Flat- and darkfield operations are necessary given the potential non-uniform parallelization of light rays, resulting in varying illumination intensities across distinct image segments. This is particularly pertinent in the context of live-cell imaging, where the dynamic movement of cells over time can contribute to these variations. Furthermore, in light of potential photon infiltration along the light path towards the camera and the persistent presence of residual signal noise, the utilization of flat- and darkfield operations becomes indispensable to ensure robust quantitative imaging. The process of image per-pixel subtraction and division was executed through the utilization of parallel instances of the MATLAB software environment.
Acquisition of images in multiple channels was conducted across 100+ regions of a plate. In each timepoint, a cell marker channel was employed for cell segmentation, subsequently serving as a basis for total image fluorescence extraction in other available channels. The automated procedure was executed using the CellProfiler software API and Windows PowerShell.
A comprehensive analysis was performed on the total cell fluorescence of over 20,000 individual cells, extracted from various timepoints and experiments, resulting in an aggregate of approximately 750,000 images. Signal alterations were subjected to fitting models at the single-cell level.
