Application of Multilayer Information Fusion and Optimization Network Combined With Attention Mechanism in Polyp Segmentation

Abstract

Colorectal cancer (CRC) is a multifaceted disease, but it can be effectively prevented through colonoscopy for the detection of polyps. In clinical practice, the development of automatic polyp segmentation techniques for colonoscopy images can significantly enhance the efficiency and accuracy of polyp detection and help clinicians to precisely localize the polyps. However, existing segmentation methods have several obvious limitations: 1) inadequate utilization of multilevel features extracted by feature encoders; 2) ineffective aggregation of high- and low-level features; and 3) unclear delineation of polyp boundaries. To address these challenges while enhancing the clarity of polyp boundaries in segmentation, we propose a novel multilayer information fusion and optimization network (MIFONet) consisting of the following components: 1) contextual and fine feature processing (CFFP) module, employed to effectively extract both local and global contextual information; 2) hierarchical feature integration module (HFIM), added to facilitate efficient aggregation of processed high- and low-level features and strengthen the association between contextual features; 3) multiscale contextual attention (MSCA) module, used to deeply integrate aggregated high-level features with low-level features; and 4) a novel refinement module composed of an adaptive channel attention pyramid (ACAP) part and a skip-reverse attention (SRA) part, with the ability to capture fine-grained information and refining feature representation. We conducted extensive experiments and comparative analysis of our proposed model with 19 popular or state-of-the-art (SOTA) methods on five renowned polyp benchmark datasets. To further validate the model’s generalization performance, we also designed three cross-dataset experiments. Experimental results demonstrate that MIFONet consistently achieves excellent segmentation performance across most datasets. In particular, we achieve 94.6% mean Dice on the CVC-ClinicDB dataset, which obtains superior performance compared with SOTA methods.