Related Experiment Video
Updated: Sep 15, 2025
Author Spotlight: Unraveling the Molecular Mechanisms in PCO and Fibrosis Following Cataract Surgery
Published on: December 1, 2023
VAULT-OCT: vault accuracy using deep learning technology-an artificial intelligence model for predicting implantable
Matthew T Hirabayashi1, Gurpal S Virdi, Taj A Nasser
1From the Parkhurst NuVision LASIK Eye Surgery, San Antonio, Texas (Hirabayashi, Parkhurst); University of Missouri Columbia School of Medicine, Columbia, Missouri (Virdi); Mason Eye Institute, Columbia, Missouri (Virdi); Tylock George Eye Care and Laser, Dallas, Texas (Nasser); Mueller Vision, Fort Worth, Texas (Nasser); Texas State University, San Marcos, Texas (Abramson).
Purpose:
To develop an accurate deep learning model, VAULT-OCT, to predict postoperative vault of phakic implantable collamer lenses (ICLs) based on preoperative optical coherence tomography (OCT).
Setting:
Parkhurst NuVision LASIK Eye Surgery, San Antonio, Texas.
Design:
Retrospective machine learning study.
Methods:
324 eyes from 162 consecutive patients who underwent ICL implantation were included. VAULT-OCT, the neural network, was trained on preoperative anterior segment OCT (AS-OCT) images paired with postoperative vault measurements for different ICL sizes. Incomplete data were excluded, and the images were consistently resized and normalized. A custom classifier was used in VAULT-OCT, and model performance was evaluated using root mean squared error on the test set, with mean absolute error (MAE) reported as the primary performance metric.
Results:
A MAE of 22.3 μm, 21.7 μm, and 98.1 μm and a SD of 13.5 μm, 17.8 μm, and 105.9 μm were achieved with 100%, 100%, and 89.1% of predictions within 200 μm, for the 12.1 mm, 12.6 mm, and 13.2 mm size, respectively.
Conclusions:
This OCT-based deep learning model, VAULT-OCT, achieved a high level of accuracy in predicting postoperative ICL vault, with most predictions falling within a clinically acceptable margin of vault, suggesting the feasibility of basing ICL sizing on preoperative AS-OCT.
