文献速递甲状腺与浅表技术与安全US

StableSeg GAN:闭环增益控制提升甲状腺结节超声分割稳定性

Improving GAN Learning Dynamics for Thyroid Nodule Segmentation.
PubMed 文献 · Ultrasound Med Biol · 2022-11-21

本研究提出一种结合监督语义分割与GAN无监督学习的混合算法StableSeg GAN,用于超声图像中甲状腺结节分割。针对GAN训练不稳定和模式崩溃问题,引入对判别器损失输出增益的闭环PID控制以稳定训练。在挑战性测试集上平均IoU达81.26%,优于DeeplabV3+,表明增益控制可兼顾低层精度与高层一致性。

要点

  • PID增益控制稳定GAN训练,避免判别器学习过快导致的模式崩溃。
  • StableSeg GAN平均IoU 81.26%,优于DeeplabV3+及无控制GAN。
  • 监督与无监督结合可同时提升分割的低层精度和高层一致性。

核心结论

  • 对判别器损失输出增益实施闭环PID控制可平滑生成器训练并避免模式崩溃。
  • StableSeg GAN在甲状腺结节超声分割中平均IoU达81.26%,优于DeeplabV3+及无控制GAN。

研究设计与样本

  • 方法学研究,提出StableSeg GAN模型,生成器为DeeplabV3+,判别器为Resnet18,采用PID控制稳定GAN学习过程。
  • 在挑战性测试集上进行甲状腺结节分割实验,与DeeplabV3+及无控制GAN比较IoU。

关键影像表现/征象

  • 增益控制使生成器训练更平滑,避免判别器相对生成器学习过快时常见的模式崩溃。
  • 监督与无监督学习风格结合鼓励低层精度和高层一致性。

诊断标准与关键数值

  • 以IoU为评价指标,StableSeg GAN在挑战性测试集上平均IoU为81.26%。

临床应用与阅片要点

  • 该算法可用于甲状腺结节超声图像的自动检测与分割,辅助医师诊断和随访。
  • 自动化工具除即时诊断外,还可支持随时间追踪恶性概率。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

Thyroid nodules are lesions requiring diagnosis and follow-up. Tools for detecting and segmenting nodules can help physicians with this diagnosis. Besides immediate diagnosis, automated tools can also enable tracking of the probability of malignancy over time. This paper demonstrates a new algorithm for segmenting thyroid nodules in ultrasound images. The algorithm combines traditional supervised semantic segmentation with unsupervised learning using GANs. The hybrid approach has the potential to upgrade the semantic segmentation model's performance, but GANs have the well-known problems of unstable learning and mode collapse. To stabilize the training of the GAN model, we introduce the concept of closed-loop control of the gain on the loss output of the discriminator. We find gain control leads to smoother generator training and avoids the mode collapse that typically occurs when the discriminator learns too quickly relative to the generator. We also find that the combination of the supervised and unsupervised learning styles encourages both low-level accuracy and high-level consistency. As a test of the concept of controlled hybrid supervised and unsupervised semantic segmentation, we introduce a new model named the StableSeg GAN. The model uses DeeplabV3+ as the generator, Resnet18 as the discriminator, and uses PID control to stabilize the GAN learning process. The performance of the new model in terms of IoU is better than DeeplabV3+, with mean IoU of 81.26% on a challenging test set. The results of our thyroid nodule segmentation experiments show that StableSeg GANs have flexibility to segment nodules more accurately than either comparable supervised segmentation models or uncontrolled GANs.

来源:Ultrasound Med Biol,查看原文。版权归原作者所有。

引用

  1. Kunapinun, Dailey, Songsaeng, et al. Improving GAN Learning Dynamics for Thyroid Nodule Segmentation. Ultrasound Med Biol. 2022. PMID: 36424307. doi:10.1016/j.ultrasmedbio.2022.09.010

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。