Li Huaidong, Li Delong, Dong Qing, Han Xue, Dong Suyu
College of Computer and Control Engineering, Northeast Forestry University, Harbin, China.
Department of Thoracic Surgery at No. 4 Affiliated Hospital, Harbin Medical University, Harbin, China.
Front Cardiovasc Med. 2025 May 15;12:1461774. doi: 10.3389/fcvm.2025.1461774. eCollection 2025.
In recent years, semi-supervised methods have been rapidly developed for three-dimensional (3D) medical image analysis. However, previous semi-supervised methods for three-dimensional medical images usually focused on single-view information and required a large number of annotated datasets. In this paper, we innovatively propose a multi-view (coronal and transverse) attention network for semi-supervised 3D cardiac image segmentation. In this way, the proposed method obtained more complementary segmentation information, which improved the segmentation performance. Simultaneously, we integrated the CBAM module and adaptive channel attention block into the 3D VNet (CBAP - VNet) to enhance the focus on the segmentation regions and edge portions. We first introduced the CutMix data augmentation mechanism to enhance 3D cardiac medical image segmentation. In this way, the proposed method made full use of the mixed regions in the images and expanded the training dataset. Our method was tested on two publicly available cardiac datasets and achieved good segmentation results. Our code and models are available at https://github.com/HuaidongLi-NEFU/TPSSAN.
近年来，用于三维（3D）医学图像分析的半监督方法得到了快速发展。然而，以往用于三维医学图像的半监督方法通常侧重于单视图信息，并且需要大量带注释的数据集。在本文中，我们创新性地提出了一种用于半监督3D心脏图像分割的多视图（冠状面和横断面）注意力网络。通过这种方式，所提出的方法获得了更多互补的分割信息，从而提高了分割性能。同时，我们将CBAM模块和自适应通道注意力块集成到3D VNet（CBAP - VNet）中，以增强对分割区域和边缘部分的关注。我们首先引入了CutMix数据增强机制来增强3D心脏医学图像分割。通过这种方式，所提出的方法充分利用了图像中的混合区域并扩展了训练数据集。我们的方法在两个公开可用的心脏数据集上进行了测试，并取得了良好的分割结果。我们的代码和模型可在https://github.com/HuaidongLi-NEFU/TPSSAN获取。
