Accuracy of multiparametric MRI in distinguishing the breast malignant lesions from benign lesions: a meta-analysis

Abstract

Background

The sensitivity of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for detecting breast cancer was high and the specificity was relatively low. However, diffusion-weighted imaging (DWI) has a high specificity in the diagnosis of malignant lesions.

Purpose

To evaluate the accuracy of the multiparametric MRI (mp-MRI) in distinguishing the breast malignant lesions from the benign lesions.

Material and Methods

A comprehensive search of the PubMed, Embase, and Cochrane Library electronic databases was conducted up to March 2020. Data were analyzed for the following indexes: pooled sensitivity and specificity; positive likelihood ratio; negative likelihood ratio; diagnostic odds ratio; and the area under the curve.

Results

A total of 2356 patients with 1604 malignant and 967 benign breast lesions were included from 22 studies. The pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, and area under the curve for mp-MRI were 0.93, 0.85, 6.3, 0.08, 81, and 0.96, respectively. The pooled sensitivity, specificity, and area under the curve for DCE-MRI alone were 0.95, 0.71, and 0.92, respectively. The pooled sensitivity, specificity, and area under the curve for DWI alone were 0.88, 0.84, and 0.93, respectively.

Conclusion

The mp-MRI did not improve the sensitivity but increased the specificity for the diagnosis of breast malignant lesions.

Keywords

Breast magnetic resonance imaging,dynamic contrast enhanced MRI diffusion-weighted imaging

Introduction

Breast cancer is the most common tumor suffered by women (1,2). An improvement in the accuracy of imaging diagnosis would help clinicians to make judgments and the patients to get timely treatment. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) has become an important tool for the detection, diagnosis, and stage of breast cancer (2 –5). The morphological and functional features of the lesions could be obtained and analyzed using DCE-MRI (6,7); however, the sensitivity of DCE-MRI for detecting breast cancer was high and the specificity was relatively low (4,6,8 –13). Thus, it may lead to unnecessary biopsy procedures (2).

Several studies have reported that the combination of diffusion-weighted imaging (DWI) and DCE-MRI could improve the specificity (2,4,6,8,10,11,13,14). DWI can reflect the diffusion of water molecules in the intracellular and extracellular spaces and be quantified by the apparent diffusion coefficient (ADC) (2,5,12,15). In some histopathological types of tumors, for example, the mean value of ADC of mucinous cancer may be higher than that of benign lesions for the content of mucin protein, of which the proportion is 2/47 and 1/75, respectively (16,17), but the diffusion of water molecules is limited in most malignant lesions due to the high cell density and the small extracellular space. Therefore, the mean value of ADC of malignant lesions is lower than that of benign lesions (18,19). This indicated that DWI has a high specificity in the diagnosis of malignant lesions. Some authors also demonstrated that the ADC can be used to differentiate malignant breast lesions from benign lesions (20 –24). Meanwhile, the value of ADC could be used as an imaging biomarker for the diagnosis of breast cancer (25 –28). Currently, the combination of DWI and DCE-MRI is defined as multiparametric MRI (mp-MRI) (29).

The aim of the present study was to retrospectively investigate whether the mp-MRI could improve the accuracy of the diagnosis of malignant breast lesions.

Material and Methods

Literature search

Studies that were published up to March 2020 were collected in PubMed, Embase, and the Cochrane Library electronic databases. To avoid the omission of studies, we used subject words plus free words in the search. The search words included breast, dynamic contrast-enhanced, DCE, DWI, diffusion-weighted, ADC, and apparent diffusion coefficient.

Data extraction

The studies that were included in the present study had to meet following criteria: the combination of DWI and DCE-MRI imaging; published in English; data obtained using a 1.5-T or 3.0-T MRI machine; sensitivity and specificity can be obtained directly or indirectly; and the final diagnosis was based on histopathological examination or follow-up. The main extracted data included the publication period, the first author, the type of study, the location of study, the age of the participants, the gold standard of diagnosis, machine type, coil type, lesion size, scan parameters, and true-positive (TP), false-positive (FP), false-negative (FN), and true-negative (TN) findings using DWI or ADC, DCE-MRI, and mp-MRI.

