Department of Medicine, Queen Mary Hospital, University of Hong Kong, Hong Kong.
Department of Surgery, Queen Mary Hospital, University of Hong Kong, Hong Kong.
Gastrointest Endosc. 2021 Jan;93(1):193-200.e1. doi: 10.1016/j.gie.2020.04.066. Epub 2020 May 4.
Meta-analysis shows that up to 26% of adenomas could be missed during colonoscopy. We investigated whether the use of artificial intelligence (AI)-assisted real-time detection could provide new insights into mechanisms underlying missed lesions during colonoscopy.
A validated real-time deep-learning AI model for the detection of colonic polyps was first tested in videos of tandem colonoscopy of the proximal colon for missed lesions. The real-time AI model was then prospectively validated in a total colonoscopy in which the endoscopist was blinded to real-time AI findings. Segmental unblinding of the AI findings were provided, and the colonic segment was then re-examined when missed lesions were detected by AI but not the endoscopist. All polyps were removed for histologic examination as the criterion standard.
Sixty-five videos of tandem examination of the proximal colon were reviewed by AI. AI detected 79.1% (19/24) of missed proximal adenomas in the video of the first-pass examination. In 52 prospective colonoscopies, real-time AI detection detected at least 1 missed adenoma in 14 patients (26.9%) and increased the total number of adenomas detected by 23.6%. Multivariable analysis showed that a missed adenoma(s) was more likely when there were multiple polyps (adjusted odds ratio, 1.05; 95% confidence interval, 1.02-1.09; P < .0001) or colonoscopy was performed by less-experienced endoscopists (adjusted odds ratio, 1.30; 95% confidence interval, 1.05-1.62; P = .02).
Our findings provide new insights on the prominent role of human factors, including inexperience and distraction, on missed colonic lesions. With the use of real-time AI assistance, up to 80% of missed adenomas could be prevented. (Clinical trial registration number: NCT04227795.).
荟萃分析显示，结肠镜检查中约有 26%的腺瘤可能会被遗漏。我们研究了人工智能（AI）辅助实时检测是否能为结肠镜检查中遗漏病变的机制提供新的见解。
首先在近端结肠镜串联检查的视频中测试一种经过验证的实时深度学习 AI 模型，以检测结肠息肉。然后，前瞻性地验证该实时 AI 模型在全结肠镜检查中的应用，在此过程中，内镜医师对实时 AI 结果设盲。提供 AI 结果的分段非盲法，当 AI 检测到而内镜医师未检测到遗漏病变时，重新检查 AI 提示的结肠段。所有息肉均切除进行组织学检查作为金标准。
共回顾了 65 例近端结肠镜串联检查的视频，AI 检测到 19/24 例（79.1%）首次检查时遗漏的近端腺瘤。在 52 例前瞻性结肠镜检查中，实时 AI 检测在 14 例患者（26.9%）中至少检测到 1 个遗漏的腺瘤，并使检测到的腺瘤总数增加了 23.6%。多变量分析显示，当存在多个息肉（调整后的优势比，1.05；95%置信区间，1.02-1.09；P<0.0001）或经验较少的内镜医师进行结肠镜检查时（调整后的优势比，1.30；95%置信区间，1.05-1.62；P=0.02），更有可能遗漏腺瘤。
我们的研究结果提供了关于人为因素（包括经验不足和注意力分散）在遗漏结肠病变中起重要作用的新见解。使用实时 AI 辅助，可以预防高达 80%的遗漏腺瘤。（临床试验注册号：NCT04227795。）
