Abstract: In recent years, research on the condition of mind cognitive impairment (MCI), which is the normal and excessive stage of Alzheimer's disease, has attracted much attention. However, the current medical MCI manual diagnosis not only has relatively large limitations in the referenced features, but also relies on manual judgment, which is prone to subjective errors. Therefore, this paper proposes an automatic diagnosis method of MCI based on random forest, hoping to determine MCI efficiently and accurately through machine learning. At the same time, in order to obtain the optimal parameters of the random forest MCI diagnosis model more efficiently, genetic algorithm was combined. The results show that the accuracy of this method is about 5% higher than that of medical manual diagnosis, and the time taken by genetic algorithm is shortened by nearly 45 times compared with grid search on the problem of obtaining the optimal parameters of random forest.