Differentiating between benign and malignant breast lesions using dual-energy CT-based model: development and...

Area of Science:

• Medical Imaging
• Radiology
• Oncology

Background:

• Accurate differentiation between benign and malignant breast lesions is crucial for patient management.
• Dual-energy CT (DECT) offers advanced imaging capabilities for breast lesion characterization.
• Incidental breast lesions detected on DECT require effective methods for assessment.

Purpose of the Study:

• To develop and validate a dual-energy CT (DECT)-based model for noninvasive differentiation of benign and malignant breast lesions.
• To assess the diagnostic performance and clinical utility of the developed DECT model.

Main Methods:

• Prospective enrollment of patients with suspected breast cancer undergoing contrast-enhanced DECT.
• Random division of breast lesions into training (70%) and testing (30%) cohorts.
• Collection of clinical, morphological, and quantitative DECT parameters; logistic regression and ROC analysis for model development and validation.

Main Results:

• A DECT-based model incorporating age, lesion shape, and effective atomic number (Zeff) demonstrated high diagnostic power.
• Area Under the Curve (AUC) values were 0.844 (training) and 0.791 (test) for differentiating lesion types.
• The model showed favorable calibration and clinical usefulness across a wide range of threshold probabilities.

Conclusions:

• The developed DECT-based model exhibits favorable diagnostic performance for noninvasively distinguishing benign from malignant breast lesions.
• This DECT model serves as a potential tool for characterizing breast lesions, including incidental findings, aiding in patient management decisions.