Breast cancer is the most common malignancy among women worldwide. In India, about 1,44,000 new cases occur annually, with nearly one in twenty-eight women affected. Early detection is essential, but conventional imaging relies mainly on morphological features that may overlap between benign and malignant lesions.1 The BI-RADS system standardizes reporting; however, BI-RADS III and IV lesions remain diagnostic gray zones that may lead to unnecessary biopsies or delayed diagnosis.2 Ultrasonography is widely used, particularly in women with dense breasts, but has limited specificity due to overlapping benign and malignant features.3 Ultrasound elastography evaluates tissue stiffness, as malignant lesions are typically harder than benign ones, thereby improving diagnostic specificity.4 However, its results may be affected by technical and lesion-related factors and should be interpreted with other imaging findings.5

Dynamic contrast-enhanced MRI (DCE-MRI) detects breast cancer with high sensitivity by assessing contrast enhancement patterns related to tumor vascularity; malignant lesions typically show rapid enhancement and washout.6 However, benign lesions may produce similar patterns, and MRI is costly with limited availability.7

The BI-RADS system standardizes breast imaging reporting, with categories 0- VI reflecting increasing malignancy risk. BI-RADS III and IV remain diagnostic gray zones, often leading to follow-up or biopsy despite many benign outcomes. Because assessment relies mainly on morphology, functional imaging techniques such as ultrasound elastography and dynamic contrast enhanced MRI help to improve diagnostic accuracy.8

Histopathology is the gold standard for diagnosing breast malignancy, and imaging modalities are evaluated by their ability to predict histological outcomes using measures such as sensitivity and specificity.9 Comparative studies within the same population provide reliable evidence for clinical decision-making.10 The present study compares the diagnostic accuracy of ultrasound elastography and dynamic contrast-enhanced MRI in predicting malignancy in BI-RADS III and above breast lesions, using histopathology as the reference standard.

This observational study was conducted in the Postgraduate Department of Radiodiagnosis, Government Medical College, Srinagar, over a period of 18 months. All enrolled patients underwent detailed clinical history taking, complete clinical examination, baseline investigations, and imaging evaluation. All data were recorded in a structured proforma. The study population consisted of female patients aged 25 years and above presenting with breast masses. Patients were selected from those attending the Outpatient Department (OPD) and Inpatient Department (IPD) from 2024 to 2026. The study was conducted after obtaining approval from the Institutional Ethics Committee of Government Medical College, Srinagar. Written informed consent was obtained from all participants prior to imaging procedures. Sample Size: All eligible patients meeting the inclusion criteria during the study period were consecutively included. Inclusion Criteria 1. Female patients aged 25 years and above presenting with breast masses. 2. Lesions categorized as BI-RADS III and above on conventional B-mode ultrasonography. 3. Breast masses measuring more than 5 mm in size, as elastography yields reliable results in lesions larger than 5 mm. 4. Patients who had been prescribed MR mammography. Exclusion Criteria 1. Lesions categorized as BI-RADS I and II on initial assessment. 2. Lesions in postoperative breasts, as fibrous postoperative changes may produce false-positive high elastography scores. 3. Cases in which MRI was contraindicated. 4. Patients with allergy to gadolinium-based contrast agents. 5. Patients with deranged renal function tests precluding contrast administration. Methodology: After obtaining Institutional Ethical Committee clearance, written informed consent was obtained from all participants prior to imaging. Ultrasound examination: After clinical examination and local palpation of the breast mass, real-time conventional B-mode ultrasonography was performed using a GE Healthcare Logiq P9 scanner with a wideband linear array probe of 7.5 MHz frequency (5-13 MHz range), with a footprint of 12.7 x 47.1 mm. The patient was positioned supine. Conventional B-mode ultrasound was performed using the grid technique, typically in the radial plane. Only lesions categorized as BI-RADS III and above were included in the study. Sonoelastography: After localization of the lesion on B-mode imaging, strain elastography was performed immediately in the same sitting using the same linear array transducer. Tsukuba Elasticity Scoring: Tsukuba elasticity scoring (scores 1-5) was applied to all lesions based on visual color coding patterns. Elastography findings were later correlated with histopathological results. Figure 1: Tsukuba Elasticity Scoring MRI Mammography: In all cases with suspicious masses, MRI mammography was performed using a 3 Tesla Siemens MRI scanner with a dedicated breast coil. After obtaining plain images, gadolinium contrast was administered intravenously at a dose of 0.1 mmol/kg at a rate of 2 ml/sec, followed by 20 ml saline flush. Subtraction images were generated by subtracting pre-contrast raw datasets from each post-contrast dataset. Time-intensity curves were obtained using the Functool application. Time-Intensity Curve Analysis: ROI was placed on the most enhancing part of the lesion. ROI size was preferably more than 3 pixels. Curves were classified as: Type I - Progressive persistent enhancement Type II - Initial enhancement followed by plateau Type III - Initial enhancement followed by washout MRI findings were correlated with histopathological diagnosis. Statistical analysis: Collected data were analyzed using appropriate statistical methods. Diagnostic accuracy parameters including sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy were calculated for ultrasound elastography and MRI mammography. Correlation between imaging findings and histopathology was assessed. Comparative analysis between modalities was performed to evaluate relative diagnostic performance. Statistical significance was determined using relevant tests, with p-values less than 0.05 considered significant.

