Breast cancer is the most common cancer globally, surpassing lung cancer, and is the fifth leading cause of cancer death. Incidence rates vary significantly worldwide, with higher rates in developed countries and lower rates in Africa and Asia. Late-stage diagnoses in transitioning countries contribute to low survival rates. In Iran, breast cancer accounts for 12.9% of all cancers, with rising incidence expected to exceed 70 per 100,000 by 2030. Factors like lifestyle changes and lack of awareness contribute to late diagnoses. Effective early detection and screening programs are crucial to reducing the burden of breast cancer in Iran.^1-3

Early diagnosis and treatment are regarded as the most effective strategies for reducing mortality rates. Nevertheless, despite the advancement of various imaging modalities, including mammography and ultrasound, as well as the widespread use of physical examination and palpation in breast cancer diagnosis, differentiation between malignant and benign breast lesions continues to pose difficulties.⁴

Mammography is established as the primary imaging modality for the diagnosis and screening of breast cancer. Nevertheless, its specificity is relatively low, and its diagnostic efficacy is diminished when assessing dense breast tissue.⁵,⁶ Ultrasound imaging was previously utilized to differentiate solid masses from cystic ones. However, with advancements in ultrasound technology, the capability to distinguish between malignant and non-malignant breast lesions has improved. Furthermore, the combination of ultrasound with breast magnetic resonance imaging (MRI) and digital mammography enhances the specificity and sensitivity of imaging.⁷

In breast MRI, both morphological features and mass perfusion are crucial for evaluating breast parenchyma. While routine MRI sequences demonstrate high sensitivity (over 90%) for detecting breast cancer, they often suffer from low specificity. This means that benign and malignant lesions may present similar imaging characteristics, complicating the diagnosis. To enhance specificity, it is essential to consider both morphological and kinetic features during interpretation. As breast MRI technology evolves, its role in diagnosis and management continues to expand, making it an indispensable tool in breast imaging.^8-11

Diffusion-weighted imaging (DWI) enhances MRI specificity by mapping water diffusion in tissues. This advanced, non-invasive technique captures the random motion of water molecules, providing critical insights into tissue microstructure and pathology. DWI is particularly valuable in diagnosing conditions like stroke and tumors, as it reveals changes in cellular environments which standard MRI may overlook. By integrating DWI into MRI protocols, clinicians can improve diagnostic accuracy and better assess disease progression.¹²

Brownian motion, the thermal movement of water molecules, causes diffusion. This movement in tissue is measured by the apparent diffusion coefficient (ADC).¹³,¹⁴

In malignant tumors, the increased proliferation of cells results in a higher cell density, which subsequently limits the diffusion of water molecules, a phenomenon referred to as restricted diffusion. As a result, malignant tumors exhibit a lower apparent diffusion coefficient (ADC) in comparison to benign tumors.^15-18 Combining diffusion with dynamic sequences can significantly enhance the sensitivity and specificity of breast MRI, making it a vital tool in breast cancer detection. This study aims to investigate the role of ADC values in differentiating between benign and malignant breast lesions in a sample of Iranian women. The purpose of this study is to evaluate the diagnostic accuracy of ADC values in breast MRI and to determine the optimal ADC cut-off value for differentiating between benign and malignant lesions. This study contributes to the existing literature on the use of ADC values in breast MRI and provides insights into the potential role of this technique in improving the diagnosis and treatment of breast cancer in Iranian women. Despite the publication of numerous studies evaluating the role of DWI MRI in differentiating breast lesions, including several meta-analyses conducted in China, there remains a gap in research focused on specific populations. The existing literature often aggregates data from diverse ethnic groups, which may not accurately reflect the ADC characteristics pertinent to Iranian women. This population-specific approach is crucial for developing tailored diagnostic protocols that consider regional variations in breast cancer pathology. Additionally, the optimal ADC thresholds for differentiating between malignant and benign breast lesions may vary across different populations and studies.

This article aims to investigate the role of ADC values in differentiating between benign and malignant breast lesions in specific populations, particularly in Middle Eastern countries such as Iran. By analyzing a well-defined cohort and correlating ADC measurements with histopathological findings, this study seeks to establish clinically relevant ADC thresholds that can enhance diagnostic accuracy and improve patient management strategies. Ultimately, our findings may contribute to more effective early detection of breast cancer and reduce the burden of unnecessary invasive procedures in this population.

A total of 93 patients with a mean weight of 70±12 kg and a mean age of 43±11 years who underwent breast MRI were included in this study, providing a comprehensive dataset for analysis. The dataset comprised 34 benign lesions (36.6%) and 59 malignant lesions (63.4%) (Figure 2), allowing for a detailed comparison between the two groups. Specifically, benign cases consisted of fibroadenoma, fibrosis, adenosis, inflammation, ductal papilloma, scar tissue, hyperplasia, and hematoma, representing a range of non-cancerous conditions.

Malignant cases, on the other hand, included ductal carcinoma in situ, invasive lobular carcinoma, invasive ductal carcinoma, and lobular carcinoma in situ, encompassing various forms of cancer. The patients' ages ranged from 20 to 75 years, with weights between 50 and 105 kg, indicating a diverse population sample.

The findings revealed a significant correlation between the minimum ADC value and the lesion type, with a P-value < 0.001 (Table 1), indicating a strong association between these two variables. Notably, a lower minimum ADC value was associated with malignant lesions, suggesting that this metric could be useful for distinguishing between cancerous and non-cancerous growths. The strength of this association highlights the potential value of ADC values in breast lesion diagnosis.

The area under the receiver operating characteristic (ROC) curve, which provides a measure of a diagnostic test's accuracy, showed a trend towards statistical significance, with a P-value of 0.000, indicating a reliable and robust result. The Receiver Operating Characteristic (ROC) curve for minimum ADC had an area of 0.949, indicating a high level of diagnostic accuracy. Furthermore, the cutoff value for differentiating malignant and benign lesions was set at 1.44×10⁻³ mm²/s, providing a clear threshold for diagnosis (Figure 3).

Using this cutoff value, 58 malignant and 34 benign tumors were identified. The proposed scheme demonstrated a true positive rate of 55/58 (94.8%) and a true negative rate of 31/35 (88.6%). The accuracy of the proposed scheme was 92.5%, with 86 out of 93 cases correctly diagnosed. The false positive rate was 4/35 (11.4%), and the false negative rate was 3/58 (5.2%) (Table 2). The sensitivity, specificity, and accuracy were estimated as 93.2%, 91.2%, and 92.5%, respectively.

The results of the proposed scheme are presented in Table 3 and Table 4, providing a clear and concise summary of its performance. The scheme demonstrated high accuracy, sensitivity, and specificity in differentiating between malignant and benign breast lesions, suggesting its potential as a useful diagnostic tool. The scheme's ability to correctly classify a high proportion of cases, while minimizing false positives and false negatives, highlights its overall effectiveness and utility in a clinical setting.

This table summarizes the distribution of benign and malignant masses based on pathological findings and ADC results.

This study investigated the use of apparent diffusion coefficient (ADC) values in breast MRI for differentiating between malignant and benign lesions. The results showed that the minimum ADC value was significantly lower in malignant lesions compared to that in benign lesions (P<0.001), showing that a lower minimum ADC value is indicative of a malignant lesion.

The study also found that the area under the curve (AUC) of the receiver operating characteristic (ROC) analysis for the minimum ADC value was 0.949, indicating a high level of diagnostic accuracy. The selected cutoff value of 1.44×10⁻³ mm²/s showed a sensitivity of 93.2% and specificity of 91.2%, which is comparable to other studies that have reported similar diagnostic accuracy for ADC values in breast MRI.

The results of this study are consistent with a large body of literature emphasizing the diagnostic utility of ADC in breast MRI. A meta-analysis by Cakir et al., which included 13,847 breast lesions, reported that malignant lesions had a mean ADC of 1.03×10⁻³ mm²/s, while benign lesions had a mean of 1.5×10⁻³ mm²/s. This supports our findings and suggests that an ADC threshold of 1.00×10⁻³ mm²/s may be effective in distinguishing between malignant and benign lesions in different populations and imaging protocols.¹⁹

In another systematic review, Dkhar et al. highlighted the correlation between ADC values and various molecular prognostic markers in breast cancer, including estrogen receptor (ER) and progesterone receptor (PgR) status. They found that ER-positive and PgR-positive tumors had significantly lower ADC values than their negative counterparts, suggesting that ADC may reflect tumor biology beyond the simple differentiation between benign and malignant lesions.²⁰ This relationship suggests that ADC values may also serve as a non-invasive biomarker for assessing tumor aggressiveness.

The effect of imaging parameters on ADC is also well documented in the literature. A meta-analysis by Zhang et al. found significant variations in ADC thresholds based on different b-values used during imaging, highlighting the need for standardization in clinical practice.²¹ Their findings suggest that optimal b-values may improve the differentiation capabilities of DWI, thus influencing the diagnostic performance metrics reported in various studies.

Similar to many studies in this field, we used minimum ADC to evaluate its role in distinguishing malignant from benign breast lesions. This study highlighted the importance of DWI and ADC maps as advanced imaging modalities that can be applied to intensify MRI specificity in detecting breast lesions (accuracy and specificity >90%). This finding is similar to that reported by Abdulghaffar and Tag-Aldee.²² In contrast, Cabuk et al. reported a specificity of 85%, an accuracy of 87%, and a sensitivity of 91%, which seems to be related to the number of patients.²³ Min et al., also reported a sensitivity of 82.8% and specificity of 90.0%¹⁴. In addition, Rinaldi et al. reported a sensitivity of 82.8% and 90.0% specificity²⁴. The study's higher sensitivity (93.2%) and specificity (91.2%) compared to previous studies may be due to optimal imaging parameters, careful ROI analysis, or differences in study population and methodology. Additionally, the small sample size may have contributed to the results, but this also raises concerns about generalizability.

We found that malignant tumors on breast MRI were characterized by a low ADC. In contrast, benign lesions are characterized by higher ADC values. In this regard, Wahab et al. showed that ADC is higher in benign lesions than in malignant lesions.⁸ Similarly, Cabuk et al. reported lower mean ADCs for malignant lesions and higher values for benign lesions.²³ Park et al., also showed significantly lower ADCs for invasive ductal carcinoma and DCIS, compared to benign lesions and normal fibro glandular tissues.²⁵

In this study, lesions with ADCs ≤ 1.44×10⁻³ mm²/s were identified as malignant, while those with ADCs above the cut-off value were considered benign. In this regard, Wahab et al. reported a cut-off point of 1.02×10⁻³ mm²/s for differentiation.⁸ Min et al., reported an ADC threshold of 1.23×10⁻³ mm²/s at b-value of 800 s/mm2.¹⁴ In addition, in a study by Mori et al., a minimum ADC below 1.1×10⁻³ mm²/s was indicative of invasive carcinoma in DCIS cases, diagnosed through biopsy.²⁶ Abdulghaffar and Tag-Aldeen reported a cut-off point of 1.25×10⁻³ mm²/s mm2/s.²²

In this study, for practical and clinical applications, the minimum ADC values were obtained directly from ADC images produced by the MRI device. ROI was then plotted on the lesion site by using the minimum ADC value. In total, three out of 58 cases were identified as false positives, while four out of 35 cases were identified as false negatives. The high true positive rate and true negative rate suggest that the proposed scheme is effective in differentiating between malignant and benign breast lesions. However, the small size of some lesions may have affected the ADC measurements, potentially leading to misclassification. Additionally, the presence of cysts containing hemoglobin or fibrosis may also have affected the ADC measurements, which could have contributed to the false positives and false negatives.

The implications of these findings are significant for clinical practice, particularly in regions with high breast cancer prevalence such as Iran. The ability to accurately differentiate between benign and malignant lesions using ADC values can facilitate earlier diagnosis and treatment interventions, potentially reducing unnecessary biopsies and associated patient anxiety. Incorporating DWI into routine clinical practice could improve diagnostic accuracy and help guide management strategies for patients presenting with suspicious breast lesions. The high diagnostic performance metrics observed in this study suggest that radiologists should consider integrating DWI into their assessment protocols to enhance specificity in detecting breast cancer.

In cases where the dynamic breast test shows false positive results, ADC can be used to screen masses and develop useful quantifiable indices. These indices can predict the outcomes and determine the treatment course without the need for invasive biopsy or dynamic MRI (with injection of a contrast medium), which is preferable for screening and diagnosis at earlier ages. The present study showed that ADC can be used as an effective parameter for differentiating benign from malignant lesions. ADC with sensitivity of 93.2% and specificity of 91.2% can be used in the differential diagnosis and the screening of lesions, in addition to routine breast MRI examinations. The study suggests that ADC values can help avoid unnecessary biopsies for suspected breast lesions, rather than serving as a screening tool.