Abstract

Cellular aging and rejuvenation remain critical areas of investigation in biomedical research, with significant challenges in objectively quantifying cellular age due to the inherent variability among cells and cell lines. This project aims to address this gap by developing advanced deep learning models capable of classifying aged (senescent) and youthful (young) cells and generating a novel, objective scale for cellular age estimation. Leveraging fluorescence microscopy data and ground truth age labels, the study employs tools like CellProfiler, CellProfiler Analyst, and deep learning frameworks to extract key morphological and intensity-based features for classification and analysis.

The project incorporates multi-channel fluorescence imaging, including DAPI (nuclear marker), Calcien-AM (viability indicator), and experimental cell identity markers, to analyze human and mouse epithelial cells. By integrating these channels and analyzing spatial features, the research will refine cell classification and establish a "visual-age score" (VAS) to estimate the cellular age. This scale will also facilitate evaluating the efficacy of chemical cocktails in reversing cellular aging.