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
Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, PMLR 300:208-216, 2026.

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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v300-song26b, title = { Dyno-Net: A Dynamic Feature Extraction Model for Gastrointestinal Polyp Detection }, author = {Song, Zijie and Wan, Jingjing and Meng, Xianchun and Hua, Qingye and Zhu, Wenjie and Chen, Bolun and SHAO, WEI}, booktitle = {Proceedings of The 29th International Conference on Artificial Intelligence and Statistics}, pages = {208--216}, year = {2026}, editor = {Khan, Emtiyaz and Li, Yingzhen and Solin, Arno and Ramdas, Aaditya}, volume = {300}, series = {Proceedings of Machine Learning Research}, month = {02--05 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v300/main/assets/song26b/song26b.pdf}, url = {https://proceedings.mlr.press/v300/song26b.html}, 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. } }
Endnote
%0 Conference Paper %T Dyno-Net: A Dynamic Feature Extraction Model for Gastrointestinal Polyp Detection %A Zijie Song %A Jingjing Wan %A Xianchun Meng %A Qingye Hua %A Wenjie Zhu %A Bolun Chen %A WEI SHAO %B Proceedings of The 29th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2026 %E Emtiyaz Khan %E Yingzhen Li %E Arno Solin %E Aaditya Ramdas %F pmlr-v300-song26b %I PMLR %P 208--216 %U https://proceedings.mlr.press/v300/song26b.html %V 300 %X 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.
APA
Song, Z., Wan, J., Meng, X., Hua, Q., Zhu, W., Chen, B. & SHAO, W.. (2026). Dyno-Net: A Dynamic Feature Extraction Model for Gastrointestinal Polyp Detection . Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 300:208-216 Available from https://proceedings.mlr.press/v300/song26b.html.

Related Material