Skip to main content

Qilu Annotated Corneal Confocal Microscopy Nerve Segmentation Dataset

A human corneal confocal microscopy dataset containing 410 source images from 88 participants, 410 filename-matched pixel-level nerve segmentation masks, 20 repeat annotations, and image-linked clinical and demographic data.

At a glance

FieldValue
Short nameqilu_ccm_nerve_segmentation
Full nameQilu Annotated Corneal Confocal Microscopy Nerve Segmentation Dataset
First published2025-11-10
Publication date precisionday
Publication date evidencezenodo.org/api
Publication date source fieldmetadata.publication_date (earliest public Zenodo version)
Publication date reviewed2026-09-11
Primary categoryconfocal
Resource rolecurrent_dataset
Dataset familyqilu_ccm_nerve_segmentation
Contained modalitiesconfocal, tabular
Taskssegmentation, classification, measurement
Primary reported quantity410 images
ClassesNot reported (Not reported)
Splitsset1, set2
Size0.06370422896 GB
Source-stated termsCreative Commons Attribution 4.0 International
Normalized termscc-by
Descriptive screening labelStandard label without an explicit NC clause; not a permission finding
Terms scopedataset_files
Access frictionanonymous_direct
Route backendZenodo
Availabilityavailable (checked 2026-08-02)
Acquisition supportstandard_platform_supported
Legacy sample-loader statusStandard loader included

Reported quantities

RoleCountUnitScopeBasisEvidence
Primary410imagesOriginal CCM images in Set1 and Set2 Set1 contains 210 images and Set2 contains 200 images.associated_publicationhttps://doi.org/10.1038/s41597-026-07418-6
Additional410annotated_imagesFilename-matched binary nerve segmentation masks One expert-reviewed pixel-level mask per source image.associated_publicationhttps://doi.org/10.1038/s41597-026-07418-6
Additional88participantsParticipants represented across Set1 and Set2 Set1 has 34 participants and Set2 has 54 participants.associated_publicationhttps://doi.org/10.1038/s41597-026-07418-6
Additional20annotated_imagesSubset with repeat annotations These are repeat labels for existing images, not 20 additional images.associated_publicationhttps://doi.org/10.1038/s41597-026-07418-6

Counts retain their source-reported units. Additional rows can describe components, paired items, or derivative copies and are not automatically added to the primary quantity.

Notes

The current Zenodo deposit contains one 68,401,895-byte archive with 410 source PNG images, 410 corresponding masks, 20 repeat annotations, and one linked workbook. The version-specific dataset DOI is 10.5281/zenodo.18779434 and the concept DOI is 10.5281/zenodo.17570502. Reference 35 in the published article lists 10.5281/zenodo.17570504, which resolves to an unrelated record; the article's Data Records section and the official deposit identify 10.5281/zenodo.18779434.

Access information and download

# Read-only preflight
eyehub download qilu_ccm_nerve_segmentation --data-dir ./data --dry-run --json

# Download, only when preflight reports supported behavior
eyehub download qilu_ccm_nerve_segmentation --data-dir ./data

Upstream page: zenodo.org/records

Source-term evidence: zenodo.org/records

Loader example

This entry includes a standard DatasetSample loader.

from pathlib import Path
from eyedatahub.datasets.registry import REGISTRY

data_dir = Path('~/.eyedatahub/data').expanduser()
ds = REGISTRY.get_dataset('qilu_ccm_nerve_segmentation')
samples = ds.load(data_dir, split='set1')
for s in samples[:5]:
print(s.sample_id, s.label, s.image_path)

Citation

@misc{qilu_ccm_nerve_segmentation,
title = { Qilu Annotated Corneal Confocal Microscopy Nerve Segmentation Dataset },
note = { Qiao Q, Cao J, Hou X. An Annotated Corneal Confocal Microscopy Dataset for Nerve Segmentation and Clinical Characterization. Scientific Data. 2026;13:1051. doi:10.1038/s41597-026-07418-6. Dataset: doi:10.5281/zenodo.18779434 },
year = { 2026 },
url = { https://zenodo.org/records/18779434 },
}

Source-stated terms

  • Raw source string: Creative Commons Attribution 4.0 International
  • Normalized category: cc-by
  • Apparent scope: dataset_files
  • Descriptive screening label: Standard label without an explicit NC clause; not a permission finding

⚠️ Source-stated terms, scope, and normalized labels are curation metadata, not legal advice or a permission finding. Review the current official source before transfer or reuse.

Similar resources by shared modality

  • ocular_chat_vqa: OcularChat-VQA: AREDS-Derived Patient-Physician Dialogue Dataset (844,000 records, cc-by-nc-sa)
  • brset_mbrset_embeddings: Embedding-Based Representations for BRSET and mBRSET (53,188 embedding vectors, unknown)
  • lmod_plus: LMOD+ Multimodal Ophthalmology Benchmark (32,633 annotated instances, unknown)
  • oculoscope: OculoScope: Fairer AI in Ophthalmology Dataset (16,530 images, cc-by)
  • dryad_r7s04: Data from: Prevalence of depression, anxiety, adjustment disorders, and somatoform disorders in patients with age-related macular degeneration in Germany (15,160 participants, cc0)
  • corn_collection: CORN: Corneal Confocal Microscope Dataset Collection (12,931 images, cc-by)
  • fairvlmed: FairVLMed: Fair Vision-Language Medical Ophthalmic Dataset (10,000 images, cc-by-nc-nd)
  • leops_erg: LEOPs Light-Adapted Electroretinogram and Oscillatory Potentials Dataset (9,743 signals, cc-by)