人工智能用于肠道准备评估
本研究利用结肠镜视频开发了两种卷积神经网络(CNN),用于客观评估肠道准备是否充分。二分类CNN区分适合与不适合波士顿肠道准备量表(BBPS)评分的视频帧,多分类CNN对适合帧进行BBPS节段评分(0-3)。验证集曲线下面积(AUC)为0.918,检测肠道清洁不充分的准确率为85.3%;测试集1对肠道准备不充分的灵敏度为100%,评分者与AI的一致性为76.7%至83.3%;测试集2各节段灵敏度均为100%,一致性为68.9%至89.7%;评分者之间与评分者-AI之间的一致性相似(κ=0.694 vs κ=0.649)。结论认为该算法基于BBPS评估肠道清洁度表现良好,与专家一致性良好,包括完整退镜结肠镜检查。
要点
- 验证集检测肠道清洁不充分的AUC为0.918,准确率为85.3%。
- 测试集1对肠道准备不充分的灵敏度为100%,评分者与AI一致性为76.7%至83.3%。
- 测试集2各节段对肠道准备不充分的灵敏度均为100%,评分者与AI一致性为68.9%至89.7%。
核心结论
- 基于BBPS的算法评估肠道清洁度表现良好,与专家一致性良好,包括完整退镜结肠镜检查。
研究设计与样本
- 使用200例结肠镜检查的73,304张图像作为训练集开发两种CNN;二分类CNN区分适合与不适合BBPS评分的视频帧,多分类CNN在26,950张经专家标注BBPS节段评分(0-3)的适合帧上训练;使用252个10秒视频片段由2名专家赋予BBPS节段评分进行验证;测试集1为30个独立10秒视频片段,测试集2为10个完整退镜结肠镜视频。
主要发现
- 验证集中算法检测肠道清洁不充分的AUC为0.918,准确率为85.3%。
- 测试集1中检测肠道准备不充分的灵敏度为100%,评分者与AI的一致性为76.7%至83.3%。
- 测试集2中各节段检测肠道准备不充分的灵敏度均为100%,评分者与AI的一致性为68.9%至89.7%。
- 评分者之间的一致性κ=0.694,评分者与AI之间的一致性κ=0.649,两者相似。
临床怎么用
- 原文未据此给出推荐。
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: A reliable assessment of bowel preparation is important to ensure high-quality colonoscopy. Current bowel preparation scoring systems are limited by interobserver variability. This study aimed to demonstrate objective assessment of bowel preparation adequacy using an artificial intelligence (AI)/convolutional neural network (CNN) algorithm developed from colonoscopy videos. METHODS: Two CNNs were developed using a training set of 73,304 images from 200 colonoscopies. First, a binary CNN was developed and trained to distinguish video frames that were appropriate versus inappropriate for scoring with the Boston Bowel Preparation Scale (BBPS). A second multiclass CNN was developed and trained on 26,950 appropriate frames that were expertly annotated with BBPS segment scores (0-3). We validated the algorithm using 252 10-second video clips that were assigned BBPS segment scores by 2 experts. The algorithm provided mean BBPS scores based on the algorithm (AI-BBPS) by calculating mean BBPS based on each frame's scoring. We maximized the algorithm's performance by choosing a dichotomized AI-BBPS score that closely matched dichotomized BBPS scores (ie, adequate vs inadequate). We tested the mean BBPS score based on the algorithm AI-BBPS against human rating using 30 independent 10-second video clips (test set 1) and 10 full withdrawal colonoscopy videos (test set 2). RESULTS: In the validation set, the algorithm demonstrated an area under the curve of .918 and accuracy of 85.3% for detection of inadequate bowel cleanliness. In test set 1, sensitivity for inadequate bowel preparation was 100% and agreement between raters and AI was 76.7% to 83.3%. In test set 2, sensitivity for inadequate bowel preparation for each segment was 100% and agreement between raters and AI was 68.9% to 89.7%. Agreement between raters alone versus raters and AI were similar (κ = .694 and .649, respectively). CONCLUSIONS: The algorithm assessment of bowel cleanliness as measured with the BBPS showed good performance and agreement with experts including full withdrawal colonoscopies.
引用
- Lee JY, Calderwood AH, Karnes W, et al. Artificial intelligence for the assessment of bowel preparation. Gastrointest Endosc. 2021. PMID: 34896100. doi:10.1016/j.gie.2021.11.041
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。