Algorithmic Interests

  • Machine and deep learning
  • Image segmentation
  • Big data, domain adaptation
  • Probabilistic modeling, Bayesian non-parametrics
  • Shape analysis and differential geometry
  • Longitudinal analysis
  • Spectral methods

Clinical Interests

  • Autism
  • Alzheimer’s disease
  • Dyslexia
  • Diabetes
  • Traumatic Brain Injury
  • Radiomics
  • Epidemiologic imaging


Age Estimation

The brain is a complex organ whose morphology varies substantially across the population. There are many causes for the morphological variation of the brain, many of which have not yet been fully understood, however several studies point towards aging as the main factor affecting brain morphology. In this project we are interested on exploring the use of imaging based biomarkers which help us to create regression models describing the healthy development of the brain. These models can in turn be used to measure brain abnormalities caused by a variety of diseases such as Alzheimer’s Disease or Autism.

Contact person: Benjamin Gutierrez Becker

Brain Segmentation
Magnetic resonance imaging (MRI) delivers high-quality, in-vivo information about the brain. Whole-brain segmentation provides imaging biomarkers of neuroanatomy, which form the basis for tracking structural brain changes associated with aging and disease. Despite efforts to deliver robust segmentation results across scans from different age groups, diseases, fi eld strengths, and manufacturers, inaccuracies in the segmentation outcome are inevitable.We introduce inherent measures for effective quality control of brain segmentation based on a Bayesian fully convolutional neural network, using model uncertainty.

Contact person: Abhijit Guha Roy



Our research is supported by the Bavarian State Ministry of Education, Science and the Arts in the framework of the Centre Digitisation.Bavaria (ZD.B) and SAP SE.