Cataract surgery innovations

This is the authors' abstract. We don't add key points for this paper.

With cataract surgery being the most commonly performed of all ophthalmic surgeries worldwide, it is not surprising that several exciting innovations continue to flood the scenes in an attempt to provide better visual outcomes and enhance patient safety. Interesting innovations in recent times include a variety of technologies involving intraocular lenses (IOLs), cataract surgery devices, cataract treatment platforms, and the role of artificial intelligence (AI) in IOL power calculations. How does the role of AI help to enhance cataract management? Engaging AI as a telemedicine diagnostic portal using slit-lamp images and fundus pictures in screening and diagnosis of cataract has been widely explored. Development of deep learning–based networks using fundus images in the diagnosis of diabetic retinopathy in eyes with cataracts is seen to be of considerable application value in clinical diagnosis.[1] Comparison of automatic and manual segmentation deep transform learning platforms showed the automatic segmentation platform to stage cataracts more quickly, while the manual segmentation platform staged cataracts with increased accuracy.[2] Several AI-assisted automatic cataract grading program-based slit-lamp and retro illumination lens clinical photographs based on lens opacities classification system III that have been validated have shown accurate and precise detection and grading of various types of lens opalescence.[3,4] However, as simulated cataracts have demonstrated significant impedance to the diagnostic capabilities of AI, it is imperative that incorporation of ocular comorbidities such as cataracts in the various AI study models will help to better equip AI for real-world clinical challenges.[5,6] Postoperative residual refractive error prediction seems to be more accurate with the latest AI-integrated IOL power calculation formulas. AI-based IOL calculation formulae such as the Hill-RBF formula and the European Society of Cataract and Refractive Surgery (ESCRS) calculator, which employs web-scraping technology to enable access to different IOL power calculation formulas, are now available. Hill-RBF formula engages AI-based pattern recognition technology combined with sophisticated data interpolation to compute accurate IOL power predictions. The ESCRS calculator, a free online tool, provides major online IOL calculators on a single platform, enabling comparison of seven calculators (Barrett Universal-II, Hill-RBF, Hoffer QST, Kane, Pearl GDS, EVO, and Cooke K6). Various new IOL power calculation formulas have been evaluated in depth in recent times for their performance in normal and variable ocular biometry setting. While the Kane formula and the Barrett Universal-II formula are applauded as better performers in comparison to others over the entire axial length range, the Kane formula has been noted to be the most accurate in long eyes and both Kane and EVO 2.0 formulas are better for short axial lengths.[7] Kane and EVO 2.0 formulas were also noted to be better predictors of postoperative refractive outcome in toric IOL implantations in comparison to other formulas (Holladay 1, Haigis, Hoffer Q, Hoffer QST [Savini/Taroni], Ladas Super, Sanders–Retzlaff–Kraff/theoretical, and T2 IOL power formulas).[8] Comparison of AI-based IOL power calculation formulas (XGBoost, Hill-RBF, and Kane) with traditional IOL formulas in highly myopic eyes showed that they were more accurate than the traditional vergence formulas. They also have been placed higher in rank than the newer generations of formulas such as Barrett and Olsen and formulas based on Wang Koch axial length modifications.[9] Adopting AI in surgical skill training in cataract surgery has been advocated to be useful in the video analysis of various cataract surgery phases, and it enhances the workflow in operating rooms by envisaging the surgical duration.[10] Another area where AI can be of potential use in cataract scenario is its application in convolutional neural network models to predict posterior capsule opacification progression and requirement for Yttrium Aluminum Garnet (YAG) laser capsulotomy.[10] Challenges in AI application to transform cataract management to enhance ophthalmic services are ethical data management and related logistics, lack of external validations, and developing reliable, real-world clinical scenario–compatible applications for populations worldwide.[10,11] Featuring among the other top picks for cataract surgery-related advances in recent times are Premium IOLs, extended depth of focus (EDOF) IOLs, and light-adjustable lens (LAL). Premium IOLs include multifocal and accommodative IOLs. Along with astigmatic correction with toric IOLs, they enable clear vision at both distance and near without spectacle aids. Non-diffractive EDOF IOLs and enhanced monofocal IOLs are now increasingly available, with many manufacturers coming into play in recent times. With multiple IOL manufactures in the foray, the advent of newer advances such as proprietary wavefront linking technology aims to provide better far and intermediate ranges with less photopic disturbances compared to traditional monofocal IOLs.[12] Of the recent advances, the other most exciting one is LAL,[13] which allows for adjustments of its power with an ultraviolet light delivery device allowing customized refractive outcomes. LAL implantation is done by standard cataract surgery, following which the ultraviolet light treatments with the light delivery device can bring about change in the shape and prescription of the lens, thereby allowing power adjustments in the follow-up clinic. While LALs make the possibility of perfect postoperative vision a reality, considerations include topography and aberrations in preoperative planning to understand indications, achieving optimal post-implantation refractive target, which is essential for accurate evaluation of residual myopia or hyperopic power and wavefront changes to effect desired refractive outcome with light treatment. LAL implantation in post-refractive cataract surgery scenarios is yet to provide consistent outcome. Defining postoperative workflow patterns with enhanced understanding of the changes in optical characteristics post-light treatment are new learning aspects required in adopting LAL into cataract practice. In conclusion, recent cataract practice pattern is constantly evolving toward more predictable refractive outcomes and enhancing post-surgery refractive outcomes with adoption of AI technologies at all phases of cataract planning and surgery.