Dyno-Net: A Dynamic Feature Extraction Model for Gastrointestinal Polyp Detection

Zijie Song, Jingjing Wan, Xianchun Meng, Qingye Hua, Wenjie Zhu, Bolun Chen, WEI SHAO

TL;DR: Dyno-Net is a dynamic feature extraction model for gastrointestinal polyp detection, integrating adaptive multi-scale fusion, dynamic convolution, and boundary refinement, achieving superior accuracy and robustness.

Abstract: Gastrointestinal polyps are precursors to colorectal cancer, underscoring the need for accurate early detection. We propose Dyno-Net, a dynamic feature extraction framework integrating multi-scale fusion (DynoFPN), adaptive convolution (DynoConv), and boundary refinement (RefineDet\_LSCSBD), achieving 23.5\% higher fusion efficiency, 17.8\% better detection of small/atypical polyps, and mean IoU improvement from 0.68 to 0.81. Experiments confirm superior accuracy and robustness over mainstream detectors, demonstrating Dyno-Net’s clinical utility.

Code Dataset Promise: Yes

Code Dataset Url: https://github.com/ZiYan416/Dyno-Net

Format Confirmation: I agree that I have read and followed the formatting instructions for the camera ready version.

Submission Number: 104