Fabian M Mager’s research while affiliated with Technical University of Denmark and other places

... For example, the clinical group may be diffuse and heterogeneous due to comorbidity and to a convergence of different pathophysiological pathways on the same symptoms. In line with this, there has been considerable effort invested in finding subgroups of patients based on cognition or symptoms over the last 50 years and many putative stratifications have been proposed [14][15][16][17][18][19][20] , yet none have advanced beyond proof-of-concept. We have suggested that one of the main reasons is that clustering algorithms always yield a result, regardless of whether clearly defined clusters are evident in the data, resulting in ill-defined and poorly replicated subtypes. ...

Reference:

Learning latent profiles via cognitive growth charting in psychosis: design and rationale for the PRECOGNITION project