High performance in risk stratification of intraductal papillary mucinous neoplasms by confocal laser endomicroscopy image analysis with convolutional neural networks (with video)

• Published in: Gastrointestinal endoscopy Vol. 94; no. 1; pp. 78 - 87.e2
• Main Authors: Machicado, Jorge D., Chao, Wei-Lun, Carlyn, David E., Pan, Tai-Yu, Poland, Sarah, Alexander, Victoria L., Maloof, Tassiana G., Dubay, Kelly, Ueltschi, Olivia, Middendorf, Dana M., Jajeh, Muhammed O., Vishwanath, Aadit B., Porter, Kyle, Hart, Phil A., Papachristou, Georgios I., Cruz-Monserrate, Zobeida, Conwell, Darwin L., Krishna, Somashekar G.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.07.2021
• Subjects: CNN; HGD-Ca; CI; LR; CAD; AI; LGD; IPMN; ROI; AUC; SD; HBM; PCL; AGA; nCLE; SBM
• ISSN: 0016-5107; 1097-6779; 1097-6779
• DOI: 10.1016/j.gie.2020.12.054

EUS-guided needle-based confocal laser endomicroscopy (EUS-nCLE) can differentiate high-grade dysplasia/adenocarcinoma (HGD-Ca) in intraductal papillary mucinous neoplasms (IPMNs) but requires manual interpretation. We sought to derive predictive computer-aided diagnosis (CAD) and artificial intelligence (AI) algorithms to facilitate accurate diagnosis and risk stratification of IPMNs. A post hoc analysis of a single-center prospective study evaluating EUS-nCLE (2015-2019; INDEX study) was conducted using 15,027 video frames from 35 consecutive patients with histopathologically proven IPMNs (18 with HGD-Ca). We designed 2 CAD-convolutional neural network (CNN) algorithms: (1) a guided segmentation-based model (SBM), where the CNN-AI system was trained to detect and measure papillary epithelial thickness and darkness (indicative of cellular and nuclear stratification), and (2) a reasonably agnostic holistic-based model (HBM) where the CNN-AI system automatically extracted nCLE features for risk stratification. For the detection of HGD-Ca in IPMNs, the diagnostic performance of the CNN-CAD algorithms was compared with that of the American Gastroenterological Association (AGA) and revised Fukuoka guidelines. Compared with the guidelines, both n-CLE-guided CNN-CAD algorithms yielded higher sensitivity (HBM, 83.3%; SBM, 83.3%; AGA, 55.6%; Fukuoka, 55.6%) and accuracy (SBM, 82.9%; HBM, 85.7%; AGA, 68.6%; Fukuoka, 74.3%) for diagnosing HGD-Ca, with comparable specificity (SBM, 82.4%; HBM, 88.2%; AGA, 82.4%; Fukuoka, 94.1%). Both CNN-CAD algorithms, the guided (SBM) and agnostic (HBM) models, were comparable in risk stratifying IPMNs. EUS-nCLE-based CNN-CAD algorithms can accurately risk stratify IPMNs. Future multicenter validation studies and AI model improvements could enhance the accuracy and fully automatize the process for real-time interpretation.