开源自动肠道准备量表的开发与验证
本研究基于人工智能开发了开源自动肠道准备量表(OSABPS),通过计算粪便像素与黏膜像素比值(F/M)评估肠道准备质量,最优评分为0。内部验证显示OSABPS与BBPS显著相关(r=-0.42,P<.001),阈值0.09以下结肠镜的息肉检出率更高(P<.001);外部验证同样显示与BBPS显著相关(r=-0.70,P<.001)。
要点
- OSABPS通过粪便与黏膜像素比值自动评估肠道准备,最优评分为0。
- OSABPS与BBPS显著相关:内部验证r=-0.42,外部验证r=-0.70。
- OSABPS>0.09的结肠镜应考虑重新检查。
核心结论
- OSABPS可自动、即时、无人工偏倚地评估肠道准备质量。
- OSABPS>0.09的结肠镜应考虑重新检查。
- OSABPS的开源性质允许免费实施。
研究设计与样本
- OSABPS基于人工智能,使用来自3家医院的20例结肠镜的50,000帧图像进行训练。
- 内部验证使用来自3家医院的1405例结肠镜,外部验证使用公共结肠镜数据库(Nerthus)的5525帧图像。
主要发现
- 内部验证:OSABPS与BBPS显著相关(Pearson r=-0.42,P<.001)。
- 内部验证:通过Youden J确定OSABPS阈值为0.09。
- 内部验证:OSABPS低于Youden J阈值的结肠镜息肉检出率更高(两比例z检验,P<.001)。
- 外部验证:OSABPS与BBPS显著相关(Pearson r=-0.70,P<.001)。
关键数值与推荐
- OSABPS评分为粪便像素(F)与黏膜组织像素(M)的比值(F/M),0为最佳评分,表示完美清洁。
- OSABPS阈值为0.09,由Youden J确定。
- OSABPS>0.09的结肠镜应考虑重新检查。
临床怎么用
- OSABPS可自动、即时、无人工偏倚地评估肠道准备质量。
- OSABPS>0.09的结肠镜应考虑重新检查。
- OSABPS的开源性质允许免费实施。
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: Insufficient bowel preparation accounts for up to 42% of missed adenomas in colonoscopy. However, major analysis programs found no correlation between adenoma detection rate and the human-rated Boston Bowel Preparation Scale (BBPS), indicating limitations of the scale. We therefore aimed to develop an Open-Source Automatic Bowel Preparation Scale (OSABPS) based on artificial intelligence that is correlated to the polyp detection rate. METHODS: The OSABPS was trained on 50,000 frames from 20 colonoscopies from 3 hospitals. It involved quantifying the presence of fecal matter within the colonoscopy frames, using an approach termed the fecal ratio-the proportion of pixels identified as feces (F) relative to those identified as mucosal tissue (M) (OSABPS = F/M)-thereby making 0 the optimal score, indicating a perfect cleansing. The Youden J was used to set the threshold because it determines the optimal balance between sensitivity and specificity. The algorithm was then tested on 1405 colonoscopies from 3 hospitals (internal validation) and 5525 frames from a public colonoscopy database (Nerthus, external validation). RESULTS: Internal validation: the OSABPS correlated significantly with the BBPS (Pearson r = -0.42, P < .001). A threshold of 0.09 OSABPS was determined using the Youden J. The polyp detection rate was higher for colonoscopies below the threshold of the Youden J (2-proportion z-test, P < .001). External validation: the OSABPS correlated significantly with the BBPS (Pearson r = -0.70, P < .001). CONCLUSIONS: The OSABPS can automatically, instantly, and without human bias assess bowel preparation quality. Colonoscopies with an OSABPS of >0.09 should be considered for reexamination. The OSABPS's open-source nature allows free implementation.
引用
- Cold KM, Heen A, Vamadevan A, et al. Development and validation of the Open-Source Automatic Bowel Preparation Scale. Gastrointest Endosc. 2024. PMID: 39557204. doi:10.1016/j.gie.2024.11.022
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。