Texture Analysis Based On Pretreatment Diffusion Weighted Magnetic Resonance To Predict The Risks Of Non-sentinel Lymph Node Metastasis In Patients With Breast Cancer

Purpose: To investigate the feasibility of predicting the risk of non-sentinel lymph node(NSLN)metastasis in patients with breast cancer by the text ure analysis based on diffusion-weighted magnetic resonance imaging(DW-MRI).Methods: Sixty patients with stage I-III sentinel-positive breast cancer who underwent preoperative breast 3-T magnetic resonance imaging(MRI)were included.Two radiologists delineated a region of interest(ROI)around the tumor outline for the largest cross-sectional area on apparent diffusion coefficients(ADC)maps derived from DW-MRI.Texture analysis was then performed to extract quantitative image parameters.The resulting parameters were correlated with NSLN metastasis by univariate and multivariate analysis.Internal validation was performed by stratified 5-fold cross validation.Results: Both independent sample T test and binary logistic regression analysis indicated that Sdga43 was predictive of NSLN status(p<0.05).The area under the ROC curve was 0.739(p<0.05).Internal validation was performed using stratified 5-fold cross-validation,suggesting that the parameter is reliable.Conclusions: Texture parameters extracted from DW-MRI could act as imaging biomarkers to evaluate the risk of NSLN metastasis in patients with early breast cancer.It may be useful for identifying the patient group who is unable to benefit from axillary lymph node dissection(ALND)in the clinical practice.