Objective: This study assessed the accuracy of three-dimensional (3-D) power Doppler ultrasound differentiating between benign and malignant breast tumors by using support vector machine (SVM).
Methods: 3-D power Doppler ultrasonography was performed on 164 patients with 86 benign and 78 malignant breast tumors. The volume of interest (VOI) in 3-D ultrasound images was automatically generated from three rectangular region of interest (ROI). The vascularization index (VI), flow index (FI) and vascularization-flow index (VFI) on 3-D power-Doppler ultrasound images were evaluated for the entire volume area, computer extracted VOI area and area opposite the VOI. Furthermore, patient age and VOI volume were also applied for breast tumor identification. Each ultrasonography in this study was classified as benign or malignant based on the features using the SVM model. All the tumors were sampled using k-fold cross-validation (k = 10) to evaluate the diagnostic performance with receiver operating characteristic (ROC) curve.
Results: The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy of SVM for classifying malignancies were 94%, 67%, 72%, 92% and 80%, respectively. The classification performance in terms of Az value for the ROC curve of the features derived from 3-D power Doppler is 0.91.
Conclusions: This study demonstrated that the vascular features of 3-D power Doppler ultrasound combined with patient age and VOI volume can effectively differentiate benign and malignant breast tumors using SVM.