Article type
Abstract
"Background: Hyperopia, a widespread vision issue that is expected to affect a significant portion of the world's population, can complicate the development of cataracts and present difficulties in determining accurate intraocular lens (IOL) power. The introduction of AI-based technologies, such as Pearl-DGS, the Hill-RBF and Kane formulas, has demonstrated potential in enhancing the precision of IOL power calculations, leading to a more efficient and effective cataract surgery process for patients with short axial length myopia.
Objective: To systematically compare and rank the accuracy of AI-based intraocular lens (IOL) power calculation formulas and traditional IOL formulas in eyes with short axial lengths.
Methods: We conducted a comprehensive literature review of studies published up to January 2024 in PubMed, Web of Science, Embase, and Cochrane Library databases. Outcomes assessed included mean absolute error (MAE), percentage of eyes with refractive prediction error (PE) within ±0.25, ±0.50, and ±1.00 diopters (D), and median absolute error (MedAE). Network meta-analysis was performed using R 4.3.0 and STATA 17.0.
Results: The analysis included 11 studies involving 1,621 eyes with short axial lengths that underwent uncomplicated cataract surgery with mono-focal IOL implantation. The top three AI-based formulas, according to the surface under the cumulative ranking curve (SUCRA) values, were Pearl-DGS and Okulix. These formulas demonstrated the lowest MedAE and superior accuracy compared to traditional vergence formulas like SRK/T, Holladay 1, Holladay 2, Haigis, and Hoffer Q, especially in terms of MAE and PE within ±0.25, ±0.50, and ±1.00 D.
Conclusions: For eyes with short axial lengths, the most accurate AI-based IOL power calculation formulas are Pearl-DGS and Okulix which outperform traditional vergence formulas."
Objective: To systematically compare and rank the accuracy of AI-based intraocular lens (IOL) power calculation formulas and traditional IOL formulas in eyes with short axial lengths.
Methods: We conducted a comprehensive literature review of studies published up to January 2024 in PubMed, Web of Science, Embase, and Cochrane Library databases. Outcomes assessed included mean absolute error (MAE), percentage of eyes with refractive prediction error (PE) within ±0.25, ±0.50, and ±1.00 diopters (D), and median absolute error (MedAE). Network meta-analysis was performed using R 4.3.0 and STATA 17.0.
Results: The analysis included 11 studies involving 1,621 eyes with short axial lengths that underwent uncomplicated cataract surgery with mono-focal IOL implantation. The top three AI-based formulas, according to the surface under the cumulative ranking curve (SUCRA) values, were Pearl-DGS and Okulix. These formulas demonstrated the lowest MedAE and superior accuracy compared to traditional vergence formulas like SRK/T, Holladay 1, Holladay 2, Haigis, and Hoffer Q, especially in terms of MAE and PE within ±0.25, ±0.50, and ±1.00 D.
Conclusions: For eyes with short axial lengths, the most accurate AI-based IOL power calculation formulas are Pearl-DGS and Okulix which outperform traditional vergence formulas."