Statistical methods

Stata 15.0 was used to analyze the pooled sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic odds ratio (DOR), and summary receiver operating characteristic (SROC) curve. Quality assessment of study and risk of bias were achieved through Review Manager 5.2. Meta-regression analysis was used to analyze research type, MRI unit, and diagnosis standard by using Meta-Disc. Student’s t-test and analysis of variance (ANOVA) were performed on field strength and the minimum or maximum b-value by using SPSS 23. P values < 0.05 were considered statistically significant.

Results

Study selection and data extraction

The flow chart is shown in Fig. 1. A total of 785 studies were obtained by searching the PubMed, Embase, and Cochrane electronic databases, of which 260 were excluded for duplication and 249 studies were excluded after reading the title and abstract. A further 103 studies were excluded for the study type (not articles), 40 studies were excluded for the language, and 111 studies were excluded for failure to meet the eligibility criteria after carefully reading the full text: failure to extract the number of true positives, false negatives, false negatives, and true negatives; no mp-MRI to diagnose breast cancer; and did not use 1.5-T or 3.0-T MRI machines. Finally, 22 studies were included in the meta-analysis (2 –4,8 –12,14 –17,29 –38).

Fig. 1.

Flow chart of summarizing the selection and exclusion of studies.

The accuracy of mp-MRI in the diagnosis of breast cancer

Table 1 showed the basic information of the studies, including the name of the primary researcher, published year, research type, MRI unit, field strength, diagnosis standard, the number of lesions (malignant/benign).

Table 1.

The key parameters extracted from the studies.

Author name	Publication year	Type of study	Location of study	MRI unit	Field strength (T)	Diagnosis standard	Lesions (malignant/benign) (n)
Yabuuchi	2008	Retrospective	Japan	Philips	1.5	Histopathologic	75 (61/14)
Imamura	2010	Retrospective	Japan	Philips	1.5	Histopathologic/follow-up	27 (16/11)
Kul	2011	Retrospective	Turkey	Siemens	1.5	Histopathologic/follow-up	84 (47/37)
Tan	2014	Prospective	Malaysia	GE	3.0	Histopathologic	44 (31/13)
Pinker	2014	Prospective	Austria	Siemens	3.0	Histopathologic	113 (74/39)
Aribal	2016	Prospective	Turkey	Siemens	3.0	Histopathologic	138 (75/63)
Si	2016	Retrospective	PR China	Siemens	3.0	Histopathologic/follow-up	57 (28/29)
Dijkstra	2016	Prospective	The Netherlands	Siemens	1.5	Histologic/cytologic/follow-up	139 (116/23)
Song	2017	Prospective	Republic of Korea	Philips	3.0	Histopathologic	107 (70/37)
Jiang	2018	Retrospective	PR China	GE	1.5	Histopathologic	275 (171/104)
Pinker	2018	Prospective	Austria	Siemens	3.0	Histopathologic	110 (69/41)
Yildiz	2018	Retrospective	Turkey	Siemens	1.5	Histopathologic	29 (6/23)
Liu	2018	Retrospective	PR China	Siemens	3.0	Histopathologic	178 (90/88)
Kim	2018	Retrospective	Republic of Korea	Philips	3.0	Histopathologic	169 (121/48)
Li	2018	Retrospective	PR China	Philips	3.0	Histopathologic	106 (47/59)
Tezcan	2020	Retrospective	Turkey	Siemens	1.5	Histopathologic	116 (79/37)
Cai	2014	Retrospective	PR China	GE	1.5	Histopathologic/follow-up	234 (149/85)
Kul-system1	2018	Retrospective	Turkey	Siemens	1.5	Histopathologic/follow-up	107 (79/28)
Kul-system2	2018	Retrospective	Turkey	Siemens	1.5	Histopathologic/ follow-up	106 (64/42)
Fan	2018	Retrospective	China	Siemens	3.0	Histopathologic	87 (67/20)
EI Bakey	2015	Prospective	Egypt	Siemens	1.5	Histopathologic	74 (36/38)
Hassan	2013	Prospective	Egypt	Siemens	1.5	Histopathologic	103 (57/46)
Bassiouny	2012	Retrospective	Egypt	Philips	1.5	histopathologic	93 (51/42)

