Mild cognitive impairment in Parkinson’s disease (PD-MCI) is a common type of PD, which has a far higher conversion rate to dementia than PD without MCI. However, the studies of neuroimaging and biochemistry markers on PD-MCI are still insufficient, and their results were inconsistent, which is possibly due to the lack of division of PD-MCI subtypes in these studies. This project plans to use a series of neuropsychological assessments to comprehensively evaluate the cognitive functions of early-stage PD patients, and from them to draw 90 PD-MCI patients, and then differentiate different PD-MCI subtypes. This project plans to conduct multi-modal MRI scanning on 90 PD-MCI patients, 30 PD patients without MCI, and 30 healthy elderly adults to acquire their grey matter and white matter structural metrics as well as functional network metrics. This project also plans to measure the levels of five biochemical molecules in the peripheral blood of all participants. This project will also track the changes of the participants in cognitive function, neuroimaging and biochemical biomarkers at a 1.5-year interval. The aim of this project is to examine the pathological characteristics of different PD-MCI subtypes in neuroimaging and biochemical biomarkers through group comparison; using correlation analysis to explore the correlations of cognitive function with neuroimaging and biochemical biomarkers in PD patients; and to investigate the specific pathological progressions of different PD-MCI subtypes in cognitive function, neuroimaging and biochemical biomarkers by the longitudinal study design. The findings could be beneficial for the early diagnosis of PD-MCI and the early warning for the conversion to PD dementia.