We used the tumor region with different thicknesses of the peritumor (surrounding tissue) to predict the metastasis status of a tumor. This study finds that the peritumor 15 pixels achieved the best performance with an accuracy of 84.8%, a sensitivity of 88.8%, a specificity of 81.3%, and an area under the curve (AUC) of receiver operating characteristic (ROC) of 0.926.

Some computer-aided diagnosis (CAD) systems used of primary tumor characteristics (such as tumor size, tumor type and grade, lymphovascular invasion, and hormonal status) to predict metastasis status in breast cancer.  But, the performance was not good. 

We developed a CAD system to determine the metastasis status in breast cancer using the tumor surrounding tissue features in 2D US images.

This study will help physicians to predict the tumor metastasis status in the future and the treatment direction of prognosis. We also need more data and clinical trials to prove the robustness of this method.