Well-annotated medical datasets enable deep neural networks (DNNs) to gain strong power in extracting lesion-related features. Building such large and well-designed medical datasets is costly due to the need for high-level expertise. Model pre-training based on ImageNet is a common practice to gain better generalization when the data amount is limited. However, it suffers from the domain gap between natural and medical images. In this work, we pre-train DNNs on ultrasound (US) domains instead of ImageNet to reduce the domain gap in medical US applications. To learn US image representations based on unlabeled US videos, we propose a novel meta-learning-based contrastive learning method, namely Meta Ultrasound Contrastive Learning (Meta-USCL). To tackle the key challenge of obtaining semantically consistent sample pairs for contrastive learning, we present a positive pair generation module along with an automatic sample weighting module based on meta-learning. Experimental results on multiple computer-aided diagnosis (CAD) problems, including pneumonia detection, breast cancer classification, and breast tumor segmentation, show that the proposed self-supervised method reaches state-of-the-art (SOTA). The codes are available at https://github.com/Schuture/Meta-USCL.

翻译：有良好说明的医学数据集使深神经网络(DNNS)能够在提取与病病有关的特征方面获得强大的力量。建立如此庞大和设计良好的医疗数据集,由于需要高层次的专门知识,成本很高。基于图像网络的示范预培训是一种常见的做法,目的是在数据数量有限的情况下更好地概括化。然而,它存在自然和医学图像之间的领域差距。在这项工作中,我们在超声波(美国)域而不是图像网对DNNS进行了预先培训,以缩小美国医疗应用的域间差距。为了了解美国基于无标签的美国视频的图像显示,我们建议采用基于新颖的基于元学习的对比学习方法,即Meta Utrasound Contrastical Learning(Meta-USCL) 。为了应对为对比性学习获得具有立体一致性的样本配对的关键挑战,我们提出了一个积极的对子生成模块以及基于元学习的自动样本加权模块。关于多种计算机辅助诊断(CAD)问题的实验结果,包括肺炎检测、乳腺癌分类和乳腺肿瘤分解(MS-CLAS),显示拟议的自我监督/CLAS-SUTAS-SUS-S-S-S-S-S-SUDAR-S-S-SUDAR-S-SUDAR-S-S-SAS-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-SU-SU-SAT-S-SAT-TAR-TAS-SAT-S-S-S-S-S-S-S-TAS-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-SAT-S-TAS-TAR-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S