| 543 | 0 | 380 |
| 下载次数 | 被引频次 | 阅读次数 |
目的：建立一种美沙酮维持治疗（methadone maintenance treatment,MMT）减量预测模型，并采用沙普利加和解释（SHAP）方法评估不同干预措施及其他临床因素对美沙酮减量的影响。方法：对两项针刺干预MMT患者美沙酮减量的临床试验进行分析，收集患者的基线资料、MMT相关信息、干预方式、减量结局指标相关数据，采用支持向量机（SVM）、K最近邻（KNN）、逻辑回归（LR）、朴素贝叶斯（NB）、随机森林（RF）和Cat Boost分类模型（CatBoost）6种机器学习算法和混合（Blending）、堆叠（Stacking）2种集成方法构建预测模型，并用SHAP方法对最优模型进行可解释性分析。结果：共251例MMT患者纳入研究，其中针刺组128例、非针刺组123例。Cat Boost模型和Stacking集成方法在测试集上表现最优，Cat Boost模型的准确率为0.780 0±0.060 8，精确率为0.500 0±0.120 0，召回率为0.818 2±0.140 2,F1分数为0.620 7±0.114 0，受试者工作特征曲线下面积（ROC-AUC）为0.857 8±0.140 2。影响针刺辅助MMT患者美沙酮减量的前5个重要特征为干预方式、体质量指数（BMI）、MMT时长、阿片类物质使用史和就业情况，其SHAP值分别为1.25、0.36、0.21、0.19和0.12。特征依赖图显示BMI、MMT时长和阿片类物质使用史与减量效果之间均呈现负相关。结论：可解释的预测模型为针刺辅助美沙酮减量治疗中需综合考虑临床因素提供了科学依据，有助于针刺临床美沙酮减量策略的改进和个性化减量方案的制定。
Abstract：Objective To construct a predictive model for the reduction in methadone maintenance treatment(MMT) and evaluate the effects of different interventions and other clinical factors on methadone reduction using Shapley additive explanations(SHAP). Methods Two clinical trials of acupuncture for methadone reduction in MMT patients were analyzed, and the baseline data, MMT related information, intervention measures, the data related to dose-reduction outcomes were collected. The predictive model was constructed by means of 6 machine learning algorithms including support vector machine(SVM), K-nearest neighbors(KNN), logistic regression(LR), Naive Bayes(NB), random forest(RF) and categorical-boosting(Cat Boost), and 2 integration methods, blending-ensemble method(Blending) and Stacking-ensemble method(Stacking). SHAP was employed for the interpretability analysis of the optimal model. Results A total of 251 MMT patients were included, 128 cases in the acupuncture group and 123 cases in the non-acupuncture group. CatBoost and Stacking performed optimally in the test set. CatBoost obtained an accuracy of 0.780 0±0.060 8, a precision of 0.500 0±0.120 0, a recall of 0.818 2±0.140 2, F1 score of 0.620 7±0.114 0, and receiver operating characteristic-area under curve(ROC-AUC) of 0.857 8±0.140 2 for the subjects. In MMT patients with acupuncture as an adjunctive therapy, the top 5 important features for methadone reduction, included intervention measures, body mass index(BMI), the duration of MMT, the history of opioid use and occupation; and SHAP values were 1.25, 0.36, 0.21, 0.19 and 0.12, respectively. The SHAP feature dependence plot showed that BMI, MMT duration and the history of opioid use presented a nonlinear negative correlation with the reduction effect. Conclusion In acupuncture as adjunctive therapy for methadone reduction, the clinical factors should be considered comprehensively; and the interpretable predictive model provides a scientific basis for it, which is conducive to the improvement of clinical strategy of acupuncture for methadone reduction and the development of personalized reduction scheme.
[1]张亚海.美沙酮临床应用的药理学基础[J].中国药物滥用防治杂志, 2012, 18(6):346-349, 363.
[2] Kourounis G, Richards BDW, Kyprianou E, et al. Opioid substitution therapy:lowering the treatment thresholds[J]. Drug Alcohol Depend, 2016, 161:1-8.
[3] Cao PH, Zhang ZH, Zhong J, et al. Effects of treatment status and life quality on anxiety in MMT patients[J]. Subst Abuse Treat Prev Policy, 2021, 16(1):9.
[4]赵受伟.针刺对美沙酮维持治疗患者生存质量的影响[D].广州:广州中医药大学, 2011.
[5] Sharkey KM, Kurth ME, Anderson BJ, et al. Obstructive sleep apnea is more common than central sleep apnea in methadone maintenance patients with subjective sleep complaints[J]. Drug Alcohol Depend, 2010, 108(1/2):77-83.
[6] Haber PS, Elsayed M, Espinoza D, et al. Constipation and other common symptoms reported by women and men in methadone and buprenorphine maintenance treatment[J]. Drug Alcohol Depend, 2017, 181:132-139.
[7] Nosyk B, Marsh DC, Sun HY, et al. Trends in methadone maintenance treatment participation, retention, and compliance to dosing guidelines in British Columbia, Canada:1996-2006[J]. J Subst Abuse Treat, 2010, 39(1):22-31.
[8]陈君,陈雯,凌莉,等.广东省社区美沙酮维持治疗剂量使用现状及其评价[J].现代预防医学, 2010, 37(9):1687-1689.