The study included 50 participants, with the highest proportion belonging to the 31–40 years (32.0%) and 41–50 years (30.0%) age groups. Right-sided lesions were observed in 50.0% of cases, followed by left-sided lesions in 48.0%, while bilateral involvement was uncommon (2.0%). BI-RADS category 3 was the most frequent imaging category (48.0%), followed by category 5 (22.0%) and category 4 (unspecified) (16.0%). On elastography assessment, Tsukuba score 3 was the most common finding (34.0%), followed by score 4 (32.0%) and score 2 (24.0%). Dynamic contrast-enhanced MRI demonstrated a predominance of Type III enhancement curves (38.0%), followed by Type II (34.0%) and Type I (28.0%) curves. These findings indicate that most lesions were classified as BI-RADS 3, exhibited intermediate elastography scores, and showed Type III kinetic enhancement patterns (table 1).

Table 1: Distribution of Age Group, Side Involvement, BI-RADS Category, Tsukuba Score, and Curve Type (n = 50)

Histopathological examination revealed that malignant lesions constituted a slight majority of cases (52.0%), while benign lesions accounted for 48.0%. Among benign lesions, fibroadenoma was the most common diagnosis, observed in 26.0% of patients, followed by phyllodes tumor (6.0%). Among malignant lesions, invasive ductal carcinoma (including variants) was the predominant diagnosis, accounting for 22.0% of cases, followed by ductal carcinoma in situ (DCIS) (14.0%). Other malignant subtypes, including invasive lobular, inflammatory, and mucinous carcinomas, each comprised 4.0% of cases, whereas medullary carcinoma was identified in 2.0% of patients (table 2).

Table 2: Histopathological Diagnosis (n=50)

A statistically significant association was observed between Tsukuba score and histopathological diagnosis (p < 0.001). Lower Tsukuba scores were predominantly associated with benign lesions, with 11 of 12 lesions (91.7%) having a score of 2 being benign. In contrast, higher Tsukuba scores were more frequently associated with malignant lesions, with 13 of 16 lesions (81.3%) having a score of 4 and all 5 lesions with a score of 5 being malignant. These findings suggest that increasing Tsukuba scores are significantly correlated with a higher likelihood of malignancy (table 3).

Table 3: Association of Tsukuba Score with Histopathology

A statistically significant association was observed between RI curve type and histopathological diagnosis (p < 0.001). Type I curves were predominantly associated with benign lesions, accounting for 12 of 14 cases (85.7%). Type II curves showed a relatively balanced distribution between benign and malignant lesions. In contrast, Type III curves were strongly associated with malignancy, with 16 of 19 lesions (84.2%) classified as malignant. These findings indicate that increasing enhancement curve aggressiveness is significantly associated with a higher probability of malignant breast lesions (table 4).

Table 4: Association of MRI Curve with Histopathology

A statistically significant association was observed between BI-RADS category and histopathological diagnosis (p < 0.001). Most BI-RADS 3 lesions were benign, with 20 of 24 cases (83.3%) confirmed as benign on histopathology. In contrast, BI-RADS 5 and BI-RADS 6 lesions were exclusively malignant, accounting for 11 and 1 cases, respectively. BI-RADS 4 lesions demonstrated an intermediate risk profile, with malignant lesions (71.4%) outnumbering benign lesions (28.6%). These findings indicate that higher BI-RADS categories are significantly associated with an increased likelihood of malignancy and show good concordance with histopathological diagnosis (table 5).

