Development and validation of a SOTA-based system for biliopancreatic segmentation and station recognition system in EUS

• Published in: Surgical endoscopy Vol. 39; no. 8; pp. 4881 - 4889
• Main Authors: Zhang, Jingjie, Zhang, Jun, Chen, Hongsuo, Tian, Feilong, Zhang, Yu, Zhou, Yi, Jiang, Zhenyu
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.08.2025; Springer Nature B.V
• Subjects: Abdominal Surgery; Algorithms; Annotations; Biliary Tract Diseases - diagnostic imaging; Classification; Coronary vessels; Datasets; Deep Learning; Endoscopy; Endosonography - methods; Female; Gastroenterology; Gynecology; Hepatology; Hospitals; Humans; Male; Medicine; Medicine & Public Health; Middle Aged; Neural networks; Pancreas; Pancreatic Diseases - diagnostic imaging; Pancreatic Ducts - diagnostic imaging; Proctology; Quality standards; Small intestine; Surgery; Ultrasonic imaging; Deep learning; Image segmentation; Biliopancreatic; Artificial intelligence; Endoscopic ultrasound; Classification
• ISSN: 0930-2794; 1432-2218; 1866-6817; 1432-2218
• DOI: 10.1007/s00464-025-11858-3

Background Endoscopic ultrasound (EUS) is a vital tool for diagnosing biliopancreatic disease, offering detailed imaging to identify key abnormalities. Its interpretation demands expertise, which limits its accessibility for less trained practitioners. Thus, the creation of tools or systems to assist in interpreting EUS images is crucial for improving diagnostic accuracy and efficiency. Objective To develop an AI-assisted EUS system for accurate pancreatic and biliopancreatic duct segmentation, and evaluate its impact on endoscopists’ ability to identify biliary-pancreatic diseases during segmentation and anatomical localization. Methods The EUS-AI system was designed to perform station positioning and anatomical structure segmentation. A total of 45,737 EUS images from 1852 patients were used for model training. Among them, 2881 images were for internal testing, and 2747 images from 208 patients were for external validation. Additionally, 340 images formed a man–machine competition test set. During the research process, various newer state-of-the-art (SOTA) deep learning algorithms were also compared. Results In classification, in the station recognition task, compared to the ResNet-50 and YOLOv8-CLS algorithms, the Mean Teacher algorithm achieved the highest accuracy, with an average of 95.60% (92.07%–99.12%) in the internal test set and 92.72% (88.30%–97.15%) in the external test set. For segmentation, compared to the UNet ++ and YOLOv8 algorithms, the U-Net v2 algorithm was optimal. Ultimately, the EUS-AI system was constructed using the optimal models from two tasks, and a man–machine competition experiment was conducted. The results demonstrated that the performance of the EUS-AI system significantly outperformed that of mid-level endoscopists, both in terms of position recognition ( p  < 0.001) and pancreas and biliopancreatic duct segmentation tasks ( p  < 0.001, p  = 0.004). Conclusion The EUS-AI system is expected to significantly shorten the learning curve for the pancreatic EUS examination and enhance procedural standardization.