Prediction of Phakic Intraocular Lens Vault Using Machine Learning of Anterior Segment Optical Coherence Tomography Metrics

Kamiya, Kazutaka; Ryu, Ik Hee; Yoo, Tae Keun; Kim, Jung Sub; Lee, In Sik; Kim, Jin Kook; Ando, Wakako; Shoji, Nobuyuki; Yamauchi, Tomofusa; Tabuchi, Hitoshi

doi:10.1016/j.ajo.2021.02.006

Cited by 44 publications

(73 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…10 A recent study showed that traditional machine learning methods are expected to improve the accuracy of ICL vault prediction. 11 However, all ICL surgery cases cannot be the gold standard for training machine learning models. Several cases may be considered as outliers with incorrect lens selection and measurement noises, and these factors can lead to the incorrect fitting of a given machine learning model and reduce the performance in the validation sets.…”

Section: Introductionmentioning

confidence: 99%

Development of a Web-Based Ensemble Machine Learning Application to Select the Optimal Size of Posterior Chamber Phakic Intraocular Lens

Kang

Ryu

Lee³

et al. 2021

Trans. Vis. Sci. Tech.

Self Cite

View full text Add to dashboard Cite

Selecting the optimal lens size by predicting the postoperative vault can reduce complications after implantation of an implantable collamer lens with a centralhole (ICL with KS-aquaport). We built a web-based machine learning application that incorporated clinical measurements to predict the postoperative ICL vault and select the optimal ICL size. Methods:We applied the stacking ensemble technique based on eXtreme Gradient Boosting (XGBoost) and a light gradient boosting machine to pre-operative ocular data from two eye centers to predict the postoperative vault. We assigned the Korean patient data to a training (N = 2756 eyes) and internal validation (N = 693 eyes) datasets (prospective validation). Japanese patient data (N = 290 eyes) were used as an independent external dataset from different centers to validate the model. Results:We developed an ensemble model that showed statistically better performance with a lower mean absolute error for ICL vault prediction (106.88 μm and 143.69 μm in the internal and external validation, respectively) than the other machine learning techniques and the classic ICL sizing methods did when applied to both validation datasets. Considering the lens size selection accuracy, our proposed method showed the best performance for both reference datasets (75.9% and 67.4% in the internal and external validation, respectively). Conclusions:Applying the ensemble approach to a large dataset of patients who underwent ICL implantation resulted in a more accurate prediction of vault size and selection of the optimal ICL size.Translational Relevance: We developed a web-based application for ICL sizing to facilitate the use of machine learning calculators for clinicians.

show abstract

Section: Introductionmentioning

confidence: 99%

Development of a Web-Based Ensemble Machine Learning Application to Select the Optimal Size of Posterior Chamber Phakic Intraocular Lens

Kang

Ryu

Lee³

et al. 2021

Trans. Vis. Sci. Tech.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Predicting the optimal ICL vault is essential, especially when considering the safety of ICL surgery. It has been demonstrated that the ideal ICL vault is identical to the central corneal thickness (approximately 500 lm) [10,15,16]. There are still ongoing concerns about selecting the optimal ICL size, even when we employ the recommended ICL sizing nomogram provided by the ICL manufacturer, based on the white-towhite distance and the anterior chamber depth measurements.…”

Section: Discussionmentioning

confidence: 99%

“…We described the details of the surgical procedures in our previous reports [4,[13][14][15][16]. In brief, on the day of surgery, dilating and topical anesthetic agents were applied.…”

Section: Surgical Proceduresmentioning

confidence: 99%

Vertically Fixated Posterior Chamber Phakic Intraocular Lens Implantation Through a Superior Corneal Incision

