Related Experiment Video
Updated: May 23, 2026
Full-Field Optical Coherence Microscopy for Histology-Like Analysis of Stromal Features in Corneal Grafts
Published on: October 21, 2022
Deep Learning for Detection of Corneal Perforation on Anterior Segment Optical Coherence Tomography in Microbial
Purpose:
To develop and evaluate deep learning models for automated detection of corneal perforation in microbial keratitis using anterior segment optical coherence tomography (ASOCT) images.
Methods:
We enrolled 150 patients with microbiologically confirmed keratitis. Contralateral healthy eyes served as controls. Four convolutional neural network models using ResNet architecture were trained and evaluated using ASOCT images to classify the presence or absence of corneal perforation at the eye level. Ground truth labels for perforation were established following consensus grading by two masked ophthalmologist graders. Models differed in inclusion of healthy controls and masking of non-corneal anterior segment anatomy.
Results:
The best-performing model (Model 1), which included healthy controls and randomly applied masking of the inferior image portion during training, achieved an AUC of 0.965 (95% CI, 0.911-0.995), sensitivity of 84.0% (95% CI, 70.0%-97.1%), and specificity of 97.8% (95% CI, 96.1%-99.3%) for detection of corneal perforation. Models including healthy controls outperformed those without, and lens masking improved discrimination.
Conclusions:
Deep learning models achieved high diagnostic accuracy for detecting corneal perforation on ASOCT imaging in eyes with microbial keratitis. These findings support the potential role of automated ASOCT analysis as a clinical decision support tool for identifying this vision-threatening complication.
Insights
Deep learning models accurately detect corneal perforation in microbial keratitis using anterior segment optical coherence tomography (ASOCT) scans. This automated analysis shows promise as a clinical tool for identifying this serious eye condition.
Area of Science:
- Ophthalmology
- Medical Imaging
- Artificial Intelligence
Background:
- Microbial keratitis can lead to corneal perforation, a sight-threatening complication.
- Early and accurate detection of corneal perforation is crucial for timely intervention and visual preservation.
Purpose of the Study:
- To develop and evaluate deep learning models for automated detection of corneal perforation.
- Utilize anterior segment optical coherence tomography (ASOCT) images for analysis.
Main Methods:
- Trained four convolutional neural network (ResNet) models on ASOCT images from 150 patients with microbial keratitis and healthy controls.
- Classified corneal perforation presence/absence using eye-level data, with ground truth from expert graders.
- Varied model training by including healthy controls and masking non-corneal anatomy.
Main Results:
- The best model achieved an AUC of 0.965, sensitivity of 84.0%, and specificity of 97.8% for detecting corneal perforation.
- Models incorporating healthy controls demonstrated superior performance.
- Image masking techniques, specifically lens masking, enhanced diagnostic discrimination.
Conclusions:
- Deep learning models demonstrate high accuracy in detecting corneal perforation from ASOCT images in microbial keratitis.
- Automated ASOCT analysis holds potential as a clinical decision support tool.
- This technology can aid in identifying vision-threatening corneal perforations promptly.
