Endoscopy-assisted lightweight diagnosis system based on transformers for colon polyp detection  

作　　者：FAN Weiming YU Jiahui JU Zhaojie 

机构地区：[1]School of Automation and Electrical Engineering,Shenyang Ligong University,Shenyang 110159,China [2]Department of Biomedical Engineering,Zhejiang University,Hangzhou 310058,China [3]School of Computing,University of Portsmouth,PO13HE,UK

出　　处：《Optoelectronics Letters》2025年第1期57-64,共8页光电子快报(英文版)

基　　金：supported by the Zhejiang Provincial Natural Science Foundation of China(No.LQ23F030001);Hangzhou Innovation Team(No.TD2022011);the Ai Ble Project co-financed by the European Regional Development Fund;the National Natural Science Foundation of China(No.52075530)。

摘　　要：The integration of endoscopy has significantly propelled the diagnosis and treatment of gastrointestinal diseases,with colonoscopy establishing itself as the primary method for early diagnosis and preventive care in colorectal cancer(CRC).Although deep learning holds promise in mitigating missed polyp rates,modern endoscopy examinations pose additional challenges,such as image blurring and atomizing.This study explores lightweight yet powerful attention mechanisms,introducing the spatial-channel transformer(SCT),an innovative approach that leverages spatial channel relationships for attention weight calculation.The method utilizes rotation operations for inter-dimensional dependencies,followed by residual transformation,encoding inter-channel and spatial information with minimal computational overhead.Extensive experiments on the CVC-Clinic DB polyp detection dataset,addressing endoscopy pitfalls,underscore the superiority of our SCT over other state-of-the-art methods.The proposed model maintains high performance,even in challenging scenarios.

关 键 词：DIAGNOSIS utilize holds 

分 类 号：R73[医药卫生—肿瘤]