et al. 2022

Self Cite

View full text Add to dashboard Cite

Introduction: To assess the 1-year outcomes of vertically fixated posterior chamber phakic intraocular lens implantation through a superior corneal incision. Methods: This pilot study comprised 78 eyes of 53 consecutive patients undergoing vertically fixated implantable collamer lens (ICL) implantation through a superior corneal incision to correct moderate to high myopia and myopic astigmatism. We prospectively determined the safety, efficacy, predictability, stability, and adverse events preoperatively, and at 1 week and 1, 3, and 12 months postoperatively.Results: The mean follow-up period was 10.4 ± 5.4 months. Uncorrected and corrected visual acuity were -0.20 ± 0.10 and -0.25 ± 0.07 logMAR, respectively, at 1 year postoperatively. At 1 year postoperatively, 98% and 100% of eyes were within 0.5 and 1.0 D, respectively, of the targeted correction. A nonsignificant change in manifest refraction of -0.01 ± 0.08 D occurred from 1 week to 1 year. The manifest astigmatism decreased significantly, from 0.69 ± 0.73 D preoperatively to 0.21 ± 0.27 D at 1 year postoperatively (Mann-Whitney U test, p \ 0.001). No visionthreatening complications occurred at any time in this series. Conclusions: According to our experience, the vertically fixated ICL through a superior incision achieved good results, without significant complications. Considering that younger patients requiring ICL surgery tend to have with-the-rule astigmatism, this surgical technique may be a viable option for reducing astigmatism without using toric ICLs.

show abstract

“…If a combination of high myopia, high vault, and poor accommodative amplitude is present, the patient may face difficulties in focusing on near objects. Currently, apart from the manufacturer's sizing method, which uses the WTW and ACD as biometric features for the lens size selection, there are alternative methods that use biometric parameters, such as the anterior chamber width, the sulcus-to-sulcus, and the crystalline lens rise, aiming to estimate the vault [23,26,27]. Reinstein et al [2] showed that when using a sizing technique based on the sulcus-to-sulcus distance, the vault range could be reduced compared to the WTW method; other studies have provided methodologies for narrowing the vault range [8,24].…”

Section: Discussionmentioning

confidence: 99%

“…Considering the 95% confidence intervals of the differences, if −18.0 DS (the most myopic ICL) was calculated using the predicted vault, the postoperative refractive error associated with vault error would be lower than 0.40 DS (when using the NK formula). While it is not possible to narrow down the vault error, at the moment of selecting the ICL power, the surgeon should consider the expected vault for the ICL size chosen (26,27) and select the ICL power by combining the predicted postoperative refraction, calculated by the manufacturer, and the expected postoperative refraction, induced by the expected lens position.…”

Section: Discussionmentioning

confidence: 99%

Posterior Chamber Phakic Intraocular Lenses for the Correction of Myopia: Factors Influencing the Postoperative Refraction

Serra

Trancón

Sierra

et al. 2021

Optics

View full text Add to dashboard Cite

Posterior chamber phakic intraocular lens implantation is a refractive technique for the correction of myopia. This study aimed to identify those factors contributing to variability in postoperative refraction. Methods: This retrospective study evaluated 73 eyes (one eye per patient) implanted with myopic implantable collamer lenses (ICL). Eyes were divided into two groups, the low myopic group (LMG) (ICL > −9.5 DS) and the high myopic group (HMG) (ICL ≤ −9.5 DS), to compare the predictability, efficacy index, and postoperative refraction between groups. The association of postoperative refraction with anatomical, demographic, and optical features was assessed through correlation analysis and investigated using ray-tracing. Results: Postoperative refraction at 3 months for the whole group was close to emmetropia at −0.02 ± 0.37 DS, the LMG tended toward myopia and the HMG, toward hyperopia. The results showed that 65% and 54% of the eyes had postoperative refraction of within ±0.25 DS, respectively, in the LMG and HMG, and in both groups, 100% were within ±1.00 DS. ICL implantation had a higher efficacy index in the HMG (1.13 ± 0.15) than in the LMG (1.04 ± 0.15). Postoperative refraction was positively associated with the vault (R = 0.408) and negatively correlated with ICL power (R = −0.382). Conclusion: The predictability and effectiveness of ICL implantation is high in a wide range of myopias. Considering the expected vault and including accurate vertex measurements would contribute to improving the predictability of the results.

show abstract

Prediction of Phakic Intraocular Lens Vault Using Machine Learning of Anterior Segment Optical Coherence Tomography Metrics

Cited by 44 publications

References 15 publications

Development of a Web-Based Ensemble Machine Learning Application to Select the Optimal Size of Posterior Chamber Phakic Intraocular Lens

Development of a Web-Based Ensemble Machine Learning Application to Select the Optimal Size of Posterior Chamber Phakic Intraocular Lens

Vertically Fixated Posterior Chamber Phakic Intraocular Lens Implantation Through a Superior Corneal Incision

Posterior Chamber Phakic Intraocular Lenses for the Correction of Myopia: Factors Influencing the Postoperative Refraction

Contact Info

Product

Resources

About