PMC-CLIP: Contrastive Language-Image Pre-training using Biomedical Documents
Foundation models trained on large-scale dataset gain a recent surge in CV and NLP. In contrast, development in biomedical domain lags far behind due to data scarcity. To address this issue, we build and release PMC-OA, a biomedical dataset with 1.6M image-caption pairs collected from PubMedCentral's OpenAccess subset, which is 8 times larger than before. PMC-OA covers diverse modalities or diseases, with majority of the image-caption samples aligned at finer-grained level, i.e., subfigure and subcaption. While pretraining a CLIP-style model on PMC-OA, our model named PMC-CLIP achieves state-of-the-art results on various downstream tasks, including image-text retrieval on ROCO, MedMNIST image classification, Medical VQA, i.e. +8.1% R@10 on image-text retrieval, +3.9% accuracy on image classification.
PDF AbstractCode
VQA-RAD
SLAKE
| Task | Dataset | Model | Metric Name | Metric Value | Global Rank | Uses Extra Training Data |
Benchmark |
|---|---|---|---|---|---|---|---|
| Medical Visual Question Answering | PMC-VQA | PMC-CLIP | Accuracy | 24.7 | # 3 | ||
| Visual Question Answering (VQA) | PMC-VQA | PMC-CLIP | Accuracy | 24.7 | # 3 | ||
| Medical Visual Question Answering | VQA-RAD | PMC-CLIP | Close-ended Accuracy | 84.0 | # 5 | ||
| Open-ended Accuracy | 67.0 | # 5 | |||||
| Overall Accuracy | 77.6 | # 4 |
