Improved Breast Cancer Detection with Computer-Aided Diagnostic System

• Enhanced Specificity and Accuracy: CAD demonstrated a notably higher specificity (92.71%) and accuracy (93.5%) when compared to conventional ultrasound (specificity: 62.5%, accuracy: 69.92%).
  
• Robust AUC: The Area Under the Receiver Operating Characteristic Curve (AUC) of CAD significantly surpassed that of conventional ultrasonography (0.9450 vs. 0.7940), signifying its superior diagnostic capability.
  
• Consistent Performance for Smaller Tumors: Even among patients with smaller masses (≤20 mm), CAD maintained its effectiveness. It exhibited higher specificity (91.43%), accuracy (90.70%), and AUC (0.8946) compared to conventional ultrasound (specificity: 51.43%, accuracy: 58.14%).
  
• Effective Downgrading: CAD successfully downgraded 31 patients initially classified as BI-RADS 4A and 4B, with subsequent confirmation that these cases were benign.
  

The findings of this study underscore the value of incorporating CAD systems into breast cancer screening and diagnosis protocols. CAD not only enhances diagnostic accuracy but also aids in the identification of benign lesions, particularly in cases initially categorised as BI-RADS 4A. This advancement has the potential to reduce the need for invasive procedures, minimising patient discomfort and healthcare costs.

Reference:

Song, P., Zhang, L., Bai, L., Wang, Q., & Wang, Y. Diagnostic performance of ultrasound with computer-aided diagnostic system in detecting breast cancer. Heliyon,2023;9(10):e20712. https://doi.org/10.1016/j.heliyon.2023.e20712