{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,7]],"date-time":"2026-09-07T19:18:51Z","timestamp":1788808731781,"version":"build-2803163510"},"reference-count":19,"publisher":"Springer Science and Business Media LLC","issue":"6","license":[{"start":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T00:00:00Z","timestamp":1778457600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T00:00:00Z","timestamp":1778457600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/100006502","name":"Defense Sciences Office, DARPA","doi-asserted-by":"publisher","award":["HR00112520300"],"award-info":[{"award-number":["HR00112520300"]}],"id":[{"id":"10.13039\/100006502","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100005405","name":"American Surgical Association Foundation","doi-asserted-by":"publisher","award":["10096783"],"award-info":[{"award-number":["10096783"]}],"id":[{"id":"10.13039\/100005405","id-type":"DOI","asserted-by":"publisher"}]},{"name":"ASSET Center Seed Grant"},{"name":"ASSET Center Seed Grant"},{"DOI":"10.13039\/100023015","name":"Advanced Research Projects Agency for Health","doi-asserted-by":"publisher","award":["D24AC00253-00"],"award-info":[{"award-number":["D24AC00253-00"]}],"id":[{"id":"10.13039\/100023015","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["2442421"],"award-info":[{"award-number":["2442421"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Schmidt Sciences","award":["G-25-67983"],"award-info":[{"award-number":["G-25-67983"]}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Int J CARS"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>\n                      <jats:bold>Purpose<\/jats:bold>\n                    <\/jats:title>\n                    <jats:p>Accurate assessment of the Critical View of Safety (CVS) during laparoscopic cholecystectomy is essential to prevent bile duct injury, a complication associated with significant morbidity and mortality. While large vision-language models (LVLMs) offer flexible reasoning, their predictions remain difficult to audit and unreliable on safety-critical surgical tasks.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>\n                      <jats:bold>Methods<\/jats:bold>\n                    <\/jats:title>\n                    <jats:p>We introduce Sum-of-Checks, a framework that decomposes each CVS criterion into expert-defined reasoning checks reflecting clinically relevant visual evidence. Given a laparoscopic frame, an LVLM evaluates each check, producing a binary judgment and justification. Criterion-level scores are computed via fixed, weighted aggregation of check outcomes. We evaluate on the Endoscapes2023 benchmark using three frontier LVLMs, comparing against direct prompting, chain-of-thought, and sub-question decomposition, each with and without few-shot examples.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>\n                      <jats:bold>Results<\/jats:bold>\n                    <\/jats:title>\n                    <jats:p>Sum-of-Checks improves average frame-level mean average precision by 12\u201314% relative to the best baseline across all three models and criteria. Analysis of individual checks reveals that LVLMs are reliable on observational checks (e.g., visibility, tool obstruction) but show substantial variability on decision-critical anatomical evidence.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>\n                      <jats:bold>Conclusion<\/jats:bold>\n                    <\/jats:title>\n                    <jats:p>Structuring surgical reasoning into expert-aligned verification checks improves both accuracy and transparency of LVLM-based CVS assessment, demonstrating that explicitly separating evidence elicitation from decision-making is critical for reliable and auditable surgical AI systems.