{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T17:49:14Z","timestamp":1792172954099,"version":"4.4.0"},"reference-count":29,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["J Med Internet Res"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Postpolypectomy surveillance intervals derived from colonoscopy findings and histopathology should be guideline-based, yet these are frequently not adhered to, contributing to under- and oversurveillance. This motivates a more structured and automated approach to improve adherence.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n We developed and evaluated an automated, implementable, guideline-aligned decision support tool that extracts colorectal polyp features from free-text colonoscopy and histopathology reports and produces auditable surveillance recommendations.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n We extracted paired colonoscopy and histopathology reports from a National Health Service (NHS) trust and designed a 4-stage prompting pipeline using open-weight large language models to separate multispecimen histopathology reports into individual containers (\u201cpots\u201d), align endoscopic findings with specimens, and normalize each pot into a structured schema. The surveillance intervals were then computed by a deterministic rule engine encoding guideline rules. We evaluated extraction performance on 100 procedures (215 pots), assessed surveillance recommendation performance on a same-site cohort (n=100), and performed external validation on a cohort from a second trust (n=100). A separate retrospective audit compared automated recommendations and observed routine scheduling against a clinician-derived reference based on full electronic health record review in 127 patients.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n Key extraction fields showed end-to-end accuracy (specimen type=0.995; location=0.981; diagnosis=0.981; excision completeness=0.949), with mean report-level pot comprehensiveness of 0.968 (SD 0.112). Overall surveillance recommendation accuracy was 0.96 (95% CI 0.90-0.98) at Guy\u2019s and St Thomas\u2019 NHS Foundation Trust (GSTT) and 0.93 (95% CI 0.86-0.97) at Sandwell and West Birmingham NHS Trust (SWB) and showed strong class-level performance (\n \n no follow-up\n F<\/jats:italic>\n 1<\/jats:sub>\n -score of 0.98 at GSTT and 0.97 at SWB; site check\n F<\/jats:italic>\n 1<\/jats:sub>\n -score of 0.93 at GSTT and 0.91 at SWB; 3-year follow-up\n F<\/jats:italic>\n 1<\/jats:sub>\n -score of 0.86 at GSTT and 0.80 at SWB; cancer alert\n F<\/jats:italic>\n 1<\/jats:sub>\n -score of 1.00 at GSTT and 1.00 at SWB). The comparison to current clinical practice showed that the tool would improve guideline adherence primarily by reducing potential undersurveillance.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n This hybrid approach using open-weight large language models for structured extraction and guideline-encoded rules for decision-making can generate accurate, lightweight, and auditable postpolypectomy surveillance recommendations. Such a tool has the potential to improve safety netting and guideline adherence in NHS colonoscopy workflows.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/102789","type":"journal-article","created":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T17:11:46Z","timestamp":1792170706000},"page":"e102789-e102789","source":"Crossref","is-referenced-by-count":0,"title":["Improving Postpolypectomy Surveillance Decisions Using an Automated Analysis of Colonoscopy and Histopathology Reports: 2-Center Validation Study"],"prefix":"10.2196","volume":"28","author":[{"given":"Agathe","family":"Zecevic","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9513-5164","authenticated-orcid":false,"given":"Sabina","family":"De Martino","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-9969-1929","authenticated-orcid":false,"given":"Joshua Shan Chen","family":"Lee","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-8219-8285","authenticated-orcid":false,"given":"Victoria","family":"To","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-9094-9592","authenticated-orcid":false,"given":"Vedant","family":"Bhardwaj","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-3393-4493","authenticated-orcid":false,"given":"Rhea","family":"Sibal","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8040-8158","authenticated-orcid":false,"given":"Nigel","family":"Trudgill","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5311-3090","authenticated-orcid":false,"given":"Shahd