Mild cognitive impairment (MCI) has become a serious threat to the health of the elderly. It is reversible in early stage. In recent years, diffusion tensor imaging (DTI), a technique that can map white matter fibers quantitatively, has been increasingly applied in clinical practice and attracted much attention. Studies have shown that patients with MCI have significant changes in DTI parameters in the hippocampus, fornix, cingulate cortex, and corpus callosum, which can be used for the early identification, disease evaluation, and prognosis prediction of MCI, and the combination of multiple parameters may improve the accuracy. DTI findings are associated with the decline of cognitive function, especially memory, for patients with MCI. Among the subtypes of MCI, amnestic MCI has a high risk of developing into Alzheimer’s disease (AD), and the lower the fractional anisotropy and the higher the apparent diffusion coefficient, the greater the possibility of MCI progressing to AD. DTI shows good sensitivity in distinguishing the white matter between AD and MCI-AD has more extensive and severe white matter abnormalities. However, the application of DTI in MCI still faces many problems and needs further exploration.