Prediction of 9 Artificial Intelligence-based Intraocular Lens Power Calculation Formulas in Long Caucasian Eyes.

Area of Science:

• Ophthalmology
• Medical Technology
• Artificial Intelligence in Healthcare

Background:

• Accurate intraocular lens (IOL) power calculation is crucial for achieving optimal visual outcomes after cataract surgery.
• Eyes with long axial lengths (AL > 26 mm) present unique challenges for traditional IOL power calculation formulas.
• Evaluating advanced AI-based formulas is essential to improve refractive predictability in these challenging eyes.

Purpose of the Study:

• To compare the accuracy of nine novel artificial intelligence (AI)-based intraocular lens (IOL) power calculation formulas.
• To assess formula performance specifically in eyes with axial lengths greater than 26 mm.

Main Methods:

• A retrospective analysis of 321 eyes with AL ≥ 26.00 mm who underwent cataract surgery.
• IOL power was calculated using nine AI formulas: 3C 2.0, Hill-RBF 3.0, Hoffer QST, Kane, Karmona, Ladas Super Formula AI (LSF AI), Nallasamy, PEARL-DGS, and Zhu-Lu.
• Accuracy was evaluated using root mean square absolute error (RMSAE), median absolute error (MedAE), and the percentage of eyes with prediction error within ±0.50 D.

Main Results:

• All formulas except Karmona showed statistical superiority over Zhu-Lu and 3C 2.0 for RMSAE.
• Pearl-DGS, Hill-RBF 3.0, and Kane demonstrated statistically better MedAE compared to 3C 2.0 and Zhu-Lu.
• Pearl-DGS (80.06%), Hill-RBF 3.0 (77.57%), and Kane (77.26%) achieved significantly higher accuracy (within ±0.50 D PE) than Zhu-Lu (66.67%) and 3C 2.0 (68.85%).

Conclusions:

• Pearl-DGS, Hill-RBF 3.0, Kane, and Hoffer QST are highly accurate AI-based formulas for IOL power calculation in long eyes (AL > 26 mm).
• These formulas offer improved refractive predictability compared to older methods in this challenging patient cohort.
• The study highlights the potential of advanced AI algorithms in optimizing cataract surgery outcomes.