Related Experiment Video
Updated: Aug 7, 2025
Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
Published on: December 15, 2014
A preliminary diagnostic accuracy study of quantitative MRI biomarkers for differentiating parotid tumor types
Aleksandar Alavanja1, Grayson W Hooper2, Adam Hasse1
1Department of Radiology, Section of Neuroradiology, University of Chicago, Chicago, IL, USA.
Background:
Differentiating among the different types of parotid tumors on imaging is useful for guiding clinical disposition, which ultimately may lead to surgical management. The goal of this study was to determine whether quantitative T2 signal characteristics and morphologic features on magnetic resonance imaging (MRI) can serve as predictive biomarkers for distinguishing between tumor types.
Methods:
A retrospective review of T2-weighted MRIs in patients with pathology-proven parotid tumors was performed. Quantitative T2 maps and surface regularity measurements of the tumors were obtained via semi-automated regions of interest (ROI). Linear Discriminant Analysis was used to populate the receiver operating characteristics (ROCs) curves for these variables. A P value of <0.05 was considered to be significant.
Results:
A total of 35 tumors (21 benign and 14 malignant neoplasms) were included in this analysis. For differentiating the benign versus malignant classes of parotid tumors, T2 signal and surface regularity combined yielded an area under the curve of 0.62 (P value: 0.2) through the ROC analysis. However, for the pleomorphic adenomas versus other types of parotid tumors, using both T2 signal and surface regularity yielded an area under the curve of 0.81 (P value: 0.007) through the ROC analysis.
Conclusions:
T2 signal and surface regularity combined can significantly differentiate pleomorphic adenomas from other types of parotid tumors and can potentially be used as a predictive imaging biomarker.
