Clinical Intelligent Diagnosis and Treatment of Hemorrhagic Stroke Based on Multiple Models

In order to effectively prevent the occurrence of hemorrhagic stroke and minimize the incidence probability, a mathematical model was established and feasible measures were proposed according to the practical needs and orientation of medical clinical practice. Solve problems with LSTM model, BP neural network model, Polynomial features function, polynomial fitting, analysis of variance model, Spielman correlation coefficient and other mathematical and neural network models. Based on the modeling of factors related to the risk of hematoma expansion in hemorrhagic stroke patients, the correlation between treatment intervention and edema progression was explored. Based on the prediction of the 90-day mRS Results of patients, the prognosis and personal history, disease history, treatment methods and imaging characteristics of patients with hemorrhagic stroke were analyzed, so as to make recommendations for clinical decision-making.