He Qingyuan, Yan Kun, Luo Qipeng, Yi Duan, Wang Ping, Han Hongbin, Liu Defeng
Radiology Department, Peking University Third Hospital, Beijing, China.
Peking University Third Hospital, Beijing Key Laboratory of Magnetic Resonance Imaging Devices and Technology, Beijing, China.
Health Data Sci. 2024 Aug 5;4:0166. doi: 10.34133/hds.0166. eCollection 2024.
MRI segmentation offers crucial insights for automatic analysis. Although deep learning-based segmentation methods have attained cutting-edge performance, their efficacy heavily relies on vast sets of meticulously annotated data. In this study, we propose a novel semi-supervised MRI segmentation model that is able to explore unlabeled data in multiple aspects based on various semi-supervised learning technologies. We compared the performance of our proposed method with other deep learning-based methods on 2 public datasets, and the results demonstrated that we have achieved Dice scores of 90.3% and 89.4% on the LA and ACDC datasets, respectively. We explored the synergy of various semi-supervised learning technologies for MRI segmentation, and our investigation will inspire research that focuses on designing MRI segmentation models.
磁共振成像（MRI）分割为自动分析提供了关键见解。尽管基于深度学习的分割方法已取得前沿性能，但其功效严重依赖大量精心标注的数据。在本研究中，我们提出了一种新颖的半监督MRI分割模型，该模型能够基于各种半监督学习技术从多个方面探索未标注数据。我们在两个公共数据集上，将我们提出的方法与其他基于深度学习的方法的性能进行了比较，结果表明我们在左心房（LA）和自动心脏诊断挑战赛（ACDC）数据集上分别取得了90.3%和89.4%的骰子系数（Dice scores）。我们探索了各种半监督学习技术在MRI分割中的协同作用，我们的研究将激发专注于设计MRI分割模型的研究。
