Related Experiment Video
Updated: May 10, 2026
11:29
Corneal Confocal Microscopy: A Novel Non-invasive Technique to Quantify Small Fibre Pathology in Peripheral Neuropathies
Published on: January 3, 2011
An Annotated Corneal Confocal Microscopy Dataset for Nerve Segmentation and Clinical Characterization
Qincheng Qiao1,2, Juan Cao3, Xinguo Hou4,5,6,7
1Department of Endocrinology and Metabolism, Qilu Hospital of Shandong University, Jinan, Shandong, 250012, China.
Scientific Data
|May 8, 2026
Summary
This study introduces a new, expert-annotated dataset of corneal images for nerve segmentation research. The dataset aims to improve the development and validation of algorithms for analyzing the sub-basal nerve plexus.
Area of Science:
- Ophthalmology
- Neuroscience
- Medical Imaging
Background:
- Corneal confocal microscopy visualizes the sub-basal nerve plexus in vivo.
- Existing datasets often lack pixel-level annotations, hindering segmentation research.
- Current resources have variable quality and annotation strategies, complicating algorithm comparison.
Purpose of the Study:
- Introduce a novel, high-quality dataset for corneal nerve segmentation.
- Provide expert-reviewed pixel-level nerve masks for 410 images.
- Facilitate training, validation, and generalisability assessment of segmentation models.
Main Methods:
- Collected 410 high-quality corneal images from 88 participants.
- Generated expert-reviewed pixel-level nerve masks for all images.
- Included demographic, clinical, and laboratory data for participants.
Main Results:
- Developed a dataset with detailed, manually curated nerve delineations.
- Ensured consistent formatting and clear provenance for all data.
- Dataset spans two subsets with distinct acquisition conditions.
Conclusions:
- The new dataset addresses limitations in existing resources for corneal nerve segmentation.
- It supports robust model development and generalisability testing.
- Enables exploration of associations between image metrics and clinical characteristics.
