The Predictive Value Of Somatic Self-rating Scale For Coronary Heart Disease In Women Based On Decision Tree Classification Model

Objective: This study provides an economical and reliable medical predictive model to assess the degree of risk of coronary heart disease.Based on the Somatic Self-rating Scale(SSS),negative emotion is discussed in hospitalized women with suspected coronary artery disease in the Department of cardiology.To evaluate the value of this model to the binary diagnosis model of coronary heart disease in women under 65 years old.Assist clinicians to make rational diagnosis and treatment decisions and reduce unnecessary consumption medical resources.Methods: Based on the Electronic medical record system of the First Affiliated Hospital of Wannan Medical College.Patients who underwent coronary angiography from January 1,2020 to December 31,2021 were retrospectively analyzed.Collect a variety of medical data,including height,weight,smoking history,examination and examination reports.The somatization symptom scale was conducted for young women aged 18-65 who underwent coronary angiography for the first time in our hospital from April to December 2021,to record whether they were diagnosed with coronary heart disease and assess the severity of coronary artery lesions.The Decision Tree classification model in Machine Learning was used to establish the model.Evaluate the value of included variables for disease classification and the performance of the model.Results: By analyzing the results of 3,791 patients who underwent coronary angiography in the cardiovascular department of our hospital in 2020 and 2021,126 data features of 2,814 patients were finally incorporated into the model establishment.Comprehensive analysis found that gender,age,basic patients with diabetes,white blood cell count,neutrophil count,fasting blood glucose in the general information between the two groups.New diagnostic markers of highdensity lipoprotein,including changes in ST-T segment of electrocardiogram,echocardiogram suggesting decreased cardiac systolic function,residual cholesterol and arteriosclerosis index,all showed significant differences and were located at the top of the decision tree.The accuracy rate of the model was 0.7050,the accuracy rate was 0.8119,and the recall rate was 0.7885.The F1 value is 0.8.133 patients were eventually included in the scale study,and SSS scores were significantly different between the coronary and noncoronary groups(p< 0.01).After the SSS score is included in the model,the SSS score is at the top of the decision tree,and the model efficiency is improved after the re-modeling.Conclusion: 1.Machine learning decision tree classification model can help clinicians quickly identify and quantify risks,avoid missed diagnosis and reduce the consumption of unnecessary medical resources.2.SSS score has important value for the establishment of coronary heart disease diagnostic model.The quality of medical decision-making can be improved by incorporating the adverse emotional effects into the medical decision-making system.