Large medical imaging datasets can be cheaply and quickly annotated with low-confidence, weak labels (e.g., radiological scores). Access to high-confidence labels, such as histology-based diagnoses, is rare and costly. Pretraining strategies, like contrastive learning (CL) methods, can leverage unlabeled or weakly-annotated datasets. These methods typically require large batch sizes, which poses a difficulty in the case of large 3D images at full resolution, due to limited GPU memory. Nevertheless, volumetric positional information about the spatial context of each 2D slice can be very important for some medical applications. In this work, we propose an efficient weakly-supervised positional (WSP) contrastive learning strategy where we integrate both the spatial context of each 2D slice and a weak label via a generic kernel-based loss function. We illustrate our method on cirrhosis prediction using a large volume of weakly-labeled images, namely radiological low-confidence annotations, and small strongly-labeled (i.e., high-confidence) datasets. The proposed model improves the classification AUC by 5% with respect to a baseline model on our internal dataset, and by 26% on the public LIHC dataset from the Cancer Genome Atlas. The code is available at: https://github.com/Guerbet-AI/wsp-contrastive.

翻译：大型医学影像数据集可以低成本且快速地通过低置信度弱标签（如放射学评分）进行标注，而获取高置信度标签（如基于组织学的诊断）则既罕见又昂贵。对比学习等预训练策略可利用未标注或弱标注数据集。这些方法通常需要较大的批量大小，而在全分辨率大尺寸3D图像场景中，由于GPU内存限制，这带来了困难。然而，对于某些医学应用而言，关于每个2D切片空间上下文的体积位置信息可能非常重要。在本工作中，我们提出了一种高效的弱监督位置对比学习策略，通过基于核的通用损失函数，将每个2D切片的空间上下文与弱标签相结合。我们利用大量弱标注图像（即低置信度放射学注释）和小规模强标注（即高置信度）数据集，在肝硬化预测任务上展示了本方法的有效性。与内部数据集上的基线模型相比，所提模型将分类AUC提升了5%，在癌症基因组图谱公共LIHC数据集上提升了26%。代码见：https://github.com/Guerbet-AI/wsp-contrastive。