Skip to main content

OphthalVQA Dataset

Ophthalmic visual question-answering benchmark dataset released as supplementary data for evaluating multimodal language models in ophthalmology.

At a glance

FieldValue
Short nameophthalvqa
Full nameOphthalVQA Dataset
First published2025-03-03
Publication date precisionday
Publication date evidenceapi.figshare.com/v2
Publication date source fieldpublished_date (Figshare version 1)
Publication date reviewed2026-09-11
Primary categorymultimodal
Resource rolecurrent_dataset
Dataset familyophthalvqa
Contained modalitiesfundus, fundus_angiography, oct, ocular_ultrasound, external_eye, text
Tasksvisual_question_answering, classification
Primary reported quantity600 question answer pairs
ClassesNot reported (Not reported)
Splitsall
Size0.03 GB
Source-stated termsCC BY 4.0
Normalized termscc-by
Descriptive screening labelStandard label without an explicit NC clause; not a permission finding
Terms scopedataset_files
Access frictionself_service_authenticated
Route backendFigshare
Availabilityavailable (checked 2026-07-21)
Acquisition supportend_to_end_tested
Legacy sample-loader statusMetadata and access only

Reported quantities

RoleCountUnitScopeBasisEvidence
Primary600question_answer_pairsRows in OphthalVQA.csvcurrent_deposit_tablehttps://doi.org/10.6084/m9.figshare.25624917
Additional60imagesJPG image files in the current depositcurrent_deposit_file_listinghttps://doi.org/10.6084/m9.figshare.25624917

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

Supplementary benchmark for ophthalmic VQA and multimodal model evaluation; inspect image provenance and splits before leakage-sensitive validation.

Access information and download

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

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

Upstream page: https://doi.org/10.6084/m9.figshare.25624917

Source-term evidence: https://doi.org/10.6084/m9.figshare.25624917

Loader status

This catalog record provides metadata and access instructions, but it does not yet include a standard DatasetSample loader. Inspect the source file structure or contribute a loader before using it in a training pipeline.

Citation

@misc{ophthalvqa,
title = { OphthalVQA Dataset },
note = { Xu P, Chen X, Zhao Z, Shi D. OphthalVQA Dataset. Figshare, 2025. doi:10.6084/m9.figshare.25624917 },
year = { 2025 },
url = { https://doi.org/10.6084/m9.figshare.25624917 },
}

Source-stated terms

  • Raw source string: CC BY 4.0
  • 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

  • eyecare_100k: Eyecare-100K: Multimodal Ophthalmology VQA Corpus (102,000 question answer pairs, unknown)
  • x_pcr: X-PCR Ophthalmology Progressive Clinical Reasoning Benchmark (18,735 rows, unknown)
  • lmod_plus: LMOD+ Multimodal Ophthalmology Benchmark (32,633 annotated instances, unknown)
  • mm_retinal_reason: MM-Retinal-Reason: Ophthalmology Multimodal Reasoning Dataset (130 question answer pairs, unknown)
  • angioreport: AngioReport Fundus Angiography Report Dataset (55,361 images, unknown)
  • ffa_ir: FFA-IR Medical Report Dataset (47,247 images, unknown)
  • multieye: MultiEYE: OCT-Enhanced Fundus Multi-Disease Benchmark (103,959 images, mit)
  • harvard_fairvision: Harvard-FairVision (AMD + DR + Glaucoma, paired SLO + OCT) (30,000 participants, cc-by-nc-nd)