Immunofluorescence microscopy is a well-established technique in bioimaging. Combined with super resolution microscopy techniques, it is able to provide single-molecule imaging and has become a powerful tool for understanding cellular pathways. However, these high-resolution techniques present some disadvantages with respect to conventional microscopy: they are costly and slow and that prevents the processing of a multitude of biological samples. Fortunately, we can use signal processing to leverage data extraction from conventional microscope images. Based on a case study, Ghaye and coworkers monitor toll-like receptor 2 proteins expressed in Caco-2 cell cultures. The authors propose a comparative study of various image segmentation algorithms for extracting regions of interest within fluorescent images. The performance of T-point, Otsu and Sauvola's segmentation algorithms are studied along with a novel approach. These algorithms are de facto tools for a deeper, cost-effective automated analysis of cells. In particular, guidelines are provided on algorithm selection for segmenting regions populated with receptors.
Image thresholding techniques for localization of sub-resolution fluorescent biomarkers
Julien Ghaye, Madhura Avinash Kamat, Linda Corbino-Giunta, Paolo Silacci, Guy Vergères, Giovanni De Micheli and Sandro Carrara
ISAC members receive Cytometry Part A as a benefit of membership in the society.