link to homepage

Institute of Neuroscience and Medicine

Navigation and service

Ji Chen

M.Sc. Ji Chen

Ph.D. student

Research topics:

  • Psychiatric disorders
  • Multi-modal MRI
  • Multivariate methods & machine learning
  • Brain-behavior relationship

Methodological topics:

  • Non-negative matrix factorization

The principal domains of my research focuses are to utilize various advanced machine learning methods to solve or to achieve a better understanding of many clinical problems in the neuropsychiatric field. Particularly, I am interested in using multi-modal fMRI data as a target to which the machine learning approaches apply. Since the psychiatric diseases, such as schizophrenia, are heterogeneous in symptomatic expressions and hard to describe, I try to search for a stable and generalizable factor-structure to conceptualize the psychopathological dimensions in schizophrenia. NMF is a non-supervised machine learning method which has been shown to provide biologically meaningful decompositions, and I thus dig into this methodology and apply it to the psychiatric field. Furthermore, the elusive phenotype and endophenotype in those psychiatric patients also trigger me to subtyping them into different biological subgroups by clustering algorithm with the combination of imaging information and clinical symptoms. Also, I would like to predict the clinical severity and disease progression from imaging data by using many prediction methods such as LASSO.


Institute of Neuroscience and Medicine (INM-7)
52425 Jülich


Phone: +49 2461 61-85334
Fax: +49 2461 61-1880