Cervical cancer remains one of the leading causes of cancer-related deaths in women. High b-value DWI with non-Gaussian modeling has made it possible to probe tumor tissue complexity, microstructures, and heterogeneity. We employed a non-Gaussian diffusion model based on continuous-time random walk (CTRW) theory to differentiate normal from cancerous cervical tissues. The CTRW parameters (Dm and β) exhibited a statistically significant difference between cancerous cervical tissue and normal. Our preliminary results illustrate the added value of high b-value DWI for cervical cancer detection, and point to a possible direction of diagnosing or staging cervical cancer using non-Gaussian diffusion models.