School of Information Science and Technology of Fudan University, 220 Handan Rd, Shanghai, 200433, China.
Key Laboratory of Medical Imaging Computing and Computer Assisted Intervention (MICCAI) of Shanghai, Shanghai, 200032, China.
Med Biol Eng Comput. 2024 Dec;62(12):3781-3799. doi: 10.1007/s11517-024-03170-4. Epub 2024 Jul 20.
Data-driven medical image segmentation networks require expert annotations, which are hard to obtain. Non-expert annotations are often used instead, but these can be inaccurate (referred to as "noisy labels"), misleading the network's training and causing a decline in segmentation performance. In this study, we focus on improving the segmentation performance of neural networks when trained with noisy annotations. Specifically, we propose a two-stage framework named "G-T correcting," consisting of "G" stage for recognizing noisy labels and "T" stage for correcting noisy labels. In the "G" stage, a positive feedback method is proposed to automatically recognize noisy samples, using a Gaussian mixed model to classify clean and noisy labels through the per-sample loss histogram. In the "T" stage, a confident correcting strategy and early learning strategy are adopted to allow the segmentation network to receive productive guidance from noisy labels. Experiments on simulated and real-world noisy labels show that this method can achieve over 90% accuracy in recognizing noisy labels, and improve the network's DICE coefficient to 91%. The results demonstrate that the proposed method can enhance the segmentation performance of the network when trained with noisy labels, indicating good clinical application prospects.
数据驱动的医学图像分割网络需要专家注释，但这些注释难以获取。通常会使用非专家注释，但这些注释可能不准确（称为“噪声标签”），从而误导网络的训练并导致分割性能下降。在这项研究中，我们专注于提高使用噪声标签训练的神经网络的分割性能。具体来说，我们提出了一个名为“G-T 纠正”的两阶段框架，由“G”阶段识别噪声标签和“T”阶段纠正噪声标签组成。在“G”阶段，我们提出了一种正反馈方法，通过使用高斯混合模型根据样本损失直方图对干净和噪声标签进行分类，自动识别噪声样本。在“T”阶段，我们采用了置信度纠正策略和早期学习策略，以使分割网络从噪声标签中获得有成效的指导。在模拟和真实世界噪声标签上的实验表明，该方法可以实现超过 90%的噪声标签识别准确率，并将网络的 DICE 系数提高到 91%。结果表明，该方法可以在使用噪声标签训练网络时提高网络的分割性能，具有良好的临床应用前景。
