X-PCR Ophthalmology Progressive Clinical Reasoning Benchmark

The version-pinned public deposit contains 18,735 ophthalmic image-text benchmark rows across CFP, external-eye, FFA, OCT, and RetCam subsets.

At a glance

Field	Value
Short name	x_pcr
Full name	X-PCR Ophthalmology Progressive Clinical Reasoning Benchmark
Publication date	2026-05-28
Date basis	Initial dataset release
Publication date precision	day
Publication date evidence	github.com/CVI-SZU
Publication date source field	Official GitHub README News entry: [2026-5-28] X-PCR dataset is released on Hugging Face
Publication date reviewed	2026-09-11
Date notes	The authors' release announcement links to Fantasy666/X-PCR on Hugging Face.
Primary category	multimodal
Resource role	derivative_dataset
Dataset family	x_pcr
Contained modalities	fundus, fundus_angiography, oct, external_eye, retcam, text
Tasks	visual_question_answering, classification
Primary reported quantity	18,735 rows
Classes	Not reported (Not reported)
Splits	all
Size	5.0 GB
Source-stated terms	Unknown
Normalized terms	unknown
Descriptive screening label	Unknown or unclear; do not assume permission
Terms scope	unknown
Access friction	self_service_authenticated
Route backend	HuggingFace Hub
Availability	available (checked 2026-07-21)
Acquisition support	standard_platform_supported
Legacy sample-loader status	Metadata and access only

Reported quantities

Role	Count	Unit	Scope	Basis	Evidence
Primary	18,735	rows	Primary quantity reported in the reviewed catalog source	legacy_catalog_field	huggingface.co/datasets

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 public deposit at commit 06a318fd852230326386e3c6514d8a11b7a6b4af has 18,735 rows. The dataset card describes a broader benchmark of 26,415 images and 177,868 VQA pairs; these units and scopes are not interchangeable.

Documented relationships

These links record source-supported lineage or overlap, not merely similar modality tags.

• This record is derived from aptos2019: Source labels in the version-pinned public X-PCR deposit identify this catalog record as upstream material. (evidence)
• This record is derived from beh: Source labels in the version-pinned public X-PCR deposit identify this catalog record as upstream material. (evidence)
• This record is derived from deepdrid: Source labels in the version-pinned public X-PCR deposit identify this catalog record as upstream material. (evidence)
• This record is derived from eye_disease_image_mendeley: Source labels in the version-pinned public X-PCR deposit identify this catalog record as upstream material. (evidence)
• This record is derived from eyepacs: Source labels in the version-pinned public X-PCR deposit identify this catalog record as upstream material. (evidence)
• This record is derived from harvard_gdp: Source labels in the version-pinned public X-PCR deposit identify this catalog record as upstream material. (evidence)
• This record is derived from jsiec: Source labels in the version-pinned public X-PCR deposit identify this catalog record as upstream material. (evidence)
• This record is derived from kermany_oct: Source labels in the version-pinned public X-PCR deposit identify this catalog record as upstream material. (evidence)
• This record is derived from mured: Source labels in the version-pinned public X-PCR deposit identify this catalog record as upstream material. (evidence)
• This record is derived from oct_c8: Source labels in the version-pinned public X-PCR deposit identify this catalog record as upstream material. (evidence)
• This record is derived from octid: Source labels in the version-pinned public X-PCR deposit identify this catalog record as upstream material. (evidence)
• This record is derived from odir2019: Source labels in the version-pinned public X-PCR deposit identify this catalog record as upstream material. (evidence)
• This record is derived from oimhs: Source labels in the version-pinned public X-PCR deposit identify this catalog record as upstream material. (evidence)
• This record is derived from rfmid2: Source labels in the version-pinned public X-PCR deposit identify this catalog record as upstream material. (evidence)

Access information and download

CLI

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

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

Python

from eyedatahub.acquisition import preflight_dataset
from eyedatahub.datasets.registry import REGISTRY

ds = REGISTRY.get_dataset('x_pcr')
print(preflight_dataset(ds, './data'))  # no download

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

BibTeX

@misc{x_pcr,
  title  = { X-PCR Ophthalmology Progressive Clinical Reasoning Benchmark },
  note   = { Fantasy666/X-PCR. Hugging Face dataset, accessed 2026-07 },
  year   = { 2026 },
  url    = { https://huggingface.co/datasets/Fantasy666/X-PCR },
}

Plain text

Fantasy666/X-PCR. Hugging Face dataset, accessed 2026-07.

Source-stated terms

• Raw source string: Unknown
• Normalized category: unknown
• Apparent scope: unknown
• 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.

Similar resources by shared modality

• eyecare_100k: Eyecare-100K: Multimodal Ophthalmology VQA Corpus (102,000 question answer pairs, unknown)
• ophthalvqa: OphthalVQA Dataset (600 question answer pairs, cc-by)
• 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)