arXiv 2311.103192024-05-17cs.CVcs.AI

转向机器监督：用于医学图像自动分割与分类的低标注成本半监督与自监督学习

Shifting to Machine Supervision: Annotation-Efficient Semi and Self-Supervised Learning for Automatic Medical Image Segmentation and Classification

Center for Data Science, New York University（纽约大学数据科学中心）；Courant Institute of Mathematical Sciences, New York University（纽约大学柯朗数学科学研究所）；NYU Grossman School of Medicine（纽约大学格罗斯曼医学院）；Colton Center for Autoimmunity, NYU Grossman School of Medicine（纽约大学格罗斯曼医学院科尔顿自身免疫中心）

Pranav Singh, Raviteja Chukkapalli, Shravan Chaudhari, Luoyao Chen, Mei Chen, Jinqian Pan, Craig Smuda, Jacopo Cirrone

备注：: Seventeen pages (incl. references), five figures, and one table. Accepted and published in Scientific Reports 14.1 (2024): 10820
发表信息：: Singh, P., Chukkapalli, R., Chaudhari, S. et al. Shifting to machine supervision: annotation-efficient semi and self-supervised learning for automatic medical image segmentation and classification. Sci Rep 14, 10820 (2024)
DOI：: 10.1038/s41598-024-61822-9

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对医学影像监督学习标注成本高的问题，提出S4MI流水线，结合自监督与半监督学习，在三个数据集上分别实现分类性能超越全监督、分割用50%标注量超全监督的效果，并开源代码。

摘要：

临床治疗的发展日益受到监督学习技术局限性的制约，这类技术高度依赖大量标注数据。标注过程不仅成本高昂，还需要临床专家投入大量时间。针对这一问题，我们提出了S4MI（医学影像自监督与半监督，Self-Supervision and Semi-Supervision for Medical Imaging）流水线，这是一种利用自监督与半监督学习前沿进展的新方法。这些技术会执行无需标注的辅助任务，因此与全监督方法相比，更易扩展机器监督的规模。我们的研究在三个不同的医学影像数据集上对这些技术进行基准测试，以评估其在分类和分割任务中的效果。值得注意的是，我们观察到在所有评估数据集的分类任务中，自监督学习的性能都显著优于监督方法。尤为突出的是，半监督方法在分割任务中表现更优，在所有数据集上使用的标注量减少50%的同时，性能超过了全监督方法。秉持为科学界做贡献的理念，我们已将S4MI代码公开，以支持这些方法的更广泛应用与进一步开发。

英文摘要：

Advancements in clinical treatment are increasingly constrained by the limitations of supervised learning techniques, which depend heavily on large volumes of annotated data. The annotation process is not only costly but also demands substantial time from clinical specialists. Addressing this issue, we introduce the S4MI (Self-Supervision and Semi-Supervision for Medical Imaging) pipeline, a novel approach that leverages advancements in self-supervised and semi-supervised learning. These techniques engage in auxiliary tasks that do not require labeling, thus simplifying the scaling of machine supervision compared to fully-supervised methods. Our study benchmarks these techniques on three distinct medical imaging datasets to evaluate their effectiveness in classification and segmentation tasks. Notably, we observed that self supervised learning significantly surpassed the performance of supervised methods in the classification of all evaluated datasets. Remarkably, the semi-supervised approach demonstrated superior outcomes in segmentation, outperforming fully-supervised methods while using 50% fewer labels across all datasets. In line with our commitment to contributing to the scientific community, we have made the S4MI code openly accessible, allowing for broader application and further development of these methods.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。