Modern methods of flow cytometry make it possible to characterize cell populations with unprecedented detail, but the traditional analysis of data using the “manual gating” method under these conditions is ineffective and unreliable. A large number of research papers have been published in recent years that describe specialized clustering algorithms for detecting and determining populations of white blood cells. However, problems related to the presence of noise and different data density remain relevant. The internal problem of cluster analysis associated with determining the preferred number of clusters, and the moment the process itself stops remains unresolved. It is proposed to use the modified UPGMC (Unweighted Pair Group Method with Centroid average) with the Markov moment of stopping the clustering process from eliminating noise and overcoming problems associated with different data density.