IEEE Trans Med Imaging. 2023 Oct;42(10):2988-2999. doi: 10.1109/TMI.2023.3273158. Epub 2023 Oct 2.
Semi-supervised learning (SSL) is a promising machine learning paradigm to address the ubiquitous issue of label scarcity in medical imaging. The state-of-the-art SSL methods in image classification utilise consistency regularisation to learn unlabelled predictions which are invariant to input level perturbations. However, image level perturbations violate the cluster assumption in the setting of segmentation. Moreover, existing image level perturbations are hand-crafted which could be sub-optimal. In this paper, we propose MisMatch, a semi-supervised segmentation framework based on the consistency between paired predictions which are derived from two differently learnt morphological feature perturbations. MisMatch consists of an encoder and two decoders. One decoder learns positive attention for foreground on unlabelled data thereby generating dilated features of foreground. The other decoder learns negative attention for foreground on the same unlabelled data thereby generating eroded features of foreground. We normalise the paired predictions of the decoders, along the batch dimension. A consistency regularisation is then applied between the normalised paired predictions of the decoders. We evaluate MisMatch on four different tasks. Firstly, we develop a 2D U-net based MisMatch framework and perform extensive cross-validation on a CT-based pulmonary vessel segmentation task and show that MisMatch statistically outperforms state-of-the-art semi-supervised methods. Secondly, we show that 2D MisMatch outperforms state-of-the-art methods on an MRI-based brain tumour segmentation task. We then further confirm that 3D V-net based MisMatch outperforms its 3D counterpart based on consistency regularisation with input level perturbations, on two different tasks including, left atrium segmentation from 3D CT images and whole brain tumour segmentation from 3D MRI images. Lastly, we find that the performance improvement of MisMatch over the baseline might originate from its better calibration. This also implies that our proposed AI system makes safer decisions than the previous methods.
半监督学习 (SSL) 是一种很有前途的机器学习范例，可以解决医学成像中普遍存在的标签稀缺问题。图像分类中的最新 SSL 方法利用一致性正则化来学习不变输入级别的扰动的未标记预测。然而，图像级别的扰动违反了分割设置中的聚类假设。此外，现有的图像级别的扰动是手工制作的，可能不是最优的。在本文中，我们提出了 MisMatch，这是一种基于一致性的半监督分割框架，该框架基于从两个不同学习的形态特征扰动中得出的配对预测之间的一致性。MisMatch 由编码器和两个解码器组成。一个解码器在未标记数据上学习用于前景的正注意力，从而生成前景的扩张特征。另一个解码器在相同的未标记数据上学习用于前景的负注意力，从而生成前景的侵蚀特征。我们沿批维度对解码器的配对预测进行归一化。然后在解码器的归一化配对预测之间应用一致性正则化。我们在四个不同的任务上评估了 MisMatch。首先，我们开发了一个基于 2D U-net 的 MisMatch 框架，并在基于 CT 的肺血管分割任务上进行了广泛的交叉验证，结果表明 MisMatch 在统计上优于最新的半监督方法。其次，我们表明 2D MisMatch 在基于 MRI 的脑肿瘤分割任务上优于最新的方法。然后，我们进一步证实，基于输入级别的扰动的一致性正则化，3D V-net 基于的 MisMatch 在两个不同的任务上优于其 3D 对应物，包括从 3D CT 图像分割左心房和从 3D MRI 图像分割整个脑肿瘤。最后，我们发现 MisMatch 相对于基线的性能提升可能源于其更好的校准。这也意味着我们提出的人工智能系统比以前的方法做出更安全的决策。
