Early detection of breast cancer is critical in determining the best possible treatment approach. Ultrasound imaging has become an important modality in breast tumor detection and classification owing to its superiority to mammography in its ability to detect focal abnormalities in dense breast tissue. This paper discusses novel Fourier-based shape feature extraction techniques that provide enhanced classification accuracy for breast tumors in the computer-aided B-mode ultrasound diagnosis system. To verify the effectiveness of the proposed technique, experiments were performed using 4,107 ultrasound images containing 2,508 malignancy cases. Experimental results showed that the breast tumor classification accuracy has specificity of 95.8%, sensitivity of 94.1%, precision and recall of 95.7%, and accuracy of 94.9%.