Background To examine the effectiveness of the use of machine learning for adapting an intraocular lens (IOL) power calculation for a patient group. Methods In this retrospective study, the clinical records of 1,611 eyes of 1,169 Japanese patients who received a single model of monofocal IOL (SN60WF, Alcon) at Miyata Eye Hospital were reviewed and analyzed. Using biometric metrics and postoperative refractions of 1211 eyes of 769 patients, constants of the SRK/T and Haigis formulas were optimized. The SRK/T formula was adapted using a support vector regressor. Prediction errors in the use of adapted formulas as well as the SRK/T, Haigis, Hill-RBF and Barrett Universal II formulas were evaluated with data from 395 eyes of 395 distinct patients. Mean prediction errors, median absolute errors, and percentages of eyes within ± 0.25 D, ± 0.50 D, and ± 1.00 D, and over + 0.50 D of errors were compared among formulas. Results The mean prediction errors in the use of the SRT/K and adapted formulas were smaller than the use of other formulas (P < 0.001). In the absolute errors, the Hill-RBF and adapted methods were better than others. The performance of the Barrett Universal II was not better than the others for the patient group. There were the least eyes with hyperopic refractive errors (16.5%) in the use of the adapted formula. Conclusions Adapting IOL power calculations using machine learning technology with data from a particular patient group was effective and promising.
This retrospective study explored the effect of the ratio of axial length (AL) to average keratometry (K) on intraocular lens power calculation in long eyes. The clinical records of eyes that had an AL of 26.0 mm or longer, and underwent cataract surgery with intraocular lens implantations, were reviewed. This study was approved by the institutional review board of Miyata Eye Hospital. Preoperative biometry data were obtained using optical low-coherence reflectometry. Prediction errors in the use of the SRK/T formulas were obtained from manifest refraction spherical equivalents one month postoperatively. Significant factors inducing prediction errors were examined using stepwise multiple regression analysis with descriptive factors of AL, K value, and their ratio (AL/K). Clinical records related to 49 long eyes of 49 patients, and 93 eyes of 93 patients with normal AL, were evaluated. Stepwise multiple regression analysis revealed that the AL/K was a significant factor increasing the prediction errors (P = 0.0003). With the regression equation, 98% of prediction errors with the use of the SRK/T formula were within ±1.00 D of differences. For our sample of 49 long eyes, the ratio of AL to K was a significant factor inducing hyperopic prediction errors with the use of SRK/T for long eyes.
Background: We aimed to evaluate the existence of accommodative microfluctuations in eyes after cataract surgery. Methods: This retrospective observational cohort study included 1160 eyes of 713 patients (mean age: 72.5 ± 8.3 years) who underwent phacoemulsification, intraocular lens insertion, and an evaluation of accommodative microfluctuations with an autorefractometer. Patients with posterior segment disorders resulting in visual acuity impairment and those with unavailable medical information were excluded. High-frequency components (HFCs), between 1.0–2.3 Hz, based on fast Fourier transform analysis of the accommodative microfluctuation data were examined at postoperative 2–3 (2 M) and 6 months (6 M). The relationships between the HFCs and patient age, manifest refraction, and axial length were analyzed. Results: Increased HFC values (>65) were observed at a constant rate after cataract surgery, with prevalence rates of 33.4% at 2 M and 34.7% at 6 M. Postoperatively, at 2 M, increased HFC values were significantly more common for eyes with axial length ≥26 mm than for those with axial length <26 mm (p = 0.0056). However, they were not significantly correlated to age or postoperative manifest refraction. Conclusions: At 2 M postoperatively, increased HFC values presented more frequently in eyes with a greater axial length; hence, the precise detection and understanding of postoperative accommodative spasms in high myopia patients is important.
The adverse effects of hard contact lenses (HCL) on the corneal endothelium have been studied in the short term; however, long-term effects remain still unclear. In this study, we analyzed the effect of long-term HCL use on corneal endothelial cell density (ECD) and morphology in healthy Japanese individuals. This cross-sectional observational study included individuals using HCL for refractive errors examined at a single specialty eye hospital. Patient age, duration of HCL usage, ECD, coefficient of variation of the cell area (CV), and rate of appearance of hexagonal cells (6A) obtained via non-contact specular microscopy were assessed. We analyzed 8604 eyes (mean age: 35.6 ± 10.0 years, 837 males, 3465 females). The mean duration of HCL usage was 14.7 ± 9.1 (range, 1–50) years. Multivariate analysis revealed that ECD significantly correlated with age (P < 0.001) but not with duration of usage; however, CV and 6A significantly correlated with both factors (P < 0.001). Univariate analysis revealed that CV and 6A correlated with duration of usage (all, P < 0.001). According to our results, CV and 6A correlated with the duration of HCL usage in ophthalmologically healthy Japanese individuals. Therefore, it is important to monitor corneal endothelial morphology in long-term HCL wearers.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.