Application of Diffusion Weighted Imaging Techniques for Differentiating Benign and Malignant Breast Lesions.

Muzhen He,Mingping Ma,Huiping Ruan,Zhongshuai Zhang

doi:10.3389/fonc.2021.694634

Muzhen He, Mingping Ma + Show 2 more

Open Access

https://doi.org/10.3389/fonc.2021.694634

Copy DOI

Abstract

To explore the value of apparent diffusion coefficient (ADC), intravoxel incoherent motion (IVIM), and diffusional kurtosis imaging (DKI) based on diffusion weighted magnetic resonance imaging (DW-MRI) in differentiating benign and malignant breast lesions. A total of 215 patients with breast lesions were prospectively collected for breast MR examination. Single exponential, IVIM, and DKI models were calculated using a series of b values. Parameters including ADC, perfusion fraction (f), tissue diffusion coefficient (D), perfusion-related incoherent microcirculation (D*), average kurtosis (MK), and average diffusivity (MD) were compared between benign and malignant lesions. ROC curves were used to analyze the optimal diagnostic threshold of each parameter, and to evaluate the diagnostic efficacy of single and combined parameters. ADC, D, MK, and MD values were significantly different between benign and malignant breast lesions (P<0.001). Among the single parameters, ADC had the highest diagnostic efficiency (sensitivity 91.45%, specificity 82.54%, accuracy 88.84%, AUC 0.915) and the best diagnostic threshold (0.983 μm2/ms). The combination of ADC and MK offered high diagnostic performance (sensitivity 90.79%, specificity 85.71%, accuracy 89.30%, AUC 0.923), but no statistically significant difference in diagnostic performance as compared with single-parameter ADC (P=0.268). The ADC, D, MK, and MD parameters have high diagnostic value in differentiating benign and malignant breast lesions, and of these individual parameters the ADC has the best diagnostic performance. Therefore, our study revealed that the use of ADC alone should be useful for differentiating between benign and malignant breast lesions, whereas the combination of MK and ADC might improve the diagnostic performance to some extent.

Highlights

Breast cancer is the most common cancer in women
apparent diffusion coefficient (ADC), D, MK, and mean diffusivity (MD) values were statistically significant in the identification of benign and malignant breast lesions (P0.05) (Table 1)
We find that the ADC, D, MK and MD values of different DWI techniques have high diagnostic value in differentiating benign and malignant breast lesions

Summary

Introduction

Breast cancer is the most common cancer in women. In 2018 there were approximately 2.1 million newly diagnosed female breast cancer cases worldwide, accounting for a quarter of female cancer cases [1]. In China the incidence of breast cancer is relatively low, but since 1990 the incidence of breast cancer has increased rapidly, especially in urban areas [2, 3]. Survival rates for breast cancer are greatly improved by early diagnosis. The main techniques used for identification of breast lesions are ultrasound, mammography, and magnetic resonance imaging (MRI). MRI offers better sensitivity and specificity than mammography and ultrasonography, especially for lesions in dense breasts [4, 5]

Methods

Results

Conclusion