PolypVision：用于结肠息肉分类与分割的三阶段层级深度学习框架
PolypVision: A Three-Stage Hierarchical Deep Learning Framework for Classification and Segmentation of Colorectal Polyps
浏览论文内容
中文总结 AI 辅助
本研究提出PolypVision三阶段层级深度学习框架，在三个公开数据集上实现结肠息肉分类与分割，性能优于或匹配现有方法，且与设备无关，以网页应用免费开放使用。
中文摘要 AI 辅助
结直肠癌（CRC）仍是全球癌症相关死亡的主要原因之一，主要源于癌前息肉。准确检测、分割及内镜下与组织学分类结肠息肉对及时临床干预至关重要。本研究提出PolypVision，一个三阶段层级深度学习框架，依次执行：第一阶段，采用带Focal Loss的EfficientNetV2-M对息肉进行腺瘤性或增生性的二分类，同时完成Paris和JNet分类；第二阶段，采用以第一阶段骨干网络为编码器的UNet++解码器进行息肉分割并推荐切除方法，以Dice和BCE损失优化；第三阶段，采用带第二阶段迁移学习的EfficientNetV2-M进行腺瘤亚型（管状、管状绒毛状、绒毛状）分类。在PolypGen、Kvasir-SEG、CVC-ClinicDB三个公开数据集上评估，PolypVision在帧分类上达到约0.99的AUC，在Kvasir-SEG上达到94.4%的检测mAP@50，性能优于或匹配现有最先进方法。梯度加权类激活映射（Grad-CAM）证实模型关注临床相关的病灶特征。该框架与设备无关，可在不同内镜成像系统上运行，无需硬件特定适配。这些结果表明，采用任务特定损失函数、基于迁移学习的层级流水线为自动化结肠息肉分析提供了一种鲁棒、设备无关且具有临床意义的方法。PolypVision作为DataBioX计划的一部分，以网页应用形式免费提供，向所有用户开放免费使用层级。
英文摘要
Colorectal cancer (CRC) remains one of the leading causes of cancer-related mortality worldwide, predominantly arising from precancerous polyps. Accurate detection, segmentation, and endoscopic and histological classification of colorectal polyps are crucial for timely clinical intervention. In this study, we present PolypVision, a three-stage hierarchical deep learning framework that sequentially performs: (Stage 1) binary classification of polyps as adenomatous or hyperplastic, with simultaneous Paris and JNet classification, using EfficientNetV2-M with Focal Loss; (Stage 2) polyp segmentation with recommended resection method using a UNet++ decoder with the Stage 1 backbone as encoder, optimized with Dice and BCE losses; and (Stage 3) adenoma subtype classification (tubular, tubulovillous, villous) using EfficientNetV2-M with transfer learning from Stage 2. Evaluated on three public datasets -- PolypGen, Kvasir-SEG, and CVC-ClinicDB -- PolypVision achieves an AUC of approximately 0.99 for frame classification and a detection mAP@50 of 94.4% on Kvasir-SEG, outperforming or matching state-of-the-art methods. Gradient-weighted Class Activation Maps (Grad-CAM) confirm that the model attends to clinically relevant lesion features. The framework is device-independent, operating across diverse endoscopic imaging systems without hardware-specific adaptation. These results demonstrate that a hierarchical, transfer-learning-driven pipeline with task-specific loss functions offers a robust, device-independent, and clinically meaningful approach to automated colorectal polyp analysis. PolypVision is freely available as a web application at https://polypvision.com, a DataBioX initiative, with a free usage tier open to all users.
