Related Experiment Video
Updated: Aug 28, 2025
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
2.9K
Deep Learning-Based System for Preoperative Safety Management in Cataract Surgery.
Gaku Kiuchi1, Mao Tanabe2, Katsunori Nagata2
1Department of Ophthalmology, Faculty of Medicine, University of Tsukuba, 1-1-1 Tennoudai, Tsukuba 305-8575, Japan.
Journal of Clinical Medicine
|September 23, 2022
Summary
An AI system enhances cataract surgery safety by verifying patient identity, eye laterality, and intraocular lens (IOL) parameters. This technology achieved high accuracy, ensuring zero false rejections or acceptances in real-world surgical settings.
Area of Science:
- Ophthalmology
- Artificial Intelligence
- Surgical Safety
Background:
- Cataract surgery requires stringent preoperative safety measures.
- Ensuring correct patient identification, surgical site, and implanted lens is critical.
Purpose of the Study:
- To implement and evaluate an AI-based system for preoperative safety management in cataract surgery.
- To assess the accuracy and effectiveness of AI in facial recognition, laterality confirmation, and IOL parameter verification.
Main Methods:
- Developed a deep-learning system using a face identification kit, YOLOv3, and VGG-16 models.
- Utilized a mobile device camera to capture patient data and compare it against a referral database.
- Tested the system on 171 patients undergoing phacoemulsification and IOL implantation.
Main Results:
- Achieved high authentication rates: 96.3% for facial recognition, 98.2% for laterality, and 88.9% for IOL parameters after repeated attempts.
- Demonstrated 0% false rejection rate and 0% false acceptance rate for all verified parameters.
- The AI system showed passable authentication rates with very high accuracy in a real surgical environment.
Conclusions:
- An AI-powered system can be effectively implemented for enhancing preoperative safety in cataract surgery.
- The system demonstrated high accuracy and reliability in critical verification steps, contributing to patient safety.
- This AI approach offers a promising solution for reducing errors in ophthalmic surgical procedures.
