Diffusion-weighted magnetic resonance imaging for characterization of focal liver masses: impact of parallel imaging (SENSE) and b value.

2008 
Purpose: To evaluate the impact of parallel imaging (sensitivity encoding [SENSE] technique) on diffusion-weighted (DW) magnetic resonance imaging, compare DW imaging techniques with 2 different b values for characterization of focal hepatic lesions, and determine apparent diffusion coefficient cutoff values. Materials and Methods: Seventy-eight patients with 86 lesions were examined with 4 different DW techniques with 2 different b values (400 and 1000 s/mm 2 ) and with/without the use of SENSE. The differences in signal-noise ratio values and image quality between DW images obtained with different techniques were compared using repeated-measures analysis of variance and Friedman test, respectively. A receiver operating characteristic analysis was applied to evaluate the apparent diffusion coefficient values as a discriminating variable to differentiate malignant lesions from benign ones; sensitivity and specificity were calculated. Results: There was no significant difference in the signal-noise ratio value and image quality between DW images obtained with b = 400 s/mm 2 without SENSE (DW400) and b = 1000 s/mm 2 with SENSE (DW1000SENSE). DW1000SENSE had the highest Az values for discriminating malignant from benign hepatic lesions (0.97) and hemangioma from metastasis (0.89). Using 1.63 × 10 -3 mm 2 /s as the cutoff value, DW1000SENSE had a sensitivity of 95.2% (40/42) and a specificity of 91.0% (40/44) for differentiating benign from malignant hepatic lesions. Using a cutoff value of 1.45 × 10 -3 mm 2 /s, DW1000SENSE had a sensitivity of 90.5% (19/21) and a specificity of 93.7% (15/16) for differentiating metastases from hemangiomas. Conclusions: Diffusion-weighted imaging with a b value of 1000 s/mm 2 and SENSE has the potential to differentiate hepatic focal lesions with improved sensitivity and specificity.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    24
    References
    33
    Citations
    NaN
    KQI
    []