Related Experiment Video
Updated: Jan 18, 2026
Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
Published on: December 15, 2014
Evaluation of parotid tumors using systemic inflammation markers and diffusion-weighted magnetic resonance images
Ahmet Ufuk Kilictas1, Ozlem Celebi Erdivanlı2, Mehmet Birinci2
1Department of Otorhinolaryngology, Konya City Hospital, Konya, Türkiye.
Background:
Preoperative differentiation between benign and malignant parotid gland tumors is challenging. Apparent Diffusion Coefficient (ADC) from Diffusion-Weighted Magnetic Resonance Imaging (DW-MRI) and systemic inflammatory markers such as neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) have emerged as non-invasive diagnostic tools.
Aims/Objectives:
To evaluate the diagnostic utility of ADC, NLR, and PLR individually and in combination for distingushing benign from malignant parotid tumors and for differentiating major histopathological subtypes.
Material And Methods:
This retrospective study included 138 patients with histopathologically confirmed parotid tumors. ADC values were obtained from DW-MRI, and NLR and PLR were calculated from preoperative CBC. A decision tree model incorporating age, gender, ADC, PLR, and NLR was developed, and diagnostic performance was assessed using ROC analysis.
Results:
Complete data was available for for 108 patients. The decision tree achieved an AUC of 0.837 for malignancy prediction, with age and PLR as key predictors. Excluding ADC did not impact model performance (AUC=0.837; p=1.0). While ADC alone did not differentiate malignancy, it was effective for subtype classification: AUC=0.891 for Warthin tumor vs pleomorphic adenoma, and 0.771 for pleomorphic adenoma vs mucoepidermoid carcinoma.
Conclusions And Significance:
Systemic inflammation markers enhance malignancy prediction, whereas ADC contributes meaningfully to histological subtype differentiation.
