Scholarlink

<p>Artificial Intelligence (AI) has increasingly been explored in connection with prevention, screening, treatment support, relapse prediction, and digital health approaches to Substance Use Disorders (SUDs). However, the current research landscape remains conceptually fragmented, and the extent to which AI-based Assistance for Combating SUDs (AI-ACSUDs) has progressed from algorithmic development toward clinically validated implementation remains unclear. This study aims to assess the global trends, methodological maturity, translational gaps, and thematic priorities in literature related to AI-ACSUDs using a mixed-methods bibliometric analysis of 4657 Scopus-indexed publications from 2020 to 2025, alongside a thematic evaluation of 2328 highly cited articles. Bibliometric and visualization tools were used to map research trends, keywords co-occurrence, research clusters, institutional contributions, emerging research areas and thematic priorities in literature related to AI-ACSUDs. Results showed increasing academic interest in this area of research, strong bibliometric visibility of Machine Learning (ML), predictive models, opioid studies, digital health, and analytical methods, as well as dominance of publications originating from North American and European research institutions. Additionally, keyword and temporal analysis revealed trends pointing towards an evolving interface between computational modeling and clinical addiction research, particularly through prediction, screening, digital intervention, and treatment-support-related topics. Nevertheless, these trends represent research interest rather than concrete proof of the clinical efficiency, safety, scalability, or readiness for implementation. Analysis indicates that AI-ACSUDs research landscape is methodologically vibrant but translationally uneven, with very limited evidence about clinically validated implementation, explainable AI, responsible governance, equitable design, and cultural consideration. This paper contributes a maturity-based interpretation to AI-ACSUDs literature by showing that technical development appears to be advancing faster than clinical validation and ethical governance. Further research should focus on prospective validation, implementation research, transparent governance, and responsible AI development for assisting SUD-related care.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Bibliometric mapping of the multidimensional landscape of artificial intelligence-based assistance for combating substance use disorders (AI-ACSUDs)

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月16日

查看全文
  • W. I. A. Gayashani,
  • Nimesha D. M. Herath,
  • A. S. Hapuarachchi,
  • Akila R. Jayamaha

摘要

Artificial Intelligence (AI) has increasingly been explored in connection with prevention, screening, treatment support, relapse prediction, and digital health approaches to Substance Use Disorders (SUDs). However, the current research landscape remains conceptually fragmented, and the extent to which AI-based Assistance for Combating SUDs (AI-ACSUDs) has progressed from algorithmic development toward clinically validated implementation remains unclear. This study aims to assess the global trends, methodological maturity, translational gaps, and thematic priorities in literature related to AI-ACSUDs using a mixed-methods bibliometric analysis of 4657 Scopus-indexed publications from 2020 to 2025, alongside a thematic evaluation of 2328 highly cited articles. Bibliometric and visualization tools were used to map research trends, keywords co-occurrence, research clusters, institutional contributions, emerging research areas and thematic priorities in literature related to AI-ACSUDs. Results showed increasing academic interest in this area of research, strong bibliometric visibility of Machine Learning (ML), predictive models, opioid studies, digital health, and analytical methods, as well as dominance of publications originating from North American and European research institutions. Additionally, keyword and temporal analysis revealed trends pointing towards an evolving interface between computational modeling and clinical addiction research, particularly through prediction, screening, digital intervention, and treatment-support-related topics. Nevertheless, these trends represent research interest rather than concrete proof of the clinical efficiency, safety, scalability, or readiness for implementation. Analysis indicates that AI-ACSUDs research landscape is methodologically vibrant but translationally uneven, with very limited evidence about clinically validated implementation, explainable AI, responsible governance, equitable design, and cultural consideration. This paper contributes a maturity-based interpretation to AI-ACSUDs literature by showing that technical development appears to be advancing faster than clinical validation and ethical governance. Further research should focus on prospective validation, implementation research, transparent governance, and responsible AI development for assisting SUD-related care.