Lung segmentation in chest X‐ray image using multi‐interaction feature fusion network

• DOI: 10.1049/ipr2.12923
• Publication Date: 2023-09-13T05:31:08Z
• Authors: Xuebin Xu; Meng Lei; Dehua Liu; Muyu Wang; Longbin Lu
• Keywords: Feature (linguistics)

Abstract Lung segmentation is an essential step in a computer‐aided diagnosis system for chest radiographs. The lung parenchyma first segmented pulmonary systems to remove the interference of non‐lung regions while increasing effectiveness subsequent work. Nevertheless, most medical image methods nowadays use U‐Net and its variants. These variant networks perform poorly detect smaller structures cannot accurately segment boundary regions. A multi‐interaction feature fusion network model based on Kiu‐Net presented this paper address problem. Specifically, Ki‐Net are utilized extract high‐level detailed features images, respectively. Then, cross‐residual modules employed encoding stage obtain complementary from these two networks. Second, global information module introduced guarantee region's integrity. Finally, decoding stage, presented, which allows interact with multiple kinds information, such as contextual branching features, fused more practical information. performance proposed was assessed both Montgomery County (MC) Shenzhen datasets, demonstrating superiority over existing according experimental results.