Using a multilayer perceptron in intraocular lens power calculation

Conditional Process Analysis for Effective Lens Position According to Preoperative Axial Length

PURPOSE

To predict the effective lens position (ELP) using conditional process analysis according to preoperative axial length.

SETTING

Yeouido St. Mary hospital.

DESIGN

A retrospective case series.

METHODS

This study included 621 eyes from 621 patients who underwent conventional cataract surgery at Yeouido St. Mary Hospital. Preoperative axial length (AL), mean corneal power (K), and anterior chamber depth (ACD) were measured by partial coherence interferometry. AL was used as an independent variable for the prediction of ELP, and 621 eyes were classified into four groups according to AL. Using conditional process analysis, we developed 24 structural equation models, with ACD and K acting as mediator, moderator or not included as variables, and investigated the model that best predicted ELP.

RESULTS

When AL was 23.0 mm or shorter, the predictability for ELP was highest when ACD and K acted as moderating variables (R2 = 0.217). When AL was between 23.0 mm and 24.5 mm or longer than 26.0 mm, the predictability was highest when K acted as a mediating variable and ACD acted as a moderating variable (R2 = 0.217 and R2 = 0.401). On the other hand, when AL ranged from 24.5 mm to 26.0 mm, the model with ACD as a mediating variable and K as a moderating variable was the most accurate (R2 = 0.220).

CONCLUSIONS

The optimal structural equation model for ELP prediction in each group varied according to AL. Conditional process analysis can be an alternative to conventional multiple linear regression analysis in ELP prediction.

Artificial intelligence applications and cataract management: A systematic review

Artificial intelligence (AI)-based applications exhibit the potential to improve the quality and efficiency of patient care in different fields, including cataract management. A systematic review of the different applications of AI-based software on all aspects of a cataract patient's management, from diagnosis to follow-up, was carried out in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. All selected articles were analyzed to assess the level of evidence according to the Oxford Centre for Evidence-Based Medicine 2011 guidelines, and the quality of evidence according to the Grading of Recommendations Assessment, Development and Evaluation system. Of the articles analyzed, 49 met the inclusion criteria. No data synthesis was possible for the heterogeneity of available data and the design of the available studies. The AI-driven diagnosis seemed to be comparable and, in selected cases, to even exceed the accuracy of experienced clinicians in classifying disease, supporting the operating room scheduling, and intraoperative and postoperative management of complications. Considering the heterogeneity of data analyzed, however, further randomized controlled trials to assess the efficacy and safety of AI application in the management of cataract should be highly warranted.

Clinical Accuracy of 18 IOL Power Formulas in 241 Short Eyes

PURPOSE

To analyze the accuracy of 18 intraocular lens (IOL) power calculation formulas in eyes with axial length (AL) ≤ 22 mm.

METHODS

We analyzed 241 eyes of 241 patients. Eighteen formulas were evaluated: Barrett Universal II (BUII), EVO 2.0, Haigis, Hoffer Q, Holladay 1 and 2, Cooke K6, Kane, LadasSuperFormula AI, Naeser 2, Olsen, Panacea, Pearl-DGS, RBF 2.0, SRK/T, T2, VRF and VRF-G. Optical biometry was performed with an IOLMaster 700 (Carl Zeiss Meditec, Jena, Germany). With lens constants optimized for the whole range of AL, the mean prediction error (PE) and its standard deviation (SD), the median absolute error (MedAE), the mean absolute error (MAE) and the percentage of eyes with PEs within ±0.25 D, ±0.50 D and <±1.00 D were calculated.

RESULTS

Post-hoc analysis of the absolute PE revealed statistically significant differences (P < .05) between some of the newer formulas (K6, Kane, Naeser 2, Olsen and VRF-G), which obtained the lowest MedAE (respectively, 0.308, 0.300, 0.277, 0.310 and 0.276 D) and the remaining ones. These formulas yielded also the highest percentage of eyes with a PE within ±0.50 D (70.54%, 72.20%, 71.37%, 70.95% and 73.03%, respectively), whereas Panacea and SRK/T yielded the lowest percentage (62.24%), with a stastically significant difference (P < .05) with respect to most formulas.

CONCLUSION

In eyes with AL ≤22.0 mm, new formulas (K6, Kane, Naeser 2, Olsen and VRF-G) offer the most accurate predictions of postoperative refraction.

Artificial Intelligence in Cornea, Refractive Surgery, and Cataract: Basic Principles, Clinical Applications, and Future Directions

ABSTRACT

Corneal diseases, uncorrected refractive errors, and cataract represent the major causes of blindness globally. The number of refractive surgeries, either cornea- or lens-based, is also on the rise as the demand for perfect vision continues to increase. With the recent advancement and potential promises of artificial intelligence (AI) technologies demonstrated in the realm of ophthalmology, particularly retinal diseases and glaucoma, AI researchers and clinicians are now channeling their focus toward the less explored ophthalmic areas related to the anterior segment of the eye. Conditions that rely on anterior segment imaging modalities, including slit-lamp photography, anterior segment optical coherence tomography, corneal tomography, in vivo confocal microscopy and/or optical biometers, are the most commonly explored areas. These include infectious keratitis, keratoconus, corneal grafts, ocular surface pathologies, preoperative screening before refractive surgery, intraocular lens calculation, and automated refraction, among others. In this review, we aimed to provide a comprehensive update on the utilization of AI in anterior segment diseases, with particular emphasis on the recent advancement in the past few years. In addition, we demystify some of the basic principles and terminologies related to AI, particularly machine learning and deep learning, to help improve the understanding, research and clinical implementation of these AI technologies among the ophthalmologists and vision scientists. As we march toward the era of digital health, guidelines such as CONSORT-AI, SPIRIT-AI, and STARD-AI will play crucial roles in guiding and standardizing the conduct and reporting of AI-related trials, ultimately promoting their potential for clinical translation.

Biometry, Refractive Errors, and the Results of Cataract Surgery: A Large Sample Study

The statistical characteristics of biometry and refractive error in a large sample of cataractous Cuban patients are presented, comparing between sexes and age groups. All patients were studied at the Cuban Institute of Ophthalmology “Ramón Pando Ferrer,” La Habana. The sample consists of 28252 eyes of 25068 patients, subjected to cataract surgery during the time period between 2006 and 2019. Their biometry was obtained using IOL Master devices; also, visual acuity, refraction, and corneal power were registered. After surgery, the visual acuity and refraction were measured. The refractive prediction error was determined. For patients with both eyes registered, anisometropia was also calculated. Age and sex were used to segment the data. The preoperative biometric parameters show highly significant differences between sexes, with male eyes being longer and with a deeper anterior chamber but with a thinner lens. Also, keratometry shows highly significant differences, with female eyes being steeper than male. Before surgery, both sexes have myopic eyes as average, with males being more myopic than females (p < 0.001). After surgery, the average spherical equivalent is −0.36 D and female eyes are more myopic than males (p < 0.001). Visual acuity with and without distance correction has a significant increase after surgery. These results are of importance not only for Cuba but also for other countries with a large Cuban population and/or similar ethnic composition, such as the USA (particularly the south of Florida), Spain, and many countries in Latin America.