Removal of out-of-plane fluorescence for single cell visualization and quantification in cryo-imaging

We developed a cryo-imaging system, which alternates between sectioning (10-40 mu m) and imaging bright field and fluorescence block-face image volumes with micron-scale-resolution. For applications requiring single-cell detection of fluorescently labeled cells anywhere in a mouse, we are developing software for reduction of out-of-plane fluorescence. In mouse experiments, we imaged GFP-labeled cancer and stem cells, and cell-sized fluorescent microspheres. To remove out-of-plane fluorescence, we used a simplified model of light-tissue interaction whereby the next-image was scaled, blurred, and subtracted from the current image. We estimated scaling and blurring parameters by minimizing an objective function on subtracted images. Tissue-specific attenuation parameters [mu (T): heart (267 +/- A 47.6 cm(-1)), liver (218 +/- A 27.1 cm(-1)), brain (161 +/- A 27.4 cm(-1))] were found to be within the range of estimates in the literature. "Next-image" processing removed out-of-plane fluorescence equally well across multiple tissues (brain, kidney, liver, etc.), and analysis of 200 microsphere images gave 97 +/- A 2% reduction of out-of-plane fluorescence. Next-image processing greatly improved axial-resolution, enabled high quality 3D volume renderings, and improved automated enumeration of single cells by up to 24%. The method has been used to identify metastatic cancer sites, determine homing of stem cells to injury sites, and show microsphere distribution correlated with blood flow patterns