Objectives To determine the appropriate parameters and evaluation method for characterizing solitary pulmonary nodules (SPNs) using quantitative parameters of diffusion-weighted imaging (DWI). Methods Thirty-two subjects with 36 SPNs underwent DWI with seven different b values (0, 50, 100, 150, 300, 500, and 1000 s/mm2). Five quantitative parameters were obtained from the region of interest drawn over each SPN: apparent diffusion coefficients (ADCs), true diffusion coefficients (DCs), and perfusion fractions (PFs), and signal-intensity ratios between lesion and spinal cord from DWI (b values: 1000 [LSR1000] and 500 [LSR500)]). All quantitative parameters and the diagnostic capabilities were statistically compared. Results SPNs were diagnosed as follow: malignant (n = 27) and benign (n = 9). Parameter comparisons for malignant and benign showed both LSRs differed significantly (p < 0.05). Applying feasible threshold values showed LSR500 specificity (88.9% [8/9]) and accuracy (77.8% [28/36]) were significantly higher than ADC, DC, and PF specificity and accuracy (p < 0.05). LSR1000 accuracy (72.2% [26/36]) was significantly higher than DC accuracy, and its specificity (88.9% [8/9]) was significantly higher than ADC, DC, and PF specificities (p < 0.05). Conclusions For quantitative differentiation of SPNs, LSR evaluation was more useful and practical than ADC, DC, and PF, and choice of b values showed little impact for the differentiation.
All Science Journal Classification (ASJC) codes
- Radiology Nuclear Medicine and imaging