These results including the pooled sensitivity, specificity, PLR, NLR, DOR, and heterogeneity were shown in Table 2. There are strong heterogeneities for sensitivity and specificity among the 22 included results (I² = 0.79, P < 0.001; I² = 0.83, P < 0.001). The pooled sensitivity, specificity, PLR, NLR, and DOR for mp-MRI were 0.93, 0.85, 6.3, 0.08, and 81. The pooled sensitivity, specificity, PLR, NLR, and DOR for DCE-MRI alone were 0.95, 0.71, 3.3, 0.07, and 49. The pooled sensitivity, specificity, PLR, NLR, and DOR for DWI alone were 0.88, 0.84, 5.5, 0.14, and 40. The pooled sensitivity and specificity of mp-MRI for the diagnosis of breast malignant lesions are shown in Fig. 2.

Table 2.

The results of pooled estimates and heterogeneity measures for all studies.

Parameter	Studies (n)	Lesions (n)	Sensitivity	Heterogeneity		Specificity	Heterogeneity		PLR	NLR	DOR
Parameter	Studies (n)	Lesions (n)	Sensitivity	Cochran QP value	I²	Specificity	Cochran QP value	I²	PLR	NLR	DOR
mp-MRI	22	2571	0.93 (0.90–0.96)	<0.001	0.79	0.85 (0.79–0.90)	<0.001	0.83	6.3 (4.5–8.8)	0.08 (0.05–0.12)	81 (53–124)
DCE alone	21	2496	0.95 (0.91–0.98)	<0.001	0.87	0.71 (0.61–0.79)	<0.001	0.88	3.3 (2.4–4.4)	0.07 (0.04–0.12)	49 (25–97)
DWI alone	18	2155	0.88 (0.84–0.91)	<0.001	0.77	0.84 (0.76–0.90)	<0.001	0.88	5.5 (3.7–8.3)	0.14 (0.11–0.19)	40 (24–66)

DCE-MRI, dynamic contrast-enhanced magnetic resonance imaging; DOR, diagnostic odds ratio; DWI, diffusion-weighted imaging; mp-MRI, multiparametric magnetic resonance imaging; NLR, negative likelihood ratio; PLR, positive likelihood ratio.

Fig. 2.

Forest plots of sensitivity and specificity of mp-MRI with 95% confidence intervals. mp-MRI, multiparametric magnetic resonance imaging.

The source of heterogeneity was analyzed by performing meta-regression analysis by removing the covariate with the largest P value one by one. In addition, the MRI unit was found to be related to the heterogeneity (relative diagnostic odds ratios (RDOR) = 1.74, 95% CI = 1.02–3.01, P = 0.042) and no threshold effect was found (R = 0.413, P = 0.05; Table 3)

Table 3.

Meta-regression (inverse variance weights, n = 22).

Variant	Coefficient	Standard error	P value	RDOR	95% CI
Covariate	3.614	0.3667	0.0000	–	–
S	–0.077	0.1290	0.5564	–	–
MRI unit	0.561	0.2596	0.0432	1.75	1.02–3.01
Threshold effect	0.413	0.050	–	–	–

CI, confidence interval; MRI, magnetic resonance imaging; RDOR: relative diagnostic odds ratios.

SROC curves using mp-MRI, DCE alone, and DWI alone are shown in Fig. 3. The pooled area under the curve (AUC) was 0.95, 0.91, and 0.92 for mp-MRI, DCE alone, and DWI alone, respectively.

Fig. 3.

SROC curves by using mp-MRI (a), DCE-MRI alone (b), and DWI alone (c). DCE-MRI, dynamic contrast-enhanced magnetic resonance imaging; DWI, diffusion-weighted imaging; mp-MRI, multiparametric magnetic resonance imaging; SROC, summary receiver operating characteristic.

The Deeks’ funnel chart suggested that there was no publication bias in the included literature (P = 0.92). However, the Egger test indicated that there is a certain degree of publication bias (Fig. 4). Neither field strength nor the minimum or maximum b-values were found to be related to the sensitivity and specificity (Table 4).

Fig. 4.

The funnel plots (a) and Egger test (b). The asymmetry in the funnel plots indicated there was no publication bias (P = 0.92), while the Egger test indicated there may be publication bias.

Table 4.

Results of different sub-analyses.

