High performance of deep learning on medical image segmentation rely on large-scale pixel-level dense annotations, which poses a substantial burden on medical experts due to the laborious and time-consuming annotation process, particularly for 3D images. To reduce the labeling cost as well as maintain relatively satisfactory segmentation performance, weakly-supervised learning with sparse labels has attained increasing attentions. In this work, we present a scribble-based framework for medical image segmentation, called Dynamically Mixed Soft Pseudo-label Supervision (DMSPS). Concretely, we extend a backbone with an auxiliary decoder to form a dual-branch network to enhance the feature capture capability of the shared encoder. Considering that most pixels do not have labels and hard pseudo-labels tend to be over-confident to result in poor segmentation, we propose to use soft pseudo-labels generated by dynamically mixing the decoders’ predictions as auxiliary supervision. To further enhance the model’s performance, we adopt a two-stage approach where the sparse scribbles are expanded based on predictions with low uncertainties from the first-stage model, leading to more annotated pixels to train the second-stage model. Experiments on ACDC dataset for cardiac structure segmentation, WORD dataset for 3D abdominal organ segmentation and BraTS2020 dataset for 3D brain tumor segmentation showed that: (1) compared with the baseline, our method improved the average DSC from 50.46% to 89.51%, from 75.46% to 87.56% and from 52.61% to 76.53% on the three datasets, respectively; (2) DMSPS achieved better performance than five state-of-the-art scribble-supervised segmentation methods, and is generalizable to different segmentation backbones. The code is available online at: https://github.com/HiLab-git/DMSPS.

中文翻译：

---

DMSPS：用于涂鸦监督医学图像分割的动态混合软伪标签监督


深度学习在医学图像分割上的高性能依赖于大规模像素级密集注释，由于注释过程费力且耗时，这给医学专家带来了沉重的负担，特别是对于3D图像。为了降低标记成本并保持相对令人满意的分割性能，稀疏标签的弱监督学习受到越来越多的关注。在这项工作中，我们提出了一种基于涂鸦的医学图像分割框架，称为动态混合软伪标签监督（DMSPS）。具体来说，我们用辅助解码器扩展主干，形成双分支网络，以增强共享编码器的特征捕获能力。考虑到大多数像素没有标签，并且硬伪标签往往过于自信而导致分割效果不佳，我们建议使用通过动态混合解码器的预测生成的软伪标签作为辅助监督。为了进一步提高模型的性能，我们采用两阶段方法，根据第一阶段模型的不确定性较低的预测来扩展稀疏涂鸦，从而获得更多带注释的像素来训练第二阶段模型。在用于心脏结构分割的 ACDC 数据集、用于 3D 腹部器官分割的 WORD 数据集和用于 3D 脑肿瘤分割的 BraTS2020 数据集上的实验表明：（1）与基线相比，我们的方法将平均 DSC 从 75.46 提高到 50.46% 提高到 89.51%三个数据集分别从 % 提高到 87.56% 和从 52.61% 提高到 76.53%； (2) DMSPS 比五种最先进的涂鸦监督分割方法取得了更好的性能，并且可推广到不同的分割主干。 该代码可在线获取：https://github.com/HiLab-git/DMSPS。