[Reality and expectation of machine learning in critically ill patients with acute kidney injury]

The prediction model constructed by machine learning can early predict acute kidney injury (AKI) in critically ill patients, which contributes to taking preventive measures to reduce incidence of AKI as soon as possible. Machine learning can also identify AKI subtypes in real time, so as to facilitate the implementation of precise treatments for their subtypes in the future, which may improve the prognosis of patients. In this article, the construction of forecasting model of AKI and identification of AKI subtypes in critically ill patients and it's precise treatment with machine learning, limitations and development trend of machine learning in the field of AKI in critically ill patients is addressed for clinical reference.