Senescence is a complex and heterogeneous process implicated in aging, tissue dysfunction, and chronic diseases. Despite its clinical significance, identifying senescent cells remains a major challenge due to the absence of a single, reliable biomarker. Conventional markers such as SA-β-gal, p16Ink4a, and p21Cip1 often vary by cell type and stimulus, limiting their generalizability. Recent advances have demonstrated that nuclear morphology can serve as a powerful proxy for senescence, enabling classification through deep learning methods. In this study, we explored the use of VisionSort™, a morphometric-based flow cytometry platform, to detect multiple forms of senescence in different cell types. By applying AI-driven analysis to both label-free and fluorescence ghost-motion imaging (GMI) data, we aim to establish a robust, scalable approach for senescence detection that circumvents the limitations of
traditional molecular assays.
