Related Experiment Video
Updated: Nov 27, 2025
10:25
Technical Approach for Infrared Tracking for Soft Tissue Navigation with a Holographic Head-Mounted Display and Preclinical Validation
Published on: September 2, 2025
305
Real-Time Surgical Problem Detection and Instrument Tracking in Cataract Surgery
Shoji Morita1,2, Hitoshi Tabuchi3,4, Hiroki Masumoto4
1Glory Ltd., 1-3-1 Shimoteno, Himeji-shi, Hyogo 670-8567, Japan.
Journal of Clinical Medicine
|December 3, 2020
Summary
This study introduces a machine learning method to objectively evaluate cataract surgery skills in real-time. The system accurately identifies surgical phases and critical structures, paving the way for standardized skill assessment.
Area of Science:
- Ophthalmology
- Computer Vision
- Machine Learning
Background:
- Surgical skill evaluation in ophthalmology is often subjective and lacks standardization.
- Variability in surgical techniques complicates objective assessment of ophthalmologists' skills.
Purpose of the Study:
- To develop a real-time method for quantifying cataract surgery techniques using machine learning.
- To establish a foundation for standardizing surgical skill assessment in ophthalmology.
Main Methods:
- Utilized InceptionV3 for surgical phase recognition and problem detection.
- Employed scSE-FC-DenseNet for segmenting critical structures like the cornea and surgical instruments during capsulorrhexis.
- Evaluated performance using Area Under Curve (AUC) and Intersection over Union (IoU).
Main Results:
- Achieved an AUC of 0.97 for detecting surgical problems.
- Demonstrated high detection rates for the cornea (99.7% at IoU >= 0.8), forceps tip (86.9% at IoU >= 0.1), and incisional site (94.9% at IoU >= 0.1).
Conclusions:
- The proposed machine learning approach provides a promising tool for objective, real-time evaluation of cataract surgery skills.
- This method contributes to the standardization of surgical skill assessment, addressing current subjective evaluation limitations.
