Multi-parameter ultrasonography-based predictive model for breast cancer diagnosis.

Area of Science:

• Medical Imaging
• Oncology
• Diagnostic Technology

Background:

• Breast cancer diagnosis relies on imaging techniques, but differentiating benign from malignant lesions can be challenging.
• Conventional ultrasonography (US), shear wave elastography (SWE), and contrast-enhanced US (CEUS) offer complementary information for breast lesion assessment.

Purpose of the Study:

• To develop, validate, and evaluate a predictive model for breast cancer diagnosis.
• To integrate data from conventional US, SWE, and CEUS for improved diagnostic accuracy.

Main Methods:

• A retrospective study of 674 patients with 674 breast lesions was conducted.
• Data were divided into training (Cohort 1), validation (Cohort 2), and independent testing (Cohort 3) sets.
• Logistic regression analysis identified risk factors, and a predictive model was established and evaluated using receiver operating characteristic curve analysis (AUC).

Main Results:

• Multivariable regression identified nine independent breast cancer risk factors from US, SWE, and CEUS findings.
• The predictive model demonstrated good diagnostic performance with AUC values of 0.847, 0.857, and 0.774 in Cohorts 1, 2, and 3, respectively.
• The model significantly improved diagnostic specificity (e.g., from 7.3% to 73.1% in Cohort 3) with minimal loss in sensitivity.

Conclusions:

• A multi-parameter US-based predictive model effectively aids in breast cancer diagnosis.
• The model enhances diagnostic specificity, potentially reducing the need for unnecessary breast biopsies.
• This tool can guide clinical decision-making for breast cancer diagnosis and treatment planning.