Purpose/Significance To explore a biomedical big data classification method for open applications, and to improve data management efficiency and strengthen data security. Method/Process Literature research and focus group discussion are used to summarize the purpose, principles and classification dimensions of data classification for open applications. A combination of linear and planar classification method is adopted to build a biomedical big data classification system, and key steps for classification implementation are proposed. Result/Conclusion The study proposes objectives of the classification of biomedical big data, and constructs a comprehensive classification system that includes factors such as subject affiliation, importance and shareability based on 3 dimensions:data characteristics, security management needs and sharing applications. The specific steps for the implementation of biomedical big data classification are proposed to help improve the efficiency and security of data management, and promote the open sharing and reuse of data.