Related Experiment Video
Updated: Jun 12, 2025
15:48
Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
Published on: December 15, 2014
22.5K
Discrimination Between Benign and Malignant Lesions With Restriction Spectrum Imaging MRI in an Enriched Breast
Stephane Loubrie1, Jingjing Zou2, Ana E Rodriguez-Soto1
1Department of Radiology, University of California San Diego, La Jolla, California, USA.
Journal of Magnetic Resonance Imaging : JMRI
|September 18, 2024
Summary
A new diffusion-weighted MRI (DW-MRI) model, BS-RSI3C, effectively distinguishes high-risk benign breast lesions (HRBL) from average-risk benign lesions (ARBL) in screening. This advanced model also shows superior performance over standard ADC in differentiating cancerous lesions from benign ones.
Area of Science:
- Magnetic Resonance Imaging
- Breast Cancer Diagnostics
- Medical Imaging Analysis
Background:
- Dynamic contrast-enhanced MRI (DCE-MRI) is used for high-risk breast cancer screening but has limited specificity.
- Distinguishing cancerous (CL) and high-risk benign lesions (HRBL) from average-risk benign lesions (ARBL) remains a challenge.
- Improved non-invasive imaging techniques are needed to enhance diagnostic accuracy.
Purpose of the Study:
- To assess the efficacy of a breast-specific diffusion-weighted MRI (DW-MRI) model, BS-RSI3C, in differentiating lesion types.
- To improve the discrimination between CL, HRBL, and ARBL within an enriched screening population.
- To evaluate the model's performance compared to traditional apparent diffusion coefficient (ADC) measurements.
Main Methods:
- A prospective study involving 187 women undergoing breast MRI.
- Multishell DW-MRI at 3.0T was performed, and data were processed to generate ADC maps and signal contributions (C1, C2, C3) using the BS-RSI3C model.
- Statistical analysis, including ANOVA and t-tests, was used to compare lesion groups, with ROC curves evaluating diagnostic performance.
Main Results:
- The BS-RSI3C model's signal contributions (C1, √C1C2, and C3) significantly differed between HRBLs and ARBLs (P<0.05).
- The model parameter C3 demonstrated the highest AUC (0.821) in distinguishing combined cancerous and high-risk lesions (CHRLs) from ARBLs.
- This parameter outperformed standard ADC (AUC 0.696), particularly for non-mass enhancement lesions (0.776 vs. 0.517).
Conclusions:
- The BS-RSI3C model successfully differentiates HRBLs from ARBLs in a screening setting.
- The model exhibits superior capability in distinguishing CHRLs from ARBLs compared to ADC alone.
- This DW-MRI model offers a promising tool for improving specificity in breast cancer screening.
