Related Experiment Video
Updated: Jun 28, 2025
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
VAULT: vault accuracy using deep learning technology: new image-based artificial intelligence model for predicting
Taj Nasser1, Matthew Hirabayashi, Gurpal Virdi
1From the Parkhurst NuVision, San Antonio, Texas (Nasser, Parkhurst); University of Missouri Columbia School of Medicine, Columbia, Missouri (Hirabayashi, Virdi); Mason Eye Institute, Columbia, Missouri (Hirabayashi, Virdi); Texas State University, San Marcos, Texas (Abramson).
A new deep learning model accurately predicts the postoperative vault of phakic implantable collamer lenses (ICLs). This AI tool ensures most predictions fall within a clinically acceptable range for ICL vault.
Area of Science:
- Ophthalmology
- Artificial Intelligence
- Medical Imaging
Background:
- Phakic implantable collamer lenses (ICLs) are used for refractive error correction.
- Accurate prediction of postoperative vault is crucial for ICL outcomes.
- Current methods for vault prediction may have limitations.
Purpose of the Study:
- To develop and validate a deep learning model for predicting postoperative ICL vault.
- To assess the accuracy of the model across different ICL sizes.
Main Methods:
- A retrospective machine learning study was conducted.
- A neural network was trained using preoperative ultrasound images, patient demographics, and postoperative vault data.
- Data from 437 eyes of 221 patients were utilized.
Main Results:
- The deep learning model demonstrated high prediction accuracy for ICL vault.
- Mean absolute errors ranged from 66.3 μm to 103 μm for different ICL sizes.
- Over 96% of predictions were within a clinically acceptable 500 μm margin for most ICL sizes.
Conclusions:
- The developed deep learning model accurately predicts postoperative ICL vault.
- The model's predictions fall within clinically acceptable margins, aiding surgical planning.
- This AI approach shows promise for improving refractive surgery outcomes.
