Related Experiment Video
Updated: Jun 24, 2025
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
An AS-OCT image dataset for deep learning-enabled segmentation and 3D reconstruction for keratitis
Yiming Sun1, Nuliqiman Maimaiti1, Peifang Xu2
1Eye Center, The Second Affiliated Hospital, School of Medicine, Zhejiang University, Zhejiang Provincial Key Laboratory of Ophthalmology, Zhejiang Provincial Clinical Research Center for Eye Diseases, Zhejiang Provincial Engineering Institute on Eye Diseases, Hangzhou, Zhejiang, China.
This study introduces a new dataset of anterior segment optical coherence tomography (AS-OCT) images for infectious keratitis research. This resource supports AI development for better keratitis diagnosis and management.
Area of Science:
- Ophthalmology
- Medical Imaging
- Artificial Intelligence
Background:
- Infectious keratitis is a leading cause of blindness worldwide.
- Anterior segment optical coherence tomography (AS-OCT) aids in assessing corneal inflammation and anterior chamber details.
- A lack of open-access annotated AS-OCT datasets hinders deep learning advancements for keratitis.
Purpose of the Study:
- To create and provide a comprehensive, open-access dataset of annotated AS-OCT images for infectious keratitis.
- To facilitate the development of AI-driven image analysis tools for keratitis.
- To support research in both 2D and 3D analysis of AS-OCT images.
Main Methods:
- Compilation of 1168 AS-OCT images from patients with keratitis.
- Inclusion of segmentation labels for corneal lesions and cornea.
- Annotation of iris for full-frame images.
Main Results:
- A novel, annotated dataset of 1168 AS-OCT images is now available.
- The dataset includes detailed segmentation masks for keratitis-affected corneas.
- Provides a foundation for training and validating AI models.
Conclusions:
- The released AS-OCT dataset is crucial for advancing AI in infectious keratitis management.
- Enables development of automated diagnostic and severity assessment tools.
- Promotes further research in 2D/3D AS-OCT image analysis for ophthalmic diseases.
