MFEFNet: Multi-scale feature enhancement and Fusion Network for polyp segmentation

• Published in: Computers in biology and medicine Vol. 157; p. 106735
• Main Authors: Xia, Yang, Yun, Haijiao, Liu, Yanjun
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.05.2023; Elsevier Limited
• Subjects: Algorithms; Attention mechanism; Benchmarking; Benchmarks; Colorectal cancer; Colorectal carcinoma; Computer networks; Deep learning; Experiments; Feature enhancement; Image Processing, Computer-Assisted; Internal Medicine; Medical research; Methods; Multi-scale fusion; Other; Polyp segmentation; Polyps; Segmentation; Semantics; Spine; State of the art; Strong associated coupler; Feature enhancement; Attention mechanism; Multi-scale fusion; Strong associated coupler; Polyp segmentation
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2023.106735

The polyp segmentation technology based on computer-aided can effectively avoid the deterioration of polyps and prevent colorectal cancer. To segment the polyp target precisely, the Multi-Scale Feature Enhancement and Fusion Network (MFEFNet) is proposed. First of all, to balance the network's predictive ability and complexity, ResNet50 is designed as the backbone network, and the Shift Channel Block (SCB) is used to unify the spatial location of feature mappings and emphasize local information. Secondly, to further improve the network's feature-extracting ability, the Feature Enhancement Block (FEB) is added, which decouples features, reinforces features by multiple perspectives and reconstructs features. Meanwhile, to weaken the semantic gap in the feature fusion process, we propose strong associated couplers, the Multi-Scale Feature Fusion Block (MSFFB) and the Reducing Difference Block (RDB), which are mainly composed of multiple cross-complementary information interaction modes and reinforce the long-distance dependence between features. Finally, to further refine local regions, the Polarized Self-Attention (PSA) and the Balancing Attention Module (BAM) are introduced for better exploration of detailed information between foreground and background boundaries. Experiments have been conducted under five benchmark datasets (Kvasir-SEG, CVC-ClinicDB, CVC-ClinicDB, CVC300 and CVC-ColonDB) and compared with state-of-the-art polyp segmentation algorithms. The experimental result shows that the proposed network improves Dice and mean intersection over union (mIoU) by an average score of 3.4% and 4%, respectively. Therefore, extensive experiments demonstrate that the proposed network performs favorably against more than a dozen state-of-the-art methods on five popular polyp segmentation benchmarks. •Proposing a polyp segmentation network MFEFNet based on CNN.•Simplifying parameters and emphasizing local information by unifying the spatial location.•Reinforcing feature engineering by decoupling, enhancing and reconstructing.•Integrating high-order features by cross-complementary information interaction modes to weaken semantic gaps.•Refining local regions by introducing efficient attention mechanisms.