Detection and quantification of fluorescent cell clusters in cryo-imaging

Abstract

We developed and evaluated an algorithm for enumerating fluorescently labeled cells (e.g., stem and cancer cells) in mouse-sized, microscopic- resolution, cryo-image volumes. Fluorescent cell clusters were detected, segmented, and then fit with a model which incorporated a priori information about cell size, shape, and intensity. The robust algorithm performed well in phantom and tissue imaging tests, including accurate (<2 error) counting of cells in mouse. Preliminary experiments demonstrate that cryo-imaging and software can uniquely analyze delivery, homing to an organ and tissue distribution of stem cell therapeutics. © 2012 Grant J. Steyer et al.