Table 5: Association of BI-RADS with Histopathology

BI-RADS ≥4 showed the highest overall diagnostic accuracy (84.0%) and sensitivity (84.6%), indicating superior ability to identify malignant lesions. Elastography using a Tsukuba score ≥4 exhibited the highest specificity (87.5%), comparable to MRI Type III curves, with an overall accuracy of 78.0%. MRI Type III enhancement curves also demonstrated high specificity (87.5%) but lower sensitivity (61.5%) and accuracy (74.0%). Overall, BI-RADS assessment provided the best balance of sensitivity, specificity, and diagnostic accuracy among the evaluated modalities (table 6).

Table 6: Diagnostic Performance of BI-RADS, Elastography, and MRI Curve Type for Detection of Malignancy (n = 50)

The significance of this study lies in addressing a critical clinical challenge in breast imaging accurate differentiation between benign and malignant lesions in indeterminate or suspicious categories. BI-RADS III and IV lesions often pose diagnostic uncertainty, leading to either unnecessary biopsies or delayed diagnosis. By evaluating the complementary roles of elastography, which assesses tissue stiffness, and DCE-MRI, which evaluates vascular perfusion and enhancement kinetics, this study contributes to improving diagnostic precision. The findings support a multiparametric imaging approach that enhances specificity while maintaining acceptable sensitivity, thereby potentially reducing unnecessary invasive procedures without compromising early cancer detection. In an era emphasizing precision medicine and evidence-based practice, the study reinforces the importance of integrating morphological and functional imaging biomarkers to optimize patient management pathways in breast cancer diagnostics. The present study found that suspicious breast masses were most common in women aged 31–40 years (32%) and 41–50 years (30%). Histopathology revealed 26 malignant (52%) and 24 benign (48%) lesions. Although malignancy was numerically higher in older age groups, no significant association was observed between age and malignancy (p = 0.31), indicating that age alone was not an independent predictor of malignant pathology within a pre-selected BI-RADS >III cohort. These findings are consistent with previous studies. Kapetas et al11 reported increasing malignancy prevalence with age but showed that diagnostic performance improved significantly after incorporating elastography and contrast-enhanced ultrasound parameters. Milon et al12 similarly observed higher malignancy rates in older women but found enhancement kinetics and early time-to-enhancement to be stronger predictors than age. Arian et al13, in a meta-analysis of 537 patients and 707 lesions, reported a pooled DCE-MRI sensitivity of 93.8% without identifying age as an independent discriminative factor. Reghunath et al14 also demonstrated that elastographic and Doppler parameters improved diagnostic confidence irrespective of age. Collectively, these findings support the concept that imaging biomarkers are more reliable predictors of malignancy than patient demographics. BI-RADS assessment showed 48% BI-RADS 3 lesions, 28% BI-RADS 4 variants, 22% BI-RADS 5, and 2% BI-RADS 6. Malignancy rates increased progressively across categories, with 16.7% malignancy in BI-RADS 3, 71.4% in BI-RADS 4, and 100% in BI-RADS 5 and 6 lesions (χ² = 28.9, p < 0.001). Using BI-RADS ≥4 as the threshold yielded sensitivity, specificity, and accuracy of 84.6%, 83.3%, and 84%, respectively. Similar findings have been reported by Ebner et al15, who demonstrated a significant increase in strain index from BI-RADS 2 to BI-RADS 5 (p < 0.0001), and by Ochoa-Albíztegui et al16, who improved diagnostic accuracy from 88.4% to 97.7% using pharmacokinetic modelling. Khattab et al11 reported a DCE-MRI accuracy of 91.8%, further validating BI-RADS-based risk stratification. Tsukuba elastography analysis demonstrated increasing malignancy rates with rising stiffness scores. Most score 2 lesions were benign (11/12), whereas all score 5 lesions were malignant (5/5). Itoh et al.71, who introduced