Interacting Cluster Point Process Model for Epidermal Nerve Fibers

• Activity Number: 21 - Spatial and Spatio-Temporal Statistics for Biomedical and Epidemiological Studies
• Type: Topic Contributed
• Date/Time: Sunday, July 29, 2018 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Guilherme Ludwig* and Nancy Lopes Garcia and Peter Guttorp
• Companies: University of Campinas and University of Campinas and University of Washington / Norwegian Computing Center
• Keywords: Spatial Point Processes; Interacting Processes; Bayesian Statistics

Abstract

We propose an interacting cluster model for the spatial distribution of epidermal nerve fibers. The model is an interacting spatial point process for the centers of nerve fiber bundles, where from each center a random number of fibers start. The center process is, however, inhibited by fiber endings that belong to different center bundles. The fibers themselves have random length and spatial orientation, following a von Mises distribution. We examine the processes that can be described with our model, how coefficient estimation can be performed via Markov chain Monte Carlo methods, and detail approaches for the implementation of efficient MCMC sampling schemes. An application to diabetic neuropathy data is presented, as well as a simulation study.