Construction of interpretable predictive model of acupuncture for methadone reduction in patients undergoing methadone maintenance treatment based on machine learning and SHAP

FAN Baochao; ZHANG Qiao; CHEN Chen; CHEN Yiming; ZHANG Peiming; LU Liming

doi:10.13703/j.0255-2930.20250110-0002

您当前的位置：

首页 >

文章列表页 >

Construction of interpretable predictive model of acupuncture for methadone reduction in patients undergoing methadone maintenance treatment based on machine learning and SHAP

Acupuncture and Intelligent Medicine | 更新时间：2025-10-11

- Construction of interpretable predictive model of acupuncture for methadone reduction in patients undergoing methadone maintenance treatment based on machine learning and SHAP
- Chinese Acupuncture & Moxibustion Vol. 45, Issue 10, Pages: 1363-1370(2025)
- 作者机构：
  
  广州中医药大学针灸康复临床医学院，针灸研究中心大数据实验室，广东广州　511400
  江苏医药职业学院
  广州中医药大学第八临床医学院
  佛山市中医院
- 作者简介：
- 基金信息：
- DOI：10.13703/j.0255-2930.20250110-0002
  CLC：
- Received：10 January 2025，
  
  Online First：14 July 2025，
  
  Published：12 October 2025
- 稿件说明：
移动端阅览
FAN Baochao, ZHANG Qiao, CHEN Chen, et al. Construction of interpretable predictive model of acupuncture for methadone reduction in patients undergoing methadone maintenance treatment based on machine learning and SHAP[J]. Chinese Acupuncture & Moxibustion, 2025, 45(10): 1363-1370.
DOI：

FAN Baochao, ZHANG Qiao, CHEN Chen, et al. Construction of interpretable predictive model of acupuncture for methadone reduction in patients undergoing methadone maintenance treatment based on machine learning and SHAP[J]. Chinese Acupuncture & Moxibustion, 2025, 45(10): 1363-1370. DOI： 10.13703/j.0255-2930.20250110-0002.

摘要

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CatBoost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。CatBoost模型和Stacking集成方法在测试集上表现最优，CatBoost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 (CatBoost)

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.

关键词

Keywords

references

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Development and research of an AI-assisted decision-making platform in treatment of insomnia with acupuncture of Tongdu Yangxin acupoint prescription

Construction of a prognostic assessment model of acupuncture intervention for lumbar disc herniation based on interpretable machine learning algorithms

An interpretable machine learning modeling method for the effect of manual acupuncture manipulations on subcutaneous muscle tissue

Current status and outlooks of acupuncture research driven by machine learning

Research progress on the regulatory mechanisms of inflammatory signaling pathways in acupuncture treatment for ischemic stroke

Related Author

WANG Chi

LIU Chengyong

WANG Xiaoqiu

LIU Enqi

SUN Juguang

LU Jin

DING Min

WU Wenzhong

Related Institution

Department of Acupuncture-Moxibustion and Rehabilitation, Affiliated Hospital of Nanjing University of Chinese Medicine

Yueyang Hospital of Integrated Traditional Chinese and Western Medicine, Shanghai University of TCM

Department of Acupuncture and Moxibustion, Xuzhou Hospital of TCM

Department of Acupuncture and Moxibustion, Nanjing Chinese Medicine Hospital

Department of Acupuncture and Moxibustion, Wuxi Chinese Medicine Hospital

⁰