Medical instrument segmentation in 3D ultrasound is essential for image-guided intervention. However, to train a successful deep neural network for instrument segmentation, a large number of labeled images are required, which is expensive and time-consuming to obtain. In this article, we propose a semi-supervised learning (SSL) framework for instrument segmentation in 3D US, which requires much less annotation effort than the existing methods. To achieve the SSL learning, a Dual-UNet is proposed to segment the instrument. The Dual-UNet leverages unlabeled data using a novel hybrid loss function, consisting of uncertainty and contextual constraints. Specifically, the uncertainty constraints leverage the uncertainty estimation of the predictions of the UNet, and therefore improve the unlabeled information for SSL training. In addition, contextual constraints exploit the contextual information of the training images, which are used as the complementary information for voxel-wise uncertainty estimation. Extensive experiments on multiple ex-vivo and in-vivo datasets show that our proposed method achieves Dice score of about 68.6%-69.1% and the inference time of about 1 sec. per volume. These results are better than the state-of-the-art SSL methods and the inference time is comparable to the supervised approaches.

中文翻译：

---

通过混合约束半监督学习在 3D US 中进行医疗器械分割

3D 超声中的医疗器械分割对于图像引导干预至关重要。然而，要训练一个成功的深度神经网络进行仪器分割，需要大量标记图像，这既昂贵又耗时。在本文中，我们提出了一种用于 3D US 仪器分割的半监督学习 (SSL) 框架，与现有方法相比，该框架需要更少的注释工作。为了实现 SSL 学习，提出了一个 Dual-UNet 来分割仪器。Dual-UNet 使用由不确定性和上下文约束组成的新型混合损失函数来利用未标记的数据。具体来说，不确定性约束利用了 UNet 预测的不确定性估计，从而改进了 SSL 训练的未标记信息。此外，上下文约束利用训练图像的上下文信息，用作体素不确定性估计的补充信息。对多个体外和体内数据集的大量实验表明，我们提出的方法实现了约 68.6%-69.1% 的 Dice 分数和约 1 秒的推理时间。每卷。这些结果优于最先进的 SSL 方法，推理时间与监督方法相当。