The diagnostic value of magnetic resonance imaging (MRI) in liver diseases has long been widely used and accepted. As a functional MRI technique, DWI is the only non-invasive imaging technique that can detect the diffusion motion of water molecules in living tissues at this stage. This provides important information for the early and accurate diagnosis of certain diseases. As the use of DWI in the diagnosis of liver diseases, especially liver cancer, has matured in recent years, DWI has shown great value in the application of liver nodular lesions and their qualitative diagnosis. In general, intrahepatic masses larger than 75px are not difficult to detect and qualitatively diagnose using the now prevailing standard radiographic imaging techniques (MRI and CT), but the overall sensitivity of these techniques is relatively lower for small masses, especially for occupying lesions smaller than 50px, and the ensuing results may lead to misdiagnosis and missed diagnoses. The lowest rate of early diagnosis of hepatocellular carcinoma smaller than 50px has been reported to be only 30%, probably because small hepatocellular carcinomas exhibit a normal portal venous blood supply pattern at this stage and fail to present with diagnostic features. In addition, most patients with hepatocellular carcinoma have mild or severe cirrhosis, and the background of diseased cirrhosis also reduces the diagnostic rate of small hepatocellular carcinoma. Because of the technical advantages of MRI (especially gradient enhancement and parallel imaging), DWI technique can be fully applied in extracranial tumor visualization. DWI has been widely used in the diagnosis of various abdominal tumors, from tumor detection, to tumor characterization, to tumor staging, to assessment of treatment outcome. Most of the liver cancers occurring on the basis of chronic viral hepatitis and cirrhosis are caused by regenerative nodules of cirrhosis undergoing atypical proliferation-carcinoma multi-stage transformation, which eventually form liver cancer nodules. It has been found that the diffusivity of water molecules in a sclerotic regenerative nodule is a marker of whether the nodule is a malignant transformation. While malignant tissues can impede the movement of water molecules due to altered microstructure inside and outside the cells, benign sclerotic nodules or tumors with low malignancy usually exhibit a diffusivity of water molecules similar to normal liver tissue. Conventional enhancement MRI is used to differentiate and diagnose tumors due to the presence of neovascularization and thus morphologic changes in the imaging film showing arterial phase enhancement and delayed phase enhancement elimination, whereas DWI is functional imaging based on cytologic changes in the diseased tissue. Studies have shown that DWI has a sensitivity and specificity of 91.2% and 82.9% for the diagnosis of hepatocellular carcinoma less than 50px, compared to 67.6% and 61% for conventional MRI morphologic imaging. In conclusion, DWI has high effectiveness in the diagnosis of hepatocellular carcinoma, especially for tumors with cirrhotic background, and functional imaging can differentiate hepatocellular carcinoma nodules from benign cirrhotic nodules, which is a useful adjunct to conventional MRI. Although DWI also has high specificity and sensitivity for tumor tissue grading, staging and evaluation of treatment effects, it still needs further improvement and observation in clinical practice. In addition, there is no consensus and uniformity in the setting of certain imaging parameters in DWI technology applied to hepatocellular carcinoma, and the quality of DWI imaging is easily affected by gas accumulation in the gastrointestinal tract and respiratory movement, resulting in serious distortion of the images, which also limits the common and routine use of DWI in clinical hepatocellular carcinoma imaging diagnosis. Rather, we should use a comprehensive clinical evaluation to achieve early diagnosis of liver cancer and improve the effectiveness of various treatments and long-term survival of patients.