Anti-senescent drug screening by deep learning-based morphology senescence scoring

We developed a novel morphology-based quantitative scoring system to evaluate the state of endothelial cells by senescence probability output from pre-trained convolutional neural network (CNN) optimised for the classification of cellular senescence, Deep Learning-Based Senescence Scoring System by Morphology (Deep-SeSMo). By using Deep-SeSMo, we screened for drugs that control cellular senescence in HUVECs using a kinase inhibitor library, and identified four novel anti-senescent drugs; terreic acid, PD-98059, daidzein, and Y-27632. We examined the mechanism by which these compounds suppress the senescence phenotype through global gene expression analysis by RNA sequencing.