Diffusion tensor imaging (DTI) is widely used to evaluate microstructural variations in brain tissue. In particular, fractional anisotropy (FA), reflecting the magnitude and orientation of anisotropic water diffusion, allows us to detect pathological events in white matter. An ex vivo DTI study coupled with histological assessment is an efficient strategy to evaluate the myelination process, i.e. demyelination and remyelination. The relationship between DTI values and myelin content or the individual cellular components such as oligodendrocytes, microglia, and astrocytes during both processes of demyelination and remyelination are not well-understood. To address this issue, we employed a cuprizone-inducible demyelination mouse model. Demyelination can be induced in this model during cuprizone exposure and termination of cuprizone exposure induces remyelination. We fed the mice cuprizone-containing chow for 4 weeks and then normal chow for an additional 4 weeks. The ex vivo DTI was performed to evaluate the white matter profiles observed by FA, mean diffusivity (MD), and radial diffusivity (RD) at both demyelinating and remyelinating time points, and then we evaluated histological properties at the same time points. The results indicated a gradual FA decrease during the cuprizone treatment (0, 2, 3, 4 weeks). A lower peak was seen at 1 week after the normal chow was resumed, with recovery to baseline at 2 and 4 weeks. MD and RD showed an opposing pattern to that of FA. These DTI values were positively or negatively correlated with myelin content regardless of the status of the white matter. The RD value was more sensitive to myelination status than FA and MD. We have clarified the temporal changes in the DTI values coupled with histological properties over both the demyelination and remyelination processes.