Analysis Of Psychological Distress Factors For Patients With Prostate Cancer And Establishment Of Prediction Model Based On The Artificial Neural Network

Research Purpose:1.Investigate the status quo of psychological distress in prostate cancer patients during treatment,analyze the risk factors of psychological distress in prostate cancer patients,and put forward nursing targeted intervention suggestions;2.Establish a prediction model of psychological distress of prostate cancer patients based on artificial neural network model,which is used to screen high-risk groups that are prone to develop into clinically significant psychological distress,and provide identification and prediction tools for early intervention.Research Methods:From January 2020 to January 2021,252 cases of prostate cancer patients who met the standards of inpatient discharge in the urology department of a third-class A hospital in Hangzhou were selected.The questionnaire survey was conducted by using the general data sheet of patients and the Distress Thermometer(DT),General Self-Efficacy Scale(GSES),Cancer Coping Modes Questionnaire(CCMQ),Perceived Social Support Scale(PSSS).Data were collected and entered.Statistical methods such as descriptive,chi-square test,Fisher’s exact probability method,Logistic regression and Spearman correlation analysis were used to analyze the data by using software SPSS23.0.Logistic regression prediction model and neural network model were established to compare their prediction efficiency.Research Results:1.A total of 273 questionnaires were sent out and 252 were effectively received with an effective recovery rate of 92.31%.The DT score of prostate cancer patients was(3.35±1.63),and the incidence of psychological distress was 47.62%.Gleason score,spouse health status,sleep status and self-rated personality type were the main influencing factors of psychological distress in prostate cancer patients(P < 0.05).The level of psychological distress was also correlated with patients’ self-efficacy,cancer coping style and social support(P < 0.05).2.The prediction accuracy of the neural network model is 84.02%,and the AUC is 0.890.The main performance indexes of the neural network model are better than the Logistic regression prediction model.Research Conclusion:1.About 47.62% of prostate cancer patients have psychological distress.Gleason score,spouse’s health status,whether patients’ sleep status has been affected since the disease,and patients’ self-rated personality type are the influencing factors of psychological distress in prostate cancer patients.2.The general self-efficacy,cancer coping style,social support and psychological distress of prostate cancer patients were correlated.3.By comparison,The prediction model of psychological distress of prostate cancer patients using artificial neural network modeling method is superior to the traditional Logistic regression prediction model and has better prediction efficiency.