---
格式版本: 2
标题: "Transferable Tool-Tissue Contact Detection from Stereo Depth in Robot-Assisted Surgery"
原文链接: "https://arxiv.org/abs/2608.18270"
发布日期: "2026-08-18"
发布时间校准状态: "found"
发布时间需复核: "否"
发布时间来源: "rule:local:strict_original_body"
发布时间证据: "**\\[v1\\]** Tue, 18 Aug 2026 19:41:16 UTC (1,428 KB)"
发布时间校准原因: "规则确认唯一严格发布时间，来源 local:strict_original_body"
发布时间校准置信度: "high"
发布时间候选数量: 18
发布时间严格候选数量: 6
发布时间原页读取状态: "source template page reused from URL open"
发布时间未找到原因: ""
发布时间校准时间: "2026-08-20T15:52:49+08:00"
发布时间仲裁状态: "skipped"
发布时间仲裁尝试次数: 0
发布时间仲裁耗时毫秒: 0
发现时间: "2026-08-20T15:48:16+08:00"
入库时间: "2026-08-20T07:52:49.171Z"
来源平台: "arXiv 学术论文搜索"
搜索渠道: "source_template"
搜索词: "https://arxiv.org/search/?query=model%20fit&searchtype=all"
匹配关键词:
  - "model fit"
  - "performance"
相关厂家:
  []
相关专家:
  []
内容类型: "网页"
抓取工具: "Free Fetch + Defuddle"
清洗工具: "Defuddle Markdown + Defuddle/Readability 正文提取"
原始附件:
  []
AI优质: "否"
AI打分: 0
AI分档: "非优质"
AI质检状态: "不通过"
AI打分理由: "内容为arXiv机器人手术论文，与超节点/AI Rack/机柜级AI基础设施完全无关，不涉及任何相关关键词或技术主题。"
AI质检模型: "ali-deepseek-v4-flash"
AI质检时间: "2026-08-20T15:59:07+08:00"
AI主题相关性: 0
AI来源权威性: 0
AI新颖性: 0
AI技术细节: 0
AI商业部署信号: 0
AI完整性: 0
AI摘要: "该研究提出利用立体深度图像的器械-组织最小距离信号，结合监督隐马尔可夫模型检测手术器械与组织的接触。"
AI摘要模型: "ali-deepseek-v4-flash"
AI摘要时间: "2026-09-07T03:21:03.686Z"
采集批次: "2026年8月20日14点19分32秒"
采集批次ID: "20260820-141932-079"
去重键: "https://arxiv.org/abs/2608.18270"
---

## Computer Science > Robotics

## Title:Transferable Tool-Tissue Contact Detection from Stereo Depth in Robot-Assisted Surgery

Authors:[Mingyeung Wu](https://arxiv.org/search/cs?searchtype=author&query=Wu,+M), [Zhonghao Zhang](https://arxiv.org/search/cs?searchtype=author&query=Zhang,+Z), [Hao Yang](https://arxiv.org/search/cs?searchtype=author&query=Yang,+H), [Alan Kuntz](https://arxiv.org/search/cs?searchtype=author&query=Kuntz,+A), [Jie Ying Wu](https://arxiv.org/search/cs?searchtype=author&query=Wu,+J+Y)

[View PDF](https://arxiv.org/pdf/2608.18270) [HTML (experimental)](https://arxiv.org/html/2608.18270v1)

> Abstract:Reliable tool--tissue contact detection can support interaction-aware control and downstream force estimation in robot-assisted surgery. Most existing methods learn a contact classifier from RGB appearance, which is hard to generalize. In this work, we use the depth image generated from a stereo pair to give more information about tool--tissue contact. For each depth frame, we localize a spatially supported minimum-distance patch around the tool boundary and reduce it to a single scalar, $-\log_{10}|d|$; this signal rises and falls in step with ground-truth contact. We formalize this observation with a fully supervised two-state hidden Markov model. We fit this model as a six-fold leave-one-session-out (LOSO) ensemble on six palpation sessions against a single silicone cup-like phantom, with the decision threshold selected from the pooled out-of-fold predictions. It is evaluated on four held-out sessions of three categories: 1. same task on same phantom; 2. same task on different phantom; 3. different task on different phantom. This model reaches held-out macro F1 $0.927$ and AUPRC $0.980$. We further compare against a reproduction of an RGB-based contact classifier from prior work. This RGB-based model achieves high performance on the first category (F1 $0.965$), but substantially lower performance on the other two, resulting in macro F1 $0.320$ across all four sessions. These results indicate that the tool--tissue distance is a strong, transferable cue for contact detection in robot-assisted surgery.

| Subjects: | Robotics (cs.RO) |
| --- | --- |
| Cite as: | [arXiv:2608.18270](https://arxiv.org/abs/2608.18270) \[cs.RO\] |
|  | (or [arXiv:2608.18270v1](https://arxiv.org/abs/2608.18270v1) \[cs.RO\] for this version) |
|  | [https://doi.org/10.48550/arXiv.2608.18270](https://doi.org/10.48550/arXiv.2608.18270) |

## Submission history

From: Mingyeung Wu \[[view email](https://arxiv.org/show-email/a78e5f13/2608.18270)\]  
**\[v1\]** Tue, 18 Aug 2026 19:41:16 UTC (1,428 KB)

[Which authors of this paper are endorsers?](https://arxiv.org/auth/show-endorsers/2608.18270) | Disable MathJax ([What is MathJax?](https://info.arxiv.org/help/mathjax.html))