迈向导管内乳头状黏液性肿瘤风险分层的自动化：人工智能借助共聚焦激光内镜超越人类专业知识。
Towards automating risk stratification of intraductal papillary mucinous Neoplasms: Artificial intelligence advances beyond human expertise with confocal laser endomicroscopy.
作者信息
Krishna Somashekar G, Abdelbaki Ahmed, Li Ziwei, Culp Stacey, Xiong Xinqi, Napoleon Bertrand, Mok Shaffer, Bertani Helga, Feng Yunlu, Kongkam Pradermchai, Luthra Anjuli K, Machicado Jorge D, El-Dika Samer, Leblanc Sarah, Tan Damien Meng Yew, Burlen Jordan, Keane Margaret G, Keihanian Tara, Ladd Antonio Mendoza, Muniraj Thiruvengadam, Visrodia Kavel H, Chen Wei, Esnakula Ashwini K, Hart Phil A, Chao Wei-Lun
机构信息
Division of Gastroenterology, Hepatology, and Nutrition, The Ohio State University Wexner Medical Center, Columbus, OH, USA.
Division of Gastroenterology, Hepatology, and Nutrition, The Ohio State University Wexner Medical Center, Columbus, OH, USA.
出版信息
Pancreatology. 2025 Aug;25(5):658-666. doi: 10.1016/j.pan.2025.05.011. Epub 2025 May 21.
BACKGROUND AND AIMS
Endoscopic Ultrasound (EUS)-guided needle-based confocal laser endomicroscopy (nCLE) enables real-time microscopic visualization of pancreatic cyst epithelium and can identify high-grade dysplasia/invasive adenocarcinoma (HGD/IC) in branch-duct (BD) intraductal papillary mucinous neoplasms (IPMNs). We aimed to compare the performance of experts (humans) with artificial intelligence (AI) in stratifying dysplasia in BD-IPMNs.
METHODS
This post-hoc analysis involved BD-IPMNs with definitive diagnoses from prospective EUS-nCLE studies (2015-2023) enrolled at a single center. Dysplasia grade was reviewed by two pathologists. Blinded EUS-nCLE experts reviewed unedited nCLE videos to classify dysplasia without and with revised Fukuoka criteria (revised-FC). The AI model, nCLE-AI, was similarly analyzed. Diagnostic parameters and AUC were compared to evaluate human and nCLE-AI performance.
RESULTS
Among 60 BD-IPMNs (mean size = 3.43 ± 1.00 cm), 23 (38.3 %) had HGD/IC. To detect HGD/IC using nCLE, interobserver agreement (IOA) among 16 nCLE experts was 'fair' (κ = 0.29, 95 % CI: 0.27-0.32), with a sensitivity of 58 %, specificity of 59 %, and AUC of 0.59 (95 % CI 0.55-0.62). Incorporating revised-FC improved the sensitivity to 72 % and AUC to 0.64 (95 % CI 0.61-0.68; p < 0.001), with similar IOA (κ = 0.36 'fair', 95 % CI: 0.33-0.38) and specificity (57 %). Comparatively, nCLE-AI achieved 87 % sensitivity, 54 % specificity, and an AUC of 0.70 (0.57-0.84). When combined with revised-FC, nCLE-AI reached 78 % sensitivity, 78 % specificity, and an AUC of 0.85 (95 % CI: 0.74-0.96, p = 0.02), significantly higher than humans with revised-FC (p < 0.01).
CONCLUSIONS
Human dysplasia classification of BD-IPMNs using nCLE showed modest IOA and accuracy. In contrast, nCLE-AI classifications combined with clinical criteria offer superior accuracy for detecting HGD/IC while eliminating interobserver variability.
背景与目的
内镜超声（EUS）引导下基于针的共聚焦激光显微内镜检查（nCLE）能够实时对胰腺囊肿上皮进行微观可视化，并可识别分支导管（BD）内导管乳头状黏液性肿瘤（IPMN）中的高级别异型增生/浸润性腺癌（HGD/IC）。我们旨在比较专家（人类）与人工智能（AI）在BD-IPMN异型增生分层中的表现。
方法
这项事后分析纳入了来自单中心前瞻性EUS-nCLE研究（2015 - 2023年）中确诊的BD-IPMN。两名病理学家对异型增生分级进行复查。不知情的EUS-nCLE专家审阅未编辑的nCLE视频，根据有无修订后的福冈标准（修订-FC）对异型增生进行分类。对AI模型nCLE-AI进行类似分析。比较诊断参数和曲线下面积（AUC）以评估人类和nCLE-AI的表现。
结果
在60个BD-IPMN中（平均大小 = 3.43 ± 1.00 cm），23个（38.3%）有HGD/IC。使用nCLE检测HGD/IC时，16名nCLE专家之间的观察者间一致性（IOA）为“一般”（κ = 0.29，95%可信区间：0.27 - 0.32），敏感性为58%，特异性为59%，AUC为0.59（95%可信区间0.55 - 0.62）。纳入修订-FC后，敏感性提高到72%，AUC提高到0.64（95%可信区间0.61 - 0.68；p < 0.00），IOA（κ = 0.36“一般”，95%可信区间：0.33 - 0.38）和特异性（57%）相似。相比之下，nCLE-AI的敏感性为87%，特异性为54%，AUC为0.70（0.57 - 0.84）。当与修订-FC结合时，nCLE-AI的敏感性达到78%，特异性达到78%，AUC为0.85（95%可信区间：0.74 - 0.96，p = 0.02），显著高于采用修订-FC的人类（p < 0.01）。
结论
使用nCLE对BD-IPMN进行人类异型增生分类显示出一般的IOA和准确性。相比之下，nCLE-AI分类结合临床标准在检测HGD/IC方面具有更高的准确性，同时消除了观察者间的变异性。
Zotero 插件