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1007\/s11548-026-03643-x","type":"journal-article","created":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T08:29:08Z","timestamp":1778488148000},"page":"1433-1438","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Sum-of-Checks: structured reasoning for surgical safety with large vision-language models"],"prefix":"10.1007","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-9326-1077","authenticated-orcid":false,"given":"Weiqiu","family":"You","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Cassandra","family":"Goldberg","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Amin","family":"Madani","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daniel A.","family":"Hashimoto","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Eric","family":"Wong","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,5,11]]},"reference":[{"key":"3643_CR1","doi-asserted-by":"publisher","first-page":"3332","DOI":"10.31661\/gmj.v13i.3332","volume":"13","author":"A Zarghami","year":"2024","unstructured":"Zarghami A (2024) Role of artificial intelligence in surgical decision-making: a comprehensive review: role of ai in sdm. Galen Med J 13:3332. https:\/\/doi.org\/10.31661\/gmj.v13i.3332","journal-title":"Galen Med J"},{"issue":"1","key":"3643_CR2","doi-asserted-by":"publisher","first-page":"11","DOI":"10.1186\/s13037-024-00393-0","volume":"18","author":"F Arjmandnia","year":"2024","unstructured":"Arjmandnia F, Alimohammadi E (2024) The value of machine learning technology and artificial intelligence to enhance patient safety in spine surgery: a review. Patient Saf Surg 18(1):11","journal-title":"Patient Saf Surg"},{"issue":"1","key":"3643_CR3","doi-asserted-by":"publisher","first-page":"331","DOI":"10.1038\/s41597-025-04642-4","volume":"12","author":"P Mascagni","year":"2025","unstructured":"Mascagni P, Alapatt D, Murali A, Vardazaryan A, Garcia A, Okamoto N, Costamagna G, Mutter D, Marescaux J, Dallemagne B et al (2025) Endoscapes, a critical view of safety and surgical scene segmentation dataset for laparoscopic cholecystectomy. Sci Data 12(1):331","journal-title":"Sci Data"},{"issue":"2","key":"3643_CR4","doi-asserted-by":"publisher","first-page":"363","DOI":"10.1097\/SLA.0000000000004594","volume":"276","author":"A Madani","year":"2022","unstructured":"Madani A, Namazi B, Altieri MS, Hashimoto DA, Rivera AM, Pucher PH, Navarrete-Welton A, Sankaranarayanan G, Brunt LM, Okrainec A, Alseidi A (2022) Artificial intelligence for intraoperative guidance: using semantic segmentation to identify surgical anatomy during laparoscopic cholecystectomy. Ann Surg 276(2):363\u2013369. https:\/\/doi.org\/10.1097\/SLA.0000000000004594","journal-title":"Ann Surg"},{"key":"3643_CR5","first-page":"505","volume-title":"Medical Image Computing and Computer Assisted Intervention\u2014MICCAI 2023","author":"S Sharma","year":"2023","unstructured":"Sharma S, Nwoye CI, Mutter D, Padoy N (2023) Surgical action triplet detection by mixed supervised learning of instrument-tissue interactions. In: Greenspan H, Madabhushi A, Mousavi P, Salcudean S, Duncan J, Syeda-Mahmood T, Taylor R (eds) Medical Image Computing and Computer Assisted Intervention\u2014MICCAI 2023. Springer, Cham, pp 505\u2013514"},{"key":"3643_CR6","doi-asserted-by":"crossref","unstructured":"Seenivasan L, Islam M, Kannan G, Ren H (2023) SurgicalGPT: end-to-end language-vision GPT for visual question answering in surgery. In: International conference on medical image computing and computer-assisted intervention. Springer, pp 281\u2013290","DOI":"10.1007\/978-3-031-43996-4_27"},{"key":"3643_CR7","doi-asserted-by":"crossref","unstructured":"Wang G, Bai L, Wang J, Yuan K, Li Z, Jiang T, He X, Wu J, Chen Z, Lei Z et al (2025) Endochat: Grounded multimodal large language model for endoscopic surgery. Med Image Anal 103789","DOI":"10.1016\/j.media.2025.103789"},{"key":"3643_CR8","doi-asserted-by":"crossref","unstructured":"Stueker EH, Kolbinger FR, Saldanha OL, Digomann D, Pistorius S, Oehme F, Van Treeck M, Ferber D, L\u00f6ffler CML, Weitz J, et al. (2025) Vision-language models for automated video analysis and documentation in laparoscopic surgery: a proof-of-concept study. Int J Surg 10\u20131097","DOI":"10.1097\/JS9.0000000000003069"},{"issue":"9","key":"3643_CR9","doi-asserted-by":"publisher","first-page":"2613","DOI":"10.1038\/s41591-024-03097-1","volume":"30","author":"P Hager","year":"2024","unstructured":"Hager P, Jungmann F, Holland R, Bhagat K, Hubrecht I, Knauer M, Vielhauer J, Makowski M, Braren R, Kaissis G et al (2024) Evaluation and mitigation of the limitations of large language models in clinical decision-making. Nat Med 30(9):2613\u20132622","journal-title":"Nat Med"},{"issue":"1","key":"3643_CR10","first-page":"101","volume":"180","author":"SM Strasberg","year":"1995","unstructured":"Strasberg SM, Hertl M, Soper NJ (1995) An analysis of the problem of biliary injury during laparoscopic cholecystectomy. J Am