Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

BIOMEDICA: An Open Biomedical Image-Caption Archive, Dataset, and Vision-Language Models Derived from Scientific Literature

About

The development of vision-language models (VLMs) is driven by large-scale and diverse multimodal datasets. However, progress toward generalist biomedical VLMs is limited by the lack of annotated, publicly accessible datasets across biology and medicine. Existing efforts are restricted to narrow domains, missing the full diversity of biomedical knowledge encoded in scientific literature. To address this gap, we introduce BIOMEDICA, a scalable, open-source framework to extract, annotate, and serialize the entirety of the PubMed Central Open Access subset into an easy-to-use, publicly accessible dataset. Our framework produces a comprehensive archive with over 24 million unique image-text pairs from over 6 million articles. Metadata and expert-guided annotations are also provided. We demonstrate the utility and accessibility of our resource by releasing BMCA-CLIP, a suite of CLIP-style models continuously pre-trained on the BIOMEDICA dataset via streaming, eliminating the need to download 27 TB of data locally. On average, our models achieve state-of-the-art performance across 40 tasks - spanning pathology, radiology, ophthalmology, dermatology, surgery, molecular biology, parasitology, and cell biology - excelling in zero-shot classification with a 6.56% average improvement (as high as 29.8% and 17.5% in dermatology and ophthalmology, respectively), and stronger image-text retrieval, all while using 10x less compute. To foster reproducibility and collaboration, we release our codebase and dataset for the broader research community.

Alejandro Lozano, Min Woo Sun, James Burgess, Liangyu Chen, Jeffrey J Nirschl, Jeffrey Gu, Ivan Lopez, Josiah Aklilu, Austin Wolfgang Katzer, Collin Chiu, Anita Rau, Xiaohan Wang, Yuhui Zhang, Alfred Seunghoon Song, Robert Tibshirani, Serena Yeung-Levy• 2025

Related benchmarks

TaskDatasetResultRank
Image ClassificationRSNA (test)
AUC56.18
59
Image ClassificationMHIST (test)
Accuracy41.78
41
Image ClassificationGeneral Domain 31 tasks
CLS Score57.97
30
Image-Text RetrievalGeneral Domain
Retrieval Score30.31
30
Medical Image-Text ClassificationMedical Specialties
Ophthalmology Performance38.47
30
Image ClassificationSIIM (test)--
30
ClassificationMRI
Accuracy78.5
20
ClassificationCT
Accuracy54
20
ClassificationX-Ray
Accuracy50.5
20
Image-to-Text RetrievalMIMIC-CXR (test)
R@10.12
20
Showing 10 of 54 rows

Other info

Code

Follow for update