Copyright: ©Author(s) 2026. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution-NonCommercial (CC BY-NC 4.0) license. No commercial re-use. See permissions. Published by Baishideng Publishing Group Inc.
World J Gastrointest Oncol. Jul 15, 2026; 18(7): 119847
Published online Jul 15, 2026. doi: 10.4251/wjgo.v18.i7.119847
Published online Jul 15, 2026. doi: 10.4251/wjgo.v18.i7.119847
Interpretable multimodal artificial intelligence model for predicting advanced neoplasia in pancreatic cystic lesions
Yavuz Özden, Ömer Yüzügülen, Ferhat Omurca, Department of Gastroenterology, Kayseri City Hospital, University of Health Sciences, Kayseri 38080, Türkiye
Author contributions: Özden Y designed the study; Özden Y and Yüzügülen Ö developed the methodology; Omurca F acquired the data; Özden Y performed the statistical analysis and drafted the manuscript; all authors revised the manuscript and approved the final version.
AI contribution statement: AI-assisted language tools were used in a limited manner for language polishing and editorial refinement during manuscript revision, including Paperpal.
Institutional review board statement: The study was approved by the Ethics Committee of the University of Health Sciences, Kayseri City Hospital (Protocol No. 602; approval date: October 13, 2025).
Informed consent statement: The requirement for informed consent was waived by the Ethics Committee due to the retrospective design and use of anonymized data.
Conflict-of-interest statement: The authors declare no conflict of interest.
STROBE statement: The authors have read the STROBE Statement—checklist of items, and the manuscript was prepared and revised according to the STROBE Statement—checklist of items.
Data sharing statement: De-identified data are available from the corresponding author upon reasonable request.
Corresponding author: Yavuz Özden, MD, Department of Gastroenterology, Kayseri City Hospital, University of Health Sciences, Şeker District, Muhsin Yazıcıoğlu Boulevard No. 77 Kocasinan, Kayseri 38080, Türkiye. yavuzozden@gmail.com
Received: February 7, 2026
Revised: February 28, 2026
Accepted: April 20, 2026
Published online: July 15, 2026
Processing time: 149 Days and 10.3 Hours
Revised: February 28, 2026
Accepted: April 20, 2026
Published online: July 15, 2026
Processing time: 149 Days and 10.3 Hours
Core Tip
Core Tip: Current guideline-based algorithms for pancreatic cystic lesions predominantly rely on morphology and show limited discrimination for advanced neoplasia. In this retrospective cohort study of 187 surgically resected lesions, we developed and internally validated an interpretable multimodal model integrating clinical variables, endoscopic ultrasound features, and cyst fluid biomarkers. The model achieved superior discrimination compared with international guideline criteria and demonstrated stable performance across key subgroups. This interpretable framework may support preoperative risk stratification, pending external validation.