Extracting information from complicated point patterns

The aims and main challenges of the project are to develop methods, models, and software to analyse and model 3D point pattern data, to construct cluster models that allow interaction between points within and between clusters, and to construct spatio-temporal models based on incompletely or only spatially observed point patterns. We will also develop inference tools to estimate parameters from hierarchically structered data with point patterns collected from several body parts, subjects, and groups. The motivation for the project comes from three extremely rich data sets: epidermal nerve fibre patterns in the outermost layer of the skin, series of images of active sweat glands, and pore structures in pharmaseutical coatings. The first two examples play an important role in understanding how the peripheral nervous system is affected by e.g. diabetes, and the last one in controlling drug release.The project will be done in collaboration with statisticians Peter Guttorp (co-applicant) and Thordis Thorarinsdottir at Norwegian Computing Center, Mari Myllymäki at Natural Resources Institute in Finland, and Tuomas Rajala at University College London, and with neurologists in Dr. William Kennedy’s group at the University of Minnesota and chemist Mariagrazia Marucci at AstraZeneca. During these four years, we will have a Ph.D. student working in collaboration with all the members of the group, having the project leader as his/her main advisor.

Partner organizations

  • University College London (UCL) (Academic, United Kingdom)
  • Norwegian Computing Center (NR) (Research Institute, Norway)
  • AstraZeneca AB (Private, Sweden)
  • Natural Resources Institute Finland (Luke) (Research Institute, Finland)
  • University of Minnesota (Academic, USA)
  • University of Washington (Academic, USA)
Start date 01/01/2019
End date 31/12/2022

Published: Thu 18 Apr 2019.