Usefulness of diffusion-weighted imaging of breast tumors: quantitative and visual assessment

Abstract

We evaluated the usefulness of quantitative and visual assessment of diffusion-weighted imaging (DWI) of breast tumors to distinguish malignant from benign tumors. The DWI findings of 106 breast lesions (15 benign, 91 malignant) were retrospectively analyzed. The mean apparent diffusion coefficient (ADC) value for each lesion was calculated using b values of 250, 500, 750, and 1000 s/mm(2) as a quantitative assessment. We visually evaluated the signal intensity of each breast lesion on the basis of a spinal signal intensity in DWI (b = 1000 s/mm(2)) and compared the mean ADC values using a threshold mean ADC +1.65 × standard deviation (SD) for malignant and benign breast lesions. Obviously strong signal intensity of the lesion relative to that of the spinal cord on DWI signifies malignancy. The mean ADC value for benign lesions (1.50 ± 0.38 × 10(-3) mm(2)/s) was significantly higher than that for malignant lesions (0.98 ± 0.19 × 10(-3) mm(2)/s), with 94.5% sensitivity, 80% specificity, and 92.5% accuracy. Sensitivity for visual assessment was 91.5%, specificity was 33.3%, and total accuracy was 82.5%. ADC values, but not visual assessment, may be useful for differentiating benign and malignant breast tumors.