Random forest (RF) is a machine learning method widely used in medical science, bioinformatics and management science to deal with classification and regression problems.RF belongs to the family of integrated learning algorithms, which are characterized as adding data sample perturbations and input property perturbations during training to handle a variety of data types.In the existing medical image analysis, RF is mainly used in image processing of medical images, diagnosis of assisted medical treatment, and exploration of the pathogenesis of certain diseases.This paper firstly introduces the basic principle of RF briefly and then focuses on the use of RF in medical imaging.Finally, the advantages and disadvantages of RF are summarized and prospected.