Related Experiment Video
Updated: Jan 18, 2026

Comparison of Agreement and Accuracy using Binocular Wavefront Optometer with Autorefractor and Phoropter
Published on: September 16, 2025
Accuracy of PEARL-DGS Formula for Intraocular Lens Power Calculation in Patients With Previous Myopic Laser Vision
Piero Zollet1,2, Federico Macario1,2, Marco Trevisi1,3
1From the Department of Biomedical Sciences, Humanitas University, Pieve Emanuele, Milan, Italy.
A new machine learning intraocular lens (IOL) formula, PEARLS-DGS, shows comparable accuracy to existing methods for cataract surgery patients with prior myopic laser vision correction. Its performance in predicting refractive outcomes is reliable, offering a viable option for refractive prediction.
Area of Science:
- Ophthalmology
- Biomedical Engineering
- Artificial Intelligence in Healthcare
Background:
- Accurate intraocular lens (IOL) power calculation is crucial for visual outcomes after cataract surgery.
- Patients with a history of myopic laser refractive surgery present unique challenges for standard IOL formulas due to altered corneal curvature.
- Novel formulas are needed to improve refractive prediction accuracy in this specific patient population.
Purpose of the Study:
- To evaluate the accuracy of the PEARLS-DGS, a novel machine learning-based open-source IOL formula.
- To compare PEARLS-DGS performance against established IOL formulas in eyes with prior myopic laser refractive surgery.
- To assess refractive prediction error and its distribution in 100 patients undergoing uncomplicated cataract surgery.
Main Methods:
- Retrospective analysis of 100 patients with a history of photorefractive keratectomy or LASIK.
- Comparison of PEARLS-DGS with Shammas, Haigis-L, Barrett True-K, ASCRS calculator average, EVO, and Hoffer QST formulas.
- Primary outcome measures included absolute refractive prediction error and cumulative distribution of error at various thresholds.
Main Results:
- EVO 2.0 demonstrated the lowest median absolute error (0.36 D), followed by Hoffer QST (0.38 D) and PEARLS-DGS (0.41 D).
- At the ±0.50 D threshold, Hoffer QST (0.65) and PEARLS-DGS (0.61) showed strong performance in cumulative distribution.
- No statistically significant differences were found between most formulas, except between Shammas and Hoffer QST at the ±0.50 D threshold.
Conclusions:
- The PEARLS-DGS IOL formula exhibits comparable accuracy to existing formulas for eyes with prior myopic laser vision correction.
- PEARLS-DGS demonstrates reliable refractive prediction error distribution, performing well against established methods like Hoffer QST.
- This study supports PEARLS-DGS as a potentially valuable tool for IOL calculations in post-refractive surgery eyes.
Related Concept Videos
Focusing of Light in the Eye
Angle Closure Glaucoma: Treatment
Open Angle Glaucoma: Treatment
Drugs such as carbonic anhydrase inhibitors, α2- and...

