Human beings are inevitably exposed to toxic substances as a result of influences of potential contamination factors in the environment,food and medicines,which poses a threat to human health.In order to effectively screen and prevent the exposure or intake of such substances,it is neces-sary to develop in vitro assays for the detection and toxicity evaluation of toxic substances.High content screening(HCS)has been recognized as an important tool for toxicity testing and risk assessment of compounds due to its high throughput and automation advantages,and has been widely used in in vitro toxicology research.In this review,we described the system components of HCS and its workflow in toxicity screening and toxicity evaluation by focusing on cases of their application in toxicity detection and evaluation studies,including the cytotoxicity,hepatotoxicity,nephrotoxicity,genotoxicity,neurotox-icity,cardiotoxicity,and developmental toxicity.In addition,the applications and developments of machine learning in HCS were explored,especially to the advantages of supervised and unsupervised machine learning strategies for high throughput image screening and data analysis.Finally,the future applications of HCS in toxicity screening and evaluation are outlined,especially in terms of binding new models and gene editing technology.