| Heart failure is caused by myocardial damage due to inflammation,myocardial infarction,hemodynamic overload,cardiomyopathy,and other causes,which results in changes in the patient’s function and myocardial structure,and finally leads to symptoms such as low filling function or ventricular pumping.The main clinical manifestations are weakness and dyspnea.Chronic heart failure is a state of heart failure that persists in patients and can be in a decompensated,deteriorating or stable state.Mood disorders such as depression and anxiety,as common comorbidities of heart failure,not only affect the quality of life of heart failure patients,but also influence their prognostic outcome.In this paper,we use several types of data mining models to conduct a statistical inference study on mood disorders in patients with chronic heart failure.In this study,559 patients hospitalized with chronic heart failure in a tertiary care hospital from January 2017 to March 2021 were collected.The "Huaxi Mood Index Scale" was used to determine the psychological status of patients after admission.The basic information of 559 patients,including age,days of hospitalization,education level,gender,marital status,and test data,were extracted from the hospital information systems to establish a heart failure research database.After standardization,cleaning and integration of the data,10 characteristic factors related to chronic heart failure were selected from more than 1300 clinical indicators by using single factor analysis,multi-factor analysis,descriptive analysis of data and combining the contents of literature and clinical experience of doctors.Logisitc regression,random forest,support vector machine,and XGBoost methods were used to model and compare the validity of the data of these 10 factors.The results showed that the random forest algorithm performed best in prediction accuracy,and three indicators of RBC distribution width SD,hospitalization days,and free thyroxine had effects on chronic heart failure disease.Finally,a Nomogram column line graph for predicting the risk of mood disorders in chronic heart failure that can be used by health care professionals on a daily basis was established. |
