Psychiatry research · 2026
Identification and validation of an explainable model of methamphetamine use disorders with anxiety symptoms: a multicenter cross-sectional study.
СтатьяМногоцентровое исследованиеВалидационное исследованиеНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Shi Sai, Su Hang, Sun Yiwen, Zhong Na, Jiang Haifeng, Du Jiang, Chen Tianzhen, Zhao Min
Аннотация
AIMS: Negative reinforcement process triggered by anxiety is an important factor in promoting the relapse behavior of substance use disorder. Objective identification of the anxiety level of methamphetamine (METH) use disorder (MUD) and other addiction behaviors becomes particularly important, yet the current understanding of this issue remains unclear.
METHODS: This multicenter cross-sectional study in China was conducted on METH patients between September 1, 2019 and September 1, 2021 for the identification and validation of the prediction model. MUD with anxiety symptoms was defined mainly based on the 7-item Generalized Anxiety Disorder Questionnaire (GAD-7) score. With 70 clinical characteristics also obtained or evaluated among the METH patients, 5 Machine Learning (ML) algorithms were used to construct prediction models. The SHapley Additive exPlanation method (SHAP) was used to rank the feature importance and explain the final model.
RESULTS: The XGBoost model performed best in discriminative ability among the 5 ML models. After reducing features according to feature importance rank, an explainable final XGBoost model was established with 12 features. Critically, METH patients with impaired inflammatory or oxidative stress factors, hormones, neuro-related factors, abnormal clinical scales, and drug use-related metrics had a higher level of anxiety.
CONCLUSION: We developed an accessible and exploratory ML model for the assistant recurrence of MUD with anxiety. This model enhances the identification of anxiety comorbidity, demonstrates the feasibility of using a parsimonious set of predictors, and supports patient stratification for personalized treatment decisions.
METHODS: This multicenter cross-sectional study in China was conducted on METH patients between September 1, 2019 and September 1, 2021 for the identification and validation of the prediction model. MUD with anxiety symptoms was defined mainly based on the 7-item Generalized Anxiety Disorder Questionnaire (GAD-7) score. With 70 clinical characteristics also obtained or evaluated among the METH patients, 5 Machine Learning (ML) algorithms were used to construct prediction models. The SHapley Additive exPlanation method (SHAP) was used to rank the feature importance and explain the final model.
RESULTS: The XGBoost model performed best in discriminative ability among the 5 ML models. After reducing features according to feature importance rank, an explainable final XGBoost model was established with 12 features. Critically, METH patients with impaired inflammatory or oxidative stress factors, hormones, neuro-related factors, abnormal clinical scales, and drug use-related metrics had a higher level of anxiety.
CONCLUSION: We developed an accessible and exploratory ML model for the assistant recurrence of MUD with anxiety. This model enhances the identification of anxiety comorbidity, demonstrates the feasibility of using a parsimonious set of predictors, and supports patient stratification for personalized treatment decisions.