Abstract:Acute respiratory distress syndrome（ARDS）is a highly heterogeneous critical illness with high morbidity and mortality. Early identification and risk assessment are crucial to improving patient prognosis. Current artificial intelligence（AI）technology，especially machine learning（ML）models，have shown significant potential in the early diagnosis，risk stratification and personalized management of ARDS. Compared with traditional scoring systems，AI models perform well in predicting mortality and optimizing clinical decision - making，especially through multimodal data fusion，which can significantly improve the prediction accuracy of the models. However，the lack of interpretability，limited clinical applicability and data privacy of AI models are still the main challenges restricting clinical application.Future research should focus on improving model transparency，optimizing clinical integration and solving ethical issues to promote the further development of AI-enabled ARDS precision medicine.