Parameters	SensitivityP value	SpecificityP value
Field strength	0.085	0.446
Minimum b-value	0.361	0.796
Maximum b-value	0.201	0.212

Discussion

MRI is a non-invasive, non-ionizing radiation and multi-directional imaging technology, and it has been widely used in clinical examination of breast lesions (11). Currently, DCE-MRI is the most sensitive imaging technology, which can evaluate tumor morphology and semi-quantitative enhancement kinetics (7). However, due to the lower specificity of DCE-MRI, DWI has been added to the DCE-MRI to improve its specificity (39). The results of this meta-analysis show that the sensitivity and specificity of mp-MRI are 0.93 and 0.85, and the specificity was higher than that of DCE-MRI alone. In addition, the AUC was 0.96, indicating that the diagnostic efficacy of the mp-MRI was reliable.

The heterogeneity was affected by the threshold effect and the non-threshold effect. Generally, many factors, such as research type, population selection, and different imaging techniques, might lead to heterogeneity by affecting the non-threshold effects. In the present study, the research type, MRI unit, and diagnosis standard were analyzed to decrease the heterogeneity for the non-threshold effect. The MRI unit was found to be related to the heterogeneity. However, due to the heterogeneity being too high, it was still necessary to expand the sample size and optimize the selection of influencing factors to further study the source of heterogeneity.

The main parameters of DCE-MRI include K^trans (volume transfer coefficient), Kep (flux rate constant), Ve (extracellular volume ratio), Vp (plasma fraction), and contrast enhancement kinetic. The contrast enhancement kinetic can be divided into three categories: 1 = persistence type I (persistent enhancement); 2 = plateau type II (initial strong enhancement and plateau phase); 3 = wash-out type III (initial strong enhancement and wash-out) (3,17). Several authors have demonstrated that the number of benign breast lesions among the lesions with type I contrast enhancement kinetic was larger than that among lesions with type II or type III contrast kinetics. It means that the lesions with type I contrast kinetic tend to be benign (36,40 –42). In addition, the number of malignant breast lesions was larger among the lesions with type III contrast enhancement kinetic than that among the lesions with type I or type II contrast enhancement kinetics (36,40 –42). This phenomenon in the tumor tissue mainly resulted from more new blood vessels and a high permeability of blood vessels, leading to contrast agents quickly entering the blood vessel cavity. These abovementioned parameters were the main parameters to distinguish the malignant breast lesions from benign lesions.

The field strength can contribute to the improvement of the quality of the image. 3.0-T MRI can increase the signal-to-noise ratio of the image and improve the temporal and spatial resolution of the image; thus, it may detect smaller lesions (2,43 –45). However, because the size of most selected lesions in the present study was >10 mm, no significant difference was found between 1.5-T MRI and 3.0-T MRI in the accuracy of the diagnosis of breast cancer. A meta-analysis has also shown that 3.0-T MRI does not have an obvious advantage in the diagnostic accuracy for breast cancer when compared with 1.5-T MRI (46).

DWI is a new technology of functional MRI, which can indirectly reflect the characteristics and changes of tissue microstructure. Due to the different structures of different tissue, the directions of diffusion of water molecules in them are also different. These directions were limited by the cell membrane and myelin sheath. Several authors reported that the mean ADC value of the malignant breast lesion was no more than 1.46 × 10⁻³ mm²/s (47). However, there is currently no uniform clinical applicable ADC value to discriminate benign and malignant breast lesions (48). Dijkstra et al. (9) reported the specificity of mp-MRI was only 34.8%. The reason may be that the value of the ADC threshold (2 × 10⁻³ mm²/s) is too large, which would increase the number of false positives, unnecessary biopsies, and even the psychological burden on patients.

At least two b-values are required for calculating the ADC value, so b-values may be used to distinguish the malignant breast lesions from the benign breast lesions. When the b-value is bigger, the ADC value is less affected by blood perfusion and can better reflect the diffuse movement of water molecules in the tissue. Compared to benign lesions, the malignant lesions can be easily detected using this technology, but the quality of the image from the technology will be low (13). Studies have shown that the optimal b-value is 800 s/mm² (13,49). However, Chen et al. (50) revealed that it did not affect the accuracy of diagnosis of benign and malignant breast tumors when the maximum b-value was in the range of 600–1000 s/mm². The previous meta-analysis results showed that the minimum or maximum b-value had no statistical significance for improving the diagnosis of breast lesions (51), and the result of the present study is consistent with them. The maximum b-value was 800, 850, or 1000 s/mm², and the minimum b-value was usually 0 or 50 s/mm² in these included studies.

