AI-powered formulas enhance accuracy of intraocular lens calculations for highly myopic patients
Artificial intelligence (AI)-powered intraocular lens (IOL) calculation formulas, particularly the Kane formula, provide significantly more accurate results than traditional formulas for highly myopic patients undergoing cataract surgery, according to a study.
This study compared 6 IOL calculation formulas: Kane, Hill-RBF, EVO, Barrett II, Haigis, and SRK/T. The evaluation metrics included the mean absolute error (MAE) and the percentage of errors within ± 0.25 diopters (D), ± 0.50 D, and ± 1.00 D.
The study’s findings indicate that the Kane formula outperforms others in accuracy. The MAE of Kane was significantly lower than that of Barrett II (mean difference = -0.03 D, P = 0.02), SRK/T (MD = -0.08 D, P = 0.02), and Haigis (MD = -0.12 D, P < 0.00001). In addition, Kane demonstrated a higher percentage of refractive prediction errors within ± 0.25 D, ± 0.50 D, and ± 1.00 D compared to SRK/T and Haigis. While no significant differences were found between Hill-RBF and Barrett, Hill-RBF performed better than SRK/T and Haigis.
Reference
Hao Y, Fu J, Huang J, Chen D. Comparing the accuracy of intraocular lens power calculation formulas using artificial intelligence and traditional formulas in highly myopic patients: a meta-analysis. Int Ophthalmol. 2024;44(1):242. doi: 10.1007/s10792-024-03227-1. PMID: 38904666.