A","family":"Mohamed","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4570-9801","authenticated-orcid":false,"given":"Angus","family":"Roberts","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1673-2663","authenticated-orcid":false,"given":"Sebastian","family":"Zeki","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,10,16]]},"reference":[{"issue":"2","key":"R1","doi-asserted-by":"publisher","first-page":"201","DOI":"10.1136\/gutjnl-2019-319858","article-title":"British Society of Gastroenterology\/Association of Coloproctology of Great Britain and Ireland\/Public Health England post-polypectomy and post-colorectal cancer resection surveillance guidelines","volume":"69","author":"Rutter","journal-title":"Gut"},{"issue":"6","key":"R2","doi-asserted-by":"publisher","first-page":"823","DOI":"10.1016\/S1470-2045(17)30187-0","article-title":"Adenoma surveillance and colorectal cancer incidence: a retrospective, multicentre, cohort study","volume":"18","author":"Atkin","journal-title":"Lancet Oncol"},{"issue":"25","key":"R3","doi-asserted-by":"publisher","first-page":"1","DOI":"10.3310\/hta21250","article-title":"The clinical effectiveness of different surveillance strategies to prevent colorectal cancer in people with intermediate-grade colorectal adenomas: a retrospective cohort analysis, and psychological and economic evaluations","volume":"21","author":"Atkin","journal-title":"Health Technol Assess"},{"issue":"8","key":"R4","doi-asserted-by":"publisher","first-page":"838","DOI":"10.1007\/s00535-010-0227-3","article-title":"Recurrence and surveillance of colorectal adenoma after polypectomy in a southern Chinese population","volume":"45","author":"Huang","journal-title":"J Gastroenterol"},{"issue":"12","key":"R5","doi-asserted-by":"publisher","first-page":"2307","DOI":"10.1136\/gutjnl-2020-323411","article-title":"Colorectal cancer risk following polypectomy in a multicentre, retrospective, cohort study: an evaluation of the 2020 UK post-polypectomy surveillance guidelines","volume":"70","author":"Cross","journal-title":"Gut"},{"issue":"6","key":"R6","doi-asserted-by":"publisher","first-page":"798","DOI":"10.1177\/2050640619841539","article-title":"Development of a national automated endoscopy database: the United Kingdom National Endoscopy Database (NED)","volume":"7","author":"Lee","journal-title":"United European Gastroenterol J"},{"issue":"Supplement_5","key":"R7","doi-asserted-by":"publisher","DOI":"10.1093\/bjs\/znac248.017","article-title":"TU2.5 Compliance of colonic polyp surveillance with British Society of Gastroenterology (BSG) and the Association of Coloproctology of Great Britain and Ireland (ACPGBI) guidelines","volume":"109","author":"Ali","journal-title":"BJS"},{"issue":"2","key":"R8","doi-asserted-by":"publisher","DOI":"10.7759\/cureus.35516","article-title":"Adherence to post-polypectomy surveillance guidelines at a large district general hospital","volume":"15","author":"Butler","journal-title":"Cureus"},{"issue":"10S","key":"R9","doi-asserted-by":"publisher","first-page":"S156","DOI":"10.14309\/01.ajg.0000950476.05709.2e","article-title":"S209 Non-adherence to colorectal polyp surveillance guidelines: magnitude and factors associated with non-adherence in an open access colonoscopy screening program at an academic institution","volume":"118","author":"Liu","journal-title":"Am J Gastroenterol"},{"issue":"1","key":"R10","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-023-02121-7","article-title":"Deep learning approach to detection of colonoscopic information from unstructured reports","volume":"23","author":"Seong","journal-title":"BMC Med Inform Decis Mak"},{"issue":"4","key":"R11","doi-asserted-by":"publisher","DOI":"10.2196\/35257","article-title":"Natural language processing for assessing quality indicators in free-text colonoscopy and pathology reports: development and usability study","volume":"10","author":"Bae","journal-title":"JMIR Med Inform"},{"key":"R12","doi-asserted-by":"publisher","DOI":"10.14309\/01.ajg.0000773508.71234.2b","article-title":"S259 Validation of a machine learning algorithm to measure adenoma detection rate from screening colonoscopies in a large health system","volume":"116","author":"Galoosian","journal-title":"Am J Gastroenterol"},{"issue":"5","key":"R13","doi-asserted-by":"publisher","first-page":"978","DOI":"10.1016\/j.gie.2021.05.036","article-title":"Automated identification and assignment of colonoscopy surveillance recommendations for individuals with colorectal polyps","volume":"94","author":"Peterson","journal-title":"Gastrointest Endosc"},{"issue":"9","key":"R14","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2023.34822","article-title":"Development and evaluation of a surveillance