The present study has several limitations. First, the different scanning parameters (such as section thickness, spatial resolution, scanning field of view, the method of delineating the region of interest, etc.) could have effects on breast cancer images. It was necessary to further optimize the imaging techniques, but there has not been a uniform standard. Second, some physiological activities including breathing motion, beating heart, and blood flow perfusion might decrease the accuracy of diagnosis. Third, the numbers of true positives, false negatives, false positives, and true positives were calculated from the sensitivity and specificity in some literature. Rounding was used when decimals appeared, which may lead to lower accuracy. Fourth, when several parameters were used in DCE-MRI, the parameter with the maximum sum of sensitivity and specificity is selected for analysis in a study, which may decrease the accuracy of data due to inconsistent use of parameters. Fifth, the types, dose, and injection speed of the contrast agents could affect the quality of DCE-MRI, which might affect pooled sensitivity and specificity.

In conclusion, although mp-MRI has comparable sensitivities, it could increase the specificity. Thus, mp-MRI may be more valuable than DWI alone and DCE-MRI alone in distinguishing malignant breast lesions from benign lesions. It is necessary to further explore the value of mp-MRI in breast diseases by increasing the sample size and using the standard parameters in the future.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Bin Wang

References

Hetta

Role of diffusion weighted images combined with breast MRI in improving the detection and differentiation of breast lesions. Egypt J Radiol Nucl Med 2015; 46:259–270.

Tan

Rahmat

Rozalli

, et al. Differentiation between benign and malignant breast lesions using quantitative diffusion-weighted sequence on 3 T MRI. Clin Radiol 2014; 69:63–71.

, et al. Use of diffusion kurtosis imaging and quantitative dynamic contrast-enhanced MRI for the differentiation of breast tumors. J Magn Reson Imaging 2018; 48:1358–1366.

Song

Park

Cho

, et al. Additional value of diffusion-weighted imaging to evaluate multifocal and multicentric breast cancer detected using pre-operative breast MRI. Eur Radiol 2017; 27:4819–4827.

Sharma

Agarwal

Hari

, et al. Role of diffusion weighted imaging and magnetic resonance spectroscopy in breast cancer patients with indeterminate dynamic contrast enhanced magnetic resonance imaging findings. Magn Reson Imaging 2019; 61:66–72.

Pinker

Baltzer

Bogner

, et al. Multiparametric MR imaging with high-resolution dynamic contrast-enhanced and diffusion-weighted imaging at 7 T improves the assessment of breast tumors: a feasibility study. Radiology 2015; 276:360–370.

Leithner

Wengert

Helbich

, et al. Clinical role of breast MRI now and going forward. Clin Radiol 2018; 73:700–714.

Zhai

Liu

, et al. MRI in the differential diagnosis of primary architectural distortion detected by mammography. Diagn Interv Radiol 2016; 22:141–150.

Dijkstra

Dorrius

Wielema

, et al. Quantitative DWI implemented after DCE-MRI yields increased specificity for BI-RADS 3 and 4 breast lesions. J Magn Reson Imaging 2016; 44:1642–1649.

10.

Kim

Kuzmiak

Kim

, et al. Diagnostic usefulness of combination of diffusion-weighted imaging and T2WI, including apparent diffusion coefficient in breast lesions: assessment of histologic grade. Acad Radiol 2018; 25:643–652.

11.

Tezcan S, Ulu Ozturk F, Uslu N, et al. The role of combined diffusion-weighted imaging and dynamic contrast-enhanced MRI for differentiating malignant from benign breast lesions presenting washout curve. Can Assoc Radiol J 2020. DOI: 10.1177/0846537120907098

12.

Cai

Peng

, et al. Diagnosis of breast masses from dynamic contrast-enhanced and diffusion-weighted MR: a machine learning approach. PLoS One 2014; 9:e87387.

13.

Abd El-Aleem

Abo El-Hamd

Yousef

, et al. The added value of qualitative and quantitative diffusion-weighted magnetic resonance imaging (DW-MRI) in differentiating benign from malignant breast lesions. Egypt J Radiol Nucl Med 2018; 49:272–280.

14.

Pinker

Bogner

Baltzer

, et al. Improved diagnostic accuracy with multiparametric magnetic resonance imaging of the breast using dynamic contrast-enhanced magnetic resonance imaging, diffusion-weighted imaging, and 3-dimensional proton magnetic resonance spectroscopic imaging. Invest Radiol 2014; 49:421–430.

15.

Fan

Chen

Cheng

, et al. Retrospective analysis of the utility of multiparametric MRI for differentiating between benign and malignant breast lesions in women in China. Medicine (Baltimore) 2018; 97:e9666.

16.

Kul

Cansu

Alhan

, et al. Contribution of diffusion-weighted imaging to dynamic contrast-enhanced MRI in the characterization of breast tumors. AJR Am J Roentgenol 2011; 196:210–217.

17.

Aribal

Asadov

Ramazan

, et al. Multiparametric breast MRI with 3T: Effectivity of combination of contrast enhanced MRI, DWI and 1H single voxel spectroscopy in differentiation of Breast tumors. Eur J Radiol 2016; 85:979–986.

18.

Partridge

Demartini

Kurland

, et al. Differential diagnosis of mammographically and clinically occult breast lesions on diffusion-weighted MRI. J Magn Reson Imaging 2010; 31:562–570.

19.

Goto

Le Bihan

Yoshida

, et al. Adding a Model-free Diffusion MRI Marker to BI-RADS Assessment Improves Specificity for Diagnosing Breast Lesions. Radiology 2019; 292:84–93.

20.

Zhao

Guan

, et al. Significance of the ADC ratio in the differential diagnosis of breast lesions. Acta Radiol 2016; 57:422–429.

21.

Bozkurt Bostan

Koc

Sezgin

, et al. Value of Apparent Diffusion Coefficient Values in Differentiating Malignant and Benign Breast Lesions. Balkan Med J 2016; 33:294–300.

22.

Jin

Zhang

Cheng

, et al. Monoexponential, Biexponential, and stretched-exponential models using diffusion-weighted imaging: A quantitative differentiation of breast lesions at 3.0T. J Magn Reson Imaging 2019; 50:1461–1467.

23.

Polat

Ozbay

Aydin

, et al. Diagnostic value of apparent diffusion coefficients to differentiate benign and malignant breast lesions. Journal of Experimental and Clinical Medicine (Turkey) 2013; 30:305–310.

24.

Ibrahim

Habib

Deif

Role of quantitative diffusion weighted imaging in characterization of breast masses. Egyptian Journal of Radiology and Nuclear Medicine 2015; 46:805–810.

25.

Zhang

Tang

Min

, et al. Accuracy of combined dynamic contrast-enhanced magnetic resonance imaging and diffusion-weighted imaging for breast cancer detection: a meta-analysis. Acta Radiol 2016; 57:651–660.

26.

Bickel

Helbich

Dubsky

, et al. Apparent diffusion coefficient-a non-invasive biomarker for the invasiveness of breast cancer. Ann Oncol 2013; 24:iii40–iii42.

27.

Bickel

Pinker

Bogner

, et al. Diffusion weighted MRI as a biomarker for breast cancer malignancy. Eur J Cancer 2012; 48:S72.

28.

Martincich

Deantoni

Bertotto

, et al. Correlations between diffusion-weighted imaging and breast cancer biomarkers. Eur Radiol 2012; 22:1519–1528.

29.

Pinker

Moy

Sutton

, et al. Diffusion-weighted imaging with apparent diffusion coefficient mapping for breast cancer detection as a stand-alone parameter: comparison with dynamic contrast-enhanced and multiparametric magnetic resonance imaging. Invest Radiol 2018; 53:587–595.

30.

Yabuuchi

Matsuo

Okafuji

, et al. Enhanced mass on contrast-enhanced breast MR imaging: Lesion characterization using combination of dynamic contrast-enhanced and diffusion-weighted MR images. J Magn Reson Imaging 2008; 28:1157–1165.

31.

Imamura

Isomoto

Sueyoshi

, et al. Diagnostic performance of ADC for non-mass-like breast lesions on MR imaging. Magn Reson Med Sci 2010; 9:217–225.

32.

Jiang

Xie

Liu

, et al. Discrimination of malignant and benign breast masses using automatic segmentation and features extracted from dynamic contrast-enhanced and diffusion-weighted MRI. Oncol Lett 2018; 16:1521–1528.

33.

Yildiz

Toprak

Ersoy

, et al. Contribution of diffusion-weighted imaging to dynamic contrast-enhanced MRI in the characterization of papillary breast lesions. Breast J 2018; 24:176–179.

34.

Liu

Zong

Wei

, et al. Differentiation between malignant and benign breast masses: combination of semi-quantitative analysis on DCE-MRI and histogram analysis of ADC maps. Clin Radiol 2018; 73:460–466.

35.

Kul

Metin

Kul

, et al. Assessment of breast mass morphology with diffusion-weighted MRI: Beyond apparent diffusion coefficient. J Magn Reson Imaging 2018; 48:1668–1677.

36.

El Bakry

MAH

Sultan

El-Tokhy

NAE

, et al. Role of diffusion weighted imaging and dynamic contrast enhanced magnetic resonance imaging in breast tumors. Egypt J Radiol Nucl Med 2015; 46:791–804.

37.

Bassiouny

Youssef

Hassan

Diagnostic performance of breast MRI with and without the addition of quantitative diffusion weighted imaging. Egyptian Journal of Radiology and Nuclear Medicine 2012; 43:311–323.

38.

Hassan

HHM

Mahmoud Zahran

El-Prince Hassan

, et al. Diffusion magnetic resonance imaging of breast lesions: Initial experience at Alexandria University. Alexandria J Med 2013; 49:265–272.

39.

Cipolla

Santucci

Guerrieri

, et al. Correlation between 3T apparent diffusion coefficient values and grading of invasive breast carcinoma. Eur J Radiol 2014; 83:2144–2150.

40.

Teama

Hassanien

Hashish

AAE

, et al. The role of conventional and functional MRI in diagnosis of breast masses. Egyptian Journal of Radiology and Nuclear Medicine 2015; 46:1215–1230.

41.

Youssef

Elahwal

HMS

Alwageeh

, et al. Role of MRI in differentiating benign from malignant breast lesions using dynamic contrast enhanced MRI and diffusion weighted MRI. Alexandria J Med 2018; 54:1–9.

42.

Mohamed

Zytoon

Amin

MA.

Diagnostic interplay of proton magnetic resonance spectroscopy and diffusion weighted images with apparent diffusion coefficient values in suspicious breast lesions. Egypt J Radiol Nucl Med 2018; 49:536–552.

43.

Matsuoka

Minato

Harada

, et al. Comparison of 3.0-and 1.5-tesla diffusion-weighted imaging in the visibility of breast cancer. Radiat Med 2008; 26:15–20.

44.

Park

Cha

Kang

, et al. The role of diffusion-weighted imaging and the apparent diffusion coefficient (ADC) values for breast tumors. Korean J Radiol 2007; 8:390–396.

45.

Woodhams

Ramadan

Stanwell

, et al. Diffusion-weighted imaging of the breast: principles and clinical applications. Radiographics 2011; 31:1059–1084.

46.

Shi

Yao

, et al. Breast Lesions: Diagnosis Using Diffusion Weighted Imaging at 1.5T and 3.0T-Systematic Review and Meta-analysis. Clin Breast Cancer 2018; 18:e305–e320.

47.

Kang

Lipson

Planey

, et al. Rim sign in breast lesions on diffusion-weighted magnetic resonance imaging: diagnostic accuracy and clinical usefulness. J Magn Reson Imaging 2015; 41:616–623.

48.

Surov

Meyer

Wienke

Can apparent diffusion coefficient (ADC) distinguish breast cancer from benign breast findings? A meta-analysis based on 13 847 lesions. BMC Cancer 2019; 19:955.

49.

Min

Shao

Zhai

, et al. Differential diagnosis of benign and malignant breast masses using diffusion-weighted magnetic resonance imaging. World J Surg Oncol 2015; 13:32.

50.

Chen

Jin

, et al. Conspicuity of breast lesions at different b values on diffusion-weighted imaging. BMC Cancer 2012; 12:334.

51.

Baxter

Graves

Gilbert

, et al. A Meta-analysis of the Diagnostic Performance of Diffusion MRI for Breast Lesion Characterization. Radiology 2019; 291:632–641.