Development of a new intraocular lens power calculation method based on lens position estimated with optical coherence tomography

Abstract

A new method is developed and validated for intraocular lens (IOL) power calculation based on paraxial ray tracing of the postoperative IOL positions, which are obtained with the use of anterior segment optical coherence tomography. Of the 474 eyes studied, 137 and 337 were grouped into training and validation sets, respectively. The positions of the implanted IOLs of the training datasets were characterized with multiple linear regression analyses one month after the operations. A new regression formula was developed to predict the postoperative anterior chamber depth with the use of the stepwise analysis results. In the validation dataset, postoperative refractive values were calculated according to the paraxial ray tracing of the cornea and lens based on the assumption of finite structural thicknesses with separate surface curvatures. the predicted refraction error was calculated as the difference of the expected postoperative refraction from the spherical-equivalent objective refraction values. The percentage error (within ±0.50 diopters) of the new formula was 84.3%. This was not significantly correlated to the axial length or keratometry. The developed formula yielded excellent postoperative refraction predictions and could be applicable to eyes with abnormal proportions, such as steep or flat corneal curvatures and short and long axial lengths.open Scientific RepoRtS | (2020) 10:6501 | https://doi.org/10.1038/s41598-020-63546-ywww.nature.com/scientificreports www.nature.com/scientificreports/ previously reported the relationship between the crystalline lens parameter and the IOL position 18 . Goto et al. 19 developed a formula to predict postoperative ACD, including the angle-to-angle (ATA) depth obtained by presented a method to accurately estimate and quantify the entire shape of the crystalline lens from OCT data, and reported that the preoperative anterior segment parameters were exceptionally valuable for estimating the IOL position. Yoo et al. 17 reported improvements in refractive outcomes by the ray-tracing method using the equatorial plane position of the lens.Paraxial ray tracing repeats simple calculations without special software or complexity. The predictive accuracy of IOL power calculations with paraxial ray tracing including the IOL positions predicted by the AS-OCT has not been clarified.Correspondingly, the aims of the present study are a) the development of a new formula based on paraxial ray tracing for IOL power calculations that utilize the postoperative IOL positions obtained using AS-OCT, and b) the validation of the accuracy of the developed formula. Scientific RepoRtS |(2020) 10:6501 | https://doi.

Use of a Machine Learning Method in Predicting Refraction after Cataract Surgery

“…The current study compared the accuracy of conventional IOL power calculation formulas with that of ML-based methods in predicting postoperative refraction. The absolute prediction errors of the conventional formulas were nearly the same as those of previous reports [ 3 , 4 , 5 ]. In addition, the Barrett Universal II formula had low prediction error, which is in accordance with previous studies.…”

“…Future studies are needed, such as using data from other races. Alternatively, the newly proposed formula using AS-OCT by Satou et al that is independent of ocular axial length may be optimal [ 4 ]. However, it is complicated to obtain detailed AS-OCT data accurately from all preoperative cataract patients.…”

“…Satou et al have recently reported a formula that uses detailed anatomical measurements of the anterior eye using anterior segment Optical Coherence Tomography (AS-OCT). Their formula shows high accuracy without being affected by the axial length of the eye [ 4 ]. In addition, several papers that used numerous intraocular lens (IOL) power calculation formulas in predicting postoperative refraction have been published, and recent reports have indicated that the Barrett Universal II formula has high accuracy [ 5 , 6 , 7 ].…”

Determining the Theoretical Effective Lens Position of Thick Intraocular Lenses for Machine Learning–Based IOL Power Calculation and Simulation

“…These methods are usually based on the parameters of optical coherence tomography, biometry, and anterior segment optical coherence tomography acquired before and after cataract surgery. 14,[28][29][30] In a work evaluating the prediction accuracy of ELP after cataract surgery using a multiobjective evolutionary algorithm, 31 the ELP was defined as the distance from the cornea to the anterior surface of the IOL at 3 months after surgery plus the distance to the presumed principal object point of the IOLs. The predicted ELP was obtained by a multiobjective evolutionary algorithm and multiple regression analysis, including a dozen parameters, and the difference between the predicted value and measured value (prediction accuracy) was compared between the two methods.…”

AI-powered effective lens position prediction improves the accuracy of existing lens formulas

“…In view of the limitations of the ELP in existing formulas, recently, more efforts have been devoted to constructing ELPs that better reflect the true location of the IOL. [5][6][7][8][9] New IOL power prediction methods have also been developed based on the new-generation ELP prediction methods, and they have shown that using a more accurately predicted IOL position helps to improve the IOL power prediction accuracy. 5 It is so far largely unexplored whether inserting a more accurately predicted ELP into existing formulas improves refraction prediction accuracy.…”