Use of neural networks to predict vault values after implantable collamer lens surgery

Abstract

Background Appropriate sizing of the implantable collamer lens (ICL) and accurate prediction of the vault are crucial prior to surgery. However, sometimes, the vault value is higher or lower than predicted, necessitating reoperation. The present study aimed to develop neural networks for improving predictions of vault values following ICL implantation based on preoperative biometric data. Methods This retrospective study included 137 eyes of 74 patients with ICLs. Linear regression and neural network analyses were used to examine the relationship between vault values at the 6-month follow-up and preoperative parameters (e.g., ICL characteristics and biometrics). Results Linear regression analysis revealed that vault values were correlated with five variables: ICL size, anterior chamber depth (ACD), angle-to-angle (ATA), white-to-white (WTW), and lens thickness (LT) (adjusted R2 = 0.411). Inclusion of more input variables was associated with better performance in the neural network analysis. The degree of fit when all 11 variables were included in the neural network model was close to 1 (R2 = 0.98). R2 values for the quaternary neural network model enrolling four input variables (ICL size, ATA, ACD, and LT) reached 0.90. Conclusions A neural network equation including the ICL size and biometric parameters of the anterior segment (ATA, ACD, and LT) can be used to predict the postoperative vault, aiding in the selection of an appropriate ICL size and reducing the need for reoperation after surgery.

Clinical Prediction of Inadequate Vault in Eyes With Thick Lens After Implantable Collamer Lens Implantation Using Iris Morphology

“…For concave shape iris, it had a tendency to demonstrate low vault value after ICL implantation; while for convex shape iris, most of eyes acquired a high vault value. With regards to the parameters about the lens, previous studies have reported that the LT and LR had significantly negative correlation with vault value in previous studies ( 10 , 11 , 19 , 20 ). However, eyes in our study obtained both high and low vault value even though they all had the LT > 4.0 mm.…”

A Case of a Low Vault after Posterior Chamber Acrylic Phakic Intraocular Lens Implantation

“…According to Kang et al [9] , these two measurements do not accurately reflect the size of the posterior chamber space in which the lens will be placed. The study by Xu et al [2] showed that although ATA distance does not use the sulcus diameter, it may be more accurate and reliable than other biometric parameters measured in the horizontal dimension. According to Igarashi et al [4] , ATA measurements had higher reproducibility than WTW measurements on AS-OCT, making them a more important predictor of a postoperative vault.…”

“…A Packer [16] meta-analysis showed that the STS-based formula is not superior to the WTW-based formula. Some additional factors, including investigator error, eye position, and anatomic changes in the ciliary sulcus, may affect the measurements [2] . In our patient, the horizontal STS size was 12.04 mm in the right eye and 11.43 mm in the left eye.…”

“…However, it is important to maintain a safe distance (vault) between a phakic intraocular lens (pIOL) and the crystalline lens. Dynamic changes in the vault [2] and age-related thickening of the lens [3] can lead to contact of the pIOL with the epithelium of the crystalline lens. At the same time, a high vault can lead to pronounced narrowing of the anterior chamber angle, which impairs outflow of the aqueous humour and may increase pigment dispersion, resulting in a risk of higher intraocular pressure [4] .…”

Predictability of the vault after implantable collamer lens implantation using OCT and artificial intelligence in White patient eyes

“…12,13 Recently, machine learning regression of the preoperative AS-OCT metrics has been shown to provide higher predictability of the ICL vault than the conventional manufacturer's nomogram, and neural networks are reported to be applicable for vault predicting. 8,[14][15][16] The aim of this study was to develop a machine learning regression model and classifier to conveniently predict the ICL vault and size in daily practice by using AS-OCT (MS-39, Costruzione Strumenti Oftalmici) metrics and to compare the predictability outcomes with the online manufacturer's nomogram.…”