CD-Net: Cascaded 3D Dilated convolutional neural network for pneumonia lesion segmentation

• Published in: Computers in biology and medicine 2024-05, Vol.173, p.108311, Article 108311
• Main Authors: Zhang, Jinli, Wang, Shaomeng, Jiang, Zongli, Chen, Zhijie, Bai, Xiaolu
• Format: Article
• Language: English
• Subjects: Artificial neural networks; Computed tomography; Convolutional neural network; COVID-19; COVID-19 - diagnostic imaging; Humans; Image Processing, Computer-Assisted - methods; Lesion segmentation; Lesions; Neural networks; Neural Networks, Computer; Pandemics; Parameters; Pneumonia; Segmentation; Social interactions; Tomography, X-Ray Computed; Transfer learning
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2024.108311

COVID-19 is a global pandemic that has caused significant global, social, and economic disruption. To effectively assist in screening and monitoring diagnosed cases, it is crucial to accurately segment lesions from Computer Tomography (CT) scans. Due to the lack of labeled data and the presence of redundant parameters in 3D CT, there are still significant challenges in diagnosing COVID-19 in related fields. To address the problem, we have developed a new model called the Cascaded 3D Dilated convolutional neural network (CD-Net) for directly processing CT volume data. To reduce memory consumption when cutting volume data into small patches, we initially design a cascade architecture in CD-Net to preserve global information. Then, we construct a Multi-scale Parallel Dilated Convolution (MPDC) block to aggregate features of different sizes and simultaneously reduce the parameters. Moreover, to alleviate the shortage of labeled data, we employ classical transfer learning, which requires only a small amount of data while achieving better performance. Experimental results conducted on the different public-available datasets verify that the proposed CD-Net has reduced the negative–positive ratio and outperformed other existing segmentation methods while requiring less data. •Cascade-wise attention can prevent the loss of scan integrality.•MPDC blocks can aggregate features of different sizes and reduce parameters•CD-Net utilizes transfer learning to extract information during pre-training•Experimental results on COVID-19 CT-MD dataset embody the superiority of our model.