Skip to main content

MM-Retinal-Reason: Ophthalmology Multimodal Reasoning Dataset

Ophthalmology-specific multimodal reasoning dataset built from 45 public datasets. Chain-of-thought reasoning traces for retinal VQA.

At a glance

FieldValue
Short namemm_retinal_reason
Full nameMM-Retinal-Reason: Ophthalmology Multimodal Reasoning Dataset
Primary categorymultimodal
Contained modalitiesfundus, fundus_angiography, oct, text
Tasksclassification, multilabel
SamplesNot reported
ClassesNot reported (Not reported)
Splitstrain, val, test
Size15.0 GB
Source-stated termsMixed (inherits from 45 source datasets) — verify per-row
Normalized termsunknown
Descriptive screening labelUnknown or unclear; do not assume permission
Terms scopemixed_components
Access frictionanonymous_direct
Route backendHuggingFace Hub
Availabilityavailable (checked 2026-07-21)
Acquisition supporttransfer_tested_partial
Legacy sample-loader statusMetadata and access only

Notes

Aggregates 45 source datasets — image licenses inherit; verify per-row before commercial use.

Access preflight and acquisition

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

# Explicit transfer, only when preflight reports supported behavior
eyehub download mm_retinal_reason --data-dir ./data

Upstream page: huggingface.co/datasets

Source-term evidence: huggingface.co/datasets

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{mm_retinal_reason,
title = { MM-Retinal-Reason: Ophthalmology Multimodal Reasoning Dataset },
note = { MM-Retinal-Reason: Ophthalmology multimodal reasoning dataset. HuggingFace, 2025 },
year = { 2025 },
url = { https://huggingface.co/datasets/lxirich/MM-Retinal-Reason },
}

Source-stated terms

  • Raw source string: Mixed (inherits from 45 source datasets) — verify per-row
  • Normalized category: unknown
  • Apparent scope: mixed_components
  • Descriptive screening label: Unknown or unclear; do not assume permission

⚠️ 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.

  • eyecare_100k: Eyecare-100K: Multimodal Ophthalmology VQA Corpus (102,000 records, unknown)
  • x_pcr: X-PCR Ophthalmology Progressive Clinical Reasoning Benchmark (18,700 records, unknown)
  • ophthalvqa: OphthalVQA Dataset (count not reported records, cc-by)
  • angioreport: AngioReport Fundus Angiography Report Dataset (55,361 records, unknown)
  • ffa_ir: FFA-IR Medical Report Dataset (47,247 records, unknown)
  • lmod_plus: LMOD+ Multimodal Ophthalmology Benchmark (32,633 records, unknown)
  • multieye: MultiEYE: OCT-Enhanced Fundus Multi-Disease Benchmark (58,036 records, mit)
  • harvard_fairvision: Harvard-FairVision (AMD + DR + Glaucoma, paired SLO + OCT) (30,000 records, cc-by-nc-nd)