TITLE:
Artificial Intelligence Assisted Cataract Surgery: A Comparative Clinical Evaluation of Conventional and AI-Assisted Surgical Approaches
AUTHORS:
Ali Tonuzi, Orjeta Tonuzi, Migena Beqiri, Luan Qafmolla
KEYWORDS:
Artificial Intelligence, Cataract Surgery, Phacoemulsification, Intraocular Lens, Optical Coherence Tomography, Refractive Accuracy, Ophthalmology
JOURNAL NAME:
Open Journal of Ophthalmology,
Vol.16 No.3,
August
25,
2026
ABSTRACT: Artificial intelligence (AI) has emerged as an important innovation in ophthalmology, providing new opportunities to improve the diagnosis, surgical planning, and management of cataract. AI-assisted technologies have demonstrated considerable potential to enhance surgical precision, optimize intraoperative decision-making, and improve postoperative outcomes. However, evidence supporting their routine clinical application remains limited. This study aimed to evaluate the impact of AI-assisted technologies on diagnostic accuracy, surgical precision, intraoperative performance, and postoperative clinical outcomes compared with conventional cataract surgery. A combined retrospective and prospective comparative clinical study was conducted involving 98 patients who underwent cataract surgery between May 2019 and April 2026. Patients were allocated into two groups according to the surgical approach employed: Group A (conventional cataract surgery, n = 55) and Group B (AI-assisted cataract surgery, n = 43). Clinical outcomes were assessed by comparing preoperative and postoperative evaluations performed at the final follow-up visit. The primary endpoint was postoperative visual acuity (VA), while secondary endpoints included refractive accuracy, intraocular pressure (IOP), intraoperative performance, surgical precision, and postoperative complication rates. Continuous variables were analyzed using appropriate parametric or non-parametric statistical tests according to data distribution, and categorical variables were compared using Chi-square or Fisher’s exact test, with statistical significance established at P Group A surgery, the Group B approach achieved significantly better postoperative visual acuity (0.51 ± 0.13 vs. 0.42 ± 0.15 decimal units, P Group B than in the Group A (87% vs. 71%, P = 0.03), indicating improved accuracy of AI-assisted intraocular lens (IOL) power calculation. Mild postoperative complications, including transient corneal edema and temporary intraocular pressure elevation, occurred in 19% of patients in the Group B and 23% in the Group A, with no statistically significant difference between groups (P = 0.64). AI-assisted cataract surgery represents a valuable advancement in contemporary ophthalmic practice by improving refractive predictability, surgical precision, and perioperative clinical decision-making. Nevertheless, AI should be considered a complementary clinical tool that supports, rather than replaces, the expertise and clinical judgment of experienced ophthalmic surgeons. Further prospective multicenter studies with larger patient populations are warranted to confirm its long-term clinical effectiveness, safety, and cost-effectiveness.