Advances for Managing Pancreatic Cystic Lesions - Integrating Imaging and AI Innovations

Sep 21, 2026

Pancreatic cystic lesions (PCLs) span benign non-neoplasms to neoplasms with malignant potential, creating major diagnostic and management challenges. Some lesions can become precursors to pancreatic cancer, so accurate differentiation is essential for optimal care. Conventional management depends on radiographic imaging and EUS-guided FNA combined with clinical and biochemical information, yet observer-dependent interpretation and complex lesion morphology produce uncertainty and variable treatment strategies. This review evaluates current diagnosis and surveillance practices and examines how AI—deep learning for pancreas/lesion segmentation and radiomics for heterogeneity analysis—can improve accuracy and risk stratification, support more objective follow-up and surgical decisions, and reduce unnecessary procedures.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages33

Reading Time1 h 23 min

AccessFree to download

Simple Summary

Abstract

Publication Information

Review Advances for Managing Pancreatic Cystic Lesions: Integrating Imaging and AI Innovations Deniz Seyithanoglu 1,2, Gorkem Durak 1, Elif Keles 1, Alpay Medetalibeyoglu 1,2, Ziliang Hong 1, Zheyuan Zhang 1, Yavuz B. Taktak 2, Timurhan Cebeci 2, Pallavi Tiwari 3,4, Yuri S. Velichko 1, Cemal Yazici 5, Temel Tirkes 6, Frank H. Miller 1, Rajesh N. Keswani 1, Concetto Spampinato 7, Michael B. Wallace 8 and Ulas Bagci 1, * Citation: Seyithanoglu, D.; Durak, G.; Keles, E.; Medetalibeyoglu, A.; Hong, Z.; Zhang, Z.; Taktak, Y.B.; Cebeci, T.; Tiwari, P.; Velichko, Y.S.; et al. Advances for Managing Pancreatic Cystic Lesions: Integrating Imaging and AI Innovations. Cancers 2024, 16, 4268. [https://doi.org/10.3390/](https://doi.org/10.3390/)[ ](https://doi.org/10.3390/)cancers16244268 Academic Editor: Yasunori Minami Received: 24 October 2024 Revised: 8 December 2024 Accepted: 19 December 2024 Published: 22 December 2024 Copyright: © 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) . 1 Machine and Hybrid Intelligence Lab, Feinberg School of Medicine, Northwestern University, Chicago, IL 60611, USA; [deniz.seyithanoglu@northwestern.edu](deniz.seyithanoglu@northwestern.edu) (D.S.); [gorkem.durak@northwestern.edu](gorkem.durak@northwestern.edu) (G.D.); [elif.keles@northwestern.edu](elif.keles@northwestern.edu) (E.K.); [alpay.m@istanbul.edu.tr](alpay.m@istanbul.edu.tr) (A.M.); [zilianghong2029@u.northwestern.edu](zilianghong2029@u.northwestern.edu) (Z.H.); [zheyuan.zhang@northwestern.edu](zheyuan.zhang@northwestern.edu) (Z.Z.); [y-velichko@northwestern.edu](y-velichko@northwestern.edu) (Y.S.V.); [frank.miller@nm.org](frank.miller@nm.org) (F.H.M.); [raj-keswani@northwestern.edu](raj-keswani@northwestern.edu) (R.N.K.) 2 Istanbul Faculty of Medicine, Istanbul University, Istanbul 38000, Turkey; [yavuztaktak@istanbul.edu.tr](yavuztaktak@istanbul.edu.tr) (Y.B.T.); [timurhan.cebeci@istanbul.edu.tr](timurhan.cebeci@istanbul.edu.tr) (T.C.) 3 Department of Radiology, BME, University of Wisconsin-Madison, Madison, WI 53707, USA; [ptiwari9@wisc.edu](ptiwari9@wisc.edu) 4 William S. Middleton Memorial Veterans Affairs (VA) Healthcare, 2500 Overlook Terrace, Madison, WI 53705, USA 5 Department of Gastroenterology, University of Illinois at Chicago, Chicago, IL 60611, USA; [cyazic2@uic.edu](cyazic2@uic.edu) 7 Department of Radiology, Indiana University, Indianapolis, IN 46202, USA; [atirkes@iupui.edu](atirkes@iupui.edu) 7 Department of Electrical, Electronics and Computer Engineering, University of Catania, 95124 Catania, Italy; [concetto.spampinato@unict.it](concetto.spampinato@unict.it) 8 Department of Gastroenterology, Mayo Clinic Florida, Jacksonville, FL 32224, USA; [wallace.michael@mayo.edu](wallace.michael@mayo.edu) * [Correspondence: ulas.bagci@northwestern.edu](Correspondence: ulas.bagci@northwestern.edu) Simple Summary: Pancreatic cystic lesions (PCLs) can range from harmless growths to precursors of pancreatic cancer, making accurate diagnosis crucial for patient care. Traditional methods for managing PCLs, such as imaging and biopsies, often depend on the skill of the doctor interpreting the images, leading to variability in diagnosis and treatment. This review highlights the challenges in diagnosing and managing PCLs and discusses the potential for artificial intelligence (AI) to improve accuracy. AI techniques, such as automated image analysis and deep learning algorithms, can provide more consistent and reliable assessments of pancreatic cysts. These tools could help doctors better identify high-risk lesions that need treatment, while avoiding unnecessary procedures for benign cysts. AI-driven methods show

FAQs

Why is accurate diagnosis important for pancreatic cystic lesions?

PCLs range from benign lesions to precursors of pancreatic cancer. Accurate differentiation is needed to determine appropriate management and avoid both missed risk and overtreatment.

What limitations affect conventional PCL management?

Conventional approaches rely on radiographic imaging and EUS-guided FNA with clinical and biochemical data, but interpretation is observer-dependent and lesion morphology is complex, leading to diagnostic uncertainty.

How can AI improve PCL diagnosis and management?

AI-driven methods such as deep learning for automated pancreas and lesion segmentation and radiomics for heterogeneity analysis can provide more objective, reproducible assessments. These can support better risk stratification and decision-making for follow-up intervals and surgery.

Sep 21, 2026

Upload by Theodora