As an effective tool for knowledge discovery, concept lattice has been successfully applied to various fields. One of the key problems of knowledge discovery is knowledge reduction. The existing work on attribute reduction has not focused on aquatic lives disease diagnosis (ALDD). This paper describes an improved incremental approach of attribute reduction in concept lattice for ALDD. Firstly, the main definition of the concept lattice is introduced. Secondly, the attributes within the framework of equivalence classes are discussed. Finally, the incremental algorithm of attribute reduction in concept lattice for ALDD is presented. Based on the algorithm, we can easily diagnose the aquatic lives diseases. The examples results validate the effectiveness of approach.