人工智能量化克罗恩病内镜下黏膜溃疡
研究基于STARDUST和SEAVUE两项3期临床试验的结肠镜视频,开发计算机视觉内镜(CVE)模型量化克罗恩病黏膜溃疡和损伤。CVE与SES-CD评分高度相关(r=0.73-0.85),且能区分临床缓解状态,效应量大于SES-CD。
要点
- CVE模型分割性能与胃肠病学家标注者相当(Dice系数0.591 vs 0.462)。
- CVE测量与SES-CD高度相关(r=0.73-0.85,P<.0001)。
- CVE溃疡测量区分临床缓解的效应量大于SES-CD(g=0.416 vs 0.290)。
核心结论
- CVE可自动量化克罗恩病溃疡和黏膜损伤,与SES-CD概念一致,能提高内镜评估的粒度和个性化。
研究设计与样本
- 对STARDUST和SEAVUE 3期临床试验的结肠镜视频进行事后计算机视觉内镜评估。
- 分割模型基于2名胃肠病学家手工标注的图像训练,预测溃疡面积、严重程度和相对大小。
- 将预测的溃疡和一般黏膜损伤空间映射至回肠和结肠以量化克罗恩病负担。
主要发现
- 溃疡语义分割模型与胃肠病学家标注者性能相当(Dice相似系数0.591 vs 0.462)。
- CVE测量与SES-CD评分高度相关(r=0.73-0.85;P<.0001),与狭窄程度相关性差(r=0.12-0.21)。
- CVE溃疡测量(28.4 vs 52.3;P=.0012)和SES-CD(6.3 vs 9.0;P=.0193)可区分研究结束时临床缓解状态。
- CVE测量效应量大于SES-CD(g=0.416 vs g=0.290)。
- CVE在SEAVUE验证队列中性能相似。
关键数值与推荐
- 临床缓解定义为克罗恩病活动指数(CDAI)<150。
- CVE溃疡测量在临床缓解组与非缓解组分别为28.4和52.3(P=.0012)。
- SES-CD在临床缓解组与非缓解组分别为6.3和9.0(P=.0193)。
临床怎么用
- CVE可辅助内镜医师更客观、细致地量化克罗恩病溃疡和黏膜损伤。
- CVE测量与SES-CD一致性高,有望用于临床试验和临床实践中的疾病活动评估。
局限性
- CVE与狭窄程度相关性差(r=0.12-0.21),可能不适用于狭窄评估。
- 研究为事后分析,需进一步前瞻性验证。
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 & AIMS: Endoscopic scoring of Crohn's disease (CD) is challenging, as mucosal disease is patchy with highly variable morphology, size, and severity. Computer vision may help quantify disease activity with similar performance as standard instruments like the Simple Endoscopic Score for Crohn's Disease (SES-CD). METHODS: Colonoscopy videos from the STARDUST and SEAVUE phase 3 clinical trials underwent post-hoc computer vision endoscopic (CVE) assessment to quantify CD mucosal ulceration and injury. A segmentation model was trained on hand annotations of images performed by 2 gastroenterologists, predicting ulcer area, severity, and relative size. Using complete endoscopic video, predicted ulceration and general mucosal injury were then spatially mapped to the ileum and colon to quantify CD burden. CVE ulceration and general injury values were compared with the SES-CD in terms of disease quantification, localization, and agreement with end-of-study clinical remission (Crohn's Disease Activity Index [CDAI] <150). RESULTS: Ulcer semantic segmentation models matched the performance of gastroenterologist annotators (Dice similarity coefficient, 0.591 vs 0.462), with neither performing better on qualitative review of disagreements. CVE measures were highly correlated with SES-CD scores (r = 0.73-0.85; P < .0001), although there was expected poor correlation with the degree of stenosis (r = 0.12-0.21). CVE ulcer measurements (28.4 vs 52.3; P = .0012) and SES-CD (6.3 vs 9.0; P = .0193) separated end-of-study clinical remission status, although CVE measures had greater effect size than SES-CD (g = 0.416 vs g = 0.290). Performance of CVE for CD was similar in the SEAVUE validation cohort. CONCLUSIONS: CVE provides a means for automated ulceration and mucosal injury quantitation that shows conceptual agreement with SES-CD. CVE offers new capabilities to improve the granularity and personalization of endoscopic disease assessment in CD.
引用
- Cai L, Wittrup E, Minoccheri C, et al. Artificial Intelligence for Quantifying Endoscopic Mucosal Ulceration in Crohn's Disease. Clin Gastroenterol Hepatol. 2025. PMID: 40835044. doi:10.1016/j.cgh.2025.05.026
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。