IEEE Trans Med Imaging. 2020 Jan;39(1):1-10. doi: 10.1109/TMI.2019.2918181. Epub 2019 May 22.
Automatic and accurate classification of apoptosis, or programmed cell death, will facilitate cell biology research. The state-of-the-art approaches in apoptosis classification use deep convolutional neural networks (CNNs). However, these networks are not efficient in encoding the part-whole relationships, thus requiring a large number of training samples to achieve robust generalization. This paper proposes an efficient variant of capsule networks (CapsNets) as an alternative to CNNs. Extensive experimental results demonstrate that the proposed CapsNets achieve competitive performances in target cell apoptosis classification, while significantly outperforming CNNs when the number of training samples is small. To utilize temporal information within microscopy videos, we propose a recurrent CapsNet constructed by stacking a CapsNet and a bi-directional long short-term recurrent structure. Our experiments show that when considering temporal constraints, the recurrent CapsNet achieves 93.8% accuracy and makes significantly more consistent prediction than NNs.
自动且准确地对细胞凋亡（程序性细胞死亡）进行分类，将有助于细胞生物学研究。目前在细胞凋亡分类方面使用的最先进方法是深度卷积神经网络（CNN）。然而，这些网络在对整体与部分关系进行编码时效率不高，因此需要大量的训练样本才能实现稳健的泛化。本文提出了一种胶囊网络（CapsNets）的有效变体，作为 CNN 的替代方案。大量的实验结果表明，所提出的 CapsNets 在目标细胞凋亡分类方面具有竞争力，而在训练样本数量较少时，其性能明显优于 CNN。为了利用显微镜视频中的时间信息，我们提出了一种由胶囊网络和双向长短期记忆结构堆叠而成的递归胶囊网络。我们的实验表明，在考虑时间约束的情况下，递归胶囊网络的准确率达到 93.8%，并且比神经网络做出的预测更加一致。
