浏览全部资源
扫码关注微信
1.石河子大学药学院,新疆 石河子 832000
2.新疆维吾尔自治区人民医院药学部,乌鲁木齐 830001
硕士研究生。研究方向:临床药学。E-mail:1440354272@qq.com
主任药师,硕士生导师,硕士。研究方向:医院药学、药物分析。E-mail:1523264450@qq.com
收稿日期:2025-02-10,
修回日期:2025-07-31,
录用日期:2025-07-31,
纸质出版日期:2025-08-15
移动端阅览
刘学涛,刘阳,李红建,等.基于机器学习构建中重度抑郁症住院患者使用SNRI类抗抑郁药的疗效预测模型[J].中国药房,2025,36(15):1936-1941.
LIU Xuetao,LIU Yang,LI Hongjian,et al.Construction of a predictive model for the efficacy of SNRI antidepressants in inpatients with moderate and severe depression based on machine learning[J].ZHONGGUO YAOFANG,2025,36(15):1936-1941.
刘学涛,刘阳,李红建,等.基于机器学习构建中重度抑郁症住院患者使用SNRI类抗抑郁药的疗效预测模型[J].中国药房,2025,36(15):1936-1941. DOI: 10.6039/j.issn.1001-0408.2025.15.21.
LIU Xuetao,LIU Yang,LI Hongjian,et al.Construction of a predictive model for the efficacy of SNRI antidepressants in inpatients with moderate and severe depression based on machine learning[J].ZHONGGUO YAOFANG,2025,36(15):1936-1941. DOI: 10.6039/j.issn.1001-0408.2025.15.21.
目的
2
运用机器学习方法构建中重度抑郁症住院患者使用5-羟色胺去甲肾上腺素再摄取抑制剂(SNRI)的疗效预测模型。
方法
2
回顾性收集2022年1月至2024年10月在新疆某三甲医院使用SNRI类药物治疗的中重度抑郁症住院患者病历资料,根据24项汉密尔顿抑郁量表评分标准的减分率,将患者分为有效组与无效组;经过LASSO回归筛选与SNRI类药物疗效相关的特征变量,应用训练集构建支持向量机、k近邻、随机森林、轻量级梯度提升机和极端梯度提升5种预测模型,使用贝叶斯优化算法调整模型的超参数,再以验证集评估模型性能,以筛选出最优模型。应用夏普利加性解释方法对最优模型进行解释。
结果
2
共收集到355例中重度抑郁症住院患者的病历资料,其中有效组285例、无效组70例,治疗有效率为80.28%。经过特征变量筛选,得到与疗效相关的5个特征变量,分别为汉密尔顿焦虑量表评分、血尿素氮、合用抗焦虑药物、饮酒史、首次发病。与其他模型相比,随机森林模型的性能表现最优,其受试者工作特征曲线下面积值为0.85,精确率-召回率曲线下面积值为0.87,准确度为0.74,召回率为0.75。
结论
2
基于5种特征变量建立的随机森林模型可用于中重度抑郁症住院患者使用SNRI类药物的疗效预测。
OBJECTIVE
2
To construct a prediction model for the efficacy of serotonin-norepinephrine reuptake inhibitor (SNRI) in inpatients with moderate and severe depression by using a machine learning method.
METHODS
2
The case records of inpatients with moderate and severe depression treated with SNRI antidepressants were collected from a third-grade class-A hospital in Xinjiang from January 2022 to October 2024; those patients were divided into effective group and ineffective group based on the Hamilton depression scale-24 score reduction rate. After screening the characteristic variables related to the therapeutic efficacy of SNRI drugs through LASSO regression, five prediction models including support vector machine, k-nearest neighbor, random forest, lightweight gradient boosting machine and extreme gradient boosting were constructed using the training set. Bayesian optimization was used to adjust the hyperparameters of these models. The performance of the models was evaluated in the validation set to select the optimal model. The Shapley additive explanations method was used to perform explainable analysis on the best model.
RESULTS
2
The medical records from 355 hospitalized patients with moderate and severe depression were collected, comprising 285 cases in the effective group and 70 cases in the ineffective group, resulting in an overall therapeutic response rate of 80.28%. After feature variable screening, five characteristic variables for therapeutic efficacy were obtained, including Hamilton anxiety scale, blood urea nitrogen, combination of anti-anxiety drugs, drinking history, and first onset of the disease. Compared with other models, the random forest model performed the best. The area under the receiver operating characteristic curve was 0.85, the area under the precision-recall curve was 0.87, the accuracy was 0.74, and the recall rate value was 0.75.
CONCLUSIONS
2
The random forest model constructed based on five characteristic variables demonstrates potential for predicting the therapeutic efficacy of SNRI antidepressants in hospitalized patients with moderate and severe depression.
陆林 . 沈渔邨精神病学 [M ] . 6版 . 北京 : 人民卫生出版社 , 2018 : 877 - 900 .
