Analysis of vault prediction in phakic implantable phakic collamer lenses: manufacturer's calculator vs theoretical formulae vs clinical practice

Monte-Carlo simulation for calculating phakic supplementary lenses based on a thick and thin lens model using anterior segment OCT data

BACKGROUND

Phakic lenses (PIOLs, the most common and only disclosed type being the implantable collamer lens, ICL) are used in patients with large or excessive ametropia in cases where laser refractive surgery is contraindicated. The purpose of this study was to present a strategy based on anterior segment OCT data for calculating the refraction correction (REF) and the change in lateral magnification (ΔM) with ICL implantation.

METHODS

Based on a dataset (N = 3659) containing Casia 2 measurements, we developed a vergence-based calculation scheme to derive the REF and gain or loss in ΔM on implantation of a PIOL having power PIOLP. The calculation concept is based on either a thick or thin lens model for the cornea and the PIOL. In a Monte-Carlo simulation considering, all PIOL steps listed in the US patent 5,913,898, nonlinear regression models for REF and ΔM were defined for each PIOL datapoint.

RESULTS

The calculation shows that simplifying the PIOL to a thin lens could cause some inaccuracies in REF (up to ½ dpt) and ΔM for PIOLs with high positive power. The full range of listed ICL powers (- 17 to 17 dpt) could correct REF in a range from - 17 to 12 dpt with a change in ΔM from 17 to - 25%. The linear regression considering anterior segment biometric data and the PIOLP was not capable of properly characterizing REF and ΔM, whereas the nonlinear model with a quadratic term for the PIOLP showed a good performance for both REF and ΔM prediction.

CONCLUSION

Where PIOL design data are available, the calculation concept should consider the PIOL as thick lens model. For daily use, a nonlinear regression model can properly predict REF and ΔM for the entire range of PIOL steps if a vergence calculation is unavailable.

Implantable Collamer Lens Procedure Planning: A Review of Global Approaches

More than 2 million implantable collamer lenses (ICLs) have been implanted worldwide. With a central port to improve aqueous flow through the ICL, the latest iteration of this phakic intraocular lens (pIOL) has been shown to have stable outcomes with very low rates of adverse events. However, correct planning and ICL size selection continue to be important to achieve an optimal vault. Shallow or excessive vaults are not complications in and of themselves but may increase the risk of complications. Historically, surgeons have relied on measurements of anterior chamber depth (ACD) and manual, caliper-measured white-to-white (WTW) distance to select the ICL size. New diagnostic and imaging technologies such as optical coherence tomography (OCT) and ultrasound biomicroscopy (UBM) provide additional opportunities for visualization and measurement of the intraocular dimensions involved in phakic intraocular lens implantation, including sulcus-to-sulcus (STS) and angle-to-angle (ATA) diameters. This paper reviews various approaches to ICL planning and sizing that have been published in the peer-reviewed literature, all of which produce acceptable results for predicting vault and size selection. Surgeons may also want to identify a methodology for patient evaluation and ICL size selection that best aligns with their personal preferences, diagnostic technology, and familiarity with analytical optimization tools.

Enhancing Vault Prediction and ICL Sizing Through Advanced Machine Learning Models

PURPOSE

To use artificial intelligence (AI) technology to accurately predict vault and Implantable Collamer Lens (ICL) size.

METHODS

The methodology focused on enhancing predictive capabilities through the fusion of machine-learning algorithms. Specifically, AdaBoost, Random Forest, Decision Tree, Support Vector Regression, LightGBM, and XGBoost were integrated into a majority-vote model. The performance of each model was evaluated using appropriate metrics such as accuracy, precision, F1-score, and area under the curve (AUC).

RESULTS

The majority-vote model exhibited the highest performance among the classification models, with an accuracy of 81.9% area under the curve (AUC) of 0.807. Notably, LightGBM (accuracy = 0.788, AUC = 0.803) and XGBoost (ACC = 0.790, AUC = 0.801) demonstrated competitive results. For the ICL size prediction, the Random Forest model achieved an impressive accuracy of 85.3% (AUC = 0.973), whereas XG-Boost (accuracy = 0.834, AUC = 0.961) and LightGBM (accuracy = 0.816, AUC = 0.961) maintained their compatibility.

CONCLUSIONS

This study highlights the potential of diverse machine learning algorithms to enhance postoperative vault and ICL size prediction, ultimately contributing to the safety of ICL implantation procedures. Furthermore, the introduction of the novel majority-vote model demonstrates its capability to combine the advantages of multiple models, yielding superior accuracy. Importantly, this study will empower ophthalmologists to use a precise tool for vault prediction, facilitating informed ICL size selection in clinical practice. [J Refract Surg. 2024;40(3):e126-e132.].

Clinical decision support system based on deep learning for evaluating implantable collamer lens size and vault after implantable collamer lens surgery: a retrospective study

OBJECTIVES

To aid doctors in selecting the optimal preoperative implantable collamer lens (ICL) size and to enhance the safety and surgical outcomes of ICL procedures, a clinical decision support system (CDSS) is proposed in our study.

DESIGN

A retrospective study of patients after ICL surgery.

SETTING

China Tertiary Myopia Prevention and Control Center.

PARTICIPANTS

2772 eyes belonging to 1512 patients after ICL surgery. Data were collected between 2018 and 2022.

OUTCOME MEASURES

A CDSS is constructed and used to predict vault at 1 month postoperatively and preoperative ICL dimensions using various artificial intelligence methods. Accuracy metrics as well as area under curve (AUC) parameters are used to determine the CDSS prediction methods.

RESULTS

Among the ICL size prediction models, conventional neural networks (CNNs) achieve the best prediction accuracy at 91.37% and exhibit the highest AUC of 0.842. Regarding the prediction model for vault values 1 month after surgery, CNN surpasses the other methods with an accuracy of 85.27%, which has the uppermost AUC of 0.815. Thus, we select CNN as the prediction algorithm for the CDSS.

CONCLUSIONS

This study introduces a CDSS to assist doctors in selecting the optimal ICL size for patients while improving the safety and postoperative outcomes of ICL surgery.