Abstract:Colorectal malignancies often arise from adenomatous polyps, which typically begin as solitary, asymptomatic growths before progressing to malignancy. Colonoscopy is widely recognized as a highly efficacious clinical polyp detection method, offering valuable visual data that facilitates precise identification and subsequent removal of these tumors. Nevertheless, accurately segmenting individual polyps poses a considerable difficulty because polyps exhibit intricate and changeable characteristics, including shape, size, color, quantity and growth context during different stages. The presence of similar contextual structures around polyps significantly hampers the performance of commonly used convolutional neural network (CNN)-based automatic detection models to accurately capture valid polyp features, and these large receptive field CNN models often overlook the details of small polyps, which leads to the occurrence of false detections and missed detections. To tackle these challenges, we introduce a novel approach for automatic polyp segmentation, known as the multi-distance feature dissimilarity-guided fully convolutional network. This approach comprises three essential components, i.e., an encoder-decoder, a multi-distance difference (MDD) module and a hybrid loss (HL) module. Specifically, the MDD module primarily employs a multi-layer feature subtraction (MLFS) strategy to propagate features from the encoder to the decoder, which focuses on extracting information differences between neighboring layers' features at short distances, and both short and long-distance feature differences across layers. Drawing inspiration from pyramids, the MDD module effectively acquires discriminative features from neighboring layers or across layers in a continuous manner, which helps to strengthen feature complementary across different layers. The HL module is responsible for supervising the feature maps extracted at each layer of the network to improve prediction accuracy. Our experimental results on four challenge datasets demonstrate that the proposed approach exhibits superior automatic polyp performance in terms of the six evaluation criteria compared to five current state-of-the-art approaches.

Using Guided Self-Attention with Local Information for Polyp Segmentation

SAEFormer: stepwise attention emphasis transformer for polyp segmentation

A lighter hybrid feature fusion framework for polyp segmentation

PPFormer: A Novel Model for Polyp Segmentation in Digestive Endoscopy

DHAFormer: Dual-channel hybrid attention network with transformer for polyp segmentation

W-MAFormer: W-shaped Multi-Attention Assisted Transformer for Polyp Segmentation

NA-segformer: A multi-level transformer model based on neighborhood attention for colonoscopic polyp segmentation

A Local Region Proposals Approach to Instance Segmentation for Intestinal Polyp Detection

Probabilistic Modeling Ensemble Vision Transformer Improves Complex Polyp Segmentation

SegT: A Novel Separated Edge-guidance Transformer Network for Polyp Segmentation

BCL-Former: Localized Transformer Fusion with Balanced Constraint for polyp image segmentation

Toward Accurate Polyp Segmentation with Cascade Boundary-Guided Attention

Colorectal Polyp Segmentation Based on Group Convolution and Transformer

Improving Polyp Segmentation with Boundary-Assisted Guidance and Cross-Scale Interaction Fusion Transformer Network

Adaptive Context Selection for Polyp Segmentation

Polyp Segmentation Network with Hybrid Channel-Spatial Attention and Pyramid Global Context Guided Feature Fusion

TMPSformer: An Efficient Hybrid Transformer-MLP Network for Polyp Segmentation

PolyPooling: An accurate polyp segmentation from colonoscopy images

Automated polyp segmentation based on a multi-distance feature dissimilarity-guided fully convolutional network

Shared Hybrid Attention Transformer Network for Colon Polyp Segmentation