Discrimination between breast invasive ductal carcinomas and benign lesions by optimizing quantitative parameters...

Area of Science:

• Radiology
• Medical Imaging
• Oncology

Background:

• Distinguishing invasive ductal carcinomas (IDCs) from benign breast lesions on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is crucial for accurate diagnosis.
• Conventional methods often rely on manual region of interest (ROI) selection and qualitative assessment of time-signal intensity curves (TICs).

Purpose of the Study:

• To propose and evaluate a novel semi-automatic method for improved differentiation of breast IDCs from benign lesions using DCE-MRI.
• To compare the diagnostic performance of the proposed semi-automatic method against conventional manual techniques.

Main Methods:

• A cohort of 142 breast lesion cases was analyzed.
• The proposed semi-automatic method involved automated lesion extraction and derivation of multiple quantitative perfusion parameters beyond the maximum slope of increase (MSI).
• Regression models were established using quantitative parameters from both the lesion region and mean TIC for diagnostic classification.

Main Results:

• The semi-automatic method achieved significantly higher accuracy (76.05%) compared to the traditional qualitative TIC categorization (59.16%) (P < 0.05).
• Quantitative regression models derived from the semi-automatic method demonstrated high diagnostic accuracies: 81.0% for lesion region parameters and 78.9% for mean TIC parameters.
• The accuracy of the conventional manual method using only MSI was 63.35%.

Conclusions:

• The proposed semi-automatic method significantly enhances the discrimination between breast IDCs and benign lesions on DCE-MRI.
• This advanced quantitative approach serves as a valuable supplementary tool, aiding clinical radiologists in subjective diagnosis and improving patient management.