摘要
目的 基于机器学习算法构建带状疱疹后神经痛(PHN)患者抑郁状态风险预测模型,为临床PHN患者抑郁状态发生的精准预测提供新的思路和方法.方法 选取2022年6月至2023年6月在陆军军医大学第二附属医院住院的PHN患者作为研究对象,按8∶2比例随机划分为训练集和测试集,以是否发生抑郁状态为结局变量,分别基于K近邻(KNN)、决策树(DT)、逻辑回归(LR)、朴素贝叶斯(NB)、随机森林(RF)、支持向量机(SVM)6种机器学习算法构建PHN患者并发抑郁状态风险预测模型.基于受试者工作特征曲线下面积(AUC)、准确度、精确度、召回率、F1分数评估模型性能,选出最优模型.结果 共纳入275例PHN患者,其中170例患者发生抑郁状态,抑郁状态发生率为61.82％.KNN、DT、LR、NB、RF、SVM模型的AUC分别为0.574、0.589、0.760、0.742、0.591、0.733,其中LR模型AUC值、准确度、精确度、召回率、F1分数最高.结论 基于LR机器学习算法构建的PHN并发抑郁状态的风险预测模型性能最优,有助于临床早期评估和预防其抑郁状态的发生.
Abstract
Objective To construct a risk prediction model for depression in the patients with posther-petic neuralgia(PHN)based on machine learning algorithm to provide a new idea and method for accurate prediction of depressive state occurrence in clinical PHN patients.Methods The inpatients with PHN in the Second Affiliated Hospital of Army Military Medical University from June 2022 to June 2023 were selected as the study subjects and randomly divided into the training set and test set according to the ratio of 8∶2,and whether or not the depressive state occurring served as the outcome variable.Based on six machine learning al-gorithms of K-Nearest Neighbor(KNN),Decision Tree(DT),Logistic Regression(LR),Naive Bayes(NB),Random Forest(RF)and Support Vector Machine(SVM),a risk prediction model for PHN patients with complicating depressive state was constructed.The model performance was evaluated based on the area under the curve(AUC),accuracy,precision,recall rate and F1 score,and the optimal model was selected.Results A total of 275 PHN patients were included,among them 170 cases developed the depressive state,and the inci-dence rate of depressive state was 61.82％.The AUC values of KNN,DT,LR,NB,RF and SVM models were 0.574,0.589,0.760,0.742,0.591 and 0.733,respectively,among which the AUC value,accuracy,precision,recall rate and F1 score of LR model were the highest.Conclusion The risk prediction model of PHN compli-cating depressive state based on LR machine learning algorithm has the best performance,which is helpful for early clinical assessment and prevention of depressive state.
维普
万方数据