Zhao Tengfei, Fu Chong, Tian Yunjia, Song Wei, Sham Chiu-Wing
School of Computer Science and Engineering, Northeastern University, Shenyang 110819, China.
Engineering Research Center of Security Technology of Complex Network System, Ministry of Education, Shenyang 110819, China.
Bioengineering (Basel). 2023 Mar 22;10(3):393. doi: 10.3390/bioengineering10030393.
Nuclei segmentation and classification are two basic and essential tasks in computer-aided diagnosis of digital pathology images, and those deep-learning-based methods have achieved significant success. Unfortunately, most of the existing studies accomplish the two tasks by splicing two related neural networks directly, resulting in repetitive computation efforts and a redundant-and-large neural network. Thus, this paper proposes a lightweight deep learning framework (GSN-HVNET) with an encoder-decoder structure for simultaneous segmentation and classification of nuclei. The decoder consists of three branches outputting the semantic segmentation of nuclei, the horizontal and vertical (HV) distances of nuclei pixels to their mass centers, and the class of each nucleus, respectively. The instance segmentation results are obtained by combing the outputs of the first and second branches. To reduce the computational cost and improve the network stability under small batch sizes, we propose two newly designed blocks, Residual-Ghost-SN (RGS) and Dense-Ghost-SN (DGS). Furthermore, considering the practical usage in pathological diagnosis, we redefine the classification principle of the CoNSeP dataset. Experimental results demonstrate that the proposed model outperforms other state-of-the-art models in terms of segmentation and classification accuracy by a significant margin while maintaining high computational efficiency.
细胞核分割与分类是数字病理图像计算机辅助诊断中的两项基本且重要的任务，基于深度学习的方法已取得显著成功。不幸的是，现有的大多数研究通过直接拼接两个相关神经网络来完成这两项任务，导致计算量重复且神经网络冗余庞大。因此，本文提出了一种具有编码器 - 解码器结构的轻量级深度学习框架（GSN - HVNET），用于细胞核的同时分割与分类。解码器由三个分支组成，分别输出细胞核的语义分割结果、细胞核像素到其质心的水平和垂直（HV）距离以及每个细胞核的类别。通过合并第一和第二个分支的输出获得实例分割结果。为了降低计算成本并提高小批量情况下网络的稳定性，我们提出了两个新设计的模块，即残差 - 幽灵 - SN（RGS）和密集 - 幽灵 - SN（DGS）。此外，考虑到病理诊断中的实际应用，我们重新定义了CoNSeP数据集的分类原则。实验结果表明，所提出的模型在分割和分类准确率方面显著优于其他现有先进模型，同时保持了较高的计算效率。
