发表机构
KAIST; Clinical Imaging Research Institute; Hokkaido University; CHA University（韩国科学技术院; 临床影像研究所; 北海道大学; 嘉泉大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究提出并验证了一个四层病灶中心的分层流程，在无需全结肠三维重建的情况下，利用公开视频实现了回访检测、病灶身份合并和局部三维重建，并优于通用基础模型。
AI 中文摘要
背景与目的：结肠镜检查记录实践保留了文本报告和静态照片，而记录视频中已有的空间信息——内镜行进的位置、观察到病灶的位置以及同一病灶是否再次被看到——在检查结束时被丢弃。本研究确定是否可以在无需全结肠三维重建的情况下，组装并验证一个以病灶为中心的空间记录。方法：构建了一个四层分层流程——（1）全局拓扑图，（2）病灶级时空轨迹，（3）按需局部三维重建，以及（4）跨重复观察的持久病灶身份——并在四个公开视频（两个具有真实深度的C3VDv2序列和两个完整的REAL-Colon检查；共40,245帧）上端到端运行。所有组件均为已发表的、单独验证的方法；其贡献在于病灶中心的组装、链接规则和评估。结果：在仅向前映射的构建下不可能实现的回访，通过入口图贝叶斯定位被检测到：在两个完整检查中分别检测到5,614和4,043次回访事件（56和68个不同节点）。在采用的阈值0.5下，病灶身份合并保持了真实标签纯度1.0，同时自动合并了231个候选对中的20个。内镜专用几何引擎在所有指标上优于通用基础模型（总体绝对相对误差（AbsRel）0.2276对0.3523）。结论：结果是部分的，但建立了一条具体的近期路径：回访检测、病灶身份和局部三维均在无需等待完整几何重建的情况下返回了定量、可复现的输出；在临床数据上验证该记录是下一步。
英文摘要
Background and Objective: Colonoscopy recording practice preserves text reports and still photographs, while the spatial information already present in the recorded video - where the scope traveled, where a lesion was observed, and whether the same lesion was seen again - is discarded when the procedure ends. This study determines whether a lesion-centered spatial record can be assembled and validated without full-colon 3D reconstruction. Methods: A four-layer hierarchical pipeline was assembled - (1) a global topological map, (2) lesion-level spatio-temporal tracks, (3) on-demand local 3D reconstruction, and (4) persistent lesion identity across repeated observations - and ran end to end on four public videos (two C3VDv2 sequences with ground-truth depth and two full REAL-Colon procedures; 40,245 frames). All components are published, individually validated methods; the contribution is their lesion-centered assembly, linking rules, and evaluation. Results: Revisits, impossible under forward-only mapping by construction, were detected by entry-map Bayesian localization: 5,614 and 4,043 revisit events (56 and 68 distinct nodes) in the two full procedures. Lesion-identity merging at the adopted threshold 0.5 maintained ground-truth purity 1.0 while auto-merging 20 of 231 candidate pairs. The endoscopy-specific geometry engine outperformed a general-purpose foundation model on all metrics (overall absolute relative error (AbsRel) 0.2276 vs. 0.3523). Conclusions: The results are partial but establish a concrete near-term path: revisit detection, lesion identity, and local 3D each returned quantitative, reproducible output without waiting for complete geometric reconstruction; validating the record on clinical data is the next step.
Comments21 pages, 12 figures, 4 tables. Code: github.com/hyunjun1121/endovision-pipeline (Zenodo DOI 10.5281/zenodo.22136766)
