Usage of Magnetic Resonance Imaging-Based Texture Analysis Features in Discrimination of Benign and Malignant Sinonasal Tumors

Aim: The objective of this study was to differentiate between benign and malignant sinonasal tumors using magnetic resonance imaging (MRI)-based texture analysis features. Materials and Method: Histopathologically proven benign or malignant sinonasal tumor patients were included in the study from MRI examinations performed between January 2013 and December 2020. Inclusion criteria included a tumor size of at least 1 cm and preoperative magnetic resonance imaging with axial T1W, axial fat-suppressed T2W, and axial T1W postcontrast sequences. After the images were transferred to a dedicated workstation, texture analysis calculations were performed. Differences between benign and malignant groups were compared. Results: The mean age of 37 patients (8 female, 29 male) included in the study was 50.8 ± 21.9 years. In our study, we found no statistically significant difference between malignant and benign sinonasal tumors in nine tissue analysis parameters obtained by MRI. Conclusion: MRI-based texture analysis needs identical MRI protocols for evaluating tumors. MRI-based texture analysis is not a useful diagnostic tool to discriminate between benign and malignant sinonasal tumors when specific pathologic types are not selected and scanning protocols are not identical.