Reliability of synthetic diffusion-weighted imaging with a high b-value for paediatric abdominal MRI

To evaluate the reliability of synthetic diffusion-weighted imaging (DWI) using a high b-value in comparison to conventional DWI for paediatric abdominal MRI.

Paediatric patients (<19 years old) who underwent liver or pancreatobiliary MRI with DWI using 10 b-values (b = 0, 25, 50, 75, 100, 200, 400, 600, 800, 1,500 s/mm²) from March to October 2021 were included in this retrospective study. Using the software, synthetic DWI using b = 1,500 s/mm² was generated automatically by selecting the b-value required as output. Conventional and synthetic DWI values for b = 1,500 s/mm² were measured at the liver, spleen, paraspinal muscle, and mass lesions, if present, and apparent diffusion coefficient (ADC) values were calculated using the mono-exponential model. Intraclass correlation coefficients (ICCs) were calculated to assess the reliability of conventional and synthetic DWI and ADC values with b = 1,500 s/mm².

Thirty paediatric patients (M:F = 22:8, mean 10.8 ± 3.1 years old) were included and four had tumours on abdominal MRI. ICC values were 0.906-0.995 between conventional and synthetic DWI and ADC with b = 1,500 s/mm² in the liver, spleen and muscle. For mass lesions, ICC values were 0.997-0.999 for both synthetic DWI and ADC images.

Synthetic DWI and ADC values obtained using a high b-value showed excellent agreement with conventional DWI for the liver, spleen, muscle, and mass in paediatric MRI.