system for follow-up after colorectal polypectomy","volume":"6","author":"Wu","journal-title":"JAMA Netw Open"},{"issue":"10S","key":"R15","doi-asserted-by":"publisher","first-page":"S1271","DOI":"10.14309\/01.ajg.0000956448.72258.9a","article-title":"S1702 Comparing large language models accuracy in following interval surveillance colonoscopy guidelines","volume":"118","author":"Brennan","journal-title":"Am J Gastroenterol"},{"issue":"2","key":"R16","doi-asserted-by":"publisher","first-page":"382","DOI":"10.1053\/j.gastro.2024.09.032","article-title":"Use of large language models to identify surveillance colonoscopy intervals-a feasibility study","volume":"168","author":"Acharya","journal-title":"Gastroenterology"},{"issue":"4","key":"R17","doi-asserted-by":"publisher","first-page":"950","DOI":"10.14309\/ajg.0000000000003864","article-title":"Use of large language models to determine the surveillance colonoscopy interval: a bi-institutional validation study","volume":"121","author":"Acharya","journal-title":"Am J Gastroenterol"},{"issue":"1","key":"R18","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1038\/s41591-024-03425-5","article-title":"The TRIPOD-LLM reporting guideline for studies using large language models","volume":"31","author":"Gallifant","journal-title":"Nat Med"},{"issue":"24","key":"R19","doi-asserted-by":"publisher","first-page":"701","DOI":"10.21105\/joss.00701","article-title":"EndoMineR for the extraction of endoscopic and associated pathology data from medical reports","volume":"3","author":"Zeki","journal-title":"J Open Source Softw"},{"key":"R20","unstructured":"National data opt-out. NHS England. URL: https:\/\/digital.nhs.uk\/services\/national-data-opt-out [Accessed 09-09-2026]"},{"key":"R21","unstructured":"Abdin M Aneja J Behl H et al. Phi-4 technical report. arXiv. Preprint posted online on Dec 12, 2024. [doi: 10.48550\/arXiv.2412.08905]"},{"key":"R22","unstructured":"Zhang Z You W Wu T Wang X Li J Zhang M . A survey of generative information extraction. In: Proceedings of the 31st International Conference on Computational Linguistics. Association for Computational Linguistics; 2025:4840-4870."},{"issue":"11","key":"R23","doi-asserted-by":"publisher","first-page":"2015","DOI":"10.1136\/gutjnl-2023-330241","article-title":"Accuracy of measuring colorectal polyp size in pathology: a prospective study","volume":"72","author":"Djinbachian","journal-title":"Gut"},{"issue":"4","key":"R24","doi-asserted-by":"publisher","first-page":"319","DOI":"10.1001\/jama.2024.21700","article-title":"Testing and evaluation of health care applications of large language models: a systematic review","volume":"333","author":"Bedi","journal-title":"JAMA"},{"issue":"1","key":"R25","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-024-01074-z","article-title":"Foundation metrics for evaluating effectiveness of healthcare conversations powered by generative AI","volume":"7","author":"Abbasian","journal-title":"NPJ Digit Med"},{"key":"R26","unstructured":"Huang Y Tang K Chen M Wang B . A comprehensive survey on evaluating large language model applications in the medical industry. arXiv. Preprint posted online on Apr 24, 2024. [doi: 10.48550\/arXiv.2404.15777]"},{"key":"R27","unstructured":"Singh A Singh N Vatsal S . Robustness of large language models to perturbations in text. arXiv. Preprint posted online on Jul 12, 2024. [doi: 10.48550\/arXiv.2407.08989]"},{"key":"R28","doi-asserted-by":"crossref","unstructured":"Moradi M Samwald M . Evaluating the robustness of neural language models to input perturbations. In: Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics; 2021:1558-1570. [doi: 10.18653\/v1\/2021.emnlp-main.117]","DOI":"10.18653\/v1\/2021.emnlp-main.117"},{"key":"R29","doi-asserted-by":"crossref","unstructured":"Dong G Zhao J Hui T et al. Revisit input perturbation problems for LLMs: a unified robustness evaluation framework for noisy slot filling task. In: Natural Language Processing and Chinese Computing. Springer; 2023:682-694. [doi: 10.1007\/978-3-031-44693-1_53]","DOI":"10.1007\/978-3-031-44693-1_53"}],"container-title":["Journal of Medical Internet Research"],"original-title":[],"language":"en","deposited":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T17:11:48Z","timestamp":1792170708000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.jmir.org\/2026\/1\/e102789"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,16]]},"references-count":29,"URL":"https:\/\/doi.org\/10.2196\/102789","relation":{},"ISSN":["1438-8871"],"issn-type":[{"value":"1438-8871","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,16]]},"article-number":"v28i2e102789"}}