You are here: CSGB About CSGB Research

Research

In the second funding period of CSGB that started in 2015 we take advantage of the established collaborative structure of CSGB and the increased international collaboration. Important research topics are

  • tensor valuations in stochastic geometry
  • modelling and statistical inference in non-Euclidean spaces
  • analysis of concrete bioimaging data

Research plan - diagram

The research is organized in six work packages and illustrated in a diagram showing the many connections between the planned projects. See the research overview.

More specifically, we focus on

Tensor valuations in stochastic geometry – we study

  • integral geometry of tensors
  • role of tensors in stochastic modelling
  • estimation of tensors from observations in digital grey-value images

Inference in non-Euclidean spaces - we consider

  • point processes on linear networks and directed graphs
  • deformation modelling and statistics of deformations
  • models for random trees

Concrete bioimaging data - we analyze new data from

  • diffusion weighted imaging
  • confocal microscopy
  • super-resolution fluorescence imaging
  • scanning and cryo-electron microscopy

Stochastic geometry

Stochastic geometry is a fusion of stochastics and geometry. Its aim is modelling and inference for random geometric structures. When using stochastic geometry methods in bioimaging, the geometric structure may be brain fibers, neurons forming minicolumns, moving protein clusters or synaptic vesicles.

Bioimaging

Bioimaging relates to methods of visualizing and quantifying biological material from images. Bioimaging is used to gain information about the 3D structure of the observed specimen from sections, projections or 3D voxel images. In CSGB, we work especially with advanced microscopy data.

Comments on content: 
CSGB static webpages

Revised 05.02.2018