{"id":3250,"date":"2025-08-04T12:33:19","date_gmt":"2025-08-04T12:33:19","guid":{"rendered":"http:\/\/www.labren.org\/mm\/?p=3250"},"modified":"2025-08-05T12:57:49","modified_gmt":"2025-08-05T12:57:49","slug":"%f0%9f%93%8a-empowering-robotic-surgery-with-sam-2-an-empirical-study-on-surgical-instrument-segmentation-%f0%9f%a4%96%f0%9f%91%a8%e2%9a%95%ef%b8%8f","status":"publish","type":"post","link":"http:\/\/www.labren.org\/mm\/news\/%f0%9f%93%8a-empowering-robotic-surgery-with-sam-2-an-empirical-study-on-surgical-instrument-segmentation-%f0%9f%a4%96%f0%9f%91%a8%e2%9a%95%ef%b8%8f\/","title":{"rendered":"\ud83d\udcca **Empowering Robotic Surgery with SAM 2: An Empirical Study on Surgical Instrument Segmentation** \ud83e\udd16\ud83d\udc68\u200d\u2695\ufe0f"},"content":{"rendered":"\n

We’re excited to share our latest research on the Segment Anything Model (SAM) 2. This empirical evaluation uncovers SAM 2’s robustness and generalization capabilities in surgical image\/video segmentation, a critical component for enhancing precision and safety in the operating room.<\/p>\n\n\n\n

\ud83d\udd2c **Key Findings**:<\/p>\n\n\n\n

– In general, SAM 2 outperforms its predecessor in instrument segmentation, showing a much improved zero-shot generalization capability to the surgical domain.<\/p>\n\n\n\n

– Utilizing bounding box prompts, SAM 2 achieves remarkable results, setting a new benchmark in the surgical image segmentation.<\/p>\n\n\n\n

– With a single point as the prompt on the first frame, SAM 2 demonstrates substantial improvements on video segmentation over SAM, which requires point prompts on every frames. This suggests great potential in addressing video-based surgical tasks. <\/p>\n\n\n\n

– Resilience Under Common Corruptions: SAM 2 shows impressive robustness against real-world image corruption, maintaining performance under various challenges such as compression, noise, and blur.<\/p>\n\n\n\n

\ud83d\udd27 **Practical Implications**:<\/p>\n\n\n\n

– With faster inference speeds, SAM 2 is poised to provide quick, accurate segmentation, making it a valuable asset in the clinical setting.<\/p>\n\n\n\n

\ud83d\udd17 **Learn More**:<\/p>\n\n\n\n

For those interested in the technical depth, our paper is available on [arXiv](https:\/\/lnkd.in\/gHfdrvj3).<\/p>\n\n\n\n

We’re eager to engage with the community and explore how SAM 2 can revolutionize surgical applications.<\/p>\n\n\n\n

Thanks to the team contributions of Jieming YU<\/a>, An Wang<\/a>, Wenzhen Dong, Mengya Xu<\/a>, Jie Wang, Long Bai<\/a>, Hongliang Ren<\/a> from Department of Electronic Engineering, The Chinese University of Hong Kong<\/a> and Shenzhen Research Institute of CUHK, and Mobarakol Islam<\/a> from WEISS – Wellcome \/ EPSRC Centre for Interventional and Surgical Sciences<\/a>, UCL.<\/p>\n\n\n\n

\"No<\/figure>\n\n\n\n
\"No<\/figure>\n\n\n\n
\"No<\/figure>\n","protected":false},"excerpt":{"rendered":"

We’re excited to share our latest research on the Segment Anything Model (SAM) 2. This empirical evaluation uncovers SAM 2’s robustness and generalization capabilities in surgical image\/video segmentation, a critical component for enhancing precision and safety in the operating room. \ud83d\udd2c **Key Findings**: – In general, SAM 2 outperforms its\u2026 Continue reading<\/a><\/p>\n","protected":false},"author":17,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"ngg_post_thumbnail":0,"footnotes":""},"categories":[4],"tags":[126,128,121,123,124,122,118,127,125],"class_list":["post-3250","post","type-post","status-publish","format-standard","hentry","category-news","tag-cuhk","tag-labren","tag-meta","tag-objectsegmentation","tag-robustness","tag-sam2","tag-surgicalai","tag-ucl","tag-weiss"],"_links":{"self":[{"href":"http:\/\/www.labren.org\/mm\/wp-json\/wp\/v2\/posts\/3250","targetHints":{"allow":["GET"]}}],"collection":[{"href":"http:\/\/www.labren.org\/mm\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/www.labren.org\/mm\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/www.labren.org\/mm\/wp-json\/wp\/v2\/users\/17"}],"replies":[{"embeddable":true,"href":"http:\/\/www.labren.org\/mm\/wp-json\/wp\/v2\/comments?post=3250"}],"version-history":[{"count":1,"href":"http:\/\/www.labren.org\/mm\/wp-json\/wp\/v2\/posts\/3250\/revisions"}],"predecessor-version":[{"id":3251,"href":"http:\/\/www.labren.org\/mm\/wp-json\/wp\/v2\/posts\/3250\/revisions\/3251"}],"wp:attachment":[{"href":"http:\/\/www.labren.org\/mm\/wp-json\/wp\/v2\/media?parent=3250"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/www.labren.org\/mm\/wp-json\/wp\/v2\/categories?post=3250"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/www.labren.org\/mm\/wp-json\/wp\/v2\/tags?post=3250"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}