College of Computer Science and Technology, Huaqiao University, Xiamen 361021, China.
Department of Medical Imaging, Guangdong Second Provincial General Hospital, Guangzhou, 510317, China.
Comput Methods Programs Biomed. 2021 Mar;200:105925. doi: 10.1016/j.cmpb.2020.105925. Epub 2021 Jan 10.
With the advancement of electron microscopy (EM) imaging technology, neuroscientists can investigate the function of various intracellular organelles, e.g, mitochondria, at nano-scale. Semantic segmentation of electron microscopy (EM) is an essential step to efficiently obtain reliable morphological statistics. Despite the great success achieved using deep convolutional neural networks (CNNs), they still produce coarse segmentations with lots of discontinuities and false positives for mitochondria segmentation.
In this study, we introduce a centerline-aware multitask network by utilizing centerline as an intrinsic shape cue of mitochondria to regularize the segmentation. Since the application of 3D CNNs on large medical volumes is usually hindered by their substantial computational cost and storage overhead, we introduce a novel hierarchical view-ensemble convolution (HVEC), a simple alternative of 3D convolution to learn 3D spatial contexts using more efficient 2D convolutions. The HVEC enables both decomposing and sharing multi-view information, leading to increased learning capacity.
Extensive validation results on two challenging benchmarks show that, the proposed method performs favorably against the state-of-the-art methods in accuracy and visual quality but with a greatly reduced model size. Moreover, the proposed model also shows significantly improved generalization ability, especially when training with quite limited amount of training data. Detailed sensitivity analysis and ablation study have also been conducted, which show the robustness of the proposed model and effectiveness of the proposed modules.
The experiments highlighted that the proposed architecture enables both simplicity and efficiency leading to increased capacity of learning spatial contexts. Moreover, incorporating shape cues such as centerline information is a promising approach to improve the performance of mitochondria segmentation.
随着电子显微镜（EM）成像技术的进步，神经科学家可以在纳米尺度上研究各种细胞内细胞器的功能，例如线粒体。电子显微镜（EM）的语义分割是获得可靠形态学统计数据的重要步骤。尽管深度卷积神经网络（CNNs）取得了巨大的成功，但它们在进行线粒体分割时仍然会产生粗糙的分割，存在许多不连续和假阳性。
在这项研究中，我们引入了一种基于中心线感知的多任务网络，通过利用中心线作为线粒体的固有形状线索来对分割进行正则化。由于 3D CNN 在大型医学体积上的应用通常受到其大量计算成本和存储开销的限制，我们引入了一种新的分层视图集成卷积（HVEC），这是 3D 卷积的一种简单替代方法，可使用更有效的 2D 卷积来学习 3D 空间上下文。HVEC 能够分解和共享多视图信息，从而提高学习能力。
在两个具有挑战性的基准上进行的广泛验证结果表明，与最先进的方法相比，该方法在准确性和视觉质量方面表现出色，但模型尺寸大大减小。此外，该方法还表现出显著提高的泛化能力，尤其是在使用相当有限的训练数据进行训练时。还进行了详细的敏感性分析和消融研究，这表明了所提出模型的稳健性和所提出模块的有效性。
实验强调了所提出的架构既简单又高效，从而提高了学习空间上下文的能力。此外，纳入中心线等形状线索是提高线粒体分割性能的一种有前途的方法。
