errors, absolute errors, and percentages of eyes with prediction errors within +/- 0.25, +/- 0.50, and +/- 1.00 diopters (D) were compared using five formulas: Hill-RBF3.0, Kane, Barrett Universal II (BUII), Haigis, and SRK/T. Seventy eyes (mean patient age at surgery, 64.0 +/- 9.0 years; mean AL, 27.8 +/- 1.3 mm) were included. The prediction errors with the Hill-RBF3.0 and Kane formulas were statistically different from the BUII, Haigis, and SRK/T formulas, whereas there was not a statistically significant difference between those with the Hill-RBF3.0 and Kane. The absolute errors with the Hill-RBF3.0 and Kane formulas were smaller than that with the BUII formula, whereas there was not a statistically significant difference between the other formulas. The percentage within +/- 0.25 D with the Hill-RBF3.0 formula was larger than that with the BUII formula. The prediction accuracy using AI (Hill-RBF3.0 and Kane) showed excellent prediction accuracy. No significant difference was observed in the prediction accuracy between the Hill-RBF3.0 and Kane formulas.