Segmentation of cultured neurons using logical analysis of grey and distance difference

Abstract

The molecular and cellular bases of neuronal cell death that underpin a wide range of neurodegenerative disorders are still not well understood. One approach to investigating neuronal death is through systematic studies of the changing morphology of cultured brain neurons in response to cellular challenges. Image segmentation methods developed to date to analyze such changes have been limited by the low contrast of cells in unstained neuronal cultures and the unimodal histograms generated by these analyses. In this paper we present new algorithms based on logical analysis of grey and distance difference of images that successfully circumvent these problems. Two key parameters of this analysis, window width and logical threshold, are automatically extracted for use in logical level technique, and spurious regions are detected and removed through use of a hierarchical window filter. The efficacy of the developed algorithms is demonstrated here through an analysis of cultured brain neurons from newborn mice.