# PairAug: What Can Augmented Image-Text Pairs Do for Radiology?

arXiv: https://arxiv.org/abs/2404.04960 · HTML: https://mlsanity.com/paper/2404.04960/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Placed in Medical vision-language models, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 36 of its 48 references that are in the map.
- First posted on arXiv 7 Apr 2024.

## Placed in

- Multimodal LLMs / vision-language models / [Medical vision-language models](https://mlsanity.com/mllm/medical-vision-language-models/)

## Builds on (most cited first, 30 of 36)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [Prompt-to-Prompt Image Editing with Cross Attention Control](https://mlsanity.com/paper/2208.01626/) (2022)
- [Decoupled Weight Decay Regularization](https://mlsanity.com/paper/1711.05101/) (2017)
- [CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison](https://mlsanity.com/paper/1901.07031/) (2019)
- [Pseudo Numerical Methods for Diffusion Models on Manifolds](https://mlsanity.com/paper/2202.09778/) (2022)
- [MedCLIP: Contrastive Learning from Unpaired Medical Images and Text](https://mlsanity.com/paper/2210.10163/) (2022)
- [Video-P2P: Video Editing with Cross-attention Control](https://mlsanity.com/paper/2303.04761/) (2023)
- [PadChest: A large chest x-ray image dataset with multi-label annotated reports](https://mlsanity.com/paper/1901.07441/) (2019)
- [Generative Adversarial Network in Medical Imaging: A Review](https://mlsanity.com/paper/1809.07294/) (2018)
- [CoTr: Efficiently Bridging CNN and Transformer for 3D Medical Image Segmentation](https://mlsanity.com/paper/2103.03024/) (2021)
- [ResViT: Residual vision transformers for multi-modal medical image synthesis](https://mlsanity.com/paper/2106.16031/) (2021)
- [Multi-Granularity Cross-modal Alignment for Generalized Medical Visual Representation Learning](https://mlsanity.com/paper/2210.06044/) (2022)
- [Does Synthetic Data Generation of LLMs Help Clinical Text Mining?](https://mlsanity.com/paper/2303.04360/) (2023)
- [Data augmentation using learned transformations for one-shot medical image segmentation](https://mlsanity.com/paper/1902.09383/) (2019)
- [Privacy for Free: How does Dataset Condensation Help Privacy?](https://mlsanity.com/paper/2206.00240/) (2022)
- [Multi-Modal Masked Autoencoders for Medical Vision-and-Language Pre-Training](https://mlsanity.com/paper/2209.07098/) (2022)
- [Medical Image Synthesis with Context-Aware Generative Adversarial Networks](https://mlsanity.com/paper/1612.05362/) (2016)
- [RoentGen: Vision-Language Foundation Model for Chest X-ray Generation](https://mlsanity.com/paper/2211.12737/) (2022)
- [CXR-CLIP: Toward Large Scale Chest X-ray Language-Image Pre-training](https://mlsanity.com/paper/2310.13292/) (2023)
- [Generalized Radiograph Representation Learning via Cross-supervision between Images and Free-text Radiology Reports](https://mlsanity.com/paper/2111.03452/) (2021)
- [Advancing Radiograph Representation Learning with Masked Record Modeling](https://mlsanity.com/paper/2301.13155/) (2023)

## Built on by (most cited first, 10 of 10)

- [Has Multimodal Learning Delivered Universal Intelligence in Healthcare? A Comprehensive Survey](https://mlsanity.com/paper/2408.12880/) (2024)
- [MAKE: Multi-Aspect Knowledge-Enhanced Vision-Language Pretraining for Zero-shot Dermatological Assessment](https://mlsanity.com/paper/2505.09372/) (2025)
- [A Survey of Medical Vision-and-Language Applications and Their Techniques](https://mlsanity.com/paper/2411.12195/) (2024)
- [MedUniSeg: 2D and 3D Medical Image Segmentation via a Prompt-driven Universal Model](https://mlsanity.com/paper/2410.05905/) (2024)
- [Can Medical Vision-Language Pre-training Succeed with Purely Synthetic Data?](https://mlsanity.com/paper/2410.13523/) (2024)
- [SigVLP: Sigmoid Volume-Language Pre-Training for Self-Supervised CT-Volume Adaptive Representation Learning](https://mlsanity.com/paper/2602.21735/) (2026)
- [NeuroSeg Meets DINOv3: Transferring 2D Self-Supervised Visual Priors to 3D Neuron Segmentation via DINOv3 Initialization](https://mlsanity.com/paper/2603.23104/) (2026)
- [PathoHR: Hierarchical Reasoning for Vision-Language Models in Pathology](https://mlsanity.com/paper/2509.06105/) (2025)
- [Discrete Diffusion Models with MLLMs for Unified Medical Multimodal Generation](https://mlsanity.com/paper/2510.06131/) (2025)
- [Domain-invariant Representation Learning via Segment Anything Model for Blood Cell Classification](https://mlsanity.com/paper/2408.07467/) (2024)

## Abstract (arXiv)

Current vision-language pre-training (VLP) methodologies predominantly depend on paired image-text datasets, a resource that is challenging to acquire in radiology due to privacy considerations and labelling complexities. Data augmentation provides a practical solution to overcome the issue of data scarcity, however, most augmentation methods exhibit a limited focus, prioritising either image or text augmentation exclusively. Acknowledging this limitation, our objective is to devise a framework capable of concurrently augmenting medical image and text data. We design a Pairwise Augmentation (PairAug) approach that contains an Inter-patient Augmentation (InterAug) branch and an Intra-patient Augmentation (IntraAug) branch. Specifically, the InterAug branch of our approach generates radiology images using synthesised yet plausible reports derived from a Large Language Model (LLM). The generated pairs can be considered a collection of new patient cases since they are artificially created and may not exist in the original dataset. In contrast, the IntraAug branch uses newly generated reports to manipulate images. This process allows us to create new paired data for each individual with diverse medical conditions. Our extensive experiments on various downstream tasks covering medical image classification zero-shot and fine-tuning analysis demonstrate that our PairAug, concurrently expanding both image and text data, substantially outperforms image-/text-only expansion baselines and advanced medical VLP baselines. Our code is released at urlhttps://github.com/YtongXie/PairAug.