Related Experiment Video
Updated: Jun 1, 2025
07:12
Author Spotlight: Revolutionizing Pancreatic Disease Understanding Through Advanced Intravital Imaging
Published on: October 6, 2023
1.8K
Artificial Intelligence in Pancreatic Intraductal Papillary Mucinous Neoplasm Imaging: A Systematic Review
Muhammad Ibtsaam Qadir1, Jackson A Baril2, Michele T Yip-Schneider2
1Weldon School of Biomedical Engineering, Purdue University, West Lafayette, IN, USA.
Medrxiv : the Preprint Server for Health Sciences
|January 20, 2025
Summary
Artificial Intelligence (AI) shows promise in improving the diagnosis of intraductal papillary mucinous neoplasms (IPMN) using medical imaging. Further research with larger, multi-center datasets is needed to fully realize AI's potential in clinical practice.
Area of Science:
- Medical Imaging Analysis
- Artificial Intelligence in Oncology
- Gastrointestinal Pathology
Background:
- Current intraductal papillary mucinous neoplasm (IPMN) management relies on imaging, which, despite high sensitivity for high-risk cases, leads to overtreatment due to low specificity.
- Artificial Intelligence (AI) offers a potential solution to enhance diagnostic accuracy in IPMN imaging.
Purpose of the Study:
- To systematically review the existing literature on AI applications in IPMN imaging.
- To identify trends, gaps, and future directions in AI-driven IPMN research.
Main Methods:
- A systematic literature search identified 1041 publications on AI in IPMN imaging, with 25 studies included in the analysis.
- Studies were analyzed based on prediction target, data type, imaging modality, cohort size, and clinical translation stage.
Main Results:
- Research in AI for IPMN imaging is rapidly growing, with most studies using CT scans and convolutional neural network (CNN) algorithms.
- The majority of models were developed on single-center datasets with fewer than 250 patients.
- Most AI models focused on differential diagnosis and risk stratification, rather than IPMN detection or segmentation.
Conclusions:
- AI holds significant potential for improving the accuracy and precision of IPMN patient stratification.
- Multicenter collaboration, diverse datasets, and a focus on clinical translation are crucial for advancing AI in IPMN management.
