A convolutional neural network-based system for identifying neuroendocrine neoplasms and multiple types of lesions in the pancreas using EUS (with videos)

Background and aims: EUS is sensitive in detecting pancreatic neuroendocrine neoplasm (pNEN). However, the endoscopic diagnosis of pNEN is operator-dependent and time-consuming because pNEN mimics normal pancreas and other pancreatic lesions. We intended to develop a convolutional neural network (CNN)-based system, named iEUS, for identifying pNEN and multiple types of pancreatic lesions using EUS.

Methods: Retrospective data of 12,200 EUS images obtained from pNEN and non-pNEN pancreatic lesions, including pancreatic ductal adenocarcinoma (PDAC), autoimmune pancreatitis (AIP), and pancreatic cystic neoplasm (PCN), were used to develop iEUS, which was composed of a 2-category (pNEN or non-pNEN pancreatic lesions) classification model (CNN1) and a 4-category (pNEN, PDAC, AIP, or PCN) classification model (CNN2). Videos from consecutive patients were prospectively collected for a human-iEUS contest to evaluate the performance of iEUS.

Results: Five hundred seventy-three patients were enrolled in this study. In the human-iEUS contest containing 203 videos, CNN1 and CNN2 showed an accuracy of 84.2% and 88.2% for diagnosing pNEN, respectively, which were significantly higher than that of novices (75.4%) and comparable with intermediate endosonographers (85.5%) and experts (85.5%). In addition, CNN2 showed an accuracy of 86.2%, 97.0%, and 97.0% for diagnosing PDAC, AIP, and PCN, respectively. With the assistance of iEUS, the sensitivity of endosonographers at all 3 levels in diagnosing pNEN has significantly improved (64.6% vs 44.8%, 87.5% vs 71.9%, and 74.0% vs 57.6%, respectively).

Conclusions: The iEUS precisely diagnosed pNEN and other confusing pancreatic lesions and thus can assist endosonographers in achieving more accessible and accurate endoscopic diagnoses with EUS. (Clinical trial registration number: ChiCTR2100049697.).