提出全域性注意機制特徵庫,輔助陰道鏡醫師對影像進行多尺度視角進行病變區域分割
透過融合多時隙影像排查低度鱗狀上皮內病變或更嚴重的病變(LSIL + 型,由 CIN1/2/3 和癌症組合)區域,準確識別需要陰道鏡活檢的患者
基於細粒度病變描述建立陰道鏡資料集,提升 CIN 分級準確率
附:相關發表論文題目與作者
題目:Feature Library: A Benchmark for Cervical Lesion Segmentation
作者:Yuexiang Li, Jiawei Chen, Kai Ma, and Yefeng Zheng
題目:Computer-Aided Cervical Cancer Diagnosis Using Time-Lapsed Colposcopic Images
作者:Yuexiang Li , Jiawei Chen, Peng Xue, Chao Tang, Jia Chang, Chunyan Chu, Kai Ma , Qing Li, Yefeng Zheng, and Youlin Qiao
題目:GRAND: A Large-scale Dataset and Benchmark for Cervical Intraepithelial Neoplasia Grading with Fine-grained Lesion Description
作者:Yuexiang Li, Zhi-Hua Liu, Peng Xuec , Jiawei Chena , Kai Maa , Tianyi Qiand , Yefeng Zhenga, You-Lin Qiao
題目:Development and validation of an artificial intelligence system for grading colposcopic impressions and guiding biopsies
作者:Peng Xue, Chao Tang , Qing Li , Yuexiang Li, Yu Shen, Yuqian Zhao, Jiawei Chen, Jianrong Wu, Longyu Li, Wei Wang, Yucong Li, Xiaoli Cui, Shaokai Zhang, Wenhua Zhang, Xun Zhang, Kai Ma, Yefeng Zheng, Tianyi Qian, Man Tat Alexander Ng, Zhihua Liu, Youlin Qiao, Yu Jiang and Fanghui Zhao