基于人工智能的肠道准备评估在筛查结肠镜检查中的验证(附视频)
本研究为前瞻性单中心观察性研究,于2021年10月8日至2022年11月9日在中山大学附属第八医院连续纳入393例接受筛查结肠镜检查的患者,采用人工智能e-BBPS与内镜医师BBPS评估肠道准备。AI判定不合格患者的>5 mm腺瘤漏诊率显著高于合格患者(35.71% vs 13.19%,P=0.0056;OR=0.2734,95%CI 0.1139-0.6565)。
要点
- AI判定肠道准备不合格患者的>5 mm腺瘤漏诊率显著高于合格患者(35.71% vs 13.19%,P=0.0056)。
- AI判定不合格患者的腺瘤漏诊率和>5 mm息肉漏诊率均显著高于合格患者。
- 研究证实AI判定不合格患者存在不可接受的>5 mm腺瘤漏诊率,为AI指导肠道再清洁提供关键证据。
核心结论
- 本研究证实,AI判定为肠道准备不合格的患者表现出不可接受的>5 mm腺瘤漏诊率,为实施AI指导肠道再清洁及未来可能标准化结肠镜筛查提供了关键证据(临床试验注册号:NCT05145712)。
研究设计与样本
- 前瞻性、单中心、观察性研究,于2021年10月8日至2022年11月9日在中山大学附属第八医院开展,连续纳入接受筛查结肠镜检查的符合条件的患者。
- AI使用e-Boston肠道准备量表(e-BBPS)评估肠道准备,内镜医师使用BBPS进行评估。
- 若BBPS和e-BBPS均判定准备充分,患者立即接受第二次结肠镜检查;否则患者接受肠道再清洁后再行第二次结肠镜检查。
主要发现
- 在393例患者中,检出72个>5 mm腺瘤,漏诊27个>5 mm腺瘤。
- AI不合格患者的>5 mm腺瘤漏诊率显著高于AI合格患者(35.71% vs 13.19%,P=0.0056;OR=0.2734,95%CI 0.1139-0.6565)。
- AI不合格患者的腺瘤漏诊率显著高于AI合格患者(50.89% vs 20.79%,P<0.001;OR=0.2532,95%CI 0.1583-0.4052)。
- AI不合格患者的>5 mm息肉漏诊率显著高于AI合格患者(35.82% vs 19.48%,P=0.0152;OR=0.4335,95%CI 0.2288-0.8213)。
临床怎么用
- 原文结论指出,本研究证实AI判定为肠道准备不合格的患者表现出不可接受的>5 mm腺瘤漏诊率,为实施AI指导肠道再清洁及未来可能标准化结肠镜筛查提供了关键证据。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND AND AIMS: Accurate bowel preparation assessment is essential for determining colonoscopy screening intervals. Patients with suboptimal bowel preparation are at a high risk of missing >5 mm adenomas and should undergo an early repeat colonoscopy. In this study, we used artificial intelligence (AI) to evaluate bowel preparation and validated the ability of the system to accurately identify patients who are at high risk of having >5 mm adenomas missed due to inadequate bowel preparation. METHODS: This prospective, single-center, observational study was conducted at the Eighth Affiliated Hospital, Sun Yat-sen University, from October 8, 2021, to November 9, 2022. Eligible patients who underwent screening colonoscopy were consecutively enrolled. The AI assessed bowel preparation using the e-Boston Bowel Preparation Scale (e-BBPS) while endoscopists made evaluations using BBPS. If both BBPS and e-BBPS deemed preparation adequate, the patient immediately underwent a second colonoscopy; otherwise, the patient underwent bowel re-cleansing before the second colonoscopy. RESULTS: Among the 393 patients, 72 adenomas >5 mm in size were detected; 27 adenomas >5 mm in size were missed. In unqualified-AI patients, the >5 mm adenoma miss rate (AMR) was significantly higher than in qualified-AI patients (35.71% vs 13.19% [P = .0056]; odds ratio [OR], .2734 [95% CI, .1139-.6565]), as were the AMR (50.89% vs 20.79% [P < .001]; OR, .2532 [95% CI, .1583-.4052]) and >5 mm polyp miss rate (35.82% vs 19.48% [P = .0152]; OR, .4335 [95% CI, .2288-.8213]). CONCLUSIONS: This study confirmed that patients classified as inadequate by AI exhibited an unacceptable >5 mm AMR, providing key evidence for implementing AI in guiding bowel re-cleansing and potentially standardizing future colonoscopy screening. (Clinical trial registration number: NCT05145712.).
引用
- Yao L, Xiong H, Li Q, et al. Validation of artificial intelligence-based bowel preparation assessment in screening colonoscopy (with video). Gastrointest Endosc. 2024. PMID: 38636818. doi:10.1016/j.gie.2024.04.015
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。