Artificial intelligence-tailored bowel preparation reduces laxative burden and promotes sustainable endoscopy

IF 25.8 1区医学 Q1 GASTROENTEROLOGY & HEPATOLOGY

Gut Pub Date : 2025-08-12 DOI:10.1136/gutjnl-2025-336200

Yang-Bor Lu, Si-Cun Lu, Yung-Ning Huang, Yu-Chieh Weng, Tung-Ying Chiang, Pei-Ting Cheng, Puo-Hsien Le, Cheng-Tang Chiu, Ankie T Cheung, Dorothy Chan, Raymond SY Tang, Xiao-Bing Cui, Hong-Zhi Xu, Shun-Tian Cai, Joseph JY Sung, Thomas YT Lam, Wei Gong

{"title":"Artificial intelligence-tailored bowel preparation reduces laxative burden and promotes sustainable endoscopy","authors":"Yang-Bor Lu, Si-Cun Lu, Yung-Ning Huang, Yu-Chieh Weng, Tung-Ying Chiang, Pei-Ting Cheng, Puo-Hsien Le, Cheng-Tang Chiu, Ankie T Cheung, Dorothy Chan, Raymond SY Tang, Xiao-Bing Cui, Hong-Zhi Xu, Shun-Tian Cai, Joseph JY Sung, Thomas YT Lam, Wei Gong","doi":"10.1136/gutjnl-2025-336200","DOIUrl":null,"url":null,"abstract":"Artificial intelligence (AI) has proven effective in assessing bowel preparation adequacy, but its role in guiding personalised laxative dosing is untested. In this prospective, multicentre, endoscopist-masked randomised trial, 1650 participants were assigned to AI-assisted or conventional groups using self-evaluation (SE) by patients during a 3 L split-dose polyethylene glycol (PEG) regimen. The AI group used significantly less PEG (mean difference −496.1 mL; p<0.001) while maintaining similar bowel cleanliness, polyp detection rates (PDR) and withdrawal times. Right colon scores were slightly higher, with less nausea and bloating reported in the AI group. AI-assisted assessment (AIA) offers an effective personalised approach to colonoscopy and warrants evaluation with other regimens. Colonoscopy plays a vital role in early detection of colorectal cancer (CRC), with adequate bowel cleansing being essential for mucosal visualisation.1 Our team previously developed an AIA tool capable of real-time evaluation of rectal effluent images, providing instant feedback and matching conventional self-assessment in quality.2 Given growing interest in sustainable practices in gastrointestinal endoscopy, and the burden of high-volume purgative regimens, a more individualised, eco-friendly approach to bowel preparation is warranted. No randomised controlled trials (RCTs) have explored real-time, AI-guided titration of purgative volume. We hypothesised that AIA could efficiently reduce PEG intake to individuals’ needs while maintaining preparation adequacy. We conducted a prospective, endoscopist-blinded RCT at Xiamen Chang Gung Hospital and Shenzhen Hospital of Southern Medical University from July 2023 through July 2024 (ChiCTR2300067499). Eligible patients aged 18–75 scheduled for colonoscopy were randomised 1:1 to either a traditional SE groupor an AIA group. All patients received 1 L of PEG the night before colonoscopy. The SE group ingested a mandatory additional 2 L, guided by photographic comparisons of rectal effluent, while the AIA group ingested a titrated volume from the remaining 2 L, stopping once the AI system returned a ‘pass’ after defecation photo …","PeriodicalId":12825,"journal":{"name":"Gut","volume":"40 1","pages":""},"PeriodicalIF":25.8000,"publicationDate":"2025-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Gut","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1136/gutjnl-2025-336200","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"GASTROENTEROLOGY & HEPATOLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Artificial intelligence (AI) has proven effective in assessing bowel preparation adequacy, but its role in guiding personalised laxative dosing is untested. In this prospective, multicentre, endoscopist-masked randomised trial, 1650 participants were assigned to AI-assisted or conventional groups using self-evaluation (SE) by patients during a 3 L split-dose polyethylene glycol (PEG) regimen. The AI group used significantly less PEG (mean difference −496.1 mL; p<0.001) while maintaining similar bowel cleanliness, polyp detection rates (PDR) and withdrawal times. Right colon scores were slightly higher, with less nausea and bloating reported in the AI group. AI-assisted assessment (AIA) offers an effective personalised approach to colonoscopy and warrants evaluation with other regimens. Colonoscopy plays a vital role in early detection of colorectal cancer (CRC), with adequate bowel cleansing being essential for mucosal visualisation.1 Our team previously developed an AIA tool capable of real-time evaluation of rectal effluent images, providing instant feedback and matching conventional self-assessment in quality.2 Given growing interest in sustainable practices in gastrointestinal endoscopy, and the burden of high-volume purgative regimens, a more individualised, eco-friendly approach to bowel preparation is warranted. No randomised controlled trials (RCTs) have explored real-time, AI-guided titration of purgative volume. We hypothesised that AIA could efficiently reduce PEG intake to individuals’ needs while maintaining preparation adequacy. We conducted a prospective, endoscopist-blinded RCT at Xiamen Chang Gung Hospital and Shenzhen Hospital of Southern Medical University from July 2023 through July 2024 (ChiCTR2300067499). Eligible patients aged 18–75 scheduled for colonoscopy were randomised 1:1 to either a traditional SE groupor an AIA group. All patients received 1 L of PEG the night before colonoscopy. The SE group ingested a mandatory additional 2 L, guided by photographic comparisons of rectal effluent, while the AIA group ingested a titrated volume from the remaining 2 L, stopping once the AI system returned a ‘pass’ after defecation photo …

