Multimodal Benchmark · Vision-Language Models · Radiology

Spatial Perception & Anatomical Reasoning in Clinical Radiology

SPARC-Rad is a manually curated benchmark dataset and evaluation pipeline with 300 image-question pairs from healthy-control TCIA studies, spanning CT, MRI, and radiography across five anatomical categories. It tests whether vision-language models can reason about radiologic anatomy as a spatial system — not just classify disease.

Tripathi S., Seker M.E., Quevada K., Enwerem E.D., Khandelwal P., Meltem E., Koca B., Faghani S., Arnold J., Daye D., Cook T. SPARC-Rad: A Multimodal Benchmark Dataset and Evaluation Pipeline for Spatial and Anatomical Reasoning in Radiology Vision-Language Models. arXiv:2608.00100.

0Image-Question Pairs
0Models Evaluated
0Imaging Modalities
0Anatomical Categories

Results

Leaderboard

Overall accuracy on 300 image-question pairs, graded by LLM-as-judge with human quality review. Expand any row to see modality and organ-category breakdowns.

Rank Model Overall Accuracy CT MRI X-Ray

Accuracy is reported with reference to the 300 image-question pairs; 95% confidence intervals are recommended for interpreting small differences in rank. See the paper for details.

Breakdowns

Top Models by Modality & Organ Category

Subgroup accuracy reveals that strong aggregate performance can mask region- or modality-specific weaknesses.

Dataset

300 Manually Curated Image-Question Pairs

Images are drawn from healthy-control studies in The Cancer Imaging Archive (TCIA). Radiology trainees designed image-grounded questions targeting anatomical identification, localization, laterality, regional recognition, device identification, counting, and inter-structure spatial relationships.

Distribution by Modality

    Distribution by Anatomical Category

    Representative Benchmark Instance

    Representative SPARC-Rad benchmark instance: a chest radiograph paired with spatially grounded questions
    Figure 1. A chest radiograph is paired with spatially grounded questions and reference answers testing device identification, localization, and counting.

    How It Works

    Evaluation Pipeline

    Every image-question pair is run through a reproducible pipeline: standardized prompting, raw-output collection, response normalization, LLM-as-judge grading, and human review.

    1

    Prompt

    “You are evaluating a radiology image. Answer the following question using only the provided image. Provide a concise answer. Question: [question]”

    2

    Collect & Normalize

    Raw outputs are stored for audit, then normalized — lowercasing, punctuation removal, abbreviation and synonym expansion — while preserving clinically meaningful distinctions such as left vs. right.

    3

    LLM-as-Judge

    A separate judging model receives the question, reference answer, accepted synonyms, and the model response to determine anatomical and spatial equivalence.

    4

    Human Review

    Radiology reviewers adjudicate ambiguous answers, especially for laterality, device localization, and relational questions. Each item receives a binary correctness label.

    Reasoning Types Evaluated

    • Anatomical identification
    • Localization
    • Laterality (left vs. right)
    • Regional classification
    • Inter-structure spatial relationships
    • Counting / device recognition

    Error Types Analyzed

    • Laterality error (right instead of left)
    • Wrong structure identified
    • Wrong body region
    • Overly vague response
    • Relational error (medial/lateral, superior/inferior)
    • Modality misunderstanding
    • Hallucinated finding

    Evaluation Workflow

    SPARC-Rad evaluation pipeline diagram
    Figure 2. Candidate VLMs receive benchmark images and questions, produce free-text outputs, and are graded against reference answers using LLM-as-judge and human review before statistical analysis.

    Get on the Leaderboard

    Submit Your Model

    We run candidate vision-language models through the SPARC-Rad evaluation pipeline on the held question set and add verified results to the leaderboard. Submissions are graded with LLM-as-judge plus human radiology review before publication.

    Request an Evaluation