Cell morphodynamics visualization from images of zebrafish embryogenesis

Abstract

Laser scanning microscopy provides high-resolution nondestructive in vivo imaging to capture specific structures that have been fluorescently labeled, such as cellular nuclei and membranes, throughout early zebrafish embryogenesis. An increasingly challenging problem biologists must face is how to effectively explore, follow, and study the thousands of cells contained in the resulting time-varying volume data that are large in space, time, and variable domain. Visual data explorations, such as direct volume rendering, have been successfully used for the analysis of volumetric data. However, visualizing large-scale time-varying fields remains a challenging problem. In this paper we present a novel Focus+Context animated volume rendering. The technique is based on the distance map of objects of interest and on a scene graph architecture. We demonstrate that distance map driven volume rendering, implemented in modern graphics hardware, is suited to generate run time and interactive representations such as ghosted rendering and cut-away rendering. The experimental results on zebrafish embryogenesis data demonstrate that the technique is suited to uncover and to analyze biological events, such as organogenesis, contained in time-varying volumetric dataset.