查看原文本刊更多论文

人工智能定制肠道准备减轻泻药负担，促进可持续内窥镜检查

人工智能（AI）已被证明在评估肠道准备是否充分方面是有效的，但其在指导个性化泻药剂量方面的作用尚未经过测试。在这项前瞻性、多中心、内窥镜医师掩盖的随机试验中，1650名参与者在3l分剂量聚乙二醇（PEG）方案中使用患者自我评价（SE）被分配到人工智能辅助组或常规组。AI组使用PEG明显减少(平均差- 496.1 mL；p<0.001)，同时保持相似的肠道清洁度、息肉检出率（PDR）和停药时间。人工智能组的右结肠评分略高，恶心和腹胀的情况较少。人工智能辅助评估（AIA）为结肠镜检查提供了一种有效的个性化方法，值得与其他方案进行评估。结肠镜检查在早期发现结直肠癌（CRC）中起着至关重要的作用，充分的肠道清洁对粘膜可视化至关重要我们的团队之前开发了一种AIA工具，能够实时评估直肠流出物图像，提供即时反馈，并在质量上与传统的自我评估相匹配鉴于人们对胃肠道内窥镜的可持续实践越来越感兴趣，以及大容量泻药方案的负担，需要一种更个性化、更环保的肠道准备方法。没有随机对照试验（rct）探索实时、人工智能引导的泻药量滴定。我们假设AIA可以有效地减少PEG的摄入量以满足个人需求，同时保持准备的充分性。我们于2023年7月至2024年7月在厦门长工医院和南方医科大学深圳医院（ChiCTR2300067499）进行了一项前瞻性、内镜师盲法随机对照试验。年龄在18-75岁、计划进行结肠镜检查的符合条件的患者按1:1的比例随机分为传统SE组或AIA组。所有患者在结肠镜检查前一晚接受1l PEG。SE组在直肠流出物的摄影比较指导下，强制摄入额外的2l，而AIA组从剩余的2l中摄入滴定量，在AI系统返回大便照片后“通过”时停止。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Gut 医学-胃肠肝病学

CiteScore

45.70

自引率

2.40%

发文量

284

审稿时长

1.5 months

期刊介绍： Gut is a renowned international journal specializing in gastroenterology and hepatology, known for its high-quality clinical research covering the alimentary tract, liver, biliary tree, and pancreas. It offers authoritative and current coverage across all aspects of gastroenterology and hepatology, featuring articles on emerging disease mechanisms and innovative diagnostic and therapeutic approaches authored by leading experts. As the flagship journal of BMJ's gastroenterology portfolio, Gut is accompanied by two companion journals: Frontline Gastroenterology, focusing on education and practice-oriented papers, and BMJ Open Gastroenterology for open access original research.