Coll Surg 180(1):101\u2013125","journal-title":"J Am Coll Surg"},{"key":"3643_CR11","unstructured":"Alapatt D, Eckhoff J, Lyu Z, Ban Y, Mazellier J-P, Choksi S, Yang K, Chiang P-H, Zorzetti N, Cannas S, et al. (2025) The sages critical view of safety challenge: A global benchmark for ai-assisted surgical quality assessment. arXiv:2509.17100"},{"issue":"5","key":"3643_CR12","doi-asserted-by":"publisher","first-page":"955","DOI":"10.1097\/SLA.0000000000004351","volume":"275","author":"P Mascagni","year":"2022","unstructured":"Mascagni P, Vardazaryan A, Alapatt D, Urade T, Emre T, Fiorillo C, Pessaux P, Mutter D, Marescaux J, Costamagna G et al (2022) Artificial intelligence for surgical safety: automatic assessment of the critical view of safety in laparoscopic cholecystectomy using deep learning. Ann Surg 275(5):955\u2013961","journal-title":"Ann Surg"},{"key":"3643_CR13","doi-asserted-by":"crossref","unstructured":"Jacovi A, Goldberg Y (2020) Towards faithfully interpretable NLP systems: How should we define and evaluate faithfulness? In: Proceedings of the 58th annual meeting of the association for computational linguistics, pp 4198\u20134205","DOI":"10.18653\/v1\/2020.acl-main.386"},{"key":"3643_CR14","doi-asserted-by":"crossref","unstructured":"Lyu Q, Apidianaki M, Callison-Burch C (2024) Towards faithful model explanation in NLP: a survey. Comput Linguist 50(2):657\u2013723","DOI":"10.1162\/coli_a_00511"},{"key":"3643_CR15","unstructured":"OpenAI: Introducing GPT-4.1 in the API. https:\/\/openai.com\/index\/gpt-4-1\/. Accessed 08 September 2025"},{"key":"3643_CR16","unstructured":"Anthropic: Introducing Claude Haiku 4.5 (2025). https:\/\/www.anthropic.com\/news\/claude-haiku-4-5. Accessed 23 February 2026"},{"key":"3643_CR17","unstructured":"Anthropic: Introducing Claude Opus 4.5 (2025). https:\/\/www.anthropic.com\/news\/claude-opus-4-5. Accessed 23 February 2026"},{"key":"3643_CR18","doi-asserted-by":"publisher","first-page":"24824","DOI":"10.52202\/068431-1800","volume":"35","author":"J Wei","year":"2022","unstructured":"Wei J, Wang X, Schuurmans D, Bosma M, Xia F, Chi E, Le QV, Zhou D et al (2022) Chain-of-thought prompting elicits reasoning in large language models. Adv Neural Inf Process Syst 35:24824\u201324837","journal-title":"Adv Neural Inf Process Syst"},{"issue":"1","key":"3643_CR19","doi-asserted-by":"publisher","first-page":"194","DOI":"10.1038\/s41597-023-02073-7","volume":"10","author":"MS R\u00edos","year":"2023","unstructured":"R\u00edos MS, Molina-Rodriguez MA, Londo\u00f1o D, Guill\u00e9n CA, Sierra S, Zapata F, Giraldo LF (2023) Cholec80-CVS: An open dataset with an evaluation of Strasberg\u2019s critical view of safety for AI. Sci Data 10(1):194","journal-title":"Sci Data"}],"container-title":["International Journal of Computer Assisted Radiology and Surgery"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11548-026-03643-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s11548-026-03643-x","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11548-026-03643-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,7]],"date-time":"2026-08-07T13:20:14Z","timestamp":1786108814000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s11548-026-03643-x"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,11]]},"references-count":19,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2026,6]]}},"alternative-id":["3643"],"URL":"https:\/\/doi.org\/10.1007\/s11548-026-03643-x","relation":{},"ISSN":["1861-6429"],"issn-type":[{"value":"1861-6429","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,11]]},"assertion":[{"value":"23 February 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"3 April 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"11 May 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"Daniel Hashimoto is a consultant for Medtronic.","order":1,"name":"Ethics","label":"Conflict of interest","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"This study used publicly available, fully de-identified surgical video data with no new human data collection or patient interaction; ethical approval was not required.","order":2,"name":"Ethics","label":"Ethical approval","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable. The study used publicly available, de-identified data.","order":3,"name":"Ethics","label":"Informed consent","group":{"name":"EthicsHeading","label":"Declarations"}}]}}