Vision-Language Pre-training (VLP) that utilizes the multi-modal information to promote the training efficiency and effectiveness, has achieved great success in vision recognition of natural domains and shown promise in medical imaging diagnosis for the Chest X-Rays (CXRs). However, current works mainly pay attention to the exploration on single dataset of CXRs, which locks the potential of this powerful paradigm on larger hybrid of multi-source CXRs datasets. We identify that although blending samples from the diverse sources offers the advantages to improve the model generalization, it is still challenging to maintain the consistent superiority for the task of each source due to the existing heterogeneity among sources. To handle this dilemma, we design a Conquer-and-Divide pre-training framework, termed as UniChest, aiming to make full use of the collaboration benefit of multiple sources of CXRs while reducing the negative influence of the source heterogeneity. Specially, the “Conquer” stage in UniChest encourages the model to sufficiently capture multi-source common patterns, and the “Divide” stage helps squeeze personalized patterns into different small experts (query networks). We conduct thorough experiments on many benchmarks, e.g., ChestX-ray14, CheXpert, Vindr-CXR, Shenzhen, Open-I and SIIM-ACR Pneumothorax, verifying the effectiveness of UniChest over a range of baselines, and release our codes and pre-training models at https://github.com/Elfenreigen/UniChest .

中文翻译：

---

UniChest：多源胸部 X 射线分类的征服和划分预训练


视觉语言预训练（VLP）利用多模态信息提高训练效率和效果，在自然领域的视觉识别方面取得了巨大成功，并在胸部X光（CXR）的医学影像诊断中显示出应用前景。然而，当前的工作主要关注对单个 CXR 数据集的探索，这将这种强大范式的潜力锁定在更大的多源 CXR 数据集混合上。我们发现，尽管混合来自不同来源的样本可以提供提高模型泛化能力的优势，但由于来源之间存在异质性，保持每个来源的任务的一致优势仍然具有挑战性。为了解决这个困境，我们设计了一个征服和划分预训练框架，称为UniChest，旨在充分利用多源CXR的协作优势，同时减少源异构性的负面影响。特别是，UniChest 中的“征服”阶段鼓励模型充分捕获多源共同模式，而“划分”阶段有助于将个性化模式压缩到不同的小专家（查询网络）中。我们在ChestX-ray14、CheXpert、Vindr-CXR、Shenzhen、Open-I和SIIM-ACR Pneumothorax等许多基准上进行了深入的实验，验证了UniChest在一系列基准上的有效性，并发布了我们的代码和预训练模型在https://github.com/Elfenreigen/UniChest 。