[9] Amato L, Davoli M, Minozzi S, et al. Methadone at tapered doses for the management of opioid withdrawal[J]. Cochrane Database Syst Rev, 2013, 2013(2):CD003409.
[10] Eklund C, Melin L, Hiltunen A, et al. Detoxification from methadone maintenance treatment in Sweden:long-term outcome and effects on quality of life and life situation[J]. Int J Addict,1994, 29(5):627-645.
[11] Ge SQ, Lan J, Yi Q, et al. Acupuncture for illicit drug withdrawal syndrome:a systematic review and meta-analysis[J]. Eur J Integr Med, 2020, 35:101096.
[12] Lu LM, Zhang YQ, Tang XR, et al. Evidence on acupuncture therapies is underused in clinical practice and health policy[J].BMJ, 2022, 376:e067475.
[13] Wen H, Chen RH, Zhang PM, et al. Acupuncture for opioid dependence patients receiving methadone maintenance treatment:a network meta-analysis[J]. Front Psychiatry, 2021, 12:767613.
[14] Wen H, Wei XJ, Ge SQ, et al. Clinical and economic evaluation of acupuncture for opioid-dependent patients receiving methadone maintenance treatment:the integrative clinical trial and evidence-based data[J]. Front Public Health, 2021, 9:689753.
[15] Lu LM, Chen C, Chen YM, et al. Effect of acupuncture for methadone reduction:a randomized clinical trial[J]. Ann Intern Med, 2024, 177(8):1039-1047.
[16] Lo-Ciganic WH, Huang JL, Zhang HH, et al. Evaluation of machine-learning algorithms for predicting opioid overdose risk among medicare beneficiaries with opioid prescriptions[J]. JAMA Netw Open, 2019, 2(3):e190968.
[17] Dong XY, Rashidian S, Wang Y, et al. Machine learning based opioid overdose prediction using electronic health records[J].AMIA Annu Symp Proc, 2020, 2019:389-398.
[18] Paparrizos J, Gravano L. K-shape:efficient and accurate clustering of time series[C]//Proceedings of the 2015 ACM SIGMOD International Conference on Management of Data.Melbourne Victoria Australia. ACM, 2015:1855-1870.
[19] Dong Y, Fan BC, Yan EL, et al. Decision tree model based prediction of the efficacy of acupuncture in methadone maintenance treatment[J]. Front Neurol, 2022, 13:956255.
[20] College of Physicians and Surgeons of Ontario. Methadone Program. Methadone Maintenance Treatment Program Standards and Clinical Guidelines[M]. Toronto:College of Physicians and Surgeons of Ontario, 2011.
[21] Heinzerling KG, Swanson AN, Kim S, et al. Randomized,double-blind, placebo-controlled trial of modafinil for the treatment of methamphetamine dependence[J]. Drug Alcohol Depend, 2010, 109(1/2/3):20-29.
[22] Rezaei F, Emami M, Zahed S, et al. Sustained-release methylphenidate in methamphetamine dependence treatment:a double-blind and placebo-controlled trial[J]. DARU J Pharm Sci,2015, 23(1):2.
[23] Pedregosa F, Varoquaux G, Gramfort A, et al. Scikit-learn:machine learning in Python[J]. J Mach learn Res, 2011, 12:2825-2830.
[24] Han DH, Lee S, Seo DC. Using machine learning to predict opioid misuse among U.S. adolescents[J]. Prev Med, 2020, 130:105886.
[25] Campo DS, Gussler JW, Sue A, et al. Accurate spatiotemporal mapping of drug overdose deaths by machine learning of drug-related web-searches[J]. PLoS One, 2020, 15(12):e0243622.
[26] Trafton JA, Minkel J, Humphreys K. Determining effective methadone doses for individual opioid-dependent patients[J].PLoS Med, 2006, 3(3):e80.
[27] Talal AH, Ding YX, Venuto CS, et al. Toward precision prescribing for methadone:determinants of methadone deposition[J]. PLoS One, 2020, 15(4):e0231467.
[28] Baldassarri SR, Beitel M, Zinchuk A, et al. Correlates of sleep quality and excessive daytime sleepiness in people with opioid use disorder receiving methadone treatment[J]. Sleep Breath, 2020,24(4):1729-1737.
[29]蓝沁舟,方芳,仓静.阿片类药物及其使用的新进[J].中国临床医学, 2023, 30(4):736-742.
[30] Christie MJ. Cellular neuroadaptations to chronic opioids:tolerance, withdrawal and addiction[J]. Br J Pharmacol, 2008,154(2):384-396.
[31] Tzschentke TM, Schmidt WJ. Glutamatergic mechanisms in addiction[J]. Mol Psychiatry, 2003, 8(4):373-382.
[32]马会芹,占乐云,许晴雯.阿片类药物镇痛效果影响因素的研究进展[J].麻醉安全与质控, 2019, 3(3):172-176.
基本信息:
DOI：10.13703/j.0255-2930.20250110-0002
中图分类号:R246.6
引用信息:
[1]范宝超,张侨,陈晨,等.基于机器学习与SHAP的针刺干预美沙酮维持患者美沙酮减量的可解释性预测模型构建研究[J].中国针灸,2025,45(10):1363-1370.DOI:10.13703/j.0255-2930.20250110-0002.
基金信息:
国家自然科学基金资助项目：821715127; 中国博士后基金项目：2024M750464
2025-07-14
2025-07-14
2025-07-14