Related Experiment Video
Updated: Mar 15, 2026
15:48
Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
Published on: December 15, 2014
23.4K
Quantitative imaging to evaluate malignant potential of IPMNs
Alexander N Hanania1,2, Leonidas E Bantis3, Ziding Feng3
1University of Texas Medical School, Houston, TX, USA.
Oncotarget
|September 3, 2016
Summary
Quantitative imaging can distinguish between low-grade and high-grade intraductal papillary mucinous neoplasms (IPMNs). This approach shows promise in identifying pancreatic cysts that require surgical resection, potentially improving patient outcomes.
Area of Science:
- Radiology
- Oncology
- Gastroenterology
Background:
- Pancreatic cysts are common, with a subset, including intraductal papillary mucinous neoplasms (IPMNs), being pre-malignant.
- Accurate prediction of malignant progression from IPMNs to pancreatic ductal adenocarcinoma (PDAC) remains a clinical challenge.
Purpose of the Study:
- To assess the malignant potential of pancreatic IPMNs using quantitative imaging features.
- To evaluate the efficacy of quantitative imaging biomarkers compared to current Fukuoka criteria for predicting IPMN grade.
Main Methods:
- Analysis of 360 imaging features (intensity, texture, shape) from CT scans of 53 patients with IPMN.
- Comparison of imaging feature performance against histopathological grading and Fukuoka criteria.
Main Results:
- Quantitative imaging identified 14 Gray-Level Co-Occurrence Matrix (GLCM) biomarkers distinguishing low-grade (LG) and high-grade (HG) IPMNs.
- A logistic regression model achieved an AUC of 0.96, with 97% sensitivity and 88% specificity for differentiating HG from LG IPMNs.
- HG IPMNs exhibited distinct imaging patterns compared to LG IPMNs.
Conclusions:
- High-grade IPMNs possess unique quantitative imaging characteristics.
- These imaging biomarkers may aid in identifying patients who would benefit most from surgical intervention, addressing a critical clinical need.
