Comparison of Magnetic Resonance Imaging (MRI) Texture Analysis Results with Pathology Results of Axillary Lymph Nodes

Introduction Materials and Methods Results Conclusion The aim of this study is to investigate the effectiveness of magnetic resonance ımaging (MRI) and texture analysis (TA) features in detecting metastatic axillary lymph nodes (ALN) in patients with breast cancer. The study included female patients over the age of 18 who underwent fine needle aspiration biopsy and/or cutting needle biopsy for ALNs in the Interventional Radiology department of our hospital between 2018 and 2020, and who had pathological ALN with dynamic contrast and diffusion breast MRI. There was a statistically significant difference between benign and malignant lymph nodes in terms of the presence of T2A fatty hilum (p=0.008). In the evaluation of MR texture analysis results; A significant difference was found between the T2 area and T2 skewness values (respectively p=0.006; 0.029), between ADC area and ADC kurtosis values (p=0.027;0.005, respectively), between postcontrast T1 area, postcontrast T1 variance and postcontrast T1 kurtosis values (respectively, p=0.036; 0.010; <0.001) in benign and malignant groups. MR features, which are significant in distinguishing between micrometastasis and macrometastasis in cases of the malignant group, mean signal intensity on precontrast T1-weighted images (p=0.048); Area, Skewness, Kurtosis in T2-weighted images of MR texture parameters; Area, Mean, Kurtosis, Sumentropy, Entropy in ADC sequences in diffusion-weighted images; Area and Kurtosis values were observed on postcontrast T1-weighted images (p:<0.001-0.015). Although magnetic resonance imaging-based texture analysis is an increasingly common application, there are not enough studies in the literature investigating MRI texture analysis for axillary lymph nodes. MRI TA, which is a noninvasive and reproducible method, contributes to other MRI methods in characterizing metastatic ALNs in the preoperative period.