Enhancing diagnostic accuracy of BI-RADS 4 breast lesions: integrating ADC and microcalcifications into the kaiser score

Abstract Background Breast cancer is considered the most commonly diagnosed cancer in the world and is responsible for a high rate of deaths among women. The malignancy risk dramatically increases in Breast Imaging Reporting and Data System (BI-RADS) 4 and 5 lesions. Therefore, this study aimed to evaluate the reliability of adding the apparent diffusion coefficient (ADC) and suspicious microcalcifications (when present) in combination with the Kaiser Score (KS) in improving the accuracy of the evaluation of magnetic resonance imaging (MRI) BI-RADS 4 lesions. Methods A total of 115 patients with 122 breast lesions categorized as BI-RADS 4 on MRI were included in the study. All patients had an MRI and a mammogram. Two observers analyzed images and calculated ADC, KS, KS1, KS2, and KS3. The diagnostic performance was calculated using receiver operating characteristic (ROC) analysis as well as interobserver agreement. Results This study involved 122 breast lesions (mean age: 48.1 years ± 10.3). The sensitivity for KS, KS1, KS2, and KS3 ranged from 85% to 91.84%, with area under the curve (AUC) values of 0.907, 0.916, 0.915, and 0.913, and accuracy rates of 90.98%, 86.89%, 91.8%, and 90.16%, respectively, for the first observer, denoting high and significant sensitivity (P < 0.001). Interobserver agreement was substantial (0.614 and 0.785) for ADC and KS, and perfect (0.822, 0.820, 0.817, and 0.852) for microcalcifications, KS1, KS2, and KS3, respectively, with the highest value for KS3 (0.85). The intraclass correlation coefficient (ICC) was moderate (0.644) for ADC and good (0.830, 0.822, 0.807, and 0.812) for KS, KS1, KS2, and KS3, respectively. Conclusion This study highlighted the value of the KS as a structured diagnostic tool in evaluating BI-RADS 4 breast lesions, particularly when combined with ADC and microcalcifications. KS3, which integrates all three parameters, provided the highest sensitivity and interobserver agreement. The findings uniquely demonstrate that microcalcifications contributed more to sensitivity than ADC when added to the KS framework, while adding ADC improved specificity and accuracy. These results support the use of multiparametric composite scoring to enhance MRI interpretation, reduce unnecessary biopsies, and improve diagnostic confidence in daily practice.