STAHL S M , GRADY M M , MORET C , et al . SNRIs:their pharmacology,clinical efficacy,and tolerability in comparison with other classes of antidepressants [J ] . CNS Spectr , 2005 , 10 ( 9 ): 732 - 747 .
LIN H S , LIN C H . Early improvement in HAMD-17 and HAMD-6 scores predicts ultimate response and remission for depressed patients treated with fluoxetine or ECT [J ] . J Affect Disord , 2019 , 245 : 91 - 97 .
SZEGEDI A , JANSEN W T , VAN WILLIGENBURG A P P , et al . Early improvement in the first 2 weeks as a predictor of treatment outcome in patients with major depressive disorder:a meta-analysis including 6 562 patients [J ] . J Clin Psychiatry , 2009 , 70 ( 3 ): 344 - 353 .
LIAO X M , SU Y N , WANG Y , et al . Antidepressant treatment strategy with an early onset of action improves the clinical outcome in patients with major depressive disorder and high anxiety:a multicenter and 6-week follow-up study [J ] . Chin Med J (Engl) , 2020 , 133 ( 6 ): 726 - 728 .
RUTLEDGE R B , CHEKROUD A M , HUYS Q J . Machine learning and big data in psychiatry:toward clinical applications [J ] . Curr Opin Neurobiol , 2019 , 55 : 152 - 159 .
中华医学会行为医学分会 , 中华医学会行为医学分会认知应对治疗学组 , 王长虹 , 等 . 抑郁症治疗与管理的专家推荐意见:2022年 [J ] . 中华行为医学与脑科学杂志 , 2023 ( 3 ): 193 - 202 .
CARPENITO T , MANJOURIDES J . MISL:multiple imputation by super learning [J ] . Stat Methods Med Res , 2022 , 31 ( 10 ): 1904 - 1915 .
WANG X C , REN H , REN J H , et al . Machine learning-enabled risk prediction of chronic obstructive pulmonary disease with unbalanced data [J ] . Comput Methods Programs Biomed , 2023 , 230 : 107340 .
谢宏宇 , 侯艳 , 李康 . 基于正则化回归的组学数据变量筛选方法 [J ] . 中国卫生统计 , 2016 , 33 ( 4 ): 733 - 736 .
崔佳旭 , 杨博 . 贝叶斯优化方法和应用综述 [J ] . 软件学报 , 2018 , 29 ( 10 ): 3068 - 3090 .
吴妍 . 基于机器学习的糖尿病早期诊断模型及可解释分析 [D ] . 重庆 : 重庆理工大学 , 2024 .
NUÑEZ N A , JOSEPH B , PAHWA M , et al . Augmentation strategies for treatment resistant major depression:a systematic review and network meta-analysis [J ] . J Affect Disord , 2022 , 302 : 385 - 400 .
YAN Y S , YANG X , WANG M , et al . Efficacy and accepta- bility of second-generation antipsychotics with antidepressants in unipolar depression augmentation:a systematic review and network meta-analysis [J ] . Psychol Med , 2022 , 52 ( 12 ): 2224 - 2231 .
李凌江 , 马辛 . 中国抑郁障碍防治指南 [M ] . 2版 . 北京 : 中华医学电子音像出版社 , 2015 : 59 - 80 .
FAVA M , RUSH A J , ALPERT J E , et al . Difference in treatment outcome in outpatients with anxious versus nonanxious depression:a STAR*D report [J ] . Am J Psychiatry , 2008 , 165 ( 3 ): 342 - 351 .
CUKOR D , RUE T , HEAGERTY P , et al . Anxiety and response to treatment of depression in people undergoing maintenance hemodialysis [J ] . Clin J Am Soc Nephrol , 2023 , 18 ( 8 ): 1075 - 1076 .
谢焕山 , 黄毅 , 黄善情 , 等 . 基于治疗药物监测的O-去甲基文拉法辛与文拉法辛浓度比值影响因素分析 [J ] . 中国临床药理学杂志 , 2022 , 38 ( 20 ): 2419 - 2422 .
BUCKMAN J J , UNDERWOOD A , CLARKE K , et al . Risk factors for relapse and recurrence of depression in adults and how they operate:a four-phase systematic review and meta-synthesis [J ] . Clin Psychol Rev , 2018 , 64 : 13 - 38 .
DAVYDOVA T V , NEVIDIMOVA T I , VETRILE L A , et al . Correlation of antibodies to neurotransmitters in the sera of women with alcohol dependence and depressive disorders [J ] . Bull Exp Biol Med , 2021 , 171 ( 6 ): 704 - 706 .
NUNES E V . Alcohol and the etiology of depression [J ] . Am J Psychiatry , 2023 , 180 ( 3 ): 179 - 181 .
ALSHEIKH A M , ELEMAM M O , EL-BAHNASAWI M . Treatment of depression with alcohol and substance dependence:a systematic review [J ] . Cureus , 2020 , 12 ( 10 ): e11168 .
TANDON N , TANDON R . Machine learning in psychiatry:standards and guidelines [J ] . Asian J Psychiatry , 2019 , 44 : A1 - A4 .
0
浏览量
27
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构