Stochastic forest model is used to analyze the prognostic factors of patients with systemic lupus erythe-matosus

YANG Qin; LIU Xiuyan; SHI Yuyuan

Abstract：Objective:To analyze the influencing factors of prognosis in patients with systemic lupus erythematosus(SLE)using the random forest model and the Logistic regression model respectively,and to provide a basis for the se-lection of treatment plans and prognosis judgment.Methods:All 182 patients with SLE diagnosed in the dermatology and rheumatology departments of Linfen People's Hospital from January 2022 to June 2023 were selected as the re-search subjects.Random forest and Logistic regression models were entered to analyze the influencing factors of the lat-er disease recovery of SLE patients and evaluate the predictive effects of the two models.Results:The relevant data of the random forest model suggest that the top five in terms of importance are complement C3,whether combined with kidney disease,infection,anti-double-stranded DNA antibody(anti-dsDNA antibody),and whether there is neuro-logical related damage.The Logistic regression model showed that the order was complement C3,whether there was neurosystem-related damage,infection,anti-dsDNA antibody,and whether there is any damage related to nervous system.Conclusion:Complement C3,whether combined with kidney disease,infection,anti-DSDNA antibody,and whether neurological damage occurs are relatively important factors affecting the prognosis of SLE patients.Both the random forest and Logistic regression models have their own advantages and disadvantages.In practical applications,the two can be used in combination.