the Tsukuba scoring system, reported sensitivity, specificity, and accuracy of 86.5%, 89.8%, and 88.3%, respectively. Similar diagnostic performance has been reported by Zhi et al17, Kumm et al18, Chang et al19, and Ebner et al15, all confirming the strong association between tissue stiffness and malignancy. DCE-MRI analysis revealed Type III washout curves as the most frequent pattern (38%), followed by Type II plateau (34%) and Type I persistent curves (28%). Type I curves were predominantly benign (85.7%), whereas Type III curves were predominantly malignant (84.2%) (χ² = 16.8, p < 0.001). These findings are consistent with Kuhl et al20, who established Type III washout kinetics as a strong indicator of malignancy. Milon et al12 demonstrated improved diagnostic performance with early time-to-enhancement parameters, while Abd Elsalam et al21 and Arian et al13 reported high sensitivity and specificity for DCE-MRI, confirming the value of enhancement kinetics in lesion characterization. Histopathology showed a balanced distribution of benign and malignant lesions. Fibroadenoma was the most common benign lesion (26%), while invasive ductal carcinoma (IDC) was the predominant malignancy (22%), followed by DCIS (14%). Similar pathological distributions have been reported by Kapetas et al11 and Milon et al12, both identifying IDC as the commonest malignant subtype. Hu et al22 demonstrated the complementary role of elastography and DCE-MRI in differentiating benign lesions from invasive carcinoma, while Ochoa-Albíztegui et al16 showed that pharmacokinetic MRI parameters effectively distinguish benign from malignant lesions, although they are less useful for predicting molecular subtypes. A highly significant association was observed between Tsukuba score and histopathology (x² = 23.5, p < 0.001). Score 2 lesions were predominantly benign (11/12; 91.7%), while score 4 lesions showed 81.3% malignancy (13/16), and all score 5 lesions (5/5; 100%) were malignant. Using Tsukuba score ≥4 as cutoff, elastography demonstrated sensitivity 69.2%, specificity 87.5%, and accuracy 78%. Itoh et al23 reported sensitivity 86.5%, specificity 89.8%, and accuracy 88.3% for elasticity scores 4 and 5 predicting malignancy, which is comparable to our specificity but higher sensitivity. Zhi et al17 showed improved specificity (78.5%) when elastography was added to B-mode ultrasound. Kumm et al18 reported sensitivity 93% and specificity 80%. Chang et al19 demonstrated AUC values of 0.928-0.943, confirming high diagnostic reliability. Ebner et al15 observed significant strain index differences between benign (mean SI 5.29) and malignant (mean SI 16.13) lesions (p < 0.0001), similar to our progressive stiffness-malignancy correlation. MRI curve pattern demonstrated significant association with histopathology (x² = 16.8, p < 0.001). Type I curves were predominantly benign (85.7%), Type II curves were intermediate (8 malignant, 9 benign), and Type III curves were predominantly malignant (84.2%). Using Type III as malignancy predictor, sensitivity was 61.5%, specificity 87.5%, and accuracy 74%. Kuhl et al20 reported sensitivity exceeding 90% for washout kinetics, while Abd Elsalam et al21 reported sensitivity 90% and specificity 80% for DCEMRI. Arian et al13 demonstrated pooled sensitivity 93.8% and specificity 68.1% for DCE-MRI, improving further when combined with diffusion parameters. Milon et al12 showed improved AUC (0.910) when early enhancement was incorporated, emphasizing that multiparametric evaluation improves sensitivity. Our moderate sensitivity suggests that relying solely on washout kinetics may miss some malignant lesions, particularly low-grade or in situ carcinomas. In the present study, BI-RADS categorization showed a highly significant association with histopathology (x² = 28.9, df = 3, p < 0.001). Among BIRADS 3 lesions (n=24), 20 (83.3%) were benign and 4 (16.7%) were malignant. In BI-RADS 4 (all variants combined; n=14), 10 lesions (71.4%) were malignant. All BI-RADS 5 (n=11) and BI-RADS 6 (n=1) lesions were malignant (100%). Using BI-RADS ≥4 as the diagnostic threshold, sensitivity was 84.6%, specificity 83.3%, and overall accuracy 84%. These findings reaffirm the established predictive reliability of the BI-RADS classification system. Ebner et al15 demonstrated a progressive increase in strain index values corresponding to increasing BI-RADS categories, with significant differentiation between benign and malignant groups (p < 0.0001). Similarly, Ochoa-Albíztegui et al16 showed that integrating pharmacokinetic MRI parameters with BI-RADS improved diagnostic accuracy from 88.4% to 97.7%, emphasizing the strong foundation provided by BI-RADS categorization. Khattab et al24 reported pre-biopsy DCE-MRI diagnostic accuracy of 91.8%, noting that accurate BI-RADS categorization prior to tissue sampling significantly improves reliability. The slightly lower accuracy (84%) observed in our study may be attributable to the relatively small cohort (n=50) and inclusion of borderline or atypical lesions within BI-RADS 3 and 4 categories. The 16.7% malignancy rate within BI-RADS 3 lesions in our study is slightly higher than the traditionally accepted <2% risk. This may reflect referral bias in a tertiary imaging center where suspicious lesions are more likely to undergo biopsy. Similar observations were reported by Reghunath et al25, who demonstrated that combining elastography and Doppler allowed reclassification of certain BI-RADS 3 lesions, improving diagnostic accuracy to 92%. In our study, BI-RADS ≥4 demonstrated the highest overall diagnostic accuracy (84%) with sensitivity 84.6% and specificity 83.3%. Elastography (Tsukuba ≥4) showed sensitivity 69.2%, specificity 87.5%, and accuracy 78%. MRI Type III kinetics demonstrated sensitivity 61.5%, specificity 87.5%, and accuracy 74%. All modalities showed statistically significant association with malignancy (p<0.01). These findings highlight the balanced performance of BI-RADS classification and the high specificity of elastography and DCE-MRI. Itoh et al23 reported sensitivity 86.5%, specificity 89.8%, and accuracy 88.3% for elastography, closely comparable to our specificity results. Kumm et al18 reported accuracy 87%, while Chang et al.74 demonstrated AUC values above 0.92 for elastographic techniques. Kuhl et al20 reported sensitivity exceeding 90% for DCE-MRI washout kinetics, whereas Abd Elsalam et al21 demonstrated sensitivity 90% and specificity 80%. Arian et al13 reported pooled sensitivity 93.8% and AUC 0.94 when combining IVIM and DCE parameters, confirming that multiparametric approaches outperform single modalities. Our relatively lower sensitivity for MRI Type III pattern (61.5%) likely reflects reliance solely on kinetic curve type without incorporating diffusion or pharmacokinetic modeling. Literature strongly suggests that combining morphological, diffusion, and perfusion parameters enhances diagnostic performance. Overall, our results confirm that elastography and DCE-MRI provide high specificity (87.5%) in predicting malignancy, while BI-RADS categorization offers balanced sensitivity and specificity. The findings support a multiparametric imaging approach for optimal characterization of suspicious breast masses. Limitations and future aspects: The sample size was relatively small (n=50), which may limit generalizability and reduce statistical power in subgroup analyses. The study was conducted in a single center, potentially introducing selection bias and limiting external validity. MRI analysis relied primarily on kinetic curve patterns without incorporating advanced quantitative pharmacokinetic modeling or diffusion parameters, which may have improved sensitivity. Operator dependency in elastography assessment may also have influenced results, as strain elastography can vary with applied compression and technical factors. The moderate sensitivity observed for MRI Type III pattern suggests that reliance on single kinetic criteria may miss certain malignancies, particularly low-grade or in situ lesions. Additionally, long-term follow-up data were not included to assess interval cancer detection or recurrence patterns.

The study supports a multiparametric imaging approach combining morphological assessment, elastographic stiffness evaluation, and contrast enhancement kinetics for optimal lesion characterization. Such integration enhances specificity, improves diagnostic confidence, and may contribute to reduction in unnecessary invasive procedures without compromising cancer detection. In clinical practice, BI-RADS classification should continue to serve as the primary framework, while elastography and DCE-MRI act as powerful adjuncts in equivocal cases. The combined application of these modalities strengthens evidence-based breast imaging and contributes to more precise and individualized patient management.

1. Badwe RA, Gupta S. Breast cancer: an Indian perspective. Natl Med J India. 2011 Jul-Aug;24(4):193-7. PMID: 22208136.
2. Liberman L, Menell JH. Breast imaging reporting and data system (BIRADS). Radiologic Clinics. 2002 May 1;40(3):409-30.
3. Checka CM, Chun JE, Schnabel FR, Lee J, Toth H. The relationship of mammographic density and age: implications for breast cancer screening. American Journal of Roentgenology. 2012 Mar;198 (3):W292-5.
4. D'Orsi C, Bassett L, Feig S. Breast imaging reporting and data system (BI-RADS). Oxford University Press, New York. 2018 Jan 26.
5. Lee CH. Problem solving MR imaging of the breast. Radiologic Clinics. 2004 Sep 1;42(5):919-34.
6. Heywang-Köbrunner SH, Hacker A, Sedlacek S. Magnetic resonance imaging: the evolution of breast imaging. The Breast. 2013 Aug 1;22 :S77-82.
7. Obenauer S, Hermann KP, Grabbe E. Applications and literature review of the BI-RADS classification. European radiology. 2005 May;15(5):1027-36.
8. Bonacho TR, Rodrigues F, Liberal J. Immunohistochemistry for diagnosis and prognosis of breast cancer: a review. Biotechnic & Histochemistry. 2020 Feb 17;95(2):71-91.
9. Wen P, Liu L, Pan L, Li X. Evaluating diagnostic significance: The utilization of elastography and contrast-enhanced ultrasound for differential diagnosis in breast lesions. Clinical Hemorheology and Microcirculation. 2024 Sep 5;88(1):81-95.
10. ElSaid NA, Mohamed HG. Sonoelastography versus dynamic magnetic resonance imaging in evaluating BI-RADS III and IV breast masses. The Egyptian Journal of Radiology and Nuclear Medicine. 2012 Jun 1;43(2):293-300.
11. Kapetas P, Clauser P, Woitek R, Wengert G, Lazar M, Pinker K, et al. Quantitative multiparametric breast ultrasound: Application of contrastenhanced ultrasound and elastography leads to an improved differentiation of benign and malignant lesions. Invest Radiol. 2019;54(4):257-264.
12. Milon A, Vande Perre S, Poujol J, Trop I, Kermarrec E, Bekhouche A, et al. Abbreviated breast MRI combining FAST protocol and high temporal resolution (HTR) dynamic contrast enhanced (DCE) sequence. Eur J Radiol. 2019;117:199-208.
13. Arian A, Seyed-Kolbadi FZ, Yaghoobpoor S, Ghorani
14. H, Saghazadeh A, Ghadimi DJ. Diagnostic accuracy of intravoxel incoherent motion (IVIM) and dynamic contrast-enhanced (DCE) MRI to differentiate benign from malignant breast lesions: a systematic review and metaanalysis. Eur J Radiol. 2023;167:111051.
15. Reghunath A, Mittal M, Chintamani C, Prasad R. Novel approach in the evaluation of ultrasound BI-RADS 3 & 4 breast masses with a combination method of elastography and Doppler. Indian J Med Res. 2021;154(3):355-366.
16. Ebner L, Bonel H, Huber A, Ross S, Christe A. Diagnostic performance and additional value of elastosonography in focal breast lesions: Statistical correlation between size-dependant strain index measurements, multimodality-BI-RADS score, and histopathology in a clinical routine setting. ISRN Radiology. 2014;2014:396368.
17. Ochoa-Albíztegui R, Sevilimedu V, Horvat J, Thakur S, Helbich T, Trattnig S, et al. Pharmacokinetic analysis of dynamic contrast-enhanced magnetic resonance imaging at 7T for breast cancer diagnosis and characterization. Cancers (Basel). 2020;12(12):3763.
18. Zhi H, Ou B, Luo BM, Feng X, Wen YL, Yang HY. J Ultrasound Med. 2007;26(6):807-15.
19. Kumm TR, Szabunio MM. Cancer Control. 2010;17(3):156-61.
20. Chang JM, Won JK, Lee KB, Park IA, Yi A, Moon WK. AJR Am Roentgenol. 2013;201(2):W347-56.
21. Kuhl CK, Mielcareck P, Klaschik S, Leutner C, Wardelmann E, Gieseke J, et al. 1999;211(1):101-10.
22. Abd Elsalam SM, Said A, Sarah M. Multiparametric approach with dynamic contrast-enhanced MRI, diffusion-weighted imaging and MRS of the breast in assessment of suspicious breast lesions. Tumori J. 2020;106(1):16-22.
23. Hu C-C, Chang T-H, Hsu H, Pen Y-J, Yu J-C. T1-weighted dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) to distinguish between concurrent cholesterol granuloma and invasive ductal carcinoma of the breast: a case report. Am J Case Rep. 2018;19:593-598.
24. Itoh A, Ueno E, Tohno E, Kamma H, Takahashi H, Shiina T, et al. 2006;239(2):341-50.
25. Khattab EM, Sobhy T, El-Shewail AM. Pitfalls of dynamic contrast enhanced MR mammography (DCE-MRM) in evaluation of post-biopsy suspicious breast lesions. Med J Cairo Univ. 2018;86:1513-1522.
26. Reghunath A, Mittal M, Chintamani C, Prasad R. Novel approach in the evaluation of ultrasound BI-RADS 3 & 4 breast masses with a combination method of elastography and Doppler. Indian J Med Res. 2021;154(3):355-366