Sun Yue, Gao Kun, Niu Sijie, Lin Weili, Li Gang, Wang Li
Department of Radiology and Biomedical Research Imaging Center, University of North Carolina at Chapel Hill, Chapel Hill, USA.
Mach Learn Med Imaging. 2020 Oct;12436:663-673. doi: 10.1007/978-3-030-59861-7_67. Epub 2020 Sep 29.
To characterize early cerebellum development, accurate segmentation of the cerebellum into white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF) tissues is one of the most pivotal steps. However, due to the weak tissue contrast, extremely folded tiny structures, and severe partial volume effect, infant cerebellum tissue segmentation is especially challenging, and the manual labels are hard to obtain and correct for learning-based methods. To the best of our knowledge, there is no work on the cerebellum segmentation for infant subjects less than 24 months of age. In this work, we develop a semi-supervised transfer learning framework guided by a confidence map for tissue segmentation of cerebellum MR images from 24-month-old to 6-month-old infants. Note that only 24-month-old subjects have reliable manual labels for training, due to their high tissue contrast. Through the proposed semi-supervised transfer learning, the labels from 24-month-old subjects are gradually propagated to the 18-, 12-, and 6-month-old subjects, which have a low tissue contrast. Comparison with the state-of-the-art methods demonstrates the superior performance of the proposed method, especially for 6-month-old subjects.
为了表征小脑的早期发育，将小脑准确分割为白质（WM）、灰质（GM）和脑脊液（CSF）组织是最关键的步骤之一。然而，由于组织对比度弱、结构极度折叠且微小，以及严重的部分容积效应，婴儿小脑组织分割极具挑战性，并且手动标注难以获取，也难以用于基于学习的方法进行校正。据我们所知，尚无针对24个月以下婴儿小脑分割的研究工作。在这项研究中，我们开发了一种由置信度图引导的半监督迁移学习框架，用于对24个月至6个月大婴儿的小脑磁共振图像进行组织分割。需要注意的是，由于24个月大的受试者组织对比度高，只有他们拥有可靠的手动标注用于训练。通过所提出的半监督迁移学习，来自24个月大受试者的标注逐渐传播到组织对比度低的18个月、12个月和6个月大的受试者。与现有最先进方法的比较证明了所提方法的优越性能，特别是对于6个月大的受试者。
