Diffusion-weighted imaging (DWI) is a powerful functional MRI technique. However, scarcely has the DWI been used to identify the expression of Cytokeratin 19 (CK19) in hepatocellular carcinoma (HCC). In this study, we aim to differentiate CK19 positive and negative HCC with diffusion parameters derived from different DWI techniques. Our results showed that diffusion parameters can be used as noninvasive quantitative imaging markers for preoperatively predicting the CK19 expression of HCCs. More importantly, the integration of different diffusion parameters could better describe the characteristics of tumor tissues，thus reaching better diagnostic performance.