Objective: To evaluate the effect of b-value distribution for diffusion kurtosis imaging(DKI)on the repeatability and role in differential diagnosis of liver lesions.Materials and methods: Thirty-seven patients with liver lesions(hepatocellular carcinomas 21,intrahepatic cholangiocarcinomas 8,liver metastases 4,hepatic cysts 4)were consecutively enrolled in this study.All patients underwent DKI(b = 0,1000,2000)and routine T1-weighted dynamic contrast-enhanced magnetic resonance imaging(MRI)on GE 3.0T MRI instrument.Nineteen of them were studied with four acquisition schemas using different numbers and combinations of b-values.The values of DKI parameters,including FA,MD,Da,Dr,FAk,MK,Ka and Kr were measured.Reliability was assessed by intraclass correlation coefficient(ICC).The values of DKI parameters of the different liver lesions were compared using one-way ANOVA.P<0.05 was considered statistically significant.Results: The repeatability of DKI parameter evaluation with various acquisition schemas using different numbers and combinations of b-values is good.The values of Dr,FAk,MK,Ka and Kr of different lesions in the liver were significantly different.The differences of FA,MD and Da were not statistically significant.Conclusion: It is feasible to reduce DKI acquisition time in clinical studies by reducing the number of b-values.The DKI parameters are helpful for the differential